Orton C., Tyers P., Vince A. Pottery in archaeology

Подождите немного. Документ загружается.

ILLUSTRATION

Introduction

Pottery is normally drawn in a highly stylised manner, although archae-

ologists soon become so used to the system that they find it remarkable that

given a sherd of pottery an illustrator unfamiliar with the conventions will

actually draw what he or she sees. Wheel-thrown or other vessels with a

central symmetry are conventionally shown with a central vertical line. On

one side of this line the cross-section of the vessel wall and the interior surface

of the pot is shown while on the other the exterior surface of the pot is shown.

The convention most widely followed is to show the interior view on the left

and the exterior on the right, but in some publications, in particular those

from eastern Europe, this is reversed.

As a means of producing a simple record of a vessel, for use by a pottery

researcher or as part of an archive for example, pottery drawing can be quick,

cheap and effective. Once a decision is made to make drawings to a publi-

cation standard, showing texture and surface treatment, illustration can

become very slow and expensive. It can become one of the most costly

elements in pottery research. It is therefore important to consider at an early

stage what exactly you will be using pottery illustrations for and what

implications this will have on your project design.

The purpose of illustrations

Pottery illustrations are intended to make comparison of vessels simpler by

reconstructing on paper as much as possible of the complete form, even if

only sherds of it have been found. Information and detail which would hinder

this comparison, such as the condition of the sherds, concretions adhering to

them or sherd edges (unless needed to show the extent of decoration), is

omitted.

At their simplest, illustrations can serve as a record of the contents of an

assemblage and be made as a matter of course as part of the recording of the

collection. Such a record is almost certain to contain redundant data which

will never be of use to anyone, but there may be cases where this approach is

justified. In particular, if it is not possible to retain the collection for some

reason and no typology is available then, as a poor second, the rims, bases

and other featured sherds can be drawn and a certain amount of information

extracted from the drawings at a later stage. The mindless illustration of

Illustration

pottery 'because it's there' is frowned upon in Britain, for the reasons given

above. Ironically, it is often the illustrations which enable early archaeo-

logical pottery collections to be re-examined and new analyses to take place.

Another strategy is to illustrate a type-series, so that only representative

examples of particular types (however they are defined) are illustrated. The

logic of this approach is that it provides some form of evidence for the

presence of a type on a particular site and ultimately forms a reference series

which can be used in the analysis of subsequent collections. The problem here

is that as one moves away from illustration as a record towards illustration to

make a particular point so it becomes less and less clear that the illustrated

material should be drawn from examples within the collection under study.

A type-series, logically, should be constructed using the most complete

example of each type, no matter where that example has been found. Thus, it

is extremely difficult to construct a type-series without undertaking a large

amount of background research. An example of this was the London Medi-

eval Pottery Corpus, which was constructed as part of the analysis of

medieval pottery stratified in large dated assemblages excavated in the City of

London, but where only a small fraction of the vessels used as type examples

actually came from archaeological contexts (Pearce and Vince 1988) - many

came instead from older museum collections.

As one moves from illustration of a site assemblage as a record through to

the construction of a multi-site or regional type-series, so the character of the

illustration itself may change. At the one end of the spectrum one needs a

simple record of the form, which may perhaps be of use in subsequent

typological analysis, while at the other one wants to convey information

about the original appearance of the vessel, as an aid to identification and

study, and the method of manufacture, since this may itself be a means of

identification, as well as being a subject of interest in its own right.

A basic decision that has to be made is whether to illustrate by means of

drawings or photographs (or both, as the two are not mutually exclusive). It

has been found that drawings give a better representation of shape and can

show both inside and outside simultaneously, which photography cannot do.

Photographs are better at showing texture, some classes of decoration and

technical details such as the means of attaching handles. They can also give

attractive 'wedding portrait'-style illustrations of large assemblages, such as

pit- and grave-groups. It used to be said that line drawings were preferable to

photographs for purposes of publication because of the difference in the cost

of reproduction. This has narrowed in recent years, but still holds in the case

of colour photography, which should be used sparingly and in very special

circumstances (for example highly-decorated polychrome pottery). Overall

we may say that drawings are better for a permanent record and for convey-

ing basic information, while photography is better at creating impressions

and conveying subtle information.

Drawing pottery

There are several works describing the method used to draw pottery recon-

structions (especially Griffiths et al. 1990). The traditional method is to mark

in pencil on paper or drawing film two perpendicular base lines, one of which

represents the vertical axis of the vessel while the other represents the top or

bottom (depending on whether the rim or base is present). The illustrator

then establishes the diameter of the rim or base, using a rim chart or direct

measurement if a complete pot. This is then marked onto the drawing. The

depth/height and diameter of other significant features on the pot are then

determined and marked on the drawing. This can be done by placing the

rim/base at the correct position on the rim chart and then using a set square

to find the correct height/diameter of each point.

Once the major features of the vessel have been accurately located on the

drawing one can either use the same method at intervals to help determine the

intermediate points or use some form of mechanical aid, such as a profile

gauge (as used by builders and DIY enthusiasts to cut floor covering to shape

around curved surfaces). With experience it will become clear that a discrep-

ancy of even a few millimetres on any measurement can make the profile look

wrong and that it is often better to use fewer measurements but more

observation when drawing the profile of a pot. A common problem with

mechanical aids such as the profile gauge, particularly in the hands of

inexperienced illustrators, is that rather more attention is paid to drawing the

device than the vessel.

The next stage is to measure the thickness of the vessel and the interior

shape of the rim or base. These are drawn on the left-hand side of the central

line. Next, sharp changes of profile can be shown as a line on the appropriate

view. The basic shape of the vessel has now been completed. If there are no

additional elements then one can start to tidy up the drawing and ink it in.

Handles are conventionally shown on the right-hand (exterior) view of the

pot. A cross-section is shown perpendicular to the handle profile with two

short lines showing the position of the section (fig. 7.1 (a)). There may be

cases where a vertical section through the handle would be informative, for

example to show the means of handle attachment. If

used,

this section would

be drawn alongside the handle profile. If the vessel has two opposed handles

then one can be drawn in profile and the other in vertical section. Three

handles would be indicated by drawing one in profile and extending the

exterior view to the left of the centre line to show the second handle in

three-quarters view (fig. 7.1 (b)). Lug handles are usually shown on the

central line.

Spouts are usually shown on the left-hand, interior, side of the drawing.

Where formed as a deformation of the rim it is usual to show two cross-

sections superimposed, one is the normal section and the other the section

across the lip. Some medieval jugs had very elaborately constructed spouts

Drawing pottery

which can require both a sectional view and a frontal view, which is usually

shown to the left of the drawing (fig. 7.1 (c)).

Decoration is difficult to illustrate satisfactorily. For some purposes it can

be important to show the plan view, especially where the vessel is an intern-

ally decorated bowl or plate. Whenever one is trying to show the decoration

on a spherical surface there will be some distortion. One way of lessening this

distortion is to 'unwrap' the decoration (fig. 7.2). This shows each element

without too much distortion but makes it impossible to see how the decor-

ation fits the shape of the pot.

You may need to use conventions to show the use of

different

colour slips,

paints and lustres. The rule seems to be not to try to fit every drawing into the

same conventions but to show stippling or hatching which makes the pattern

clear and to provide a key alongside the drawing.

The use of drawings to show surface texture is not universally accepted,

one suspects because few archaeologists have the necessary illustrative skills.

Cost is also an important factor, but many drawings are carefully shaded to

show the curvature of the pot (redundant since we know the pot is round, and

if it were not then special attention would be drawn to the

fact).

Even worse is

the painstaking stippling which achieves nothing except to make the pot

appear to be made of expanded polystyrene. If the time devoted to such

ritualistic exercises were spent on a realistic representation of surface texture,

the standard of illustration would be raised enormously. Similarly, drawings

can be used to show constructional features, such as the coils of hand-built

vessels, and surface treatment such as paddle-and-anvil and knife-trimming,

all of which help the reader to understand the pot (fig. 7.3).

Another difficult question is whether and how to reconstruct in a drawing

the missing parts of a vessel. Some authorities say that you should only draw

what is present, no more (Blake and Davey 1983, 42), but this seems

unnecessarily purist provided you make it clear what you have done. There

are in fact two distinct problems, which one might call 'vertical' and 'horizon-

tal' interpolation/extrapolation. The vertical sort, usually extrapolation, is

needed when the profile of the drawn vessel is incomplete. It may be reason-

able, using knowledge of similar vessels, to extend the drawn profile beyond

the limits of the actual pot, to give an estimate of the shape of the whole thing.

Fig. 7.1. Illustrating handles. These four vessels are all examples of thirteenth- or fourteenth-

century Surrey Whiteware found in London. The standard convention of

drawing

the handle

in profile on the right hand side of the drawing has been adapted to best portray the

manufacturing details and decoration, (a) Two-handled storage jar in which the body has been

pushed into the handle and the resulting depression inside the pot filled with an extra wad of

clay, (b) Jug on which the rod handle is decorated with stabbed holes and two

'ears'

of clay best

portrayed by a vertical view, (c) A pipkin whose handle has been inserted through the body

and the resulting hole sealed by an extra wad of clay, (d) Jug similar to (c) but whose furrowed

and stabbed handle is best illustrated by a horizontal view. Scale: 1/4

92 Illustration

To be honest, you must show which parts are 'real' and which are extrapo-

lation; this can be done, for example, by not blacking-in the extrapolated

section, using broken lines, or by the use of discrete marks beside the drawn

profile (Gillam 1957).

The horizontal problem arises when, at any depth below the rim, not all of

the horizontal circumference of the pot is present. Strictly speaking, the

problem is one of interpolation because the circuit of the pot is closed (that is

if you go round far enough you return to your starting point). The problem

does not really exist for undecorated pots since they can be assumed to be 'the

same' all the way round and drawn accordingly. Decoration which is 'hori-

zontal' in nature, such as burnished zones, cordons, horizontal lines (applied,

grooved, and so on) present few problems, since radial symmetry can again be

assumed. But how far can you go with more complex decoration? This

depends on the regularity of the pattern and the extent of one's knowledge

about the class of pottery concerned. Once again, the golden rule is to make it

clear to your readers how much exists and how much has been 'made up',

leaving them to judge how reasonable the attempt was. Similar consider-

ations apply in the addition of features which you can reasonably assume to

have been present on more complete examples of certain classes of vessel,

such as handles, feet and spouts.

Since pottery illustrations can be both laborious and repetitive there have

been several attempts to partially or completely mechanise the method. By

and large these attempts have not been successful, or at least have not

superseded manual methods to any great extent. One which may have a

certain appeal, but is not to be recommended for use with museum acces-

sions, is the use of a large circular saw to slice through every rim. One half is

then placed onto paper or film and the outline drawn around (Holladay

1976). There are a number of published descriptions of simple gadgets which

some practitioners have found valuable aids to pottery drawing (e.g. Terrell

and Osborne 1971; Trump 1972; Edwards 1974). With the wider availability

Drawing pottery

Fig. 7.3. The limits of objective recording. The same vessel drawn by four different illustrators.

(Drawn by John Newton, Paul Tyers, Give Orton and Sunil Nandha)

of computers others have sought to eliminate paper and pencil altogether.

Turner et al. (1990) describe a system based on a modified 'mouse' with

attached probes to input profile data directly onto the computer. Such a

system may be a quick and efficient method of converting profiles into a

computer-readable form for some form of shape analysis (see p. 1571 but the

trend in publication drawings should be away from the 'machine-made" look

and towards more realistic renderings.

94 Illustration

Photography

Photography can be very useful in recording pottery. Photographs of vessels

in situ may aid interpretation and can form the most vivid means of demon-

strating the function of a vessel. Individual sherds can be recorded by

photography as a source of reference in research or as a means of illustrating

lectures or workshops. Photographs should be taken of sherd families before

restoration as a final record of their size and shape. Complete vessels should

also be photographed as a matter of course. If they have any monetary value

then the photograph will act as a means of registering their identity and there

will always be calls for photographs to illustrate articles and leaflets as well as

to form elements in museum displays.

Close-up photography is very useful for recording surface treatment and

details of manufacture as well as providing an aid to identification if the

broken sherd edges are photographed. Micro-photography is used as an

adjunct to fabric analysis while Scanning Electron Microscope images can

also be preserved as photographs and, again, form a graphic means of

presentation of evidence (fig. 7.4).

The use of a scale in photography is important. Full-size, two-metre,

ranging rods will be too large for most site shots and a one-metre or

half-metre rule is better. For studio shots a ten-centimetre scale may be

suitable whilst a common coin will act as an instantly recognisable scale for

shots destined to be used in lectures. Here too, scales are frowned upon by

non-specialist publications, and even by some specialist ones. You may

therefore have to augment your photographic record with the height or other

maximum dimension of one of the vessels in the shot. The distortion in size in

group shots is so great that no scale is going to be of value in taking precise

measurements from such a photograph. For close-up work you may want to

have the photograph produced at a set scale: life-size, twice life-size, half scale

or whatever. This is best achieved either by photographing the sherd on top

of a grid or with a scale rule laid along one edge of the shot. In the former case

you would have to crop the photograph for publication whilst in the latter the

scale can be kept outside the printing area. It is difficult to supply a scale in

close-up photography. For use in lectures and other publicity a common

object, such as the human hand, can be very effective but for record work it is

better to have a small rule, perhaps part of an etched metal ruler, which you

photograph before starting on the objects themselves. This will be perfectly

acceptable if the camera and stage are fixed in position. The same method can

be used for micro-photography and you can purchase microscope slides

incorporating minute graticules. There is no easy way of putting a scale into a

micro-photograph without a two-stage process of taking a photograph,

determining the scale and marking it on a print and then re-photographing it.

SEM photographs have the scale superimposed on them automatically.

On-site photography is notoriously difficult to get right, since the subject

photography

mi 15KV X180 UD48

Fig. 7.4. SEM photomicrograph of Stamford ware crucible. Fabric SI (BMRL sample ?5835l

Poorly sorted grains of quartz (light grey) are embedded in a fired clay matrix sh

elongate cracks (black). The small white grains are iron oxides. All the Photomicrographsare

orientated with their long axes parallel to the surfaces of the sherds and all were taken at the

same magnification, the long axis representing 1mm. (Courtesy of the British Museum

Department of Scientific Research)

and angle of the shot are predetermined, no matter what the natural lighting.

It is also not always possible to wait for suitable conditions, since photo-

graphy holds up further work on that part of the site. Studio photography

can use lamps but if necessary sherds and vessels can be photographed in

natural light, preferably out of strong sunlight so as to minimise shadows.

Various types of background are used in studio shots. Whatever back-

ground is chosen, it should have a matt surface, so as not to cause reflections

and glare, and not be distracting from the subject of the shot. This is not to

say that it must be a plain white background. Coloured backgrounds and

textured backgrounds can be used to great effect, especially in presenting

pottery vessels for the general public. One problem with most backings is that

the need to use two lamps can create double shadows which are distracting

and slightly

confusing.

Sherds can be photographed on a sheet of glass so that

the shadows are out of the field of view. As mentioned above, it

also

96 Illustration

possible to use a gridded background, for example a sheet of graph paper,

which will make reproduction to scale easy. The result is, however, hardly

photogenic.

Most camera lenses cannot be focussed closer than c. 0.5m and at that

distance it is often possible to include several sherds on the one exposure.

This can be useful, for example when making record shots or black and white

prints which can be cropped for publication. For slides, however, it is best to

think carefully about the inclusion of more than one sherd or, at least, of

more than one type of vessel, in a shot. You may spend valuable lecturing

time explaining that the audience should ignore particular sherds and even

then it is quite often the irrelevant points which are picked up for later

discussion of your presentation. You will find the photographing of frag-

mentary vessels very frustrating. On the one hand they cannot just be laid flat

on the backing and photographed straight on whilst on the other they cannot

easily be photographed so as to show their form. Rims and body sherds are

difficult

than those with bases, which can be photographed upright. It is

sometimes possible to support rims by arranging blocks of wood under a

sheet until the rim is horizontal.

Complete vessels should ideally be photographed so as to show in one shot

the general form of the vessel, the handle and the spout. This will normally

mean that the vessel is positioned so that the rim is just visible as an ellipse.

The handle will be at 45° towards the camera and one will be able to see the

inside of the spout. On vessels where the spout is decorated, or where the

decorative scheme is based around the spout, this arrangement will have to be

reversed and if both the spout and the handle are significantly decorated you

may need two shots to record the vessel. Vessels decorated with figurative

scenes or other complex designs may need a series of shots to record the

design or it may be that you can use a combination of photography and line

drawing to portray the vessel. An ideal method of recording such vessels

would be using a video camera. The pot could be slowly rotated on a

turntable so that every aspect of the vessel is recorded. The main dis-

advantage of video recording at present is the difficulty in obtaining a hard

copy of the image and its poor quality when compared with still photo-

graphy.

Photographing groups of vessels produces another set of problems, similar

in many ways to those encountered by photographers at weddings. Tall

vessels should be positioned at the back and you may have to build up a stage

so that these vessels are not completely masked by those in front. With a

sizable assemblage you will find it difficult to get both the front and back rows

of vessels in focus at the same time and the positioning of vessels is so critical

that you need two people to take the photograph, one of whom is looking

down the view-finder calling out instructions to the other, who will be making

the final adjustments to the layout. The same strictures over the positioning