The central problem faced by any student of the fragments from Reculver, Kent, lies in deciding, firstly, whether or not all of the pieces for which a claim to have originated at Reculver (Reculver 1–3 and Canterbury Old Dover Road 1) has been advanced are in fact from that site; secondly, if such a claim can be substantiated, whether or not they derive from a single monument, and lastly, how that monument may be reconstructed. In reaching any conclusions about these difficult issues a number of factors must be considered: the style of the decoration of the fragments; the technique of carving; the comparative size and stone types of the various drums; and the documentary evidence for a cross at Reculver provided by the account of the early sixteenth-century antiquary John Leland, and by the late thirteenth-century register of archbishop Winchelsea. The detailed analyses of these problems are presented in the discussion of the various pieces in the catalogue, where it is argued that, despite their differing find-spots, the sections of shaft Reculver 1a–e, Canterbury Old Dover Road 1, and probably also the lost fragment Reculver 2, all derive from the same monument, whereas the relationship between that monument and the one described by Leland (and perhaps also referred to in Winchelsea's register) remains an open question. It is further proposed that the lost cross-head from the site, Reculver 3, probably derives from a different monument to the one probably represented by nos. 1–2. Any conclusion on these issues remains to an uncomfortable degree subjective; nevertheless, the broader questions of the stylistic context and date of the fragments which form the subject of the present chapter, will proceed on the basis of the conclusions outlined above.
The fragments from Reculver have excited much interest among twentieth-century art historians who, from the time of Peers onwards, tended to consider them to be of an early date (Peers 1927b), and such a view has also been advocated in a recent major study (Kozodoy 1976; idem 1986). There seems little doubt that the lack of publication both of Anglo-Saxon sculpture and of works in other materials from south-east England has handicapped Kozodoy in her study. This apparent lack of comparative material led her to dismiss the suggestion that there could be any local context for the Reculver fragments. Instead she compared the iconography, figural and drapery styles of the fragments first with Late Antique and early Byzantine models. These comparisons are often sound, and on occasions compelling, but Kozodoy failed to confront the fact that much of Anglo-Saxon art both in content and style is based on Late Antique exemplars, either directly, or through intermediaries. An alternative view of the date and stylistic context of the stones is presented here.
The Reculver fragments are normally regarded as of late seventh-century date (Peers 1927b, 255; Brown 1903–37,
Peers's argument is open to several objections. Firstly, the opus signinum floor need not be primary as he supposed. Excavations at Reculver by B. J. Philp have revealed that an identical flooring is present in one of the added eighth-century porticus (Wilson and Hurst 1969, 161), and work at St Pancras's Church, Canterbury, has revealed a very similar floor to that at Reculver datable no earlier than the mid eighth century (Rigold 1977, 163). The base may still have been primary even if the floor were laid round it later, however. Secondly and more seriously, the function of this feature as the base of a cross has been challenged, Taylor having reinterpreted it as the base of an altar (Taylor 1973, 52–8). Finally, even if the base in front of the triple arcade could be unequivocally accepted as a cross-base, there is no acceptable way of proving that the surviving fragments of sculpture formed part of the cross which originally stood on it (see Reculver 1a–e, Discussion). And even if it could be proved that Leland's cross and the surviving fragments are one and the same, there is no reason to assume that it is of the same date as the surviving base. It could equally have been a later monument placed on an earlier base.
It is evident that the line of reasoning used by Peers to establish a late seventh-century date for the Reculver fragments is unsound. This being so it is necessary to examine the whole problem of dating the fragments afresh. This is a course which presents grave difficulties. Only Reculver 1e exhibits a range of decorative motifs, with its use of a plant-scroll inhabited by human busts combined with interlace (Ills. 119–20). The rest of the fragments are decorated purely with figural ornament (Ills. 108–18). The figures are technically highly accomplished, and in a classicising style, but a classicising element, represented by the use of figural scenes, is present in English art throughout the pre-Conquest period. Initially it derived from late Classical works introduced into England with the Augustinian mission, but these works, and late Classical works imported subsequently, were copied throughout the pre-Conquest period by Anglo-Saxon artists, as were earlier native works drawing in part on Classical sources, and imported continental works of art which drew on similar sources. For this reason the topic of figure decoration in pre-Conquest art, and, in consequence, the position of the Reculver fragments, is extremely difficult to approach. There are, however, several discrete groups of works (almost exclusively manuscripts) in southern England which exhibit classicising impulses with which the Reculver fragments might be compared. First, there are imported late Classical works, including manuscripts and panel paintings, reaching England in the seventh century in the wake of the Augustinian mission. Of these only the St Augustine's Gospels is a possible survivor. Second, there is a group of eighth-century manuscripts produced in Kent, of which the Vespasian Psalter, the Codex Aureus, and the Codex Bigotianus survive. The Barberini Gospels may also have been produced in Kent (Wilson 1984, 91), or at least in southern England in the late eighth or early ninth century (Budny and Graham-Campbell 1981, 16), although Brown prefers a provenance in Mercia or Northumbria (Brown 1991, 13; Webster and Backhouse 1991, no. 160). Third, classicizing tendencies are also observable in Canterbury works of the early ninth century represented by the luxurious manuscript BL MS Royal I. E. VI. Fourth, there are pre-Winchester style works of the late ninth and early tenth centuries, notably the St Cuthbert stole and maniple, the Aethelstan Psalter, and the fragment of wall painting from the New Minster excavations at Winchester. All of these are Winchester products. Last, there are Winchester style products of the late tenth and eleventh centuries. Such manuscripts and works in other materials were produced in numerous southern English centres including Canterbury, and Winchester itself. Both pre-Winchester style works of the late ninth and early tenth centuries and Winchester style work proper were themselves heavily influenced by Carolingian art, and this might have formed another or additional source influencing the Reculver fragments. Precise comparisons between the Reculver sculptures and all of these works are, however, hampered by the poor condition of many of the Reculver pieces. This has rendered the decorative programme of the monument or monuments indecipherable, and has obliterated many of the stylistic details.
IMPORTED WORKS OF THE SEVENTH CENTURY
Classicising art was re-introduced into England with the advent of the Augustinian mission. Bede, for example, records that St Augustine brought with him from Rome a panel painting depicting the figure of Christ (Bede 1969, 75 (
From this Italian Gospel fragment, in which only two decorated folios remain, 125r and 129v, it is difficult to form any accurate picture of art in Kent in the seventh century, even if the manuscript was in England at that time. The manuscript does, however, at least provide one of the possible models available in Kent. In broad terms the manuscript does have points of comparison with some of the Reculver fragments. The seated figure of St Luke on fol. 129v is within an architectural frame, consisting of columns with Corinthian capitals supporting a flat entablature, a feature seen also on Reculver 1d (Ills. 111–12) and Canterbury Old Dover Road 1 (Ills. 108, 110), although the sculptured figures are standing and not seated. Moreover they form part of a range of figures, not a single, isolated portrait as in the St Augustine's Gospels. Like Reculver 1b–c the St Augustine's Gospels employs a series of narrative scenes, twelve filling fol. 125r, and a further six flanking the seated figure of St Luke on fol. 129v. As on Reculver 1a (Ills. 113–15) the scenes are divided by plain borders, although those at Reculver probably appeared very different when their metal fittings were in place. There, however, the comparison breaks down, for the iconography of the St Augustine's Gospels is not reflected in the Reculver fragments, although only two figure-decorated pages from the manuscript survive.
In style also the St Augustine's Gospels and Reculver 1a–d and the Canterbury fragment are far apart. In the St Augustine's Gospels the figures are small and dumpy, and only sketchily drawn, with the emphasis on the action of the scene. The garments, although rendered naturalistically, are not fully worked up, depending on line, not modelling, to define the folds in the cloth. The figures on the Reculver stones, in contrast, are more carefully and skilfully drawn, with the folds of the draperies defined not only by line, but also by careful modelling. This is a particularly remarkable feature given the intractability of stone as a medium compared with colour on parchment. If there are tenuous points of comparison between the illustrations of the St Augustine's Gospels and those on Reculver 1a–d and the Canterbury fragment, there are none with 1e (Ills. 119–20). This employs frames filled with interlace, and interlace is not used in the St Augustine's Gospels, or in any other late Classical manuscript. The plant-scroll inhabited by human busts on stone 1e from Reculver is a Hellenistic form in origin, later incorporated into Roman art (Toynbee and Ward-Perkins 1950, 1–43), but again is not paralleled in the St Augustine's Gospels. Where plants do occur in the Gospels they are semi-naturalistic, as are the plants developing from the flat entablature on either side of the arch head over the seated figure of St Luke. Again the figure style of the manuscript differs from that of the sculpture. At Reculver the bust in the roundel has naturalistically-treated draperies which are well modelled, while the hand raised in blessing across the body is unnaturally enlarged and extended to emphasise the gesture. In the St Augustine's Gospels the figure of Christ blessing the bread in the Last Supper scene on fol. 125r has exactly the same pose as the bust on stone 1e. Here, however, the draperies have their features indicated by dark lines, and are not fully modelled; moreover, the hand raised in blessing is normally proportioned.
There is, then, a marked divergence in iconography and style between the Reculver fragments and the St Augustine's Gospels, but it is possible that there existed in Kent works such as the Vienna Genesis (Wellesz 1960) or Cotton Genesis (Weitzmann 1977, pls. 21–2), both of which are more fluent and classicising in style than the St Augustine's Gospels. Indeed it has been argued that Aelfric's Pentateuch, a Canterbury manuscript made in the eleventh century, depends in part on such a model (Temple 1976, no. 86, ills. 265–72, fig. 34; Brown 1991, 52). If this was the case, and they inspired sculpture of such outstanding quality as the Reculver fragments, then it might be expected, given the durability of the material, that other sculptures of similar inspiration and quality would survive. In fact, there is no sculptural context for the Reculver fragments in the seventh century. As noted above (see Chap. V), the only early sculptures in south-east England are in Kent, namely, the Reculver columns (Reculver 4a–b), the inscription from St Martin's Canterbury, and the Sandwich stones (nos. 1–2), but all of these sculptures are simple in form and employ an incised technique which is not comparable with the sophisticated relief techniques used at Reculver. Equally there is no southern English work of the seventh century in any medium which can compare in complexity of design, or subtlety and skill of execution, with the Reculver fragments.
Since there are no surviving southern English works with which the Reculver sculptures can be compared, it is necessary to look further afield for comparative material; specifically to Northumbria where two seventh-century figure-decorated manuscripts survive, namely, the Codex Amiatinus and the Lindisfarne Gospels. In addition, some information may be gleaned from the silver plate on the front of the portable altar of St Cuthbert (Battiscombe 1956, 326–35).
The Codex Amiatinus was made to the order of Ceolfrith, Abbot of Monkwearmouth-Jarrow 689–716 (Bruce-Mitford 1978, 2–7), and is possibly a copy of the Codex Grandior, a fifth-century Italian manuscript made at Cassiodorus's monastery of Vivarium (ibid., 8–9). There are three pages with figure decoration. On fol. Vr is a miniature of Ezra (ibid., pl. 2), fol. 796v is filled with a Christ in Majesty (ibid., pl. XIII), and on fol. VIr is a roundel containing the bust of God the Father (ibid., pl. XI.1). There are disparities in style between the Ezra figures and the other figural scenes, but all the figures are treated in a naturalistic, classicising manner, with the draperies carefully delineated, and on all of them there is some attempt at modelling. This naturalistic treatment, however, is not carried through wholly successfully. With all the figures, despite the modelling, the folds of the draperies are delineated by hard, dark lines. For the most part these draperies reflect the shape of the body beneath, but on the figure of the seated Christ on fol. 796v the folds over the legs have been reduced to a pattern which has lost touch with the shape of the body beneath. This occurs again with the seated figure of Ezra on fol. Vr, although here it is not only the folds which are stylized, but the form and colour of the garment. On the upper part of the body the outer garment is coloured red, and the under-garment green. On the lower part of the body these colours are arbitrarily reversed. The draperies on the Reculver fragments, in contrast, are well modelled without the dependence on line betrayed by the Codex Amiatinus, and they consistently reflect the shape and movement of the body beneath. Unlike the figures in the Codex Amiatinus, which are short and wooden, those on the Reculver fragments are well-proportioned and fluently drawn, whether in repose or in movement. Any relationship must, therefore, be dismissed.
The tendency of the Codex Amiatinus to depend on line to depict the draperies is accentuated in the Lindisfarne Gospels, made at Lindisfarne c. 698 (Alexander 1978, no. 9, 35–40, ills. 28–46), and like the Codex Amiatinus copying a south Italian model, at least in part (ibid., 36–7). In each of the Evangelist portraits the draperies of the figures are not modelled. Instead hard, thick, dark lines are used to delimit the folds. Moreover, the draperies no longer reflect the shape of the body beneath, instead the folds are arranged to form surface patterns. This tendency to reduce the draperies to pattern finds its ultimate expression in the St Luke portrait on fol. 218r of the Lichfield Gospels, where the shape of the body is wholly replaced by a symmetrical patterning (ibid., no. 21, 48–50, ill. 82). As with the Codex Amiatinus, the approach to figure drawing used in the Lindisfarne Gospels has little to do with the concepts underlying the figure drawing of the Reculver fragments. The Lindisfarne Gospels does use interlace of the quality found on stone 1e, but it does not use the plant ornament combined with it at Reculver. The Codex Amiatinus lacks both interlace and plant ornament. Any relationship with the Reculver fragments can, therefore, be rejected.
In addition to the figure-decorated manuscripts, a seventh-century date has been argued for the silver plate on the front of the portable altar from the grave of St Cuthbert (Battiscombe 1956, 334). The sheet is decorated with the seated figure of St Peter (ibid., fig. 330, pl. XIX), identified by semi-vertical inscriptions, perhaps paralleling the putative arrangements of metal strips which may have borne inscriptions on stone 1a from Reculver (see Reculver 1a–e, Discussion). Unfortunately, only fragments of the sheet survive, but the drapery style is highly naturalistic, unlike that in the manuscripts, and much closer in spirit to Reculver 1a–d. There must, however, be doubts about the date of the silver sheet on the altar, and an eighth-century date has been proposed by Webster, with further additions later in the century (Webster and Backhouse 1991, no. 99). Okasha certainly places the silver sheet much later than the core on the grounds of their differing scripts (Okasha 1971, no. 35, pl. 35).
If there is doubt over the date of the silver sheet on the St Cuthbert altar, and little comparison between the Reculver fragments and the Codex Amiatinus and Lindisfarne Gospels, there is equally no early sculpture in northern England with which the Reculver fragments can be compared. As Cramp has pointed out, the only seventh-century sculpture in Northumbria seems to be architectural, such as the turned balusters from Monkwearmouth and Jarrow (Cramp 1984,
If there are no manuscripts or sculptures in England in the seventh century with which the Reculver fragments can adequately be compared, it remains a possibility that they are themselves exotic in a southern English context: not imports, but the products of imported workmen, such as those which Benedict Biscop brought from Gaul to build the monasteries at Monkwearmouth and Jarrow. If so, then the sources of the Reculver sculptures would have to be sought on the continent, either in Italy or in Gaul. Lively contacts between Kent and Gaul in the sixth and seventh centuries are attested by the grave-goods from the many rich pagan Anglo-Saxon cemeteries in the county, as well as by the marriage of king Aethelberht of Kent to the Frankish princess Bertha recorded by Bede (Bede 1969, 73–5 (
In this period Gaul had a number of flourishing regional sculptural traditions, represented by the accomplished sarcophagi of south-western Gaul (James 1977,
Is Italy a possible source of inspiration for the Reculver fragments? Again there were clearly close links between southern England and Italy in the very late sixth and seventh centuries, beginning with the Augustinian mission. These links were renewed with the despatch from Rome of further missionaries to England including Justus and Mellitus (Bede 1969, 105 (
Sixth- and seventh-century Italian art is diverse, drawing on late Classical traditions, on Ostrogothic and Lombard sources, and on Byzantine art. Within these traditions there is accomplished Classical figural art of the type used at Reculver, as in the Maria Regina scene in the presbytery at Sta Maria Antiqua in Rome (Hubert et al. 1969, 107, pl. 122). Such accomplished figural art, however, is not translated into stone, and among the many surviving sarcophagi, church fittings, and architectural sculptures produced in Italy, figural scenes are rare. Instead simple plant and animal ornament was used, and invariably worked in a flat, low-relief technique. This has little relationship to the elaborate, high relief, almost three-dimensional styles which are employed at Reculver (ibid., 250–7, pls. 277–81). It remains possible, however, that portable pictures in the style of the Sta Maria Antiqua frescos were brought to England to provide a source for the Reculver figural scenes.
In conclusion, the Reculver fragments do not fit easily in terms of style, iconography, or technique into the pattern of seventh-century art either in England or on the continent. It is evident, therefore, that the traditional seventh-century date for the fragments must be rejected.
SOUTHERN ENGLISH ART OF THE EIGHTH CENTURY
If a seventh-century date for the Reculver fragments can be rejected, the next recognisable phase of southern English art in which Classical sources were of major importance is in the eighth century. Fortunately, three decorated manuscripts from Kent survive from this period; the Vespasian Psalter (Alexander 1978, no. 29, 55–6, ills. 143–6; Webster and Backhouse 1991, no. 153), the Stockholm Codex Aureus (Alexander 1978, no. 30, 56–7, ills. 147-59; Webster and Backhouse 1991, no. 154) and the Codex Bigotianus (Alexander 1978, no. 34, 60, ills. 166–8; Webster and Backhouse 1991, no. 155). In addition the Barberini Gospels may have been made in southern England (see above). The manuscripts were probably made in that order, the Vespasian Psalter in the second quarter of the eighth century, the Codex Aureus in the mid eighth century, the Codex Bigotianus in the second half of the eighth century, and the Barberini Gospels in the late eighth century. All betray the use of Classical sources for at least part of their ornament. In the Vespasian Psalter, the Codex Aureus and the Barberini Gospels this is suggested by the use of accomplished figural decoration. A Greek Psalter of the early sixth century was a major source for the illustrations of the Vespasian Psalter (Webster and Backhouse 1991, no. 153), while Alexander has proposed the derivation of the decoration of the Codex Aureus from the St Augustine's Gospels, discussed above (Alexander 1978, 56). The Codex Bigotianus lacks figural decoration, but on fol. 137r the initial Q incorporates a naturalistically drawn dolphin (ibid., ill. 166), a characteristic Classical motif. There is, for example, a closely comparable dolphin with the same three-element tail in the Jewish Catacomb, Villa Torlonia in the Via Nomentana, Rome, dated to the third century (Du Bourguet 1972, pl. p. 17).
The use of figure decoration in the Vespasian Psalter, the Codex Aureus and the Barberini Gospels allows comparisons of iconography and style with the Reculver fragments which, with the exception of stone 1e, are exclusively figure-decorated. The Vespasian Psalter contains an introductory miniature, now displaced on fol. 30v, of king David playing the harp with musicians, dancers and secretaries in attendance (Alexander 1978, no. 29, ill. 146). Two other miniatures are placed within the large initials introducing Psalms 26 and 52 on fols. 31r and 53r (ibid., ills. 143–4). These depict David and Jonathan, and David rescuing a lamb from a lion. A third miniature, now lost, depicting Samuel, probably introduced Psalm 1 (ibid., 55). In broad terms these scenes can be described as narrative, and thus compare with the narrative scenes on stones 1b–c from Reculver (Ills. 116–18), one of which is also drawn from the Old Testament. There, however, the comparison ends, for the subject- matter of the Vespasian Psalter miniatures is not repeated at Reculver. Equally, the standing ranges of figures (probably Apostles) employed at Reculver have no place in the decoration of the Psalter. Figures of the Apostles do occur in the Codex Aureus in the two surviving author portraits prefacing the Gospels of St Matthew and St John on fols. 9v and 150v (ibid., ills. 147, 153) and, as at Reculver, each Apostle is viewed in isolation within an architectural frame. This feature has, however, no evidential value, as the convention originated in late Classical manuscripts and persisted throughout the Anglo-Saxon period.
There is also little resemblance in style between the Vespasian Psalter and Codex Aureus on the one hand and the Reculver fragments on the other. In the Vespasian Psalter the garments of the figures are less naturalistically rendered than on the Reculver fragments. On the frontispiece of the Psalter, for example (Alexander 1978, no. 29, ill. 146), the hems of the garments are treated as hard straight lines. Folds in the garments are rendered by thick, dark lines, and there is a marked tendency to reduce the folds to symmetrical patterns which only inadequately reflect the shape of the body beneath; a feature particularly noticeable in the figure of king David. Where there are edges of drapery which fall vertically in a series of folds, then the folds are treated with absolute regularity. These stylistic traits in the figure drawing recur in the Codex Aureus. In the miniature of St Matthew, for example (ibid., no. 30, ill. 153), the saint wears an overgarment which has been reduced to a series of hard dark lines, highlighted in white, which make a formalised pattern having little relationship with the saint's body beneath. This same patterning is seen in the garments of St John (ibid., ill. 147) which are arranged almost symmetrically, and where the vertical folds of drapery, as in the Vespasian Psalter, are rendered with absolute regularity. In addition, in both the Vespasian Psalter and the Codex Aureus, there is only the most rudimentary attempt at modelling, which leaves the figures flat and two-dimensional. The figure style of the Vespasian Psalter and Codex Aureus thus bears little relationship to that of the Reculver fragments, although on stone 1a (Ills. 113–15) one of the figures does have the edge of his garment falling in a series of regular folds, as in the St John portrait of the Codex Aureus (Alexander 1978, no. 30, ill. 147). However, this is only an isolated point of comparison, and the Reculver figures are imbued with an entirely different spirit. The figures are wholly naturalistically treated. They are fully modelled, and sometimes almost fully free-standing and three-dimensional. Their garments fall naturalistically in a way which reflects the shape, posture and action of the body. Hems are gently folded, not like the hard straight lines of the manuscripts.
Stone 1e from Reculver stands apart from the other fragments in the complexity of its ornament (see Reculver 1e, Discussion), which embraces a figural subject in combination with interlace and plant ornament (Ills. 119–20). In this variety of ornament it is broadly comparable to the decoration of the Vespasian Psalter. There, however, the plants are a minor element in the decoration, occurring only on fol. 30v (Alexander 1978, no. 29, ill. 146) and do not resemble the simple plant-scroll used on Reculver 1e, but rather are bush-scrolls with thick, fleshy leaves. The only point of comparison between them and the Reculver plant-scroll is in the use of groups of rounded-ended appendages developing from the underside of the branches, although this feature also occurs widely on later plant ornament. Unlike the Vespasian Psalter the Codex Aureus lacks plant ornament, except for the short sprays on either side of the seated figure of St Matthew on fol. 9v (ibid., ill. 143), and the plant spray flanking the figure in the roundel on top of the left-hand column of the same folio. It does, however, make lavish use of interlace, and the interlace decoration on the chair of St Matthew is closely comparable in form to the interlace framing on Reculver 1e.
The Codex Aureus also employs roundels containing human busts, a feature broadly comparable to the busts in the plant-scroll on stone 1e from Reculver (Ills. 119–20). Such busts occur in the Codex Aureus on the tops of the columns flanking the seated figure of St Matthew on 9v, and they also occur in similar positions in the Canon Tables, as on fol. 6v, where the outer piers of the Canon Table arcades have busts within roundels at their head and foot. This feature appears in manuscript art in Canon Tables deriving wholly or in part from the Apostolic type, and elsewhere in manuscripts which draw on such sources. They occur over a very restricted date range in Anglo-Saxon or Anglo-Saxon-influenced art from the mid eighth century to the mid ninth, in the Maaseik and Trier Gospels, the Codex Aureus, BL MS Royal I. E. VI, and the Book of Cerne (Budny 1984, 462–4; Webster and Backhouse 1991, no. 165). The busts in the Codex Aureus share the same flat drawing style without modelling, and with a tendency to show the folds of the garments as symmetrical patterns with hard dark lines. This contrasts with the handling of the busts on Reculver 1e, which are treated naturalistically with the garments falling in a manner which reflects the shape of the body beneath.
The late eighth- or early ninth-century Barberini Gospels also employs figural ornament, in the Evangelist portraits on fols. 11v, 50v, 79v, and 124v, and in the tympana of the Canon Table arcade on fol. 1r, where each column is headed with the zoo-anthropomorphic symbol of an Evangelist. The style of the figure drawing is rather different from that employed in the Vespasian Psalter and the Codex Aureus: more naturalistic and much closer in spirit to that employed at Reculver. In the Evangelist portraits in every case the figure is naturalistically posed with the feet properly positioned and correctly foreshortened as at Reculver. The garments reflect the shape of the body beneath and fall naturalistically with the folds being indicated both by line and modelling, the hems are gently folded and, as on fol. 11v, often paralleled by a coloured line. There is, however, a residual tendency to surface patterning, as on the overgarment of St Matthew on fol. 11v. Despite this the figure and drapery style lies very close to that at Reculver stones 1a–d and the Canterbury fragment. There are fewer parallels which can be drawn with stone 1e. Busts within plant-scroll do not occur in the Barberini Gospels, and where plant ornament is used, as on fols. 18r and 79v, it employs multi-lobed leaves and delicate flowers, not the pointed leaves used at Reculver. The Barberini Gospels does, however, use interlace-filled frames, as, for example, on the frames of the Evangelist portraits on fol. 11v, and in the Canon Table arcade on fol. 1r. In the arch heads half patterns are used, as on Reculver 1e.
In summary there is a wide divergence in both content and style between the Vespasian Psalter and the Codex Aureus on the one hand, and on the other the Reculver fragments, including stone 1e. There are, however, much closer links between the figure style of the Barberini Gospels and Reculver 1a–d and the Canterbury fragment, although Reculver 1e seems to have little relationship with the decorative content of the Barberini Gospels.
As with the seventh century, so in the eighth, it is impossible to find a sculptural context in southern England into which the Reculver fragments would fit. There are no stone sculptures from Kent which can be dated to the eighth century, and the only non-architectural sculptures in south-east England which may be eighth-century works are the shafts from Elstow, Bedfordshire, and Bedford St Peter, and the works of the south-western type at Winchester (e.g. Prior's Barton 1), Steventon, and South Hayling, all in Hampshire. These are on the periphery of the region and their decoration, with animals developing into, or enmeshed in, interlace, are very different in conception and spirit from the classicising Reculver fragments (see Chap. V).
In the eighth century, the closest group of figural sculpture to Reculver geographically is to be found in east Mercia, where figural panels occur at Peterborough (Cramp 1977, fig. 58b), Fletton (ibid., figs. 56a–b), and Castor (ibid., fig. 57b) in Northamptonshire, and Breedon, Leicestershire (ibid., figs. 55, 57a, 58a and c, 59 a–c). Ranges of figures also occur on the sides of the Hedda stone at Peterborough cathedral (ibid., fig. 57c). At Fletton (Kendrick 1938, pl. LXXIV) and Breedon (Cramp 1977, figs. 50–3) these figural panels occur in conjunction with narrow friezes decorated with plant, animal, interlace, pelta ornament, and (at Breedon) frets. There is also a fragment of a similar frieze at Ely, Cambridgeshire (St John's Farm, unpublished). There are generalized parallels between some of these sculptures and the fragments from Reculver. The Hedda stone, for example, employs on each of the long sides ranges of standing figures beneath arches supported by columns, a disposition which echoes that on the Canterbury fragment and Reculver 1d, although there the columns carry a flat entablature (Ills. 108, 110–12). Similar ranges of standing figures occur also at Breedon (Cramp 1977, figs. 59 a–c), Castor (ibid., fig. 57b), and Peterborough (ibid., fig. 58b). Apart from the fact that they are on panels, not on a cross-shaft as at Reculver, these figure sculptures are also on a larger scale, and lack the narrative scenes employed on stones 1b–c from Reculver. Stylistically there are two broad divisions within the figure sculptures of east Mercia. At Fletton (ibid., figs. 56 a and b), Breedon (ibid., figs. 57a, 58a) and Castor (ibid., fig. 57b) the figures are treated naturalistically, as are their draperies which have the main folds of the garments falling naturally in a way that reflects the shape of the body, and gently folded hems. However, the surface of the draperies is covered with fine parallel grooving varying in direction within each subdivision of the garment. This is an effect which is not employed in southern English manuscript art and which is absent also from the Reculver fragments. Where fine grooving does occur at Reculver it is used to point up the details of the folds of the garments, not as a mere surface enrichment as on the east Mercian examples.
At Breedon the second figure style is represented by the panel with the angel (Cramp 1977, fig. 58c). Here again the figure is represented naturalistically, but the draperies are reduced to a series of parallel folds and loops forming a surface pattern. Between the folds are deep and wide U-shaped channels which have the effect of hard dark lines, particularly when the sculptures are side lit. The treatment of the draperies provides a good parallel in stone for the style employed in the southern manuscripts, particularly in the Vespasian Psalter and the Codex Aureus, and which occurs also in the early ninth-century Book of Cerne, as on the symbol of St Matthew on fol. 2v and the imagines clipeatae on the heads of the arches on fols. 2v, 12v, 21v, and 31v (Alexander 1978, no. 66, 84–5, ills. 310–15). Indeed, Cramp has suggested that the Breedon angel and the stylistically-related sarcophagus fragments (Cramp 1977, figs. 59 a-c) are of ninth-century date (ibid., 211–18). As noted above, this figure style is in many ways the antithesis of that employed on the Reculver fragments. As with the first east Mercian figure style this material provides no adequate context against which the Reculver fragments might be viewed. The only stylistic link lies in the turning up of the edges of the hem so as almost to form a volute, as on the Canterbury fragment (Ill. 109), on the Breedon angel, and to a lesser degree on fol. 2v of the Book of Cerne (Alexander 1978, no. 66, ill. 312).
This conclusion applies also to the unusually-decorated stone 1e from Reculver where the figure sculpture follows the small-scale, naturalistic style employed on the rest of the fragments. The plant-scroll and interlace employed in conjunction with the figure sculpture on stone 1e (Ills. 119–20) can in general terms be paralleled in east Mercia, particularly on the friezes from Breedon (Cramp 1977, figs. 50, 51, 52 a–c; Jewel 1986, pls. XLII, XLIII, XLV, XLVII–LIII), but detailed analysis reveals that the structure and leaf type of the Reculver scroll differs markedly from that at Breedon. There the stems of the scroll are narrow and wiry with the diverging stems developing from trumpet bindings. Where leaves are used they are trefoils or ovoids, and they are used in conjunction with berry bunches. These are absent from stone 1e where the leaves are single-lobed, pointed, and hatched to simulate veins. A distinctive motif is the leaf bent back under its own stem. All these features are absent at Breedon. Trumpet bindings do occur at Reculver, but they are treated less as three-dimensional cones, as at Breedon, and more as organic parts of the plant. Some of the scrolls at Breedon are inhabited by birds, animals, or centaurs, but none has protamoi, as at Reculver.
ART IN KENT IN THE EARLY NINTH CENTURY
Only two manuscripts which were probably made in Kent in the early part of the ninth century survive, BL MS Cotton Tiberius C. II (Alexander 1978, no. 33, 59–60, ills. 134, 165; Webster and Backhouse 1991, no. 170) and BL MS Royal I. E. VI (Alexander 1978, no. 32, 58–9, ills. 160–4; Budny 1984; Webster and Backhouse 1991, no. 171). Tiberius C. II is a copy of Bede's Historia Ecclesiastica. The decoration is confined to the initial letters and continuation lettering at the opening of each of the five books, on fols. 5v, 34v, 60v, 94r and 126r. The first of these, the initial B, is the most elaborate. The letter is panelled, each panel containing plant, interlace, animal, or fret decoration. The centre of the letter is divided into four fields, each filled with a single animal (Wilson 1984, pl. 111). The decoration of this initial is clearly of the same type and in the same style as the decoration of the Canon Tables of Royal I. E. VI, where the panelled decoration is almost identical in layout and decorative content.
BL MS Royal I. E. VI is a fragment of a luxurious Bible of which only the Gospels partially survive. Of the original decorated pages only the Canon Tables, fols. 4r–6r, and the incipit page of St Luke's Gospel on fol. 43r survive, together with small decorated initials on fols. 42r and 68r. The openings of the gospels of St John, St Matthew, and St Mark are lost, but an offset indicates that the opening of St John's Gospel closely resembled that of St Luke (Budny 1984, 442-5, fig. 10). At least three full-page miniatures are also lost. The former presence of these is indicated by their facing pages which are purple stained and carry inscriptions in gold and silver display lettering. These inscriptions serve to identify the lost scenes as the Lamb of God surrounded by the symbols of the four Evangelists opposite fol. 1v; the Baptism of Christ prefacing the Gospel of St Mark opposite fol. 30r; and the Annunciation to Zachariah prefacing the Gospel of St Luke opposite fol. 44r. The presence of numerous other illustrations can be inferred (ibid., 688–727, fig. 13). The manuscript bears the late thirteenth-century press mark of St Augustine's abbey, Canterbury, although the folio which has this mark is not original to the manuscript (ibid., 199–213). Despite this there seems little doubt that the manuscript was made at St Augustine's (loc. cit.) probably in the period 815–845 (ibid., 756).
Comparison between the Reculver fragments and these manuscripts presents an almost insuperable problem. As noted above, all but stones 1a and c from Reculver are decorated solely with figural ornament, but BL MS Cotton Tiberius C. II lacks figure decoration, and while BL MS Royal I. E. VI once possessed an extensive programme, only a bust within a roundel on the incipit page of St Luke's Gospel survives. Unfortunately on this page the normal portrait of the Evangelist beneath the arch, employed for example in the Codex Aureus (Alexander 1978, no. 30, ills. 147, 153), has been replaced by the opening of the text. Despite this severe handicap there is an evident comparison between the handling of the bust in Royal I. E. VI and the figure sculptures at Reculver. The manuscript bust is naturalistically drawn. The figure is well proportioned apart from the hand raised in blessing, where the two raised fingers are unnaturally extended. The figure's garments are also naturalistically drawn, reflecting the shape of the body. Both the figure and its draperies are carefully modelled, without the reliance on line evident in the earlier manuscript art of Kent.
This naturalistic treatment is carried through onto the Evangelist symbol, a half-length bull, in the tympanum of the arch beneath the bust. Again the animal is naturalistically drawn and well proportioned with a combination of thin, delicate line to pick out the detail and skilful modelling to yield a three-dimensional effect. The revolutionary change which overtook figural drawing in Kent between the mid eighth century and the early years of the ninth is underlined by a comparison between this Evangelist symbol and those of the Codex Aureus. The St Luke portrait is lost, but the St John portrait survives (ibid., pl. 147). As in all the drawing of the Codex Aureus the symbol of St John, the eagle, is depicted as two-dimensional, and the drawing of the feathers has turned them into a regular surface pattern picked out by hard, thick, dark lines. The outspread wings of the eagle can be compared directly with those of the bull in Royal I. E. VI. In the Codex Aureus the feathers are formalised and regular with the usual hard, dark outlines and thick white highlights. In Royal I. E. VI the feathers are outlined with thin lines, and the white highlighting has yielded to more subtle modelling using accomplished gradations of shading. The effect is to make the wing life-like. In the Codex Aureus the eagle symbol is set against a uniform plain coloured ground. In Royal I. E. VI the ground is treated impressionistically to give a feeling of depth and space.
The handling of the figure and animal drawing in Royal I. E. VI is clearly comparable to the manner of handling the figure sculpture on the Reculver fragments, even if direct comparisons are difficult to draw. The figures on the Reculver fragments are portrayed wholly naturalistically; they are well proportioned and naturalistically posed with the folds of their garments reflecting the shape of the bodies beneath. The folds of the draperies are carefully delineated without the tendency to surface patterning seen in eighth-century works such as the Codex Aureus. As in BL MS Royal I. E. VI, the Reculver figures are fully modelled and, on the Canterbury fragment, heavily undercut (Ills. 108–10) to give the three-dimensional effect to which Royal I. E. VI also aspires.
Stone 1e from Reculver has more concrete links with Royal I. E. VI. Like that manuscript it employs busts within roundels, although here within a plant-scroll (Ills. 119–20). The busts on 1e are stylistically close to that in the manuscript. Like it they are naturalistically drawn and fully modelled with the draperies of the garments fully reflecting the shape of the body beneath. Completely absent are the hard, formalised patterns of eighth-century manuscript and sculptural draperies, seen, for example, in the busts within roundels in the Codex Aureus (ibid., pls. 153, 155).
Stone 1e also has links with BL MS Royal I. E. VI in that, like the manuscript, it employs a mixture of motifs: figural, plant, and interlace; a mixture which, as noted above, is typical of Anglo-Saxon art of the late eighth and early ninth centuries. There are no specific comparisons between the interlace employed on 1e and that of the manuscript, but there are exact parallels in the plant ornament. In particular the fan-shaped groups of rounded-ended extensions developing from the upper tendril of the stem on 1e can be paralleled in the Canon Tables of Royal I. E. VI, as, for example, on fol. 4r. The use of the pointed leaf turning back under its own stem, seen both on stone 1e and the lost stone (Reculver 2), is also paralleled on fol. 4r. Indeed, so close is the comparison that Budny has concluded that the same 'model or type of model' must lie behind both works (Budny 1984, 630–1). These are both features which occur also in ninth-century Anglo-Saxon metalwork. The leaf turning under its own stem, employed on Reculver 1e (Ills. 119–20) and in BL MS Royal I. E. VI, is used, for example, on the Fuller brooch (Fig. 15a; Wilson 1964, no. 153, 211–15, pl. XLIV; Backhouse et al. 1984, no. 11) and one pair of the brooches from Pentney, Norfolk (Fig. 15b; Wilson 1984, pl. 120, top), as well as on a strap-end from the Cuerdale hoard (Wilson 1964, no. 13, 128-9, pl. XVIII). The use of groups of foliate elements which expand towards rounded ends is paralleled on the sword-pommel from the River Seine, Paris (Fig. 24c; ibid., no. 66, 166–7, pl. XXIX), and on the Fuller brooch, where similar elements emerge from the cornucopias held by the figure of Sight in its centre (Fig. 24f; ibid., pl. XLIV).
There are, then, very close and specific comparisons between the decoration of BL MS Royal I. E. VI and stone 1e from Reculver, and striking, if generalised, comparisons between the handling of the figure decoration in the manuscript and that on stones 1a–d from Reculver and the Canterbury fragment. Certainly if the Reculver fragments can be regarded as of ninth-century date there is a sculptural context in south-east England at this period into which they will fit. For example, the mixture of animal, plant and interlace ornament seen on Reculver 1e is mirrored on the stone ring from Godalming, Surrey (no. 1; Ills. 81–91), on which the plant-decorated field employs the same pointed leaf turned under its own stem (Ill. 86). The Godalming piece must be of early ninth-century date as it compares directly with Royal I. E. VI which, in the Canon Table arcades, has the same type of panelled ornament (Alexander 1978, no. 32, ills. 162–4). As at Godalming the panels are filled with animal, plant or interlace ornament, often alternating with each other; very similar decoration is employed also in BL MS Cotton Tiberius C. II (ibid., ill. 165; Wilson 1984, pl. 111) (see above).
There are also other sculptures which employ small-scale classicising sculpture, discussed more fully in Chap. V. Frustratingly, there is nothing of this type from Kent itself, but there are pieces from the Winchester area. A small-scale figure, only 25 cm high, occurs on the grave-marker from Whitchurch, Hampshire (Ill. 483). Unfortunately this is heavily weathered but appears to be in the same naturalistic, classicising style as the figure sculptures from Reculver. At Romsey, Hampshire, is another small-scale figure sculpture, this time a panel decorated with a crucifixion scene (no. 2). The figures are, however, treated in a very different manner from those at Reculver and Whitchurch, in a flat, unmodelled style with the draperies picked out by incised lines and having little relationship with the bodies beneath (Ills. 453, 455). This is very much in the tradition of the late eighth-century east Mercian material discussed above, but the use of acanthus ornament closely comparable to that on the St Cuthbert stole and maniple places the piece firmly in the late ninth or early tenth century. Perhaps this piece represents the continuity of older figural styles into the tenth century, or perhaps it is simply the work of a less accomplished sculptor who strove for the effects seen at Whitchurch or Reculver but lacked the technical competence to achieve them.
If, as argued above, the classicising figure sculpture at Reculver and Whitchurch belongs to the same artistic milieu as the classicising figure drawing of BL MS Royal I. E. VI, then the manuscript seems to pinpoint the sources of this new impulse in ninth-century art, sources which appear to lie both in late Classical and Carolingian art (Budny 1984, 795–800). The importance of Carolingian sources in the art of the ninth century is something which has been argued for other areas of Anglo-Saxon England, particularly Northumbria. There are, unfortunately, no identifiable surviving Northumbrian manuscripts made after the mid eighth century, but there are a number of sculptures which can be distinguished by their use of small-scale, classicising figure decoration combined with technical excellence. These include the Otley, Easby, and Rothbury shafts, the cross-head at Hoddom, and the slab at Hovingham. At Otley the so-called 'Angel Cross' employs on one of the main faces a series of three vertically-placed panels each of which presumably originally contained a human bust within a single element of a medallion plant-scroll; only one now survives (Cramp 1970, taf. 41, A, 42, 1; Collingwood 1927, fig. 52). This provides the closest parallel in Anglo-Saxon art to the plant-scroll with busts used on stone 1e from Reculver (Ills. 119–20), and is a feature unparalleled outside these two sites. Other busts within roundels occur in Northumbrian art of the ninth century, most notably at Hoddom, Dumfriesshire, where the cross-head has a central roundel containing a human bust holding a book. The bust is rendered in a technically highly accomplished manner (Collingwood 1927, fig. 51). The figure is naturalistically rendered with carefully delineated robes reflecting the shape of the body beneath, a treatment which closely resembles that at Reculver, and the bust in BL MS Royal I. E. VI. A very similar bust in the centre of a cross is known also from Little Ouseburn, Yorkshire.
Other features of the Reculver sculptures are also reflected in Northumbrian art of the ninth century. At Masham, Yorkshire, for example, there is a round shaft decorated with superimposed ranges of decoration, some of which consist of rows of figures beneath arches (Collingwood 1927, fig. 13), a layout which closely reflects that of Reculver 1a–d and the Canterbury fragment, and a similar range of figures is employed on the rectangular panel from Hovingham (Lang 1991, ills. 494–9). Here, as at Reculver, the figures are small-scale and highly classicising, but more detailed analysis is rendered impossible by their poor state of preservation. That there was any direct link between these Northumbrian sculptures and the art of south-east England is, perhaps, unlikely, but what can be suggested is that Northumbria was subject to the same artistic impulses which affected south-east England, and that these impulses lead to the creation of similar works in two geographically widely separated areas.
THE ART OF WINCHESTER IN THE EARLY TENTH CENTURY
The fragments of large-scale figure sculpture found in excavation at Winchester (nos. 75–84) cannot now be related to the type of elaborate sculptural scheme described as adorning the tenth-century tower of New Minster (Quirk 1961), and (with the exception of no. 75) is probably all later than the early tenth century. Moreover, between the middle of the ninth century and the late tenth century no figure-decorated work of art in any medium survives from Kent. The nature of southern English art is indicated by the three major representatives of the figural art of early tenth-century Winchester which do survive: the St Cuthbert stole and maniple (Battiscombe 1956, 375–432, pls. XXIV–IX, XXXII–IV), the Aethelstan Psalter (BL MS Cotton Galba A. XVIII (Temple 1976, no. 5, ills. 15–17, 30–3)), and the Corpus Christi Vita Cuthberti (Corpus Christi College Cambridge MS 183 (ibid., no. 6, ills. 18–19, 29; Wilson 1984, pl. 203)). The evidence of these is supplemented by a fragment of wall painting from the New Minster excavations at Winchester (Biddle 1967b; Biddle and Kjølbye-Biddle 1990b; Wilson 1984, 155–6, pl. 204). These artifacts are of special relevance to any evaluation of Reculver.
Of the major works the earliest in date are the St Cuthbert stole and maniple. The stole bears the names of Queen Aelflaed of Wessex and Frithestan, bishop of Winchester. It must, therefore, have been made between the consecration of Frithestan in 909 and the death of Aelflaed in 916 (Freyhan 1956, 409). The name of Frithestan also appears on the maniple (loc. cit.), which given its similarity in style and technique with the stole must have been made at the same date and probably by the same hands. The stole is decorated with a series of standing figures, identified by inscriptions as the Old Testament prophets. The figures are placed vertically (a layout dictated by the shape of the object) and are separated by sprays of acanthus ornament. In broad terms this use of a series of compartmentalised figures corresponds with the decoration on stones 1a and 1d from Reculver and the Canterbury fragment.
In the poses of the figures there are also close comparisons between the figures on the stole and some of those on the Reculver fragments. Each of the figures on the stole stands on a rocky ground and is half-turned to one side or the other in regular alternation. The figures are bearded with the hair falling down behind. The figure's left hand holds either a palm leaf, as is the case with Hosea, Isaiah, Daniel, Amos, and Jonah, or holds a book as do Joel, Nahum, and Zachariah. In each case the figure's right hand is raised in blessing. Each figure is clad in a long undergarment with an overgarment bound at the waist and falling diagonally across the body. On the Reculver fragments the left-hand figure on stone 1d (Ills. 111–12) closely resembles the prophets on the stole. Like them the figure is half-turned to one side and, as on the stole, the range of figures may have alternated (the right-hand figure on stone 1d is half-turned to the other side) but with only two figures surviving this may be mere chance; it does not occur on stones 1a and the Canterbury fragment which also have ranges of figures. The figure on stone 1d, like the figures on the stole, is also bearded with his right hand raised in blessing, but the left hand holds a scroll rather than a book. Again the figure is clad in an undergarment, with an overgarment bound at the waist and falling diagonally across the body. The other figures from Reculver are less easy to parallel on the stole. The figures on the Canterbury fragment (Ills. 108–10) are too badly damaged for meaningful comparisons to be drawn. On stone 1a the figures are frontally placed (Ills. 113–15), but if the metal fittings on the borders between them carried identifying inscriptions, as suggested elsewhere (see Reculver 1a–e, Discussion), then they would have been comparable to the vertically-placed inscriptions on the stole. The rocky ground employed on the stole can also be paralleled at Reculver where a similar ground is used on stone 1b (Ill. 116).
The comparison of style between the stole and the Reculver fragments presents more difficulties than comparisons of iconography, difficulties which arise out of both the poor state of preservation of the Reculver fragments, and from differences in medium. The figures on the stole are treated in a naturalistic, classicising manner. The figures are naturalistically posed with the left leg slightly flexed as if to take the weight of the body. The feet are also carefully drawn with the right foot foreshortened and the left foot in profile. Like the figures the garments are naturalistically rendered, reflecting faithfully the shape and movement of the body beneath. With the figure of Daniel, for example, where the left arm is raised to hold the palm leaf the loose sleeve of the garment is shown as having fallen back to the elbow, leaving the forearm bare. There is no working up of the figure using shading and highlighting to give solidity to the bodies, but this is a painter's technique which is inappropriate in embroidery, where instead there must be a dependency on line.
As noted elsewhere (see Reculver 1a–e, Discussion) the Reculver stones exhibit an almost identical naturalistic style, although with sculpture the solidity of a body can be faithfully reproduced and there is, therefore, less dependence on line. One particular stylistic trick is common to the stole and to the Canterbury fragment, the turning up of the hemline to each side in an unnaturalistic manner. This occurs with the figures Amos, Jonah, Joel, and an incomplete, unnamed prophet, and on the left-hand figure on the Canterbury fragment. Here the hem is turned up (particularly to the left) and curled back (Ill. 109), and is not as elaborately folded as is usually the case on the stole, but the centre of the hem is an elaborate double fold which precisely parallels that on the figure of Amos on the stole (Ills. 108, 110). There can be little doubt that this use of the billowing hem on the stole is a forerunner of the elaborate treatment accorded to the hem in works of the Winchester style, and the occurrence of this feature at Reculver may be of significance in dating. Unfortunately, it is only an isolated occurrence. The hems of the figures on stones 1a and 1d do not survive, and the feature is absent from 1b–c.
There are, then, some links both in iconography and style between the St Cuthbert stole and elements of the decoration on the Reculver fragments. Similar links can also be identified with the Aethelstan Psalter (BL MS Cotton Galba A. XVIII). This is a ninth-century Carolingian psalter probably made in the region of Liège (Temple 1976, no. 5, 36). To it have been added in England a metrical calendar (fols. 3r–14v), compotus matter (fols. 15r–19r) and a Greek litany (fols. 178r and 200r) (loc. cit.). Four miniatures have also been added, including two Majesty scenes (fols. 2v and 21r; ibid., ills. 32–3), the Ascension (fol. 120v; ibid., ill 31), and the Nativity originally prefacing Psalm 1.8 The additions were probably made in Winchester in the reign of king Aethelstan (924–39) (ibid., 37).
Iconographically, the closest link with the Reculver stones is provided by the first Majesty scene (Temple 1976, ill. 32). This has Christ seated in a mandorla adored by choirs of the angels and prophets. The figures are divided into four ranges separated by plain coloured bands, a disposition which echoes the superimposed ranges of figures which originally decorated the Reculver cross. Unlike the Reculver figures, those in the Psalter are not compartmentalised, but instead are partially overlapping. Compartmentalised figures do, however, occur in the calendrical material added to the Aethelstan Psalter where, in the bottom left-hand corner of some folios, there is a single standing figure within a plain rectangular border (ibid., ill. 15). Some of these figures compare closely with those on the St Cuthbert stole and maniple. The use of plain borders to contain figures can be paralleled on stone 1a from Reculver (Ills. 113–15), although elsewhere architectural frames are employed. In the Aethelstan Psalter the dividing frames of the Majesty scene are used to carry inscriptions (ibid., Ill. 32), something which may have occurred at Reculver. The Aethelstan Psalter even has curved inscriptions on the mandorla of the second Christ in Majesty scene (ibid., ill. 33), something which may be postulated on stone 1b from Reculver (Ill. 116) where the lost curved metal fitting may have carried an inscription.
There are also some specific iconographical links between the Reculver stones and the Aethelstan Psalter. For example, the figures on the lower range of the first Christ in Majesty scene are all partially turned to one side. The right hand of each figure is raised in blessing and the left hand holds a book. Each figure is clad in a long undergarment with an overgarment bound at the waist and falling diagonally across the body. Some of the figures are bearded with their hair falling down behind. These figures are closely comparable to those of the prophets on the St Cuthbert stole, and also to the left-hand figure on stone 1d from Reculver (Ills. 111–12).
Stylistically, there are also parallels between the Aethelstan Psalter and the Reculver stones. The Aethelstan Psalter uses a naturalistic, classicising style of figure drawing, although one hand in the figure drawing is more accomplished than the other. In every case the figures are naturalistically posed with the draperies also rendered naturalistically and faithfully reflecting the shape of the body beneath. In each case the emphasis is not on line but on modelling to portray the figures in a three-dimensional manner. Stylistically, this precisely reflects the aims of both the Reculver fragments, discussed at length above, and the figures of the St Cuthbert stole and maniple.
The fragment of wall painting from the New Minster excavations at Winchester, is small, but again is figure-decorated, although little survives apart from several superimposed busts (Wilson 1984, pl. 204). Wormald has suggested that these can be compared directly with the superimposed ranges of busts flanking the second Christ in Majesty scene in the Aethelstan Psalter (Temple 1976, no. 5, ill. 33), a scene which he had argued, before the discovery of this fragment, was probably derived from a wall painting. There are no specific comparisons which can be made between this fragment and the Reculver stones, but it is useful in providing corroborative evidence for the dating and Winchester provenance of the manuscript.
The last of the major Winchester works of the early part of the tenth century is the Corpus Christi Vita Cuthberti. This was probably made in Winchester in the reign of king Aethelstan (Temple 1976, no. 6, 38) and incorporates on fol. 1v a framed frontispiece showing king Aethelstan presenting the book to St Cuthbert (Wilson 1984, pl. 203). The manuscript can probably be identified as the book given by king Aethelstan to the community of St Cuthbert at Chester-le-Street in 937 (Temple 1976, 38). Unfortunately this work has no iconographic parallels with the Reculver fragments, although the presentation scene is probably based on a late Classical exemplar transmitted through Carolingian intermediaries (loc. cit.). Similarly the framing relies on Carolingian antecedents and is possibly derived from manuscripts of the Court School (loc. cit.). Stylistically also the Vita Cuthberti has little in common with the Reculver fragments except that the figures are rendered naturalistically, although their poses are wooden and the folds of the garments tend to be formalised. This handling reflects that of the Reculver fragments although no specific comparisons can be made.
In summary there are both general and specific points of comparison in both iconography and style between the Aethelstan Psalter and St Cuthbert stole on the one hand, and some of the Reculver fragments on the other. There are no specific comparisons which can be made with the Vita Cuthberti, or the New Minster wall painting fragment, although these clearly belong to the same artistic milieu as the St Cuthbert stole and maniple and the Aethelstan Psalter. It seems likely, however, that all these works draw on ninth-century sources and represent the end of ninth-century traditions, before the Winchester style was developed and overtook them. Certainly the fragmentary wall painting from the New Minster excavations was discovered in foundations which can be dated to c. 903, placing the painting firmly in the ninth century (Biddle and Kjølbye-Biddle 1990a). These Winchester products may, therefore, represent the end of a tradition which began with works such as BL MS Royal I. E. VI. That these works are later in date than the Reculver fragments is confirmed by their use of acanthus foliage. This is found on the corners of the frame enclosing the second Christ in Majesty scene in the Aethelstan Psalter (Temple 1976, ill. 33), and similar acanthus is used to separate the scenes on the St Cuthbert stole and maniple (Battiscombe 1956, pl. XXXIV). Acanthus is absent from the New Minster wall painting, but occurs in the frame of the presentation scene in the Vita Cuthberti, although incorporating features of vine-scroll, such as berry bunches, and birds and animals inhabiting the foliage (Wilson 1984, pl. 203). On stones 1e and 2 from Reculver, in contrast, the plant form is the earlier form of vine- or plant-scroll (Ills. 119–20, 122).
THE WINCHESTER STYLE
In the tenth century a new phase of classicising art developed in southern England, the so-called 'Winchester style' (Kendrick 1949, 1–54; Temple 1976, 12–17). This was primarily a manuscript style, but is found on works in other materials, especially in ivory, but also in metal. Here the emphasis of the decoration is on the figural scenes, often contained in lush acanthus-filled frames. Interlace and animal ornament, hitherto such important components of Anglo-Saxon decoration, were relegated to a relatively minor role, occurring most commonly in initials. The style was not, however, uniform and monolithic, and different styles of drawing can be discerned (Temple 1976, 12–22). One, drawing on various Carolingian sources (ibid., 17), is exemplified by an outline drawing of St Dunstan at the feet of Christ (Oxford, Bodleian Library MS Auct. F. 4. 32, fol. 1r, made at Glastonbury c. 950 (ibid., no. 11, ill. 41)). Here the figures are calm, monumental, and solid, although clad in light fluttering draperies (ibid., 17). The second style probably derives from Carolingian manuscripts of the Rheims school (ibid., 17), and in it figures are lively and excited, the focus being on the action of the scene. Gestures are exaggerated and draperies flutter around the figures with their edges folded in masses of small zig-zags (ibid., 18–22). The earliest example of the style may be in the Presentation Scene of the New Minster Foundation Charter (ibid., no. 16, ill. 84). This is often dated to c. 966 (ibid., 18), but is probably a later work. These two styles, the calm and monumental, and the dramatic, clearly interacted and towards the mid eleventh century fused to form a new style characterised by an impressionistic manner allied with fuller modelling (ibid., 22–4), as in the Judith of Flanders Gospels (ibid., nos. 93–4, ills. 285–6, 289).
The Winchester style clearly developed in manuscripts, and nearly a hundred such manuscripts survive, but it did translate easily into ivory, as with the figures of the Virgin and St John from St Omer (Beckwith 1972, no. 25, pls. 57–8) and a plaque from Winchester (ibid., no. 16, pl. 39). Works in the Winchester style in other materials are less common. In metal there is the engraved decoration around the altar in the Musée de Cluny (Backhouse et al. 1984, no. 76, pl. 76), on the reverse of the Brussels cross (ibid., no. 75, pl. 75), on the Heribert crosier (Beckwith 1972, no. 30, pls. 80–1), and a cast figure from Colliergate, York (Backhouse et al. 1984, no. 267, pl. 267). Only the major figural styles which have a specific relevance to sculpture are discussed with special reference to Reculver.
In stone sculpture the rich decoration which once adorned the Old Minster after its extensive alterations in the late tenth and early eleventh centuries now survives only in fragments, though scraps of acanthus ornament (no. 71; Ill. 615) hint at a relationship to manuscript styles, and the remains of large-scale figural decoration hints at the elaboration of what has been lost (nos. 76–84; Ills. 620–2, 625–38). The more complete works which do survive should probably be seen as provincial reflections of the lost major works in the capital itself. Two angels from Bradford on Avon, Wiltshire (Rice 1952, pl. 7) compare closely in style with the manuscripts and ivories, while the crucified figure of Christ from Romsey, Hampshire (no. 1), closely relates to similar scenes in the manuscripts, such as BL MS Harley 2904 (Temple 1976, no. 40, ills. 140–2). The lush acanthus employed in the manuscripts is virtually absent from stone sculpture, but there are fragments of fully-developed Winchester-style acanthus from Avebury, Wiltshire (Cramp 1975, 189), and Peterborough cathedral (Backhouse et al. 1984, no. 137, pl. 137).
Against this background the Reculver fragments sit rather uneasily, although Talbot Rice has suggested a tenth-century date for them (Rice 1952, 97–8). The plant-scroll employed on stone 1e cannot be paralleled in the late tenth and eleventh centuries, when acanthus was the predominant form of foliage. The interlace borders which contain the plant-scroll on stone 1e are equally difficult to parallel among works of the Winchester style. Reculver 1e (Ills. 119–20) and the closely allied no. 2 (Ill. 122) cannot belong to this period.
Turning to the figure-decorated drums, the figures at Reculver are still and composed, particularly on 1a and d (Ills. 111–15), and the fragment from Canterbury (Ills. 108, 110). Even on stones 1b–c, which are narrative, there is a stillness which is quite at variance at least with the second of the Winchester styles, as a comparison between the Ascension scene on stone 1b and the Benedictional of St Aethelwold (Temple 1976, no. 23) demonstrates. On stone 1b, although Christ is climbing a hill, the scene is static, without any upward movement (Ill. 116). In the Benedictional, in contrast, the figure of Christ leaps into the air in a swirl of draperies, reaching forwards towards the Hand of God; the entire scene is filled with movement and life. It is possible that the Reculver figures belong to the more monumental figure style, but even here the draperies are handled very differently, with fluttering ends flying away from the body, and elaborate zig-zag folds with a distinctive broken profile. All this is very far from the naturalistic, but calm and still effect conveyed at Reculver. It seems clear, then, that the Reculver fragments must belong to the period before the Winchester style became so all-embracing, before say c. 950.
CONCLUSION
In order successfully to place the Reculver fragments it is necessary to find a period in which certain criteria can be met: when round shafts were made and used; when the combination of interlace, plant-scroll and naturalistic figures employed on stone 1e was popular; and when a naturalistic, classicising figure style was current.
The form of the shaft, of circular section and drum-built, may serve to suggest a terminus post quem for the making of the Reculver monument. As noted elsewhere (see Reculver 1a–e, Discussion) round shafts only seem to have been introduced in the ninth century, and were used, if infrequently, in southern England thereafter, although the making of specifically columnar monuments is rare enough to urge caution in pressing this argument too far. This possible terminus post quem is supported by the decoration of stone 1e, which mixes naturalistic figure sculpture, plant-scroll and interlace. This mixture of motifs is typical of the art of the late eighth and ninth centuries (Budny and Tweddle 1984, 78). It is seen, for example, in manuscript art in BL MSS Cotton Tiberius C. II (Alexander 1978, no. 33, ills. 134, 165) and Royal I. E. VI (ibid., no. 32, ills. 161–4), although both works also employ frets and animal ornament. In other materials the use of figural ornament in conjunction with interlace and plant-scroll is unparalleled, but the mixture of different motifs is common. The Maaseik embroideries, for example, combine plant, animal, and interlace ornament (Budny and Tweddle 1984, 78-84, ills. I–VI), and a similar mixture is often encountered in metalwork, for example, on the brooches from Pentney, Norfolk (Wilson 1984, pl. 120). Foliate decoration was itself an introduction into southern English art only in the late eighth century (Budny and Graham-Campbell 1981, 11), but the precise form of the foliage on stone 1e from Reculver can only be paralleled in the ninth century, particularly in Royal I. E. VI, and in some pieces of metalwork decorated in the Trewhiddle style.
The absence of comparative material means that the figure style of fragments 1a–d and the Canterbury fragment is more difficult to parallel, but specific comparisons can be drawn, especially with BL MS Royal I. E. VI, and the Barberini Gospels. Moreover, it is in this period, the early ninth century, that small-scale classicising sculpture was introduced elsewhere in southern England, most notably in Hampshire. The terminus ante quem must lie at the latest in the mid tenth century. The ornament of the fragments has no affinities with works decorated in the Winchester style. The figure style is very different and much less agitated than that employed in the Winchester style. The foliage is plant-scroll, not the acanthus used universally in the Winchester style. The only parallels which can be drawn between the Reculver fragments and tenth-century works are with products of the early part of the century, before the Winchester style had fully formed and become so universal. In particular, parallels can be drawn with the St Cuthbert stole and maniple (Freyhan 1956, 409-32, pls. XXIV–XXV, XXXIII–XXXIV), and with the Aethelstan Psalter (Temple 1976, no. 5, ills. 30–3). Both works employ a very similar figural style to that employed at Reculver.
The date of the Reculver fragments probably therefore lies between the early ninth and early tenth centuries. This presents a problem, as the early tenth-century works noted above are usually regarded as innovative, representing a new phase in Anglo-Saxon art, the precursors to the Winchester style (Backhouse et al. 1984, 19). Given the lack of late ninth-century works from southern England it is, however, possible that these works constitute a continuation of ninth-century traditions, before the Winchester style developed and overtook them. The discovery of the fragmentary wall painting from the foundations of the New Minster in an archaeological context dated to c. 903 (Biddle 1967b; Wilson 1984, 155, pl. 204), places it at least in the ninth century and may confirm that the closely-related Aethelstan Psalter represents late ninth-century artistic traditions. If this conclusion can be accepted, it reconciles some of the difficulties in dating the Reculver sculptures, and places them firmly in the ninth century, perhaps, given the close links with BL MS Royal I. E. VI, in the earlier part of that century.
NOTES
8. The latter leaf is now detached and in Oxford, Bodleian Library, MS Rawlinson B. 484, fol. 85r; ibid., ill. 30.
