GENMTD-L Archives

Archiver > GENMTD > 1998-03 > 0889044535


From: Richard Pence <>
Subject: CLASS: BARD and POE - When Do You Know You Have "Enough Proof"?
Date: Wed, 04 Mar 1998 12:48:55 -0800


Understanding Sources, Citations, Documentation and "Proof"
An On-Line Genealogy Class in soc.genealogy.methods

Class Leaders: Richard Pence, Cheryl Singhal and Lesley Robertson
Editor: Margaret Olson

Lesson Four: POE (Preponderance of Evidence) and BARD (Beyond a Reasonable
Doubt): When Do You Know You Have "Enough Proof"?
[Original module written by Richard Pence and reviewed and augmented by
Lesley Robertson and Cheryl Singhal.]

In our previous lesson, we discussed the strengths and shortcomings of
various types of evidence. Among other things, it was noted that primary
records do not always exist for the events you are researching - and even if
they do, these records can not always be relied on for complete accuracy or
the full story.

How, then, do you know when you have "proof" of a specific "fact" about an
individual?

Again, countless books have been written on this topic and there are, as you
might guess, varying opinions. Some even claim that in genealogy you can
never "prove" anything. They say, for instance, that you can never be
positive that the child's father is the person named as the father on a
birth record. (It used to be said that the only thing you can be sure of is
who the mother was; unfortunately with today's laws even that can be in
doubt - but I will leave that problem to future generations!) In this sense,
then, "proof" is not the standard; the worth of the documentation you have
assembled is what you must rely upon. And, as previously noted, not all
"documentation" is equal. Therefore you must rely on an accumulation of
source materials to weave your web.

Often the worth of genealogical conclusions are compared with legal concepts
such as "beyond a reasonable doubt" (BARD) or "preponderance of evidence" (POE).

The former is generally applied in criminal cases, while the latter is a
test in civil cases.

Conviction in a criminal case requires that 12 people unanimously believe
that there is _no_ reasonable doubt that the accused committed the crime he
or she is charged with. While you may aspire to this level of certainty, it
is generally a level above what can usually be attained in genealogy.

On the other hand, "preponderance of evidence" (POE) in a civil case can
mean that evidence on the side of the plaintiff only slightly outweighs the
evidence on behalf of the defendant. A jury in a civil case can vote by a
majority of one in favor of one side or the other. This is a test that is
generally below that sought in genealogy. (Note that the Board for
Certification of Genealogists recently announced it is abandoning
"preponderance of evidence" as a concept in its certification process.)

My own view of genealogy, shared by many others and borrowed from some of
them, is that the level needed to substantiate a line lies somewhere between
BARD needed for criminal conviction and the POE that tips the scale only
slightly one way or another in a civil case. I have heard this level
described as "clear and convincing."

It can more properly be compared with the level of evidence needed to
convince a judge of legal heirship. In fact, while looking for some
definitions in preparation for this article, I ran across the following in a
newsletter for title attorneys in Pennsylvania. Citing state court
decisions, the newsletter noted:

A claimant seeking to show that he or she is a legal heir to an estate
would have to provide "a fair preponderance of credible, trustworthy
and satisfying evidence. The proof should be clear, precise and definite."

The evidence most often presented to the Department [of Revenue] is
a family tree connecting the decedent to the petitioner, along with
birth certificates, death certificates and marriage certificates
that establish the connection.

The courts have also determined that when a claimant to an estate comes
forward, not only must they prove that they are an heir, but that no
other heirs are available to inherit.

Leaving aside the logical impossibility of "proving a negative" outlined in
the preceding paragraph (see later), there are a couple of things I like
about the above definition. Instead of the simple POE, where the scales can
balance only slightly one way or the other, there must be a "fair"
preponderance of _credible, trustworthy and satisfying_ evidence. Further,
the proof must be _clear, precise and definite._

The first set of qualifiers - "credible, trustworthy and satisfying" - are
the key.

How "credible" is, for example, a county history? Are Uncle Carl's notes
"reliable"? Is the information in someone else's compilation "satisfying" in
terms of being free of ambiguity and doubt?

The second set of qualifiers - "clear, precise and definitive" - puts a
still more demanding (and perhaps too demanding) level of "proof" upon the
genealogist. That is, the evidence must be "clear and convincing" - it must
be absent any unresolved conflicts.

Notwithstanding the above-cited court decision requiring a demonstration
that there are no other heirs, this concept defies the dictum of "the
scientific method" that says negatives cannot be proved. While it may be
possible to prove that other persons seeking to show heirship are not, in
fact, heirs, this does not eliminate the possibility that there are other
unknown (to both the claimants and the courts) persons who may well be.

Similarly, genealogists should be wary of approaching a problem in terms of a
process of elimination. Although a sometimes fruitful approach in terms of
focusing research efforts, the mere fact that persons A, B and C are
eliminated by factual circumstances from being the father of person E, it
does not necessarily follow that the remaining known contender, D, is E's
father. The most obvious reason for this is that there may be as yet
undiscovered contenders. And, of course, there may be unknown facts for
Person D that also eliminate him from contention.

In using concepts such as POE or BARD, you should establish a working
hypothesis and then gather as much evidence as possible to test this
hypothesis. Often as you collect the "facts" you will be "clearly convinced"
one way or the other.

Note that this concept does not require that every doubt be banished, as in
BARD. But it does require one essential thing not needed under the every-day
definition of POE: Your hypothesis cannot be considered sustainable if there
are unresolved or unexplained contradictory evidence.

What does "unresolved" or "unexplained" mean in this context?

Most importantly, keep in mind that the primary purpose in building your
family history is to establish _links_ from one generation to the next.
While it is always satisfying to "fill in all the blanks" on a family group
sheet or a demanding computer screen fully and completely, that's not the
purpose of the exercise. Thus, if there are conflicting dates for a great
grandfather in various sources, that is not an "unresolved" or "unexplained"
conflict; such discrepancies are almost the rule rather than the norm. A
conflict in dates is irrelevant unless they create an anachronism.

But a challenge to the thesis, such as a document stating that your quarry
is the son of A rather than B, as stated in your hypothesis, demands that
there be a plausible, rational - "explainable" - reason for this conflict.
If there isn't, then the hypothesis must either be reformulated or abandoned
in favor of a new one.

A "Clear and Convincing" Exercise:
William and John Griffin of Highhope County, Ohio

I have put together an imaginary case study that you can use as an exercise
in proving or disproving a relationship based on accumulated evidence. See
if you think the hypothesis below is supported by the evidence given. I hope
you will share your analysis with the group so that others can have the
benefit of your analysis.

[I hope that I haven't provided unintentional contradictions in developing
the case study; this effort has convinced me that some puzzles may be easier
to solve than they are to create!]

OK, class, listen up! Since we can't take roll, we will have to base your
final grade on how well you do with this problem! Assignments are due by
Sunday! And no fair peeking at someone else's answers!

Working Hypothesis: John Griffin, born 1823 KY, and in Highhope
County by 1850, is the son of William Griffin, born 1799 KY,
and in Highhope County by 1840.

Here is what you have found:

These three families are listed in Broad Brush Twp. in the 1840
census of Highhope County:

-- William Griffin, head of household, one male 30-40, one male
15-20, one male 10-15, one male 5-10 and one under
5, one female 60-70, one female 20-30, one female under 5.

-- John Griffin, one male age 40-50, one male 10-15, two
males under 5; one female 40-50. one female 5-10.

-- George Griffin, head of household, one male age 20-30,
and one female, age 20-30.

The 1850 census of Highhope County shows these Griffin families, all in
Broad Brush Twp.:

-- Wm., 51, born KY; Clara, 40, born OH; Robert, 18, KY;
Samuel, 13, OH; and Polly, 12, OH.

-- John Sr., 53, born KY; Elizabeth, 49, born KY; John, age 22, KY;
George, 19, KY; and William, age 17, OH; Sarah Griffin, 79, KY.

-- John Jr., age 27, born KY; Sarah Ann, 23, born OH; Martha, 2, KY;
Margaret, 3 months, OH.

In 1860, there were only these two Griffin families in the county:

-- Clara, age 49, OH; Samuel, 23, OH; Alice, 19, OH.

-- John, 36, born KY; Sarah, 30, OH; Patsy, 12, OH; Margaret
10, OH; William, 5, OH; Henry, 3, OH; John, 2, OH; "Baby" (M),
2 months, OH.

Clara was in Broad Brush Twp.; John in Clark Twp.

Only the family of John Griffin can be found, again in Clark Twp., in
1870 and this listing contains no new information.

Highhope County was organized from Upper County in 1837; the
marriage records. as published in an index, reveal these Griffin
marriages:
-- Clara to Adam Matthews, 1862; no bondsman listed, by Rev. A.J. Hart
-- George to Betsy Meyers, 1839; bondsman William Griffin
-- John to Sally A. Brown, 1848; bondsman John Griffin, by Rev. A.J. Hart
-- John to Sarah Forman, 1854; bondsman John Griffin; by Rev. John Price
-- Mary to Charles A. Davenport, 1859; bondsman, Harris Davenport, by
Rev. A.J. Hart
-- Robert to Minnie Kurtz, 1854; bondsman John Griffin Jr., by Rev. A.J. Hart
-- Samuel to Alice Long, 1859; bondsman John Griffin, by Rev. A.J. Hart

William Griffin wrote his will in July of 1858 and it was proved at the
October term of court that year in Highhope County. In it he names "my
now wife, Clara" and "my son Samuel and my daughter Polly," who
are to receive his farm in Broad Brush Twp. "when my wife is through
with it" and the same two are two divide "the land I bought from John
Griffin." This is the only estate record found under the name Griffin.

Pertinent deeds in Highhope County include: John Donovan to
William Griffin and John Donovan to John Griffin, both in Broad
Brush Twp., 1837; William Griffin to George Griffin to George
Griffin, "the land I purchased from Aaron Wilson," Broad Brush
Twp.; George Griffin "to my brother John," Broad Brust Twp.,
1849; John Griffin to William Griffin, 1856, Broad Brush Twp.
(same land as previous deed); Henry Forman to John Griffin,
1856, Clark Twp.

The 1881 history of Highhope County has bios of two of the above
individuals:

-- Mary Davenport, wife of A. C., Broad Brush Twp.; a native of
the county, daughter of William and Clara Griffin; her father was a
native of Fertile County, KY, came here in 1833, mother this state.
Their children are listed as Louise and Alexander C., Jr. She has one
brother, Samuel, living in California. Baptist.
-- John Griffin, farmer of Clark Twp., native of KY, came to this
county with his family when about 10 years of age; parents not
named; his "only surviving brother," not named, lives "in the
west." Details of his wife and children are given, no new info.

Mary G. Davenport, in her family Bible, handwritten copy sent
you by a descendant, gives her father William Griffin as being born in
"January - 1799" and gives dates of birth and death for "Clara Mathews."
Their children are listed as Louise and Alex Davenport. Under deaths
there is an entry that reads "Robert Griffin died Sept. 6th, 1911, in
Mudslide, Calif., age 79"

Greenlawn Methodist Cemetery, Clark Twp., Tombstone Inscriptions:
-- John Griffin, "father," 1823-1901
-- Sarah J. Griffin, "mother," 1830-1919
-- "Infant son of J. and S.J. Griffin, 1860"

Baptist Cemetery, Broad Brush Twp., Tombstone Inscriptions:
-- Albert Griffin, 1839-1849
-- "Infant son of J. & S.A. Griffin, 1851"
-- "Infant daughter of J. & E., died 6 July 1842"
-- James Griffin, 1838-1849
-- Patsy Griffin, "mother", rest illegible
-- Sarah Griffin, 1771-1851
-- Sarah A. Griffin, 1827-1851
-- William Griffin 1799-1858

A 1985 letter you received from a Davenport descendant in Highhope
County quotes from a letter addressed to Louise Carpenter, dated 1900
in California, and signed only "Cousin Esther and Aunt Wilhelmina." The
letter mentions "your brother Alexander" and "your mother Polly."

[In doing this exercise, it may be helpful to place the above individuals
into conjectured family groups and to classify the various pieces of
evidence as "Pro Hypothesis," "Con" and "Neutral."]

Evaluation

Keeping in mind that the test is "a fair preponderance of evidence that is
credible, trustworthy and satisfying," let's see your response to these
questions:

1. Does the accumulated evidence support the hypothesis?
2. What are the main reasons for your conclusion?
3. Are there any "unresolved" conflicts in the evidence? What are they?
4. If you don't believe the hypothesis is "proved," is it flatly "disproved"
or is it "not proved but worth pursuing"?
5. Did you detect any clues for further research in the case study? List.

Or: Answer by writing a brief narrative describing the history of this
Griffin family.

Further Study

Two excellent books on genealogical evidence and citations are:

Noel C. Stevenson, _Genealogical Evidence_ (Aegean Park Press, 1989)

and

Donald Lines Jacobus, _Genealogy as Pastime and Profession_ (Baltimore:
Genealogical Publishing Co. 2nd ed., rev., 1968, repr. 1996. Softcover. 120
pp.). [This is one of the first books I read on the topic of genealogy and
it had more than a little influence on my approach to the "pastime."]

Both books are available through most genealogy book stores and can often be
found in general book stores, such as Barnes and Noble.

One that I haven't read - and which may present differing views on the topic
of this lesson, is:

Christine Rose, _What IS the Preponderance of the Evidence?_ (Rose Family
Association. 1996. softcover. 15pp.)

------------------------------------------------------------------
This ends the class, except for responses to your questions and comments.
Thanks for participating!
-- Richard Pence, Lesley Robinson, Cheryl Singhal
And our thanks to Alicia for ramrodding and Margaret for her upcoming
summary.

Richard A. Pence, 3211 Adams Ct, Fairfax, VA 22030-1900

Pence Family History Home Page:
http://www.pipeline.com/~richardpence/index.ht

This thread: