Orton C., Tyers P., Vince A. Pottery in archaeology
Подождите немного. Документ загружается.
Forming
119
120 Making pottery
and thus indirectly the moulds, has engendered a vast and apparently endless
literature aimed at identifying the web of relationships between potters,
mould-makers, styles and workshops, their chronology and organisation.
Wheel throwing
The process of throwing pots on a wheel is perhaps that most commonly
associated in popular culture with the pottery production process. There are
allusions in classical and later literature to the process of wheel-throwing,
particularly the rise and fall of the pot on the wheel, the fluid changes of shape
under the hands of the potter (e.g. Horace, Ars poetica, 21-2). This aspect of
potting is well described by Rye and Evans, in a passage from their study of
traditional pottery manufacturing in Pakistan:
To any observer who has not attempted to make pottery, forming operations
often appear to happen so effortlessly that it can mistakenly be assumed that
'anyone can do it'. This illusion is created partly by the plastic nature of the clay,
which when soft deforms at the slightest touch, and partly by the skill and
experience of the potter in controlling such a sensitive material.
(Rye and Evans, 1976,131)
Despite the clear evidence of their widespread use in the products, the
archaeological remains of potter's wheels are slight and fragmentary at all
periods. Various flywheels, spindles and sockets recovered from excavations
on kiln sites have been identified as parts of potter's wheels but these are often
difficult to interpret, and there is some risk of confusion with, for instance,
mills and other rotary devices (Peacock 1982, 55-8).
Much of our detailed information on their structure and operation comes
from contemporary illustrations or ethnographic sources. The history, con-
struction and use of the potter's wheel has been discussed by several authors
(Foster 1959; Rieth 1960; Scheufler 1968; Lobert 1984) who draw on illustra-
tions of these devices in use, culled from a range of ethnographic and
historical sources. Two broad classes of wheel are distinguished by most
writers.
The simple wheel, also referred to as the stick wheel or single wheel
- a single flywheel rotating on a central pivot (fig. 10.2(a)).
Depending on the method of attachment between the wheel and
the pivot, the flywheel may either come to rest at the horizontal
or at an angle. The balance of the flywheel maintains an even,
level rotation. The top surface usually has a small depression
near the perimeter at one point and a short stick inserted in this
socket rotates the wheel to start it up, or to speed it up when it is
slowing down. The potter (or the assistant) is usually sitting or
squatting beside the wheel while doing this, but may, more
rarely, use a longer stick from the standing position.
Forming
10.2. The two basic types of potter's wheel, (a) the single wheel, and (b) the double wheel
(Rye 1981, fig. 58)
The pit wheel (or kick wheel or double wheel) is composed of two
flywheels attached to a long central spindle and supported by a
lower socket and an upper bearing
(fig.
10.2(b)). The upper wheel
is the throwing surface and is usually light and of relatively small
diameter. The lower wheel is larger and heavier, and stores the
122 Making pottery
Fig. 10.3. A potter in Srinagar (Kashmir) (a) using a single wheel, and (b) the underside of the
wheel, showing the pivot set in the ground. (Photos: Paul Tyers)
Forming
123
124 Making pottery
momentum which is applied by the potter who kicks the wheel,
or, in modern wheels, by a motor.
Amongst modern traditional potters the single wheel is particularly char-
acteristic of the Indian subcontinent, but the kick wheel is widespread
throughout both Europe and Asia. The discussion of pottery wheels in
Pakistan by Rye and Evans (where both types are found) is particularly
valuable (Rye and Evans 1976, 116-18). In the Greek and Roman world it is
probable that both single and double wheels were employed (Peacock 1982,
55-8). In medieval and later Europe the simple wheel was often in the form of
a cartwheel mounted horizontally with the head attached to the centre, and a
form of kick-wheel where the upper and lower flywheels were connected by a
series of vertical struts was employed (McCarthy and Brooks 1988, figs.
8-10).
The different forms of wheel are not all completely interchangeable, each
equally suitable for the manufacture of all types of vessel. The mechanics of
the pottery wheel dictate, to a certain extent, the forms of vessel that can be
produced, or rather the character of the wheel will reflect the requirements of
different vessel types. Two important features are discussed by van der Leeuw
in a consideration of the type of potter's wheels employed in the manufacture
of sixteenth-century pottery at Haarlem in the Netherlands (van der Leeuw
1976, 124-5 and appendix D):
The relationship between the weight of the wheel, its momentum
and its optimum speed
—
a large heavy wheel will maintain its
momentum for a longer period and revolve at a slower speed and
resist friction from the potter's hands (and other sources) better
than a small light wheel.
The 'linear' speed, the speed at which the clay passes between the
fingers of the potter, increases with distance from the centre of
the wheel.
According to Hulthen (1974, 69) the minimum linear speed required to
produce wheel-thrown pottery is 0.7m per second. A large heavy wheel, such
as the single cartwheel type, will revolve at a slow speed, but for relatively
long duration. The slow speed makes it suited to the manufacture of large-
diameter vessels, for the linear speed will increase with the diameter of the
vessel. A faster lighter wheel would not be suitable for large diameter vessels
as the linear speed would be too high. Conversely, a faster wheel would be
preferable for the forming of smaller diameter vessels, where the linear
velocity would be adequate a short distance from the centre of the wheel. In
Pakistan the simple stick wheels are often used for throwing small vessels 'off
Forming
125
the lump' (Rye and Evans 1976, 116-18) and are unsuitable for the manufac-
ture of larger diameter vessels
(fig.
10.3). Thus, it may on occasion be possible
to predict the character of the wheel employed for the manufacture of a
particular group of pottery, even in the absence of direct archaeological
evidence such as fragments of the wheel structures themselves (van der Leeuw
1976, 126).
Composite techniques
The broad distinctions made between hand-made and wheel-thrown pottery
should not distract us from the recognition that many vessels are formed by a
combination of techniques. In the simplest cases some alterations or addi-
tions may be made to the basic form created by the primary technique. A
spout may be created by deforming the circular mouth to produce a 'pinched'
effect,
the body shape may be altered by 'folding' or parts of the vessel may be
cut away. Pots may have handles, spouts, footrings or other features attached
by hand after the primary forming stage. These subsidiary features may
themselves be hand-made or wheel-thrown and in some cases the method of
attachment varies significantly between potters and may be a valuable indica-
tor of evolving manufacturing traditions (Pearce 1984). Whatever the precise
alteration, these actions may mask or erase, at least over a small part of the
final vessel, the traces of the underlying technique.
More complex combinations are also possible. Beating the wall of a vessel
between a 'paddle' and 'anvil' (usually a rounded stone) to make it thinner is
a common technique in modern India for the manufacture of large water jars
and items of similar dimensions. The starting point may be a vessel that is
either hand-formed or wheel-thrown and the effect of the beating is to
produce a vessel of larger capacity, usually more globular in shape and with a
higher volume-to-weight ratio. Several descriptions of the details of the
technique are available (Miller 1985, 222-6; Saraswati and Behura 1966,
55-8). The finished vessel will exhibit traces of both the primary forming
process, such as wheel marks, on the rim and upper shoulder, and the beating
action, usually on the body. The latter leaves characteristic surface markings
and finish (Rye 1981, 84-5). In fragmentary material it would be important to
recognise the connection between the sherds from the two parts of the vessel.
Moulds are often employed in combination with other techniques. Vessels
with a mould-formed lower part and a hand-formed or wheel-thrown upper
part are known. Early Roman samian ware bowls are formed primarily in
moulds but have rims and footrings added later. Samian production was a
large-scale enterprise involving many potters and workmen each responsible
for a particular stage of the procedure. Some workshops maintained quality
and quantity control by the use of small stamps impressed on the surface of
the pots by those carrying out the different operations (summaries of samian
ware production and the use of stamps are given by Johns 1971 and Buhner
126 Making pottery
1979). An understanding of the different stages of manufacture of such
vessels is particularly important as the stamps often refer to only one stage of
a complex procedure.
Surface treatments
There is no precise dividing line between some secondary forming procedures
and surface treatments. Trimming or scraping serve as much to smooth the
irregularities left by coil or slab construction and join the different parts
together as to alter the appearance of the vessel. Some surface treatments can
only be applied when the vessel has dried to the 'leather-hard' stage while
others may be used while the vessel is still plastic.
One of the most common surface treatments is burnishing, where the
leather-hard pot is rubbed with a smooth pebble or other tool. This compacts
the surface and leaves a series of
facets
and a slight lustre on the surface. The
burnishing may also have some effect on the heating effectiveness of ceramic
vessels (SchifTer 1990) by reducing the passage of liquids through the vessel
while it is being heated over a fire. There are many other possible types of
surface decoration. Many involve penetrating, compressing or cutting the
vessel surface with a tool. This may be a simple narrow-ended implement (a
point) or something more complex such as a comb or a stamp (see p. 85).
Drying
Prior to firing, the finished vessel must be dried. This eliminates the water
mechanically combined to the clay particles. Drying may take place in
free
air
or in specially heated drying sheds. The vessel will shrink during drying which
may set up stresses in the pot, resulting in drying cracks. The shape and
position of these cracks reflects to some extent the procedures employed in
the manufacture of the pot (Rye 1981, 66, fig. 46) - the S-shaped crack on the
base of wheel-thrown vessels is one of the clearest examples.
Drying also has the effect of concentrating dissolved salts and fine clay
particles towards the surface of the pot, drawn there by the movement of
water through the wall. On enclosed vessels this effect is particularly marked
on the outer surface. This in turn may affect the colour of the vessel during
firing. It is important to distinguish these effects from slips, paints and other
surface treatments.
Firing
The purpose of firing is to transform the clay minerals into a new material,
ceramic. The necessary chemical and physical changes start to take place in
some clays at about 550-600°C (Rice 1987, 90-3) and pottery which does not
reach this temperature during firing will eventually disintegrate when immersed
in water, even if it takes up to a year to do so (Skibo and Schiffer 1987, 85).
Two categories of pottery firing procedure are recognised:
Firing
127
Open firings, also referred to as clamp firings or bonfire firings,
where the pots and the fuel are in immediate contact and are
arranged in a stack on the ground or in a shallow depression.
Kiln firings, where the pottery and fuel are separate - the pot
usually in a chamber which is heated by the hot gases and flames
from the fuel.
Open firing
Open firings are common amongst traditional potters and would
undoubtedly have been responsible for a substantial proportion of the mater-
ial recovered from archaeological contexts. Many excellent ethnographic
descriptions of the procedures are available (e.g. Gruner 1973; Rye and Evans
1976,165-7; Arnold 1978, 351-7; Miller 1985,228-32; see also Kramer 1985,
81). The usual method is to stack the pots over a layer of
fuel
and mix more
fuel in and around the vessels, perhaps covering the entire stack with more
fuel and/or a layer of waste sherds from earlier firings (fig. 10.4). The fuel is
usually lit from below or at one end of the stack and burns through. Perhaps
the most notable characteristics of an open firing are the fast rise in tempera-
ture during the initial stages and its short duration. It may only take a few
minutes for the temperature to reach its maximum and fired pots can be
removed from the fire shortly thereafter (Rye 1981, 102, table 3 records
details of eleven such firings), although some other open firings take almost
two hours to reach their maximum and the stack may not be opened until
eight to ten hours later (Miller 1985, 230, fig. 71).
In the simplest open firings a very small number of vessels, perhaps only
one or two, are fired in a small shallow depression in the ground. The ash
remaining in the firing pit after the pots are removed may be collected for use
as fertiliser (Rye and Evans 1976, 165) and the identification of such features
in the archaeological record as evidence of pottery manufacturing would be
rather problematic (Sullivan 1988, 33). Larger firing areas which are
repeatedly reused are often thickly covered in broken and burnt pottery,
although very little of this is distorted or fused. Again their identification
might not be entirely clear, particularly if the area is later reused as a rubbish
dump or midden. In fact, open-firing areas are only very occasionally identi-
fied in the archaeological record (Peacock 1982,67, with references) although
the assertion that a particular pottery type cannot have been produced in an
area 'because no kilns have been found' is heard with surprising frequency.
Although undoubtedly a simpler technique than the use of kilns, open
firings are perfectly adequate for the firing of many coarse wares and have the
advantage of flexibility. Greater or lesser quantities of pots may be fired at a
single session simply by making the stack larger or smaller. The potter is not
confined to the cycle of producing and firing a complete kiln load. Although
often associated with hand-formed coarsely tempered wares, open firings are
over layers of firewood on the ground. (Photo: Paul Tyers)