Die Life Pubblico Deposited
- From mhodder@worldpath.net Sun Sep 10 05:04:33 2000
Return-Path: <mhodder@worldpath.net>
Received: (qmail 17755 invoked from network); 10 Sep 2000 12:04:33 -0000
Received: from unknown (10.1.10.142) by m4.onelist.org with QMQP; 10 Sep 2000 12:04:33 -0000
Received: from unknown (HELO unix.worldpath.net) (206.152.180.10) by mta3 with SMTP; 10 Sep 2000 12:04:33 -0000
Received: from hxdfb (pm41-112-161.worldpath.net [209.187.112.161]) by unix.worldpath.net (8.9.3/8.9.3(WPI)) with ESMTP id IAA02200 for <colonial-coins@egroups.com>; Sun, 10 Sep 2000 08:04:31 -0400 (EDT)
Message-ID: <200009100805500304.003E1F92@worldpath.net>
X-Mailer: Calypso Version 3.10.01.00 (1)
Date: Sun, 10 Sep 2000 08:05:50 -0400
To: colonial-coins@egroups.com
Subject: Die Life
Content-Type: text/plain; charset="us-ascii"
From: "Mike Hodder" <mhodder@worldpath.net>
To all:
The question of how many coins could an average die pair strike
has been wrestled with in other fields besides colonial American.
The subject has been a particularly thorny one for students of
Roman republican coins. It might be instructive to look at the
literature and see what others have come up with. An easy place
to start is in AJN 9 (1997), the review article by T.V. Buttrey
pp. 113-135. The notes will lead elsewhere. Buttrey's conclusions
are, IMHO, right on. An average of coins struck per die can be
created easily for any coinage. Large sample sizes (35+ per die)
lend an insubstantial credibility to the average generated. All
such averages are intrinsically fictive because our data cannot
control all the variables involved in striking coins, especially
those made on a screw press. At best, the exercise is academic
and as long as it is understood that the averages are
guesstimates and not data, then no harm done. However, if the
averages generated are applied as data for studies on mint
histories, circulation patterns, and economic impacts of the
coinages, then those studies will have been built on unsound
foundations.
A similar caution should be applied to studies that attempt to
discover a rate of survivability per variety based upon current
estimates of the surviving population. The variables involved in
what allowed one part of a mintage to survive and not another are
not always controllable. We just don't know in every case why
there are so many of one variety and not of another. To assume
that what we see today in our populations is a random and
representative sample of the original whole is unsupportable by
any evidence we have. . The very fact that most of the coins in
our databases come from collectors' holdings shows that somewhere
down the line those coins were subject to some selection process
whose criteria have nothing to do with randomness.
When we start working up average coins struck/die or survival
rates we're in the field of theoretical numismatics. Unlike the
other theoretical sciences we don't have the tools with which to
test these numismatic theories against objectively established
standards. Conclusions drawn, therefore, must remain in the realm
of theory and not be applied to the real world as facts.
Regards
Mike Hodder - 2000-09-10
- 1
