TMG-L Archives

Archiver > TMG > 2000-01 > 0947620733


From: "Mills" <>
Subject: Re: TMG-L: Citations and Sureties in Shared Datasets (methodology?)
Date: Tue, 11 Jan 2000 13:58:53 -0600


Robin, you directed your inquiry to me, but frankly I'm not the best person
to answer it for you. I could isolate particular comments and respond, but I
can't offer you the insight you seek for setting up a genealogical model
that effectively uses *sureties* for data sets to flow back and forth
between researchers.

I have tried not to comment in the past on sureties, for basically two
reasons:

(1) Sureties are a function of certain computer programs--none of which use
it exactly the same, to my knowledge--and Evidence! is not directed toward
any particular program. Hence, advice on the use of sureties in TMG should
come from TMG users, not me.

But . . . and I realize this could create a firestorm of dissension from
those who think I'm totally wrong and, given that I have a 650-page book
manuscript that has to be finished in 2 weeks, it's a *mighty* bad time for
me to fan a firestorm <g>, but I'll say my say and then leave the debate to
y'all . . . .

(2) I am simply not convinced that the use of sureties is a positive good
for genealogy. For all the reasons you discuss in your message -- and many
others -- my opinion of sureties is this: At best, they are arbitrary and
volatile. At worst, they mislead more people than they help.

Most arguments for sureties that I've seen on this list from highly skilled
genealogists convey the sense that sureties are beneficial to them *on a
personal and individual level.* In carefully contemplating the surety issue,
they have developed their own concept of where logical stops might be put
along that slippery scale from very reliable to highly questionable.
Therefore, when they personally consult a source, they can assign a number
that is meaningful *to them* -- a number to serve as a flag to remind
*themselves* that they need (or maybe don't need) to find a better source.

However, I'm speaking here of genealogy's elite (which includes you, of
course, as one of the cornerstones of this user list). My concern is for the
use of sureties by -- and the impact of sureties upon -- the "great
unwashed" who are not experienced in evidence analysis and really don't want
to get involved with "the trappings of scholarship." This group, after all,
does constitute the greatest number of people who say they are "doing
genealogy."

You are right, the Internet is changing the universe of genealogy. Major
sites now get a million or more hits a day/week/month. Meanwhile, major
annual conferences consider themselves lucky to draw 2,000 -- major
institutes, 200. Among scholarly journals in the U.S., only two (NGSQ and
the Register) can boast a circulation as high as 18,000 while TAG hovers
near 2,000 and TG at 500 -- yet Betterway's new genealogy magazine to be
marketed wherever the curious public buys magazines was launched with a
print run of 100,000.

Given these dynamics, how much trust can be put in surety values assigned by
the masses who never study manuals on evidence analysis or attend
conferences and institutes to be educated in research methodology and record
interpretation -- but, nonetheless, are exceedingly active in exchanging
data, publishing their databases on their websites, and copying data
trustingly from the databases of anybody and everybody?

The quality genealogists on this list exercise considerable discretion, I am
sure, in the people they choose to work with cooperatively on a family
project. In such cases, they can and do work out, among themselves,
guidelines for assigning those sureties.

But establishing surety as an essential component of a universal
"genealogical data model" creates a Pandora's box, I fear. Any such model is
going to be the "standard" by which computer programs are designed. Computer
programs are virtually the first thing bought by those masses who come into
genealogy knowing zilch about the resources they'll use. Because software
ABC has a field in which a surety number is supposed to be typed, they'll
feel they have to enter something. Because they lack the experience to
properly evaluate sources, their decisions will often be unsound. But they
will still publish their data on the web for, literally, a world of other
naive researchers, who will see a surety value of X and assume they can
trust it because X means "the best."

Genealogy is a wonderful mix of yen and yang. Of humans who are both
left-brained and right-brained. Of people who approach research as an art
and those who see it as a science. Individuals who view *historical*
research as an ephemeral world of interpretation and others who would like
to see everything--even history--quantified in a nice orderly fashion.
Despite my admiration for the mental acuity of the latter, my own background
in history just doesn't convince me that we can build data models with
numerical values that are reliable when applied to and used by the masses.

None of the above, of course, addresses your more specific question about
the use of the "royal we." In part, you wrote:

<We need to be careful about the choice of singular for "you" and the royal
"we" in the context of _seeing_ something before citing it. If I am
maintaining a data set that is shared by a group of researchers, then I have
the problem with the model where the surety and the citation depends on who
enters that evidence with sureties and who publishes or amends that record
or its sureties.>
<There is no problem here if the "you" and "we" can be agreed by the
researchers and any journal editor as if they are acting as if they are one
person. If this group-personage concept cannot be accepted by the publishing
fraternity then a record entered by one researcher must have its surety
?down-graded? and citation modified when that record is transferred to
another equal researcher.>

I agree with you on all points -- as I am interpreting them <g>. While
editors do use the "royal we," that usage does *not* mean that the
publishing fraternity agrees on any concept. (Incidentally, when I use the
"we" in the context of the NGS Quarterly, I am not invoking the "royal we."
I refer to we the editor, co-editor, review editor, and editorial board.) In
fact, in the publishing world I doubt that one could ever truly say that
researcher/writers and editors at any publication "act as if they are one
person." Almost every publication -- scholarly or popular -- carries a
policy statement to the effect that the opinions of the writers are not
necessarily those of the editor or the publisher. This is as it should be in
a free world that guarantees freedom of thought.

So, this leaves us with your statement "If this group-personage concept
cannot be accepted by the publishing fraternity, then a record entered by
one researcher must have its surety ?down-graded? and citation modified when
transferred to another equal researcher" . . . ."

You are correct -- and, in my view, this statement argues my own case that
sureties are ambiguous, volatile, and even detrimental to sound genealogical
research. Hence, I cannot see that sureties should be a part of a source
citation at all.

A citation, in my view, is a straight-forward, detailed identification and
description of the source. As part of a *reference note,* that citation may
be (and often should be) accompanied by a commentary that reflects the
researcher's evaluation of that source--a *verbal* assessment of that
source. But I just don't see how the absolutely infinite variations in the
nature and reliability of sources can be adequately, comprehensibly,
meaningfully, and reliably expressed by including in our citations a digit
from 0 to 5!

Elizabeth Shown Mills, CG, CGL, FASG


- ---------------------------------------------------------------------------
TMG-L The Internet Mailing List for The Master Genealogist
To unsubscribe: Send an e-mail to with 'UNSUBSCRIBE TMG-L'
or... if you get the digest version: 'UNSUBSCRIBE TMG-L-DIGEST'
- ---------------------------------------------------------------------------

This thread: