Uppdaterad: 12 feb 2019
Among colleagues especially interested in the treatment and use of temper in ancient pottery, the term and idea of sieving the temper material appears from time to time. This is primarily relevant for materials that are crushed or in other ways fragmented before they are used as temper and may be even for sand. The basis for discussing whether sieving is a part of the pottery technology of past potters is of course that sieving is documented in a number of present-day pottery craft traditions (f.x. Rye 1981, 37). Most often the sieving of the dried, crushed raw clay is just as, or more common. This is done in order to remove stones that are too large and the majority of the roots and other organic material to achieve a clay as homogeneous and clean as possible.
When it comes to sieving crushed temper (or sand), the objective is to get an even better control over the grainsize-distribution of the temper than what can be achieved by just crushing. When crushing a rock (as the most obvious example) you will get grains of a range of sizes. The range will be dependent on which type of rock you have chosen. The only size you may control by crushing is the maximum size and to some extent the relationship between the larger grain sizes as finer crushing will reduce this difference. Sieving a crushed material, you divide it into two portions – one larger than and one smaller than the mesh width. In theory at least. If the mesh width is 1 mm there will be a small portion of grains with one dimension < 1mm but oblong grains with one dimension > 1 mm that will pass through the sieve. That is of little or no consequence for the result from the potters view but for the analyst wanting to establish that sieving took place, it is important to know.
Back to the potter that has just sieved his temper and achieved two fractions. One fraction has a fairly accurate maximum grain size of 1 mm but is otherwise spread across a number of different sizes smaller than that. The other fraction has a lower limit of 1 mm and an upper limit determined by the initial crushing of the rock. While maximum grain size of the temper (together with amount of temper) is important for the workability of the tempered clay and for the forming methods used by the potter, it is hard to see any practical meaning of controlling the minimum grain size. The most logical is therefore that the potter applies sieving to determine a reasonably exact maximum grain size for the temper and thus uses the temper grains that went through the sieve.
So far so good. The next question is whether we may establish the use of a sieve by analysing a ware made in the past? If the sieve was used as suggested above – to better the control of the maximum grain size – this activity could only be argued if analyses of several different ceramic objects from the same production come up with the same maximum grain size. Even then we may only say that it is likely because even a meticulous crushing could produce the same result.
If – on the other hand – we hypothesize that the upper fraction after sieving was used that would mean that we should record an absence of the smallest pieces of rock fragments (below a fixed size) in the temper. Once again, this absence should be clearly shown and recorded in several samples before the use of sieving could be argued.
Ref. Rye, O. S. 1981. Pottery Technology. Principles and Reconstruction. Manuals on archaeology 4.