Hoards — Artisans' and Husbandmen's Tools — Domestic Appliances — CUTLERY, &C.
The products of the Romano-British ironsmith were severely utilitarian
and rarely exhibit ornamentation. In this he differed from his medieval
and modern successors, as also from his contemporary bronze-smith. It
would seem that the capability for ornamental work was practically
unknown or disregarded, nevertheless it was used for an immense number
of purposes. The iron used was wrought, not cast; the only known
example of the latter is a statuette found in a slag-heap of the era at
Beauport near Hastings.1
Three noteworthy hoards of iron — one found at Great Chesterford2 in 1855, and two at Silchester in 1890 and 19003 — will give the reader an insight into the ironmongery of the era.4.
The first two were in rubbish-pits and the last in a well, and each
apparently consisted of a smith's tools and stock-in-trade. The latter
comprised tools and other articles, finished and unfinished, such as
were used by carpenters, farriers, shoemakers, husbandmen and others,
domestic appliances, and all sorts of oddments that are best described
as 'scrap' — unconsidered pieces of iron collected by the possessors or
received in exchange for goods supplied and services rendered. In the
aggregate the total number of items in these hoards was nearly three
hundred, and included the following:— 22 hammers of various shapes and
sizes, 10 axes, 3 adzes, 3 tongs of different
types, 9 socketed chisels, 5 socketed gouges, 3 files or rasps, 1
plane, 1 centre-bit, 1 saw, 1 farrier's tool of a type known as the
'boutoir' in France, 2 dividers or compasses, 4 paring-knives, 2 heavy
anvils and an anvil bed, 3 shoemakers' anvils, 2 nail-makers' tools, 2
sates, 1 drift, 1 wringer or hand-lever, 1 shears, 1 turf-cutter, 10
plough coulters, 12 scythes, 17 mowers' anvils, 2 forks (?), 8 large
bars of uncertain use, 1 axle-tree (?), 10 felloe-bands, 2 or 3
axle-boxes (?), 1 small wheel, several shoes for staves or poles,
several knives and choppers, 1 large gridiron, 1 square girder, 5 or 6
padlocks and 3 keys for the same, 1 lamp, 1 millstone rynd, 8 shackles,
2 horse-shoes and a 'hipposandal,' several bucket-handles and hoops, 3
lengths of chain and a curious object with chains attached to it, 1
large ring, 7 hinges, and 4 holdfasts, the residue consisting of hooks,
pieces of straps, bands, and other fragments.
The heavier hammers of the era, as Fig. 55, A, resemble our
sledge-hammers, and B, a frequent lighter form, has its 'cross-paned'
end blunt as in our joiners' hammers. Both examples are from
Silchester, and the former is probably a smith's hammer and the latter
a carpenter's. Hammers of the latter form with the 'cross-paned' end
sharp were probably masons' walling-hammers. The shaft-holes are often
small, and Sir John Evans conjectured that compound hafts with iron
ends were used for these. He also observed instances in which the face
of the hammer was 'steeled' by a plate of steel welded to it.5
The Silchester example, C, is an unusual combination of hammer and
light pick, and is probably a mason's tool. D, also from Silchester,
combines hammer and adze, and resembles a tool used by modern
wheelwrights and coopers.
Fig. 55. Hammers and Axes. (All 1/3)
The two Silchester axes,
F and G, represent the ordinary Roman forms. The former approximates to
the present American felling axe, and the latter to the Kent axe. These
axes vary considerably in size and weight, and doubtless served all the
industrial purposes of their modern successors. Other shapes are rarely
found. One at Lydney resembles some of the Saxon battle-axes in its
crescentic form and long cutting-edge. One in the Guildhall, H, with a
spike behind, is certainly a butcher's pole-axe. The tool, E, from Lakenheath, Suffolk,
may be described as an axe-adze, and is not uncommon. The adzes of the
era are in general similar to the modern, and Fig. 56, A, from
Ardoch, is a typical example. Occasionally they are wider, or are
gouge-shaped, a form specially useful for shaping the staves of tubs
and barrels. It is hardly possible to draw a line between adzes, hoes,
and mattocks. Both axes and mattocks are combined with picks, but
neither quite resemble the pick-axes and pick-mattocks of to-day.
Examples of the former have been found at Newstead, B, and of the
latter at Aldborough, C, and elsewhere. An implement smaller than the
last, but with two sharp prongs behind, has been found at Lydney, Rough
Castle, and Caerwent. The Roman pick, pure and simple, seems to have
had a single arm like the medieval.
Fig. 56. Picks, Mattocks, Sickles, Spuds, etc. (All 1/3)
Of smiths' tongs of the
simpler sort, Fig. 57, A, is a good example from Silchester.
A large variety with the points of the grip turned up at right
angles, and the one again turned so as to overlap the other, has been
found at Silchester and Newstead. The pincers, B, in the
Guildhall, would be indispensable to both carpenters and farriers.
Files and rasps have been sparingly found, and D is a small
Guildhall example of the latter. A larger flat rasp, with a
cranked tang and coarsely serrated on one face, obtained from the first
Silchester hoard, is seemingly a carpenter's tool, as also for similar
rasp with a straight tang at Aldborough. Drills with tapering square or
flat butts are fairly common, and indicate that braces or kindred
appliances were in general use, but as no example has come down to us,
they were probably of wood, like our old-fashioned braces.
G, H, and I — a rimer, a gouge-bit, and apparently
a large centre-bit — are certainly carpenters' tools, the first two in
the Guildhall and the last from Chesterford. F is a metal drill,
also in the Guildhall. J is described as a shoemaker's awl in
Roach Smith's Illustrations of Roman London. It has a wooden handle and bronze ferrule; and a similar tool has been found at Bar Hill.
Fig. 57. Pincers, Drills, Chisels, Gouges, etc. (All 1/3)
Five examples of chisels and gouges are shown, and of these, M, from
Housesteads, is probably a mason's chisel, the rest, K, L, N, O, all
from Silchester, being carpenters' tools. These are of two varieties, the socketed to receive wooden handles,6 and those with expanded solid heads, but several in the Guildhall have
tangs. The plane associated with the Silchester hoard of 18907 was of wood which has perished; but the iron sheathing of the face and sides indicates that it was a jack-plane 13 ¼ ins. long and 2 ¼ ins.
wide. The blade still remains in position between two transverse rivets
extending from side-plate to side-plate, that behind threading a lead
roller against which it rests. It was probably secured by a wooden
wedge between it and the rivet in front. The remains of two planes of
similar size have been found at Caerwent, but they apparently lacked
The 'paring-knives' of the Silchester hoards seem to represent the
modern joiners' bench-knives. Each has a convex cutting-edge, 8
or 9 ins. long, and a straight back with a projecting stop at one
end and the remains of a long handle at the other. Saws are rarely
found perfect. Many seem to have been similar to small billet saws.
A tapering hand-saw, 20 ins.
long, was found at Great Chesterford, and a very small one with
deer-horn handle, at Newstead. The iron tool, Fig. 57, E, with
spatula-like ends, is one of several in the Guildhall, and is regarded
as a modelling tool, and another has one end only flattened. In the
same museum are a hollow punch for making holes in leather, and several
trowels with tangs for handles, all closely resembling the modern.
are by no means rare. They were of iron or of bronze, and the latter
sometimes had iron points. Occasionally the rivet has a slot for a
wedge-shaped cotter by which the joint could be so tightened as to
become practically rigid.
Fig. 68, H, is an ornamented bronze example from Tingewick, Buckinghamshire.
The smith's anvil of the Great Chesterford hoard is a rectangular block of iron with a projecting tabular face, 7
by 5 ins., and a stout tapering tang below for insertion into a
wooden block. The Silchester anvil resembles the modern in having a
conical beak at one end, and it has a similar tang to the foregoing.8 Although modern in appearance, it is of a form that goes back to the Bronze Age.9
A small anvil similar to those used by goldsmiths was found at
Rushmore. The Silchester shoemakers' anvils resemble those still in
use, and were supported on stems with shouldered tangs for wooden
blocks. The mowers' anvils are from 7 to 11 ins. long, the upper third about
1 in. square in section, and tapering below to a point. The
shoulder is perforated for one or two strips of iron with their ends
horizontally coiled to form supporting brackets. In the Caerwent
example (Fig. 57, C) the brackets are of a single strip. Similar
anvils are still used for beating the edges of scythes upon, in France,
Spain, and Italy, and they are made at Birmingham for exportation to
South America. They are driven into the ground and flat stones or
pieces of wood are placed under the brackets to give them a firm
The scythes of the Great Chesterford hoard were remarkable for their shape and length, which was little
short of 7 ft. Like the modern scythes, they had a stiffening
ridge at the back, but they differed in their curve. This, instead of
being gentle throughout, made a rapid bend at about
17 ins. from the butt, causing this recurved portion to be turned
somewhat in the direction of the point. This portion was narrow and
ended in a turned-up tang for insertion in the sneed. There must,
however, have been some additional means for securing these large
blades to their handles. Several scythes found at Newstead were shorter
and wider, and their curves less accentuated towards the butt, thus
approximating to the modern. A scythe found at Bokerly by General
Pitt-Rivers, 2 ft. 5 1/2 ins. long, differed again in its sickle-like shape and in having a socket for the sneed.
Curved knives of various shapes and sizes, and evidently used in
agriculture, are of common occurrence. The larger of the form of
Fig. 56, D, from Silchester, and E, F, from London, are certainly
sickles, and the smaller may have been pruning-hooks. The small size of
the Roman and the prehistoric sickles is due to the ancient custom of
cutting the ears of corn from off the straw, handful by handful.
A socketed tool less curved than the last and about 1 ft. long, found at Caerwent, may be described
as a bill-hook, and was probably used for slashing
off branches. The socketed tools from Rushmore, G and H, are usually
described as spuds; they may, however, have been respectively the
points of a wooden mattock and pick.
The 'hippo-sandal' of the Silchester hoard of 1890 is a not uncommon
object both in this country and in France. It has a remote resemblance
to a slipper, with a portion of each side of its sole turned up to form
a wing or clip, an ascending tongue with a loop at one end, and the
other slightly rising and terminating in a loop or hook. In all these
details, however, it varies considerably, and sometimes in lieu of the
second, the wings are developed and coalesce with a loop at the
junction. Others again may be regarded as half-'sandals,' being
narrower, with one side straight and lacking the clip. Two of these —
a right and a left — would make a complete 'sandal.' These
articles have been regarded as lamp-stands, as skids for wheels, as
shoes for the ends of the pole-car or sledge, but the prevailing
opinion is that they were temporary shoes for horses with injured hoofs
or when going over stony ground. Those in halves may have been for oxen.10
Horse- and ox-shoes are found on Roman sites, and they differ from the
modern chiefly in their smaller size, which is explainable by the
well-known fact of the small size of the Romano-British horses and
oxen. Horse-shoes with undulating or slightly scolloped sides are
rather characteristic of the era.
The gridiron of the same hoard is about
17 by 18 ins. It consists of a rectangular frame, with bars
arranged longitudinally and transversely (the central one expanding
into a circle), resting on four legs and with a ring at each end.
Gridirons are rarely found with Roman remains. There are two in the
Lewes Museum, each about
1 ft. square, with four legs, parallel bars, and a straight
handle. The curious object with chains in the Great Chesterford hoard
was certainly a pot-hanger.11
The swivel-piece was large and ornamented with a large ring on the
summit to receive the supporting beam or bracket; and from it depended
a chain, at
the foot of which was
attached two shorter chains, each ending with a hook. There are the
remains of a similar hanger in the Cirencester Museum. An iron folding
tripod, 4 ft. 3 ins. high, found at Stanfordbury, Bedfordshire, had a swivelled pot-hanger suspended from its summit.12 It was associated with a pair of fire-dogs and several bronze cooking utensils.
Other fire-dogs have been found associated with Roman remains at Mount Bures13 near Colchester, Capel Garmon, Denbighshire,14 and near Barton, Cambridgeshire.15 They were all of one type, consisting of two uprights about 2 1/2 ft.
high, connected below with a horizontal bar, and resting on four feet,
each pair of feet being formed of a curved piece attached to the bottom
of the upright. Each upright terminated above in an ox's head with long
horns, and facing outwards. The Capel Garmon dog was an elaborate
example with the uprights ornamented with series of semicircular loops.
These fire-dogs, having double fronts, were adapted for central
hearths, in this respect unlike the medieval and later, which usually
had single uprights and hook-like projections on the inner sides of
these. These were probably placed between the dogs, and held in
position by being threaded on the horns, the hooks serving as supports
for horizontal spits, and the intervening portions for hanging toasters
and other cooking appliances. Bars resting upon the horizontal members
of the dogs would usefully support pans and other cooking utensils.
Knives are almost invariably found on our Roman sites, but it is only
where they have escaped the extreme effects of oxidization that their
good quality and finish can be appreciated. Many have bone handles, but
as most are without, it may be inferred that the majority were of wood.
They were attached by three methods — by a narrow tang inserted into or
passing through the handle; by a plate-tang, the handle being in two
halves, one on either side of the plate, and riveted through it; and
less frequently and only in the larger knives, by a socket, into which
the handle was inserted. Occasionally the handle is of
iron, blade and handle being in a single piece.
Some are of bronze, usually in the form of an animal or terminating in
an animal's head. Three prevailing shapes of blades can be
distinguished, examples of which are shown in Fig. 58, A to
I. Those with the curved blades of B, C, and I, and of the rarer
shape A, all from London, are specially notable for their careful
finish. The handles of these are mostly of bone, ornamented with
incised lines or circles, and the plates to which they are riveted
usually end in semicircular loops or rings. In rare instances these
knives are tanged for tubular handles, as in E, from Rushmore. Knives
of this type apparently answer to our table-knives, be they are smaller
as a rule, and the handles rarely exceed 3 ins. in length.
Straight-bladed knives with tangs are more common. They vary
considerably in shape and size, but G, from Rushmore, is representative
of the majority. Occasionally they are leaf-shaped, or the back is
straight from base to point. The handles were usually of wood, bone
handles as in D, a large knife from Lydney, being uncommon; and,
to judge from the length of the tangs, they were generally long. We can
hardly class with either group the little knives, A and H,
both from London. The first is a rare example stamped with the maker's
name — OLONDVS F., and its handle appears to have been of wood. The second has an iron handle, and was possibly a surgeon's implement.
Fig. 58. Knives and Shears. (J, K, L, 2/3; the rest, 1/3)
Large knives with triangular blades of the shape of F have tangs
or sockets, and there is a good example of the former with a bone
handle from Arncliffe in the British Museum. They are probably
butchers' knives, and this is corroborated by the fact that the knives
carved on altars, with other sacrificial implements, are of this shape.
Clasp- or pocket-knives are mostly of the forms of J and K.
The former is from Caerwent and has a turned cylindrical bronze handle.
The latter is of more frequent occurrence, and its handle is of
openwork bronze, representing a hound catching a hare. In the next
example, L, from Lydney, the handle is of bone, ornamented with
incised circles, and furnished with a ring for suspension. The Roman
clasp-knives lacked the back spring and the nail-groove of the modern.
Steels were certainly used, for Mr. Roach Smith figured one found
in London, of a shape resembling that of a modern flat file, with a
bronze handle in the form of a horse's head.16
The usual implement for the purpose, however, was the whetstone, of
which examples are almost invariably found on Roman sites. Any
convenient piece of hard sandstone an
inch or two in width, and from 4 to 8 or more in length, or a
long siliceous pebble, served the end; but sometimes the stone was
neatly shaped, and this especially so in the case of small whetstones,
which often had a hole for suspension.
Shears are not uncommon, and three examples are given,
M, N, O, the first being of bronze from Caerleon, and the
other two of iron, from Rushmore and Aldborough respectively. The last
is perhaps the more usual form. The other two resemble the modern
shears in the circular sweep of their heads — an arrangement which
materially increases their elasticity. The rounded notch at the base of
each cutting edge of N for cutting cord or twigs is a convenient
feature, but the bronze example, M, which possibly is medieval, is
unusual in having four on each side. The shears are of all sizes from
about 5 ins. to 1 ft. or more, and the smaller were certainly
domestic appliances used as our modern scissors, for although the
Romans were acquainted with scissors on the lever principle, very few
examples have been found in this country, and it is doubtful whether
they are Roman at all.
The list of the contents of the Great Chesterford and Silchester hoards
by no means exhausts the varied uses of iron. Nails, straps, holdfasts,
clamps, sheaths and sockets for door pivots, hinges of various types,
hasps, bolts, latches, and joints for tree-pipes, indicate its
extensive use in building-construction. The first are invariably found
on the sites of buildings and often in abundance, and most closely
resemble those of modern joiners and carpenters.17
Iron padlocks, large keys, lock-escutcheons, chains with links as
varied as the modern, bridle-bits and other details of horse-harness,
swivels, shackles, goad-heads, swords, daggers, spear, arrow and bolt heads, and chain and
plate-mail, have all been found, some plentifully, others rarely.
Besides these, the excavations of large sites, as those of Silchester
and Caerwent, have yielded many iron objects and fragments, the
purposes of which are uncertain or quite unknown.
The Author's Notes:
1. Proc. Soc. Ant. 2, xiv, p359.
2. Arch. Jour. xiii, p1.
3. Archaeologia, lii, p742; liv, p139; lvii, p246.
4. Liger, La Ferronnerie is a useful book of reference.
5. Archaeologia, liv, p145.
6. One of four chisels found at Newstead has a haft of deer-horn.
7. Figured in Archaeologia, liv, 151.
8. These two anvils are figured in Arch. Jour. xiii, plate i; and Archaeologia, liv, p142.
9. Evans, Ancient Bronze Implements, p182.
10. Brit. Arch. Assoc. l, p251; Archaeologia, liv, p154. See also Essex Arch. Soc., 1, p108.
11. Arch. Jour. xiii, plate ii. See also Archaeologia, lvi, p242.
12. Collect. Antiq. ii, plate xi.
13. Ib. ii, p25.
14. Arch. Camb. 3, ii, p91.
15. Archaeologia, xix, p57.
16. Illustrations of Roman London, p141.
17. A series of these is illustrated in Rom. Brit. Buildings and Earthworks, chap. xi.