Table of Contents
Introduction
Wrist-clasps were used to fasten the cuffs of women’s clothing in the early Anglo-Saxon period. They are found almost exclusively in the Anglian culture-province (East Anglia, Cambridgeshire, the East Midlands, North and East Yorkshire).
Occasionally the top of the slit in the cuff was covered with a triangular metal fitting called a gusset plate.
Hines 1993 is the definitive typology, with new finds almost always fitting in, and it’s easy to use. The only occasional problem that might crop up is allocating a clasp to form B18 or B20; these can be very similar, and the boundary between the two perhaps a little subjective.
Over half the PAS’s records of early Anglo-Saxon wrist-clasps come from the two counties of Norfolk and Suffolk. FLOs here will have Hines 1993, but for anyone else with an outlying example, a quick summary of the Hines forms is given below, in separate classes.
PAS object type to be used
Wrist-clasps are currently termed SLEEVE CLASP in the mda thesaurus. Although this terminology seems unlikely to catch on in the wider world, we have to use it on the PAS database too. Also use SLEEVE CLASP for gusset plates. Do feel free to use the term ‘wrist clasp’ in the description.
PAS object classification to be used
If possible, the Hines form should be given in the Classification field in the following format: Hines form B12 or Hines class C. If you are recording a gusset plate, use gusset plate in the Classification field.
Hines has three levels of classification. The top level is classes (A, B and C) and below this there are forms (e.g. B12, B18a, C2). Some forms are further subdivided into types (e.g. Barrington type, Mildenhall type). This means that we are at risk of using different words (class, form and type) for the same concept (different designs of clasp). To avoid this, make sure that you use class A, B or C where you cannot sub-divide further, and form B1, B2, etc where you can.
Terms to use in the description
The Hines form is necessary but not sufficient – the wrist-clasp must be fully described as well, using the terminology defined by Hines. The two halves of a single clasp should be known, following Hines, as the hook-piece and the catch-piece; don’t use other terms such as male and female, or hook and eye.
The only place where we part company with Hines is in the use of the word ‘spiggot’ for a T-shaped lug. Don’t use ‘spigot’ either; T-shaped lug is easier to understand, and to spell.
Date
The date range of all wrist-clasps and gusset plates is given by Hines (1993) as late 5th or 6th century.
More recently, Penn and Brugmann included some types of wrist-clasp in their chronological study of selected East Anglian grave-good types (Penn and Brugmann 2007, 28-9). They included clasps of class A, form B7 and form B12, as well as some clasps with bars and plates. It is not made clear which Hines forms were included in this last category, but only rectangular shapes (i.e. B13 and B19) are illustrated.
Class A and form B12 clasps are allocated to to Penn and Brugmann’s Phase FA1 (2007, fig. 5.21). Phase FA1 starts “soon after AD 450” and “the transition from Phase FA1 to FA2 has been dated to the late 5th century”; the next phase, FA2, “probably began as early as c. AD 480”, so this is probably to be taken as the date of the end of Phase FA1 (Penn and Brugmann 2007, 71).
Hines’s analysis of grave-groups including B12 clasps (1993, 49) shows that the latest examples are found with cruciform brooches that would now be allocated to Martin’s Group 3 (e.g. Morningthorpe 253) or Group 4 (e.g. Sleaford 116) (find a summary of the chronology of these brooches here). Similarly, Hines’s dating of class A clasps convincingly runs “up to the end of the Migration Period in Anglian England” (1993, 11). The restriction of class A and form B12 clasps to this early date therefore seems unwarranted, and may be a result of Penn and Brugmann’s small sample size.
Form B7 clasps decorated with bosses are allocated to Penn and Brugmann’s Phase FA2a, and form B7 clasps without bosses to their Phase FA2b (2007, fig. 5.22 and 5.24). Clasps with plate and bar (perhaps forms B13 and B19) are allocated to Phase FA2a-b. No suggested calendar dates are given for the FA2a-FA2b boundary, but Phase FA2 ends in “the mid 6th century at latest” (Penn and Brugmann 2007, 71).
From this it will be seen that Penn and Brugmann‘s general conclusion is that the dating of wrist-clasps should be pushed back a little, to the century c. 450-c. 550 AD.
Hines 1993 and Penn and Brugmann 2007 have not been completely reconciled, so it seems reasonable to give all wrist-clasps a wide date-range, of say c. 450-570 AD.
Class A clasps
Class A clasps are made from wire and are usually of silver. They have the centre of the wire formed into a hook or catch, and the ends rolled up into a spiral. There is no evidence for attachment, and they were probably sewn on to the garment.
The only certain example of a class A wrist-clasp on the PAS database is SF-BC7072. Another possible example is DENO-089A26.
Class A is not further sub-divided, so there are no separate forms, but there may be a gradual development to larger spirals over time.
Class B clasps
Class B is made up of clasps with a plate, or a bar, or both. Most are roughly rectangular, and the plate can be made from sheet or cast. Cast plates often curve slightly around the wrist.
There are twenty forms within Class B. Some of these are very rare; the commonest on the PAS database are forms B12 and B20, with B18 close, then B7 and B13.
All other Class B forms are found on the PAS database in low numbers, except for B5 and B6. There is only one example each of several forms: B2, B3, B4, B9 and B15.
Definitions of all Class B forms found on the PAS database
B1: plate riveted to the sleeve using button-headed rivets (no examples of a complete clasp-half have been recorded on the PAS database yet, just the rivets)
B2: plate riveted to the sleeve using rivets joined by a bar
B3: plate and buttons, not riveted to the sleeve but held by a removable pin
B4: plate and bar, held to the sleeve by a removable pin (rectangular or T-shaped bar)
B7: plate only, sewn to the sleeve (rectangular plate)
B8: plate only, sewn to the sleeve; decorative mock-rivets
B9: bar only, sewn to the sleeve (bar with two roundels flanking a central lozenge)
B10: bar only, held to the sleeve by a removable pin
B11: bar only, sewn to the sleeve around the bar (T-shaped bar)
B12: bar only, sewn to the sleeve through pierced or T-shaped lugs (rectangular bar, sometimes with central extensions making it look T-shaped)
B13: plate with separately applied bar (rectangular plate)
B14: plate with separately applied bar (plate with projecting pierced lugs on rear edge)
B15: plate only, sewn to the sleeve (plate with projecting pierced lugs on rear edge) – this differs from B16 in that the projections must all be perforated
B16: plate only, sewn to the sleeve (plate with shaped rear edge) – this differs from B15 in that not all the projections are perforated
B17: plate with separately applied bar (plate with shaped rear edge)
B18: bar only, conjoined roundels along rear edge
B19: plate and bar cast in one piece (rectangular plate)
B20: plate and bar cast in one piece (plate with shaped rear edge or projecting pierced lugs)
Form B18 is defined as having a bar with a row of conjoined ‘knobs’ (normally roundels) cast in one along the rear edge. The different types are defined below, but first it is necessary to define the difference between form B18 and form B20.
Some examples of Hines form B20 (e.g. FAKL-4F5333, SWYOR-3B3366 and NMS-546DD8) look very like examples of form B18. Hines explains the difference: the B20s “differ in principle in not having perforations between all the conjoined knobs or lugs behind the bar” (Hines 1993, 64). In other words, all the roundels or lobes behind the bar on a form B18 have to be distinct, not merged into a plate as on the B20s.
B18a has triangles on or in front of its roundels; the only good example on the PAS database is SWYOR-859C26.
B18b has lugs (pierced or T-shaped) projecting from its rear edge behind the roundels; it can also have a second row of holes between the roundels and the bar (e.g. DUR-8F73B6).
B18c has sewing holes between the roundels and bar, and can have roundels on both sides of the bar (e.g. SF-33DF73). B18d has rectangular knobs, giving a crenellated appearance. B18e is a small but homogeneous group, with a wide relief-decorated bar with flat ends, and triangular knobs joined by a rear bar.
B18f is defined as having a wide relief-decorated bar with flat ends, with roundels on the rear edge joined by a bar. Hines illustrates two examples (1993, fig. 117) but on both of these the rear ‘bar’ is more a set of small pierced lugs; when the bar has no flat ends (as on SF-20A577), this form can merge with form B18c and it can be very hard to distinguish the two.
No certain example of a form B18g is recorded on the PAS database. The hook-piece of a form B18g invariably has a triangular extension to one end, a characteristic that is also sometimes found on other forms (e.g. form B18d, see Hines 1993, fig. 115b). The effect is that a triangular gusset plate has been cast in one piece with the clasp-half. The catch-piece of a form b18g has triangular knobs joined by a rear bar, so looks a little like a B18e, but without any bar along the front.
Form B18h is known from one example only, from Icklingham (Hines 1993, fig. 121) A clearer drawing of both clasp-halves can be found in West 1998, fig. 52.9), showing that both clasp-halves have triangular extensions to one end (contra Hines 1993, 62).
It is odd that form B19 – with rectangular plate and bar cast in one piece – is not more common. Form B20 is a far larger group; it also has a plate and bar cast in one piece, but has a shaped rear edge or projecting lugs. Although the group is diverse, Hines gives no formal sub-division; those illustrated here are therefore a random selection to show the variety.
Forms B18 and B20 are easily confused; some examples of Hines form B20 (e.g. FAKL-4F5333, SWYOR-3B3366 and NMS-546DD8) look very like examples of form B18. Hines explains the difference: the B20s “differ in principle in not having perforations between all the conjoined knobs or lugs behind the bar” (Hines 1993, 64); in other words, their knobs or lugs have merged into a proper plate. Having said this, though, it is still not easy sometimes to make the decision; so if you are trying to find a parallel to a B20, do look through the B18s as well.
Class C clasps
Class C is defined as clasps that do not fit into either Class A or Class B. At the time Hines wrote, all Class C clasps had zoomorphic relief ornament in Salin’s Style I. This still appears to hold true for PAS-recorded examples.
Form C1 is the commonest form. These are perhaps best described as m-shaped, with a pair of symmetrical curving birds’ heads. Form C1 is divided by Hines into several sub-groups, to which he has given names. One of these, the Norwegian type, is not recorded on the PAS database.
The Barrington type is the commonest type of C1 clasp. The relief decoration is fairly simple, with plain lines making up the elements of the birds’ heads, and no other decoration.
The next four of Hines’s types – the Mildenhall type, the Great Chesterford type, the Saxonbury/Bidford type and the Central Midlands type – are more complex in their decoration. On all, the curving necks are filled with transverse ridges, relief S-shapes or running spirals. The Mildenhall type has no other decoration (e.g. LANCUM-BA1A24); the other types have relief animal ornament that fills the V shape between the two necks, and runs along the rear edge. The detail of this ornament varies between the types, but unless the right area survives it may not be possible to assign a precise type. We do not have many good examples of these types on the PAS database.
Among the three ‘miscellaneous’ examples given by Hines (1993, 69-70) is a distinctive example from Fonaby, Lincs. There are now several other examples of this on the PAS database, so we can now call it the Fonaby type.
The other versions of Class C are not common. Form C2 is only found in Scandinavia. Form C3 is rectangular, and we have a few fragmentary examples on the PAS database, recognisable mainly because of the rectilinear frames around their hooks or catches. They also have masks on top of their hooks, and heart-shaped elements behind the hook or catch.
C4 is defined as having an outward-facing mask to three sides of a rectilinear frame around the base of the hook or catch (Hines 1993, 71). We have one good example on the PAS database, SWYOR-A3A885, illustrated below, and NMS-F2229C, with no illustration.
C5 is, oddly, a plate-and-bar type (these are normally Class B clasps) but the ornament includes a symmetrical pair of birds’ heads. Again we only have one example on the PAS database, NMS-380C00, and again there is no illustration.
Hines also includes a couple of clasps from the Morningthorpe cemetery in Norfolk, as a Class C ‘Miscellaneous’ group. The first is from Morningthorpe grave 153 (Hines 1993, 73, fig. 141a) and can be paralleled on the PAS database at LVPL-F15235.
The second is from Morningthorpe grave 360 (Hines 1993, 73, fig. 142). We now have several examples on the PAS database, and the Morningthorpe parallel can be quoted to allow them to be found easily. They are T-shaped, and the PAS examples include YORYM-07B0E3, NMS-740954 and NLM-D53EB4.
Idiosyncratic and difficult-to-place clasps
There are a very few wrist-clasps that do not appear to fall neatly into Hines’s classification. YORYM-893431 is an example from Class B; LEIC-EEF020 is very unusual, but probably a Class C.
SF-EB0CD1 is also an interesting example, with a large rectangular bar filled with Style I decoration; although the decoration might throw a recorder off track towards Class C, the bar-only construction has led John Hines to allocate this clasp to his form B12.
Gusset plates
Gusset plates are much rarer than wrist-clasps, and should be recorded as SLEEVE CLASP with ‘gusset plate’ in the classification field. They have fragile edges, and are usually found incomplete; the curve of the plate around the arm helps to identify them. Some are illustrated in Hines 1993, 74, figs. 143 and 144.
Those recorded on the PAS database seem to fall into two main shapes, with concave edges and with straight edges. Here is a selection.
Examples
Search for all examples of wrist-clasps
Search for all examples of gusset plates
