Herrington's original observation back in the 1950s was that clay
tobacco pipe stems appear to decrease in diameter over time. This
only works on pre-Revolutionary War English dominated sites and does
not work in the Hudson Valley where Dutch pipes were still the norm
(John McCashion, New York State Historical Society, Personal
Communication.
Harrington, J. C.
1951 Tobacco Pipes From Jamestown. Quarterly Bulletin of the
Archaeological Society of Virginia 5(4):1-8.
1 953 Dating Trade Pipes - Some "Cans" and "Can'ts". Eastern States
Archaeological Federation Bulletin 12:8.
1954 Dating Stem Fragments of Seventeenth and Eighteenth Century Clay
Tobacco Pipes. Quarterly Bulletin of the Archaeological Society of
Virginia 9(1):10-13.
1955 A Comment on "A Critique and Rebuttal of the paper - "Dating
Stem Fragments of 17-18 Century Clay Tobacco Pipes" by John F.
Chalkley. Quarterly Bulletin of the Archaeological Society of Virginia
9(4). (2 pages)
1990 Dating Stem Fragments Of Seventeenth And Eighteenth Century Clay
Tobacco Pipes. Archaeological Society of Virginia Quarterly Bulletin
45(3):123-128
At that time, drill bits in U.S. Customary measurement were widely
used because they were both widely available and inexpensive and
offered an easy way to for various researchers to compare information.
Calipers back then, only came with vernier scales then and an amazing
number of folks had difficulty reading them.
In addition to the limited Geographic areas where clay pipe stem
dating is valid, there are other prole ms with assigning mathematical
values to the stem bore.
Tools could be used for generations in a single, long-lived &
muti-generational pipe making shop. This would include the wires for
piercing the holes down the length of stem. I do not think that every
manufacturer in all of the English pipe making cities changed their
stem piercing wires at the same time.
Back when I was a graduate student, I conducted an experiment where I
took five long stemmed Goedewaagen pipes and broke the stems into
about 2 inch lengths (5 pieces plus shank still attached to bowl for
each - 11 measurements for each pipe). While most of the interior
bore diameters were identical, I did get three that were larger than
the average. I attribute this to the effect of the displaced clay as
it is pushed down the bore by the wire.
Binford also says that his formula assumes a constant rate of
accumulation through the period of sample building. A mighty tough
thing to prove archaeologically.
Just some pipe thoughts on a pleasant fall afternoon.
:-);
Smoke
On 10/29/07, Alasdair Brooks <[log in to unmask]> wrote:
> The lack of replies - other than Silas Hurry's helpful e-mail- over whether anyone's converted Harrington and Binford clay pipe approaches to metric measurements, leads me to strongly suspect that the answer is 'no, not really'. While Silas has very kindly agreed to get a copy of the relevant original GWS transcript to me when he has a spare moment, that still won't basically solve the problem that metric drill bits increase in different increments than their Imperial cousins. A straight metric conversion only becomes possible if we can convert 64th" / .4 or 128th" /.2mmincrements to formulae that permit measurement in .5mm.
>
>
> On that basis, I've put together a very preliminary rough draft metric Harrington pipe stem dating table. I stress that I have -not- tested it. Please considerthis a hypothesis only.
>
>
>
> The basic principle at work is the observation that the ratio between Imperial (or 'English' if Americans prefer) drill bit increments and metric drill bit increments is essentially 4:5: where Imperial drill bits increase by 0.016" / .4mm per increment, a .5mm increment increase in a metric drill bit equates to an increase of 0.020".
>
>
>
> The different period increases on the Harrington table are 30 years, 30 years, 40 years, 30 years, and 50 years. Assuming a direct ratio translation, this gives us 37.5 years, 37.5 years, 50 years, 37.5 years, and 62.5 years on their metric equivalents. Taking the casual liberty to round up 37.5 to 40, and round down 62.5 to 60, thispotentially gives us:
>
>
>
> 1590 - 1630 = 3.5mm
>
> 1630 - 1670 = 3mm
>
> 1670 - 1720 = 2.5mm
>
> 1720 - 1760 = 2mm
>
> 1760 - 1820 = 1.5mm
>
> This is a bit crude, and won't be as 'accurate' as Harrington's figures(or the St. Mary's City GWS figures) because there are fewer increments, but it strikes me as at least adecent starting point for testing and to use as a preliminary rough training model in countries which use metric. I have a small late 17th- to early 18th-century assemblage in front of me from Norwich (UK) that I'll use as a preliminary test in the coming weeks. It looks like I can source 64th of an inch drill bits as well (there just weren't any in the office), so I should be able to generate 'control' figures based on the original formula to test against the new metric
>
>
> On the basis of the above 4:5 core ratio, it's hypotheticallypossible to convert the Binford dating formula by replacing the latter's'38.26' with '47.825' (38.26 / 4 = 9.565and 47.825 / 5 = 9.565). I'll try and test this with the Norwich pipes as well since I also have the control of being able to date the latter's dates from contextual and other data.
>
>
>
>
> I thought I'd try and open this up to discussion and ask if anyone mighthave anythoughts on the above? Particularly, am I mad, or is this potentially useful and worth continuing with?(not that the two are necessarily mutually exclusive)
> Alasdair
>
--
Smoke Pfeiffer
Remember: When seconds count, the police are only minutes away!
|
