A History of the County of Oxford: Volume 1. Originally published by Victoria County History, London, 1939.
This free content was digitised by double rekeying. All rights reserved.
III. Bronze Age
Early Bronze Age
With the passing of the Neolithic epoch the picture becomes clearer. For although, as already noted, no settlement-site of the western or Windmill Hill type has as yet come to light in the county, it can only be chance that makes this so. At any rate, there is ample evidence to prove the permeation of Oxfordshire by representatives of the eastern or Peterborough culture, and that indeed only shortly in advance of the coming of the beaker-folk, or, as will be seen, in actual association with them. The former of these intrusive cultures—for initially Oxfordshire must have lain within the sphere of its western counterpart—has been established already for some years past owing to the fortunate recovery of three intact bowls of its distinctive pottery, soapy in texture and often highly decorated, from the Thames at Mongewell, two of which (Pl. III d) are now in the collection of Mr. G. W. Smith of Reading, (fn. 1) and the third in the British Museum, while other imperfect material was found in 1910 up the Cherwell just across the Northamptonshire border at Astrop. (fn. 2) Within the last three years new light has been thrown upon its relationship to other cultures in the Thames Valley by discoveries at Cassington (Pl. III a) and at Linch Hill, Stanton Harcourt. (fn. 3) From three areas near the former village sherds have been found in circular pits, some black and highly decorated, others redder in colour and with careless or superficial decoration, suggestive of a technical advance and an artistic deterioration in the old native wares that finds some corroboration in sherds of even more slovenly and decadent appearance, which were associated in one of the pits with a single sherd of finely potted beaker-ware. A group of chance finds at Stanton Harcourt, including portions of a large bowl of the Mongewell type (Pl. III b), comprises forms like some from Astrop that herald the overhanging-rim urn of the Bronze Age. Possibly earlier than these is a single sherd decorated with curvilinear lines found in a fire-hole ¼ mile east of Asthall Barrow (Pl. III c). (fn. 4)
The close association of this culture with that of the beaker-folk, already long suspected and recently proved by discoveries in Wiltshire, has received interesting confirmation from the further discovery at Cassington of a small cemetery of the beaker period. (fn. 5) In this contracted burials were explored in round or oval pits excavated some 3 ft. in the gravel. Several were accompanied by beakers, all, save possibly one, of Abercromby's A—C type; these were mostly large and clumsily made, but fully decorated, two with lozenge panels (Pl. V c), while another is provided with a handle and was accompanied by a copper or bronze awl or tattooing-needle. One remarkable feature was the wide range of skull-types among the interments. These varied between an index of 65 and one of 87, with a predominance of dolichocephalic or sub-dolichocephalic types instead of the brachycephalic form that is generally expected with beaker interments. Remains of eight skeletons in disorder came from a possible hut-site, and with them was a sherd of black Peterborough ware. A short distance away other beakers came to light, one of A—C form, the other of B type with the characteristic horizontal lines of comb ornament, and with it was a shale ring, 13/16 in. in diameter, with a low flange surrounding its wide perforation on both sides (Pl. V d).
With these Cassington beakers may be compared others discovered in the county. From Yarnton and Summertown lower down the river come beakers that are very closely related in form and decoration to the two last-mentioned specimens from Cassington. Two from Summertown bear witness to occupation of the gravelspur along which north Oxford now stretches for over two miles, and additional evidence is furnished by two beakers found in Polstead Road, on the same spur, ½ mile south. Of these latter, one is a small A—C example with lozenge ornament and zigzag bands separated by reserved bands (Pl. V a), while the other and those from Summertown belong to the B type. Two of them, now in the British Museum, together with a barbed flint arrow-head, were found with a burial. (fn. 6)
Higher up the Thames, about ¾ mile along the road from Eynsham to Stanton Harcourt, (fn. 7) a child's skeleton in a circular pit was furnished with a handled beaker ornamented with lozenges (Pl. V b), like one from Brixworth, Northants., and with in-turned rim comparable to that of one specimen from Cassington and of a fragment of another from Wytham on the Berkshire bank opposite Cassington. The Linch Hill site at Stanton Harcourt has also yielded a group of four, all of A—C type, either plain or simply decorated, while farther west at Hardwick near Standlake Down was found in excellent preservation a degenerate A beaker with vertical zigzag lines impressed with a square-toothed comb. (fn. 8)
For south Oxfordshire records of beakers are scarcer. One was found at Clifton Hampden in 1864 with a skeleton, lying east and west. (fn. 9) The fragment preserved is of a thin-walled vessel of B type, to which also one found by the Thame at Drayton St. Leonard must have belonged.
Any tumulus that covered these burials—for such must one and all of these beakers denote—must have been destroyed by cultivation; none at any rate has been recorded. It might indeed have been an insignificant pile, since at Cassington the filling of the pits was the gravel that had been removed in the original excavation. But that normal Bronze Age round barrows exist there can be no possible question, though, as the evidence from Anglo-Saxon times must warn us (see infra, p. 365), not all round barrows in Oxfordshire are necessarily of the earlier date. This even applies to several on Chadlington Downs and on the confines of Wychwood. Two, one of them 8 ft. high, which were levelled some time before 1857 in a field called Little Disslings at Chalford, 1 mile north-west of Enstone, were found to contain an enormous mass of black and red ashes and charred earth, among which in one case were remains of charred metal. The account of this reads more like that of the Saxon barrow at Asthall, but one explored by the Hon. Harold Dillon at Spelsbury Down Farm, (fn. 10) while also revealing a layer of burnt earth 30 ft. in diameter, contained stones, some of them recorded as weighing 200 lb., presumably the remains of a large cist. More is known about barrows exposed in the disafforestation of Wychwood. Two explored by Sir Henry Dryden at Rowstedge (or Roustage) Copse, 3–4 ft. high and respectively 39 and 63 ft. in diameter, were composed of blocks of limestone, like a third at Round Hill, near Chalford, (fn. 11) and in the larger was a stone cist 3 ft. by 1 ft. in size, but no relics are recorded. In another between Rowstedge and South Lawn was a larger stone cist in which were found a perforated stone hone, sherds, and ashes, and below the pavement was a flint arrow-head. Another opened by government employees also contained a cist with ashes and burnt bones. Of two explored by Rolleston and others in 1872 at South Lawn, about 1 mile north of Swinbrook, an account is preserved in the Rolleston MSS., now in the Department of Human Anatomy at Oxford. One had been previously mutilated, when several small Roman coins ranging from Constantius to Valentinian II(?), charcoal, sherds, and bones came to light. It measured 21 yards in diameter and stood about 5 ft. high, and was again composed of stones. In it the later explorers found burnt human bones, many Constantinian minimi at about 4 ft. deep, and at the bottom a sherd decorated with incised lines. A pit at the centre yielded small sherds and flints, some of them flakes with serrated edges as if intended for saws. Some of the sherds appear to be sub-Neolithic in character.
Only the last gives any reliable indication of its relative date. A tradition that in 'The Squire's Clump', a large barrow at Sarsden, Lord Ducie found two skeletons, each in a little chamber in a sitting posture, accords better with what is known of early Bronze Age burials, but taken as a whole the evidence available in regard to such barrows in Oxfordshire is at present deplorably meagre.
Among casual finds of stone implements are some that may normally be assigned to this period. The possibility of persistent manufacture of stone axes has already been noted. Two specimens seem more particularly to suggest it. One, a pretty axe in chipped brown flint, found at Howbery Park, near Crowmarsh Gifford, resembles in outline some early axes of bronze, and the same is true of one found at Great Tew, one of the few stone axes recorded from north Oxfordshire. A remarkable double-ended implement of chipped flint, c. 7 in. long, slightly curved longitudinally and with incurved sides, may also have affinities with bronze types. It came from the Thames at Long Wittenham, (fn. 12) and like one from Yorkshire (fn. 13) may, as suggested by Sir John Evans, have served as an adze. In any case there is no question about the date of several perforated stone axe-hammers and mace-heads. Of the former there is one of greenstone with faceted butt from the Thames near Folly Bridge, Oxford (Pl. II r), while from Dorchester comes the half of a beautiful example in green-veined hornblende-gneiss of the rare bolster type, and a variant of this type from the Thames at Reading. (fn. 14)
Early copper or bronze implements are distinguished by their scarcity. All that are known, apart from the Cassington awl, are flat axes from the Thames at Long Wittenham and Wallingford, and from Beckley (Pl. V 1 a). Technical improvement in the provision of flanges and the initial stages of a stop-ridge is exemplified in two chisels, one from Culham (Fig. 6), the other from Dorchester. (fn. 15) A perforated horn haft found with human remains and pottery at Cockshoot Hill, Wychwood, is compared by Thurnam to one with a small bronze axe inserted in it in the Stourhead collection (fn. 16) in Devizes Museum, but the perforation at the middle rather suggests a tanged implement, if indeed it held a bronze implement at all.
Whatever be the true age of the leaf-shaped flint arrow-heads, it appears that the barbed and tanged variety (Pl. III e, two lower rows), does not antedate the Bronze Age. Not only these, but also large quantities of scrapers, cores, flakes, and other signs of flint-working are found on the limestone between the Cherwell and the Evenlode and again round North Stoke and Mongewell. The majority of the arrow-heads are of the so-called 'beaker' type, (fn. 17) small, with curved edges and with barbs and tang of equal length. The more elongated form with longer tang (fn. 18) seems to be that normally associated with cremation, though it does occur with burial, e.g. at Summertown, as also the still more elaborate type, (fn. 19) exemplified by specimens found at Dorchester and at Hensington near Woodstock, with broad-ended tangs (Pl. III e, middle of third row).
Dr. Grahame Clark has shown that the discoidal flint knife with ground edge belongs in the main to the beaker period. (fn. 20) Four examples are recorded from the county, the fine piece from the Thames at Wallingford, (fn. 21) a neat black specimen from Benson Lock, another from Beggar's Bush Hill, Benson, and a fourth from Tackley.
Middle Bronze Age
If one fact emerges from the archaeological evidence reviewed above, supported by an equal wealth of material from the Berkshire bank between the river and the Downs, it is that human occupation of the upper Thames basin was steadily becoming more intensive, especially on the lightly covered gravel terraces bordering the river. (fn. 22) Unfortunately it is difficult as yet to distinguish accurately the occupation-sites of the early part of the Middle Bronze Age from about 1500 b.c. The process of change from inhumation to cremation, which becomes the sole funerary rite in the late Bronze Age, still remains obscure. That the latter began earlier than is generally believed seems incontestable. Plenty of evidence exists of its use in the early Bronze Age, and the Neolithic ancestry of the collared urn and its gradual evolution towards the bucket-shape of the late Bronze Age seems to postulate a long period of time. Leaving this on one side for the moment we may note here the occurrence of another ceramic form, the food-vessel, whose descent from the Neolithic bowl has been demonstrated by Mr. Reginald Smith. (fn. 23) Exceedingly common in northern districts, like Derbyshire and Yorkshire, it is rare in southern England, and from Oxfordshire only four examples can be cited, one from Park Town, Oxford, a poor version of a well-known northern type, found with two skeletons; one from Standlake, still more decadent and suggesting influences from the collared urn, while a third from Yarnton has lost even more of the original features (Pl. VII b). The fourth, an unusual piece, has alternating broad ribs and furrows, the former ornamented either with diagonal square-toothed comb-lines or with incised chevrons (Pl. VII a), a decorative feature curiously recalling one from Ireland. (fn. 24) It was found in 1824 at a place called Brismere, near Oddington, on the northern edge of Ot Moor, and is one of the rare prehistoric objects found in north-east Oxfordshire. It may be surmised that some of the taller round barrows may contain cremation-burials belonging to this period, since a partially explored barrow adjoining the Saxon cemetery at Caldecote, Abingdon, contained both crouched burials of brachycephalic skeletons and also cremation, while in the floor of the barrow was found an atypical food vessel. (fn. 25) The total absence of true brachycephaly among the skulls of late prehistoric and protohistoric times in the Thames region indicates its gradual extinction, an additional reason for attributing burials such as these to an early post-beaker period.
It is possible that such barrows mark the turning-point at which cremation with or without urns becomes general. Apparently the earliest type of vessel associated with cremation-burials is the collared urn, like the fine example from Cowling's Piece, Stanton Harcourt, (fn. 26) the upper part of which is decorated with 'maggot' ornament (Pl. VII c). The process of evolution is illustrated by one from Stadhampton, in which the collar shows signs of losing its definition and the ornament is incised, or by an imperfect specimen from Chadlington on which incised linear ornament is employed.
By the close of this stage the population of the Thames valley must have increased considerably. Air-photography, and pre-eminently the thorough aerial survey of the district by Major G. W. G. Allen, has revealed that in some localities there were large settlements represented by groups of circular ditches measuring as much as 90–120 ft. in diameter, and even larger still. Alike at North Stoke, at Eynsham (Pl. IX a), at Dorchester, and above all at Stanton Harcourt and Standlake, extensive collections of such ring-ditches have been observed. At Stanton Harcourt so numerous are they that scarcely a single modern field in an area nearly 2 miles long between Linch Hill and Beard Mill, the area within which stand 'the Devil's Quoits', (fn. 27) is without one or more of these ditches. At certain points at the eastern and western ends of this tract there are compact groups like that at Eynsham. Many, though not all, as will be noted later, of these ring-ditches have proved to be sepulchral enclosures, and as such may be compared with a close counterpart on a smaller scale in the 'kringgreppels' of Drenthe and Groningen in north-east Holland. (fn. 28) Two have already been investigated, one at North Stoke and another at Fullamoor Farm, Clifton Hampden. (fn. 29) The former, a double-ringed example, yielded a cremation and burials of three infants at the centre, the latter a parcel of cremated bones only, in both cases without associated relics.
Evidence of their date is in part provided by a large cremation-urn found in an inverted position behind Freelands House, Donnington, on the southern outskirts of Oxford (Pl. VII d). Sixteen and a half inches in height and 16¼ in. in diameter, it tends towards a biconical shape, has a moulding at the rim and upon the carination, between which are two curved false handle-mouldings. It belongs to a class exemplified by one from Bulford, Wilts., (fn. 30) which contained a cremation with amber beads and a fluted leaf-shaped razor, and is therefore probably contemporary with a smaller, plain, biconical vessel with two perforated lugs with which was associated a plain razor of the same type in a central cremation-pit of a double-ringed disk barrow at Radley, Berks. (fn. 31) Even more conclusive is the recent discovery in another disk barrow at Radley of a segmented faience bead, which confirms a date at the close of the second millennium, b.c.
This Middle Bronze Age is attested by a wealth of isolated finds of bronze implements. The difficulty of assigning individual specimens to this epoch has indeed been stressed by Mr. Kendrick; (fn. 32) nevertheless, it will be simpler to include here those which, though they may actually be products of or hoarded in the Late Bronze Age, typologically certainly originated in the Middle. This holds good of the palstave with expanding blade, examples of which without loops have been found at Rycote, near Thame (Pl VI 1 b), at Coggs, near Witney, at Freeland, near Eynsham, and at Dorchester, Stanton Harcourt, and Tackley. Its later persistence is illustrated by two interesting hoards both found on the outskirts of Oxford. (fn. 33) The first, from Leopold Street on the borders of the ancient Cowley Marsh, contained 7½ examples all cast in the same mould; with them were two palstaves from other moulds, and also a clumsily shaped socketed axe. These would appear to have been the stockin-trade of a merchant-founder, the same man whose everyday equipment may be represented in the second hoard (Pl. VIII 1) found in Burgess's Meadow, an enclosure lying on the eastern edge of Port Meadow on a line that leads across the meadow to an ancient ford of the Thames at Binsey. In it were broken implements, two looped and socketed lance-heads, a bar of bronze, a crude knife, a tanged chisel, a socketed hammer, and, lastly, a palstave from the same mould as those found at Leopold Street. That the implements were actually cast in Oxford cannot be proved, but others cast in the same mould were traded far afield. One such, now in the British Museum, has the Isle of Thanet as its provenience, while another in private hands found at Vernham's Dean, Hants, was recognized by Mr. Crawford.
Two varieties of spear-heads fall to be catalogued here. The first, usually of small size, is that of the Oxford hoard. Such are known from Aston, Standlake, the Thames at Reading and Long Wittenham, and a still smaller specimen from Minster Ditch (Pl. VI 3 e), on the western boundary of the county opposite North Hinksey, Berks. Spear-heads of larger size, with the loops incorporated in the base of the blade, come from Wendlebury, near Bicester, and others dredged from the Thames at Reading, Wallingford, Little Wittenham Lock, and Sandford Lock (Pl. VI 3 c), from which also were recovered two so-called rapiers, one 14½ in. long with rounded butt (Pl. VIII 2 a), the other 15½ in. with square-ended butt and narrower blade, like a third from the river at Reading. Finally, from the Thames come short swords, one 18½ in. long, with two rivets, found at Sandford, another with four rivets at Reading, and a third with six at Benson, 22 in. long (Pl. VIII 2 c). (fn. 34) With these goes the short sword from the Cherwell, with its grip and rounded pommel cast in one piece with the blade (Pl. VIII 2 b).
Late Bronze Age
The period of roughly five hundred years preceding the middle of the first millennium b.c. witnessed the beginning of great changes in Britain, particularly in its southern half that corresponds to Sir Cyril Fox's Lowland Zone. But the changes did not come at once, though the influences under which they eventually crystallized soon began to make themselves felt, and their effects manifest themselves beforehand in the archaeological material. This comes out, for example, in the adoption of new types of implements and pottery, some of them foreign, and above all in the evidence of intensified trade with the mainland. Along with all these, however, there can be seen a retention of characteristics that are essentially native. The palstave still persists, but a new, purely British type makes its appearance in a heavy, almost parallel-sided form furnished with a loop and decorated with vertical linear mouldings below the stop-ridge, a translated survival of a V-moulding on earlier types. This implement is known in the county by good examples from Dorchester, from Pot's Stream, a backwater of the Thames opposite North Hinksey, Berks. (Pl. VI 1 c and d), and by a third found along with a lump of bronze at Wardington, 4 miles north-east of Banbury. (fn. 35) Their date is to be gauged by the association of the type in an early hoard of the period at Nettleham, Lincs. (fn. 36)
Sir John Evans introduces his examination of the looped variant of the palstave with one from Dorchester, (fn. 37) that had evidently been subjected to long use and ground down from a larger implement, like another from North Stoke. Under Garsington is recorded a double-looped palstave (Fig. 7) such as is found rarely in these islands, and then mostly in the west, suggesting foreign intercourse, perhaps with the Spanish peninsula where twin loops are common. If it was actually found in association with the socketed axe (Fig. 8), which strongly recalls a French type, the suggestion receives reinforcement.
The first appearance of the socketed axe is more difficult to prove, but apparently it falls about the same time. Simple types with a squarish socket and little or no ornamentation are not particularly common. One included in the Leopold Street hoard has already been noticed. Others come from Beckley, Dorchester, Holton(?), and the Thames at Reading, and from a backwater of the Thames between Oxford and Iffley (Pl. VI 2a). Of examples with vertical mouldings, perhaps imitating those of the latest palstaves, four have been recovered from the Thames, one at Reading, two at Wallingford (Pl. VI 2c), and one at Sutton Courtenay; others come from Garsington along with a two-looped palstave and some bronze rings.
A large example of the Middle Bronze Age type of spear-head, but without loops and secured to the shaft by a rivet only, has also been found in the Thames at Reading, a second with remains of its ashen shaft pierced by a wooden rivet at Wallingford (Pl. VI 3d), and smaller pieces have been found at Sandhill, Headington, and at Culham, the last (Fig. 9) with a decorated socket. (fn. 38) Another large example, dredged up opposite Moulsford, with incipient barbs at the base of the blade (Fig. 10), seems to herald the more familiar 'Speen' type. (fn. 39) A gruesome relic of the period is part of a human pelvis transfixed by the broken head of a bronze spear from a burial at Queenford Mill, Dorchester.
The earlier forms of tanged sword, the so-called U-type with leaf-shaped blade, is represented by one from St. John's Priory, Banbury (Pl. VIII 2d); its successor, the V-type, by one from the Thames at Reading, and a still later development approximating to a continental Hallstatt form from the river at Henley. Other specifically British objects may here be noted: a socketed sickle from the Thames at Reading, and a sword-chape of elongated V-form, 12 in. long, from Day's Lock, Dorchester, which also yielded a repoussé bronze buckler (more correctly perhaps the central plate affixed to a leathern shield), 13½ in. in diameter. It has a large central boss surrounded by two rings of smaller bosses. A second buckler, only 9½ in. in diameter, with a slightly conical boss surrounded by a single ring of largish bosses, was found in the Thames at Swinford Bridge, Eynsham (Pl. VIII 3).
Towards the close of the period foreign imports and novelties come increasingly to the fore. The large influx of late Bronze Age foreigners from mid-eastern France into the eastern counties from Kent to Cambridgeshire, their presence well attested by numerous hoards and isolated finds of bronze implements, receives only a mild echo in the upper Thames area. The most characteristic continental type, the winged axe, is not known, but its successor, the socketed axe with vestigial wings, occurs at Dorchester (Pl. VI 2b), and also certain small forms, either with oval socket as at North Aston, Mapledurham Lock, and Minster Ditch opposite North Hinksey (Pl. VI 2d), or with one octagonal in section. One of the latter type was associated with tanged chisel, socketed gouge, a double-bladed (maple-leaf) razor (Fig. 11), (fn. 40) and a socketed knife (cp. a second specimen from the same part of the river (Pl. VI 3a), in the Thames at Wallingford, forming a small hoard containing types all characteristic of the immigration. These more or less casual occurrences probably came up river largely by way of trade, that brought also from the south coast two miniature socketed axes, 3½ in. long, a form of currency known by many hundreds from north-western Gaul, one to Watlington, the other lost in the River Cherwell at Magdalen Bridge, Oxford (Pl. VI 2e).
It is from the south that the upper Thames received the full force of immigration, not as later in the Iron Age one of conquest, but rather a gradual and peaceful accession of settlers. This influx is signalized by the appearance of new ceramic shapes, large vessels of barrel-form with an applied finger-imprinted moulding, degenerating later into others of flower-pot shape with or without finger-imprints at one-third of their height. These vessels and smaller counterparts of them served as receptacles for cremation-burials and foodofferings in the large urn-fields of continental type that spread from Dorset and Hampshire to the upper Thames. At Standlake in 1857 Stephen Stone discovered one of the classic examples of these urn-fields. While investigating several of the group of ring-ditches on Standlake Down, he encountered one in which he found, not only in the ditch itself, but also in the ground inclosed by it, a large number of cremation-burials, some in pockets in the gravel, others contained in urns of flower-pot shape, and at one point in the ditch an extensive patch of carbonized wood and ashes, which he considered to have been the ustrinum or site of the pyre where the funerary rites were performed. To judge, however, by other examples of these urn-fields in Wiltshire and by one at Long Wittenham, Berks., it would seem that the connexion of the Standlake urn-field with the ring-ditch is abnormal. The burials were apparently confined to the southern half of the ring, and no other ring-ditch examined at Standlake yielded similar results, nor in recent years has any other of the many such ditches opened in Oxfordshire and Berkshire produced similar pottery. It is probably a coincidence that the Standlake ring-ditch was used as such a cemetery.
This review of Bronze Age Oxfordshire may be closed with a record of two more discoveries. One, a piece of ring-money, probably like a palstave found at Wallingford, (fn. 41) Irish in origin, was picked up on a farm at Combe near Woodstock by Mr. John Joslin in 1911. It is not, like most such rings, of pure gold, but like some others is made of a copper rod plated with gold striped with bands of paler alloy. The other is a huge, globular bronze cauldron, (fn. 42) 18 in. high and 24 in. in diameter (Pl. X a). This remarkable vessel, found in the bed of the Cherwell near Shipton-on-Cherwell in 1929, is built up of sheets of bronze riveted together, its bottom renewed and its sides patched in several places, marking its value and long use. The upper part of the body, neck, and rim are hammered out of two sheets of metal; the neck is corrugated to lend it additional strength. The ring-handles are solid castings and have been secured to the vessel by an ingeniously contrived pair of moulds that allowed each staple to be cast over the rim and on to both sides of the neck in such a way as to grip the corrugation of the neck immovably. The cauldron represents the earliest stage in the evolution of a group of such vessels, imitated from some foreign form like the Greek dinos (cauldron), but themselves peculiar to the British Isles, no parallels being known from Gaul or central Europe. They are commoner in Ireland and Scotland than in England, although the Shipton example stands high among its class in point of date. A fairly close estimate of this is provided by the use of a similar technique for securing the handle on another class of bronze buckets similarly distributed in the British Isles. These —some Irish specimens exactly—resemble in form the bronze situlae found in graves in north-eastern Italy, Carniola, and Austria between the 7th and 5th centuries b.c.