Artifacts as time markers
Pipe stem dating
The clay pipe industry expanded rapidly as tobacco smoking gained popularity in both England and America. Historical archeologists excavating English colonial sites often find pieces of white clay smoking pipes on their sites. In the 1950s J. C. Harrington studied the thousands of pipe stems excavated at Jamestown and other colonial Virginia sites, noticing a definite relationship between the diameter of the pipe stem bore—or hole—and the age of the pipe of which it had been part. The earliest pipes, dating to about 1600, had stems with 9/64-inch diameter bores. By 1800 this diameter had decreased to 4/64 of an inch. This change in diameter may have occurred because pipe stems became longer through time, requiring a smaller bore. Louis Binford later devised a mathematical formula to refine Harrington's method (Deetz 1996:27). This dating technique only applies to pipe stems manufactured in England between approximately 1590 and 1800.
Historical archeologists do not rely on pipe stem fragments as the only source for determining a site's history. They also consider historical documents and other material culture recovered from the site—such as ceramics, glass, metal artifacts, faunal and botanical samples, and features—to determine its occupation and use.
Parts of clay smoking pipes. (Heather Hembrey, University of Maryland)
Try it yourself
Pipe stem dating
You have recovered sixty-three pipe stem fragments from Verysignificantsite.
You wish to analyze these pipe stems to determine when your
site was most heavily occupied. According to J. C. Harrington's
initial studies, the time periods and average bore diameters
are as follows (Deetz 1996:28):
You have measured the pipe stem bores from your site and find the following distribution:
|Number of stems||Bore diameter (in 64ths of an inch)|
Look at the number of stems you recovered and determine which bore diameter is represented by the most pipe stems. Match the most frequently occurring diameter with Harrington's bore diameter chart above.
Based on Harrington's table, when does this distribution suggest that Verysignificantsite was mosth eavily occupied?