Fresh from the Wash

At nearly every site in use by the British in the late 16th century or after, from here to Ireland to the east coast of the United States, there is one artefact type that is often found in quantity: clay pipe fragments. The material is often English white clay, usually still white or beige after recovery and a quick wash, but local production material and use could lead to variation. They are almost never found intact, but rather as separate stems or bowls. Both fragments however can help us get a rough date on the artefact and perhaps even a range for the surrounding assemblage.

Just as an aside, (you know we love digressions on this blog), the prevalence of pipe fragments on colonial sites in Virginia has actually confirmed historical documents and proved wrong architectural conservators’ assertions in a heated series of arguments witnessed by one of our staff members. A certain structure was being surveyed by an interdisciplinary team of experts to prepare for excavation, conservation, and restoration at living history museum. The archaeologists and historians posited the structure had a porch during the mid-18th century, based on trial trenches for the former and primary sources for the latter. The primary sources referred to one of the more colorful residents sitting on the porch smoking and shouting hellos to other townsfolk as they passed by. The architects swore up and down that the building phases all went against the stylistic conventions for porches for the period. Finally, a trench that spanned the front façade of the building was opened, and, when it reached the expected depth of the 18th-century street level, the perfect outline of a porch and its supports was visible almost entirely due to ash deposits and pipe fragments.

Dating

If you are lucky enough to find a fragment with a maker’s mark, that could be a first major clue to production site and time range. On other sites, members of our staff have been able to literally in the field search local pipe maker mark databases and get a date based on the style of the mark. (This database from a governmental heritage organisation based in Canada was the key to unlocking a particularly plain maker’s mark from Bristol in 1668 at a dig along the Welsh border.)

There is also a noticeable evolution of the bowls from the beginning of English production in the 1580s through the 19th century. Adrian Oswald mapped the changes in bowls to create a seriation of stylistic and functional elements from rims of the bowl to the heels and spurs for setting the pipe down upright.

In the mid-20th century, a technique for dating the pipe stems was devised by JC Harrington who looked at a huge collection of colonial and early republic stems from excavations at Jamestown and Williamsburg in Virginia, USA. He measured the diameter of the bore-hole using standard imperial drill bits with increments of 64ths of an inch and found a relationship between size and date of production. In the decades since, multiple archaeologists and mathematicians teamed up to develop increasingly more precise formulae. The graph is giving us statistical information in that there is variation in bore-hole size for each of the date range, where the longer bar is the higher number at a specific diameter which is the mode of the sample (the increment with the highest prevalence in the entire set). The best use of this technique would involve a large sample size in a secure context, and even its inventor noted the inadequacies of his model. In general, however, the earlier stems, produced just before the turn of the 16th century, have wider bores. Over time, the bores are narrower perhaps as a function of the longer stems that came into fashion.

A bit clearer graph designed by McMillan after Harrington.
McMillan, L.K. (2016). An Evaluation of Tobacco Pipe Stem Dating Formulas. Northeast Historical Archaeology, 45(45), Article 3. https://orb.binghamton.edu/neha/vol45/iss1/3

Both bowl and stem methods of dating should be used if the assemblage of pipe fragments allows, and these results should be cross-referenced with seriation models for pottery, architecture, and other dateable finds and features.

Unfortunately, the two pipe stem fragments we found so far this summer were in the topsoil and thus unstratified. They were clearly in a very disturbed layer and cannot give us precise chronological information, and one would never base an entire thesis on the presence of one or two pipe stem data points, but they are useful for teaching. We broke out the old drill bits and determined that both are most likely from 1720-1750!

For more information, we highly recommend this primer from the National Pipe Archive.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s