File:Iron Age bracelets positioned roughly as found (FindID 741751).jpg

English: Treasure Report for H.M Coroner

2015 T656 - Vessel and bracelet hoard, from near Wakefield, West Yorkshire

To be known as the 'Brookfield Hoard' (named after the finders, at their request)

PAS no.: SWYOR-1494DB

Circumstances of Discovery

On 13th September 2015, a group of Iron Age objects were found by [name redacted] while metal detecting at a 'Coil to the Soil' rally in North Elmsall, near Wakefield in West Yorkshire (grid reference provided). Amy Downes, Finds Liaison Officer for the Portable Antiquities Scheme in West Yorkshire, was able to attend the site within two hours of the discovery. The finds included a copper alloy strainer vessel, a pottery vessel, and five copper alloy bracelets - a single wire piece and two pairs in more robust designs. The finder reported that the strainer was found inverted on top of the upright pottery vessel, with the five bracelets underneath the pot. The four larger bracelets were interlinked at the bottom of the deposit, with the more delicate bracelet on top of them. These finds were lifted on the day of discovery. The pottery vessel collapsed during lifting.

Other finds, including Roman sestertii, were made in the same area by detectorists taking part in the rally, but these did not seem to be directly associated with the strainer hoard (see 2015 T658).

Archaeological Investigation

The day after the discovery, 14th September 2015, a team from West Yorkshire Archaeology Advisory Service (WYAAS) and Archaeological Services WYAS (ASWYAS) attended to carry out a small-scale excavation (ASWYAS Report No. 2804). A test pit measuring 2.15m x 1.50m was excavated by hand by archaeologists Ian McIntyre (ASWYAS) and David Hunter (WYAAS). This test pit was an expansion of several smaller pits dug on the previous day by members of the detecting club. A metal detector was used to comprehensively scan the test pit, spoil and surrounding area.

The test pit revealed a feature, thought to be a linear terminus or a pit. The upper fill contained large stones, possibly suggesting a packing material. It contained one sherd of Romano-British pottery along with animal bone, fired clay and iron nails. The lower fill contained Romano-British pottery sherds and one small piece of burnt bone. The feature was not excavated to its base. In total, the additional finds recovered as part of the archaeological excavation comprised 23 sherds of pottery, eleven bone fragments, four pieces of fired clay, an iron nail, and a possible piece of iron slag.

Due to the complexity of the underlying geology, and the digging which had occurred the previous day, it was not possible to ascertain with certainty whether the strainer hoard was from the same feature as the archaeological finds. However, given the rarity of Roman pottery in West Yorkshire, it is likely that the two could be linked (Hunter and Sanderson, pers. comm.). However, because they cannot be conclusively demonstrated to be part of the same find, the archaeological finds are not included in this Treasure report.

Description

1.Copper alloy strainer

A round-bottomed hemispherical bowl made of thin, copper alloy sheet. The wide, projecting rim has a separate binding-strip attached around the edge.

The bottom of the strainer bowl is decorated with small perforated holes, just over a millimetre in diameter, forming a decorative pattern centred on the base of the strainer. There are two concentric outer rings of holes and, in the centre, a triskele radiates outwards from a small central circle. The design covers around a third of the depth of the bowl, although this is much of the visible surface area when looking directly down into the strainer. The remaining surface area of the upper bowl is smooth and undecorated.

The strainer bowl is rimmed with a wide, flat flange. The bowl and flange were made from one sheet of copper alloy, either beaten into shape or turned on a lathe. Around the circumference of the flange a separate copper alloy binding strip has been added.

The strainer is mostly complete but has suffered plough damage to around a third of its side prior to discovery, in this area the rim has been torn away and the flange is heavily damaged. There is also some damage to the decoration on the base.

Parts of the outer side of the base have a rich, smooth, dark green patina. The rest of the surface is more corroded, both inside and outside the strainer, with areas of lighter green concretions.

There are also three detached fragments: two binding fragments and one section of flange. One binding fragment has a clean edge suggesting that this was the point on the circumference where the two ends of the binding strip met.

Measurements:

Maximum diameter of bowl (including flange): approx. 200 mm; width of binding: 5mm; depth of binding: 4 mm; thickness of binding in cross-section: 0.5-1.0 mm; width of flange (including binding): 25 mm; diameter of bowl (excluding flange): 151 mm; depth of strainer: 64 mm.

Diameter of outer circle of perforated holes: 95 mm; diameter of inner circle: 89 mm; diameter of central circle inside triskele: 20 mm.

Weight: 291g

Smallest binding fragment: Length: 24.5mm, Weight: 0.89g

Longest binding fragment: Length: 133mm, Weight: 12.72g

Flange fragment: Maximum remaining length: 48mm, Maximum remaining width: 29mm, Thickness in cross-section: 0.2-0.4mm, Weight: 3.63g

Parallels:

Hemispherical strainer vessels are rare finds. The closest parallel to the Brookfield hoard strainer is one from the Langstone hoard (NMGW-9C0216; Gwilt et al. Unpublished Treasure Report). Found near Newport, South Wales, in 2007, this hoard consisted of a strainer vessel and two metal bowls. A tankard was also found nearby. The Langstone strainer is similar in many respects to the Brookfield find: both are decorated with a triskele design, and have wide flanges. However, the Langstone strainer is slightly smaller, and has handles attached to both sides, whereas Brookfield has no handles. Langstone also has a simple up-turned rim, rather than a separate binding strip. The rendering of the triskele also differs, with the Langstone triskele depicted through the 'absence of perforations' in the base of the bowl, contrasting with the perforated design of Brookfield (Gwilt et al. Unpublished Treasure Report).

In his write-up of the Langstone hoard, Gwilt (Gwilt et al. Unpublished Treasure Report) lists several other comparable finds based on the corpus of broad-flanged strainers without handles published by Tomalin (1989). Three are from vessel hoards in Wales (finds from Manorbier and Cogyan Cave), one from a hoard at Helmsdale in Sutherland, north Scotland, and two single finds have been recovered: one from Marston Moor near York, and one from Thorpe, in Suffolk. The strainers from vessel hoards are associated with Romano-British style objects, such as trulleii and shallow bowls, which Tomalin (1989) used to suggest a date in the later first or early second century AD.

However, the Langstone strainer was found in conjunction with two Iron Age style bowls, of earlier date. As the strainer fits both bowls, it is believed all three functioned as a set. Gwilt uses this association to argue for pushing the dating of these strainers back to earlier in the first century AD. For the Langstone find, he suggests a date of AD 40-60 for manufacture, and burial in the period AD 40-75, around the time of the Roman campaigns in Wales (AD 47-78) (Gwilt et al. Unpublished Treasure Report).

Gwilt's dating is also borne out by his observation of the designs on the southern strainers, which are typical of conquest-period Celtic art. The Welsh strainers often combine indigenous decorative traditions with new Romano-British style designs, incorporating various combinations of triskeles, running-scrolls and geometric border decoration of infilled triangles. The Thorpe strainer, a single find from Suffolk, does not feature a triskele, but nevertheless does have a 'triple' element to its design. Most of the bowl was destroyed during gravel extraction, but enough survives that the design can be reconstructed to view the perforated outline of three circles at the top of the bowl. This potentially echoes the triskele design on the Langstone (Gwilt et al. Unpublished Treasure Report) and, now, the Brookfield strainer.

The Brookfield strainer, with its triskele design, seems to fit best with this group of early strainers identified by Gwilt, suggesting a date of manufacture around the mid-first century AD. The two other northern strainers, from Helmsdale, Sutherland (Curle 1931) and Marston Moor Yorkshire (Tomalin 1989) are probably of later date. Their bowls are decorated with Romano-British style flower patterns, and their flanges are either punched (Marston Moor) or impressed (Helmsdale) with motifs. These two strainers likely date to around AD 80 -120 (Gwilt et al. Unpublished Treasure Report).

2. Copper alloy ribbed bracelet

Penannular copper alloy bracelet, D-shaped in cross-section with flatter side on interior. Terminals are flat sections through the bracelet with no additional decoration. Terminals more oval than the D-shaped cross-section of the rest of the bracelet. This may suggest that continued wear changed the overall cross-section at the back. Smooth on interior surface with a central line, suggesting possible rolled construction. Outer surface is decorated with ribbed decoration. Raised transverse ribs (approx. 3mm wide) alternate with lower ridges of beaded decoration (approx. 1mm wide). There are 38 bands of decoration. Green patination, with some evidence of wear and polishing on both the interior and exterior near the middle of the bracelet, predominantly on one side.

Measurements: Max diameter 85 mm; internal diameter 66 mm; Terminals: 12 mm by 9.5 mm and 14 mm by 9 mm; width in cross-section 8.5-10 mm; weight: 158.3g

3. Copper alloy ribbed bracelet

Penannular copper alloy bracelet, oval in cross-section with slightly flatter side on interior. Terminals are flat sections through the bracelet with no additional decoration. Outer surface is decorated with ribbed decoration. Raised transverse ribs (approx. 3mm wide) alternate with lower ridges of beaded decoration (approx. 1mm wide). There are 39 bands of decoration. Green patination, with little evidence of wear, although there is some polishing on external surface near one of the terminals.

Max diameter 85 mm; internal diameter 65 mm; Terminals: 12 mm by 9 mm and 13 mm by 9mm; width in cross-section 8.5-10 mm; weight: 154.8g

4. Copper alloy heavily worn, ribbed bracelet

Penannular copper alloy bracelet, lentoid in cross-section at the terminals, swelling to a more D-shaped cross-section at the back. Terminals are flat sections through the bracelet with no additional decoration. Outer surface is decorated with almost erased, heavily worn ribbed decoration. Raised transverse ribs (approx. 3-4mm wide) alternate with lower ridges of (?)beaded decoration (approx. 1mm wide). Green patination.

Max diameter 73.5 mm; internal diameter 65.5mm; Terminals: 9.5 mm by 4 mm and 10 mm by 4 mm; width in cross-section 3.5-5.5 mm; weight: 51.1g

5. Copper alloy heavily worn, ribbed bracelet

Penannular copper alloy bracelet, lentoid in cross-section at the terminals, swelling to a more D-shaped cross-section at the back. Terminals are flat sections through the bracelet with no additional decoration. Outer surface is decorated with almost erased, heavily worn ribbed decoration. Raised transverse ribs (approx. 3-4mm wide) alternate with lower ridges of beaded decoration (approx. 1mm wide). Green patination.

Max diameter 71 mm; internal diameter 63 mm; Terminals: 9.5 mm by 4.5 mm and 10 mm by 4 mm; width in cross-section 3.5-5.5 mm; weight: 50.8g

Parallels

There are no direct parallels for the four Brookfield hoard bracelets. However there are close similarities to other finds from Britain, several of which are from graves. The design of raised transverse ribs and alternating beaded decoration is most similar to two bracelets found at the Kings Barrow, North Staffordshire (Lock et al., 2014, 86). There are also close parallels to bracelets from South Ferriby in Lincolnshire (May, J. 1976. fig 66, 5). Another bracelet similar to the South Ferriby example was found at Mount Batten, near Plymouth. These bracelets all have a circular shape and a hinging mechanism, either mortice and tenon or catch, which is absent from the Brookfield bracelets.

These examples were all viewed as having evolved from Halstatt and early La Tène continental bracelet types of the fourth to third century BC (May 1976, 125; Lock et al, 2014, 85). Cunliffe (1988, 61-2) suggested that the Mount Batten bracelet might be an import, owing to its high lead content.

There are certainly close continental parallels for these bracelet types, including examples in the British Museum's Morel Collection from North Eastern France (Stead and Rigby 1999, pl 112). These ribbed bronze bracelets also have undefined terminals but none have the beaded separators found on the Brookfield bracelets.

However, there is nothing to suggest that the Brookfield bracelets are themselves imports. They could equally be of local manufacture. Indeed, the closest parallels to both form and design of Brookfield hoard bracelets are those found in Arras burials of Yorkshire dating to the late third to first centuries BC. Arras W43 is circular and without a hinging mechanism. The transverse ribs are separated by undecorated cuts instead of the beaded ridges seen at Brookfield. The inner surface of the bracelet is smooth and the terminals are slightly off centre when they meet (Arras W.43, Stead 1979, p74 fig 28.2). (Stead 1965, p52 fig 29. 3) Arras 5b (Stead 1965, p51 fig 28.1) also has undecorated dividers for the transverse ribs but the transverse ribs are more ovoid in shape.

6. Slim copper alloy wire bracelet with wound fastening

Circular copper alloy bracelet with square cross-section, broken into three pieces. At one point on the circumference, the two ends meet in a form of clasp or fastening, now broken. Possibly this has been constructed by wrapping the tapering ends of the bracelet around each other. The shortest piece comprises a short arc of the main section of the bracelet. The longer two pieces are separated at the clasp or fastening. On the longer piece all that remains of the fastening is the tapering end of the square-section wire, protruding approximately 6 mm from a tightly wrapped coil of rectangular-section wire. There are four complete turns to the coil. The broken tapering end originally joined a length of wire protruding from the other side of the fastening. This appears to form a flattened loop before coiling tightly around the opposite end of the square-sectioned wire bracelet. There are also four complete surviving turns. On this side, the tapering length of the square-section wire which protrudes from the coil curls back on itself, forming a U-shape. Whilst not all of the wire has been preserved, the remaining parts suggest that this was a permanently secured fastening, decorated in the middle with a series of loops. Only parts of two loops have survived but it is possible there were more.

Some areas of the surface have a smooth dark green patina but others have lighter green corrosion product.

Measurements: Maximum diameter 95 mm; width in cross-section of bracelet 2.6-2.9 mm; diameter in cross-section of the coiled fastening 6.2 mm;

Shortest piece: length 50 mm; weight 2.2g.

Shortest piece with a clasp: length 73 mm; weight 4.9g.

Longest piece, with clasp: length 94 mm, weight: 8.0g.

Total weight: 15.1g.

Parallels:

The closest parallel for this bracelet comes from Vindolanda (Birley and Greene p139, 140 cat. SF9025). On this complete object, the joining section (missing on the Brookfield example) is an elaborate series of loops, which had been broken at one end, and repaired in antiquity. The Vindolanda bracelet does not have the square section of the Brookfield example but is otherwise very similar in form. The bracelet is identified as a 'very elaborate style of a Type 8 bracelet, South Shields, Tyne and Wear' (Birley and Greene p139, 140 cat SF9025, Type 8 to be found in Allason-Jones and Miket 1984: 126). The bracelet is dated to period IV (c. AD105 to c. AD122) at the fort.

Another potential parallel was found at Mount Batten, Plymouth, the body was made of twisted wire but it had a similar catch mechanism. This example was dated only as 'Roman?' (Cunliffe 1988, p62 fig 33.64).

7. Pottery vessel

Heavily fragmented globular vessel, narrowing towards the neck, and with wide, everted rim. Black sandy fabric with some buff/greyish areas, some white inclusions.

Rim can be largely reconstructed as can large parts of the body. The base appears to be heavily fragmented or missing.

The size of the strainer relative to the pot means that the strainer could have rested flat on the top. However this does not appear to have been the case. On one body sherd there is an area of greenish corrosion that corresponds to one of the areas of damage on the strainer, where the top of the bowl meets the inner edge of the flange. This suggests that the strainer had been lying inverted at an angle across the top of the pot, touching the side of the pot rather than the rim. The location of this area of damage on the strainer, directly opposite the plough damage, suggests that the strainer could have been knocked into this position by the plough.

A soil sample from inside the pot has been preserved. There are no obvious inclusions of animal bone or cremated bone but this would benefit from further study.

Measurements: Rim diameter: approx. 170 mm; widest part of pot: approx. 180 mm

Weight: approximately 910g.

Weight of soil sample from inside pot: approx. 600g.

Discussion

The Brookfield hoard most likely dates to the first century AD, based upon the objects within the hoard and the dating of similar hoards. However it may have been buried as late as the early second century.

Its location in north-central England is unusual, as only one other object hoard has been found in the area from the period 800 BC - AD 100. The only other object hoard known from this part of Yorkshire, the Honley hoard, was discovered c.12 miles away. The Honley hoard was buried at a similar time, around the end of the first century AD, but the hoard is very different in nature. It contains coins, a brooch, and strap fittings amongst other items. However, the region sees several coin hoards buried from the Iron Age onwards.

Generally the strainers are assumed to have been used for preparing and serving wine, or perhaps other alcoholic beverages. Strainers may also have had a medicinal purpose; an example included with the 'Doctor's' grave in Colchester had residue of Artemisia, commonly used for healing/remedies (Crummy 2007). The inclusion of a complete pottery vessel within an object hoard is unusual and may suggest that the strainer and pot were used together.

The five bracelets demonstrate a range of different influences; there are two pairs of Iron Age bracelets, and one later thinner-style piece most likely dated to the Roman period. One pair of Iron Age bracelets been heavily worn perhaps suggesting an 'heirloom' pair, and a less worn, potentially newer pair. Despite their differing patterns of wear, the bracelets are almost identical and closely resemble styles found in Iron Age graves in Yorkshire and Lincolnshire, potentially dating to the third or second century BC. At least 200 years earlier than the latest objects in the hoard. This contrast is heightened by the fifth Roman-style bracelet whose closest parallel was found at the Roman fort of Vindolanda. This combination of types is mirrored by the form and decoration of the strainer. The strainer form is Roman but the perforated pattern suggests Iron Age influences.

The composition of the Brookfield hoard is unusual; the mixing of personal ornaments and vessels is more commonly seen in grave goods, rather than hoards. However, the excavation of the surrounding area found no evidence of human bone, cremated or otherwise. Whilst the bracelets fit well with general Yorkshire/Lincolnshire distribution of the ribbed bracelet types, it is unusual to find them in a hoard context. The strainer itself is an unusual object, but fits with a pattern of similar finds from around the edges of the core Romano-British heartlands of the south-east. Overall, the Brookfield hoard can be seen as part of a wider conquest-period phenomenon which sees the mixing of local and Roman styles and ideas about dress and food preparation. The unusual aspect, in this case, is the hoard composition and the great time depth represented.

Conclusion

These finds represent a prehistoric base-metal assemblage, since at least 4 of the objects (the ribbed bracelets) can be securely assigned an Iron Age date, and the assemblage was probably buried around or before the time of the Roman conquest of the area. The group therefore qualifies as Treasure under the stipulations of the Treasure (Designation) Order 2002.

Potential Acquisition:

Wakefield Museum Services hope to acquire this find. They would also be the museum service to take receipt of the archive and finds from the archaeological investigation. For the purposes of future study, it would be useful for this material to be held by the same institution.

Julia Farley,

Curator of Iron Age Collections,

British Museum

Rachel Wilkinson

CDA Student,

British Museum and University of Leicester

16th January 2018