Pottery for Beginners

Basic Pottery Groups

If you have found a piece (or sherd) of pottery and want to know more about its age and the people who made it there are a few simple steps you can take to narrow it down. Consult the glossary if the terms are unclear.

  • First, how hard is the body? Pottery is either an earthenware, stoneware, or porcelain. Earthenwares are soft and have porous bodies, where stoneware and porcelain have hard non-porous bodies. If there is any doubt you can either put water on the edge to see if it is absorbed, or simply give it the tongue test. If it sticks to your tongue, it is an earthenware. If you can crumble a piece off of the edge, its probably an earthenware.
  • Second, it either has a glaze or it doesn't. Native American pottery in South Carolina is unglazed. A glaze is a separate body coat. It is a visibly different coating like paint, but fired to the vitrification point. Some glazes are soft enough to scratch with steel or rock, but not with your fingernail or a soft object. Paint or slip can be scratched with a soft object usually. This is sometimes seen on Native American pottery, but is relatively rare in South Carolina.

So if you have a soft, unglazed pottery sherd the next indicator is its surface treatment. Native American pottery is hand made, most often, by rolling out coils of clay and winding them up to form the body. They are then pressed together by hand, or “malleated” by applying a paddle. The most common surface treatments are plain, textile marked or stamped. Within each of these there are a number of variations, and sometimes additional marking such as incised lines, and impressed marks called “punctation” are seen. This is where it starts to get complicated, because the different variations are important indicators of age.

Even “plain” pottery comes in a variety of flavors ranging from barely smoothed to polished to a high gloss called burnishing. Plain wares are common throughout, but highly burnished wares are more common during the late prehistoric period and after European contact (about 900AD through the 1730s).

Incising and punctate marks are often seen on otherwise plain surfaces (image). These decorations are common on early period (3000-500BC) and late period wares (500-1520), but not so often during the middle period (500BC to 500AD).

Textile marked wares are usually marked with textile wrapped paddles that leave distinct individual string marks (cord marked), with weaved cloth (fabric impressed) or net. Net differs from fabric in that instead of being tightly weaved, the individual threads are tied together at each intersection. Sometimes the knots are clearly visible. These surface treatments are most common in the middle period.

Cord marked wares vary internally. Although it is unusual in South Carolina, cord can be impressed in recognizable patterns using one or two strands to make geometric shapes. This is more common in the far north (see Holmes 1903). More often here a paddle is wrapped with cordage (the threads wound together to form string) and used to smooth the surfaces and press together the coils that make up the body. The size of the cords, tightness of the winding, direction of the twist (ie, left to right or right to left), spacing of the cords on the paddle, and the way they were applied to the body are all significant.

Fabric marking is usually seen in one of three variations. Fabric is wrapped around a flat paddle and used to smooth the surface, as with cordage; weaved using hard, non-fabric material, such as sticks or river cane for one of the weave elements (called weft and warp); or impressed directly into the surface so that weave of the fabric is clearly visible. With the first it simply appears that the surface was more or less randomly blotted with cloth, and with the second the rigid elements leave distinct linear impressions. The third form is less common and is usually in the form of a rough fabric that often leaves distinct marks that sometimes have overlapping elements. This is usually seen on late period Pee Dee pottery.

The paddles used to malleate the coils are often carved to form stamped impressions. The earliest of these are either simple, linear marks that may not even be actively carved- a rough piece of split wood could form some of these marks in fact. Carved intersecting lines forming square checkered impressions came into use during the early part of the middle period, by about 500BC. More complicated designs came in later, and elaborately stamped wares are the hallmark of the late period.

With the surface treatment narrowed down you can move to the next step. Although some surface treatments can be fairly specifically dated others, such as cord marking for instance, were used for thousands of years and we need to look closely to find the clues that tell us their age.

For this we need to look at the makeup of the body fabric or “paste.” Clay molecules are tiny and flat, where silt and sand grains are larger and rounded. Clay forms when the parent rock weathers and is washed away by erosion. It becomes suspended in flowing water, but when the flow stops all of the sediment in the water drops out, with the heavier elements falling first. The lighter, flatter clay particles float a little longer than the sand. Perfectly pure clay is hard to find and doesn't usually hold up very well because of shrinkage. Naturally sandy clays are more common, and potters often “temper” the clay to make it more stable and workable by adding sand of varying sizes, crushed rock and shell, crushed sherds or prefired clay, or fiber.

Fiber tempering is found on the very earliest pottery - Stallings Ware - which dates between about 3000 and 1000BC. Fiber tracks are visible on the surface, and the holes it left behind when the pot was fired are visible in the cross section. This pottery is usually plain, but sometimes it is punctated with solid or hollow sticks or shells or incised with a sharp object.

Lithic tempered wares include natural sand and crushed rock. For geologists sand comes in a range of size grades referred to as very fine, fine, medium, coarse, very coarse, granular and pebble following what is known as the Wentworth Scale. The very fine to medium sizes are tiny grains less than .5mm across. This may in fact be a natural part of the clay and pottery made with inclusions this size is often called “temperless.” Sand tempered wares were made in all periods and there is no single trend that can be pointed to that is consistently diagnostic of a given type and time.

Larger inclusions can be natural, but in general, because of the way clay deposits are formed in sedimentary environments the larger pieces fall out of the solution first and form separate layers from the finer pieces. Occasional large inclusions may be unintentional but usually if there is a uniform mix of coarse to very coarse or granular sand it is intentionally added. That said, it seems that pottery of all time periods in some areas such as the Midlands and Fall Line in general, have a high density of large sand and other inclusions in their bodies. As a broad generalization early wares tend to be temperless or to include medium to coarse sand; earlier middle period wares tend to have coarse to granular inclusions; and later wares tend to include more fine to medium sand. There is much variation however.

Crushed rock temper usually includes quartz, felspar or limestone/marl. This is fairly uncommon on pottery in South Carolina, so it can help to date ceramics. Only three defined types with crushed rock temper are found: Yadkin,Hamps Landing, and Wando. Yadkin is tempered with quartz and felspar, while the others are tempered with crushed limestone or marl. In the case of Hamps Landing and Wando the limestone that was used has often dissolved and leached out because of the acidity of the soil and ground water, leaving angular voids.

This is also the case with shell tempered wares, though the voids tend to be flat and sometimes show the layers of the shell's body. Shell tempered wares are rare in South Carolina and tend to be very late, even historic period ceramics. Yadkin is thought to be a middle period ware, dating between about 500BC and 500AD, though few reliable carbon dates have been generated. Hamps Landing has been dated to around 1000BC, though it too has only been found in a few datable contexts so far. This type has only been found in Horry County. Wando seems to be limited to the Charleston area and has been dated to the latter part of the middle period, between 600 and 1100AD.

Prefired clay or “grog” can sometimes take the form of chunky crushed sherds whose surfaces are visible, or can appear to be fine bits of clay. With the sherd tempered wares the chunks often protrude form the surface, or are only covered with a skin of clay so the surface appears lumpy. Three defined types have this tempering, though the differences between two, Hanover and Wilmington, are hard to pin down. Both are sherd tempered and are common along the coast from North Carolina to north Florida. People working south of Charleston tend to call it all Wilmington and those to the north call it Hanover (see distribution map). Dates vary widely, ranging from about 100AD into the 1500s. Surfaces tend to be fabric impressed or cord marked. St Catherinesis tempered with fine bits of crushed clay and tends to have a hard, compact body and to be marked with distinct, tightly twisted cordage. This is a late period ware that dates between about 800 and 1300AD. It is found mostly on the south coast and in Georgia.

In the lab an archaeologist might go into far more detail and look at a wider variety of variables but if you can identify the surface treatment and paste inclusions (or the lack thereof) you can usually make a fairly safe estimate of the age of a piece of pottery. More detailed pottery type descriptions and images are found in the Period sections.