...what are yours?

Genealogy demands proof, family history not so much.

With the scarcity of records in the Colonial VA and NC areas prior to
1750, it is rare for a family to have recorded proof of lineage so
what do you do?

Do you stop where the records stop or do you create an unarguable
scenario that can neither be proved nor disproved?

Unarguable meaning the only RECORDED families in the area and the
timing and proximity is exactly what it should be for ancestor and
descendants.

On the other hand how do we know the brother took in his deceased
brother's family and records make it appear they were his children?

It might be interesting to see if we have a consensus of solutions -
not that there is right or wrong if that's how you want to do your
thing.

I know there are more than 4 people here - don't be shy.

Hugh

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

A good question.

There are a few approaches I can take based on past career. That career
has been half in science and half in IT. The science half has been in
two fields, one palaeoecology and the other forensic science, both
dealing with attempting to reconstruct the past based on whatever
fragmentary materials were available.

The forensic science half deals with two standards of proof: beyond
reasonable doubt for criminal cases and balance of probabilities in
civil cases. When I started out in family history I went on a course so naturally I asked the lecturer which of these applied. The question was
never answered. Obviously if one were having to give expert opinion in
a legal case the relevant standards would apply with the proviso that
it's up to the tribunal to make an overall judgement, the expert can do
no more than give what evidence is available and their own interpretation.

That leaves me with the general scientists approach for which the
default position should be "I don't know for sure". I can collect
evidence and make whatever interpretation best explains it all. That
becomes my current hypothesis. I have to accept that subsequent
evidence might contradict it in which case I have to examine the
evidence again, possibly reject anything that's misleading and come up
with a revised hypothesis. If further evidence agrees with the
hypothesis then I can regard it as being strengthened. In fact
scientific method demands that I should look for material which has the potential to contradict the hypothesis.

Here are a couple of examples which illustrate both aspects of this:

Years ago,before I became interested in family history I read a
published diary of a C18th local apothecary who, at one stage, was
secretary of a book club. A John Goddard was fined 6d for coming late -
a Goddard who was interested in books but not a good timekeeper - he
must be an ancestor. Yes, I know, we shouldn't work forwards but in
fact I worked back to him as a 5xggfather. Working on his family I had
a number of children who were born or baptised in different locations,
the first in one parish, locality unknown, the remainder born in two
locations in another. However hanging these children together as
members of the same family was a hypothesis which could be false. I
eventually found a will which had the potential to prove it false if it
were so but, in fact, all the expected children were listed there.

The second hypothesis was received family history about later
generations, namely that 2xggfather emigrated with the rest of the
family but ggfather, aged 14, refused to go. This one broke up
gradually under accumulating evidence. A half-sister didn't emigrate,
she married and lived locally. Two brothers emigrated to Australia
arriving about a month apart. The older half-brother emigrated to
Chicago and the last brother simply disappeared after the 1851 census.
The father didn't emigrate; he died here. I have his death certificate
and it may well be that the emigration story is an attempt to combine
facts into a fable to disguise the reality that he had committed
suicide. This hypothesis fell apart under testing to be replaced by a
couple of others, one being that the missing brother probably emigrated
(if anyone has a Joseph Goddard who emigrated from Yorkshire in 1851 or
later I'd be interested to hear) and the other being that the
half-brother offered to take ggfather with him as ggfather was living
with him in 1851.

The IT half of my career has led me to make little use of genealogical packages. ISTM that the lure of a recognised data structure, the tree,
has inveigled developers into using this as the basis of their data
store. As the family tree is a statement of an hypothesis, and one that
might have to be replaced, using it as the framework on which to store
the evidence requires too much prejudgement and might make changes of
mind needlessly difficult. I prefer a mixure of RDBMS and spreadsheets
as a means (not ideal but not enough to prod me into developing
something better) for organising data into timelines.

Ian

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Tue, 28 Apr 2020 13:00:50 +0100, Ian Goddard
<ianng@austonley.org.uk> wrote:

A good question.

There are a few approaches I can take based on past career. That career
has been half in science and half in IT. The science half has been in
two fields, one palaeoecology and the other forensic science, both
dealing with attempting to reconstruct the past based on whatever
fragmentary materials were available.

The forensic science half deals with two standards of proof: beyond >reasonable doubt for criminal cases and balance of probabilities in
civil cases. When I started out in family history I went on a course so >naturally I asked the lecturer which of these applied. The question was >never answered. Obviously if one were having to give expert opinion in
a legal case the relevant standards would apply with the proviso that
it's up to the tribunal to make an overall judgement, the expert can do
no more than give what evidence is available and their own interpretation.

That leaves me with the general scientists approach for which the
default position should be "I don't know for sure". I can collect
evidence and make whatever interpretation best explains it all. That >becomes my current hypothesis. I have to accept that subsequent
evidence might contradict it in which case I have to examine the
evidence again, possibly reject anything that's misleading and come up
with a revised hypothesis. If further evidence agrees with the
hypothesis then I can regard it as being strengthened. In fact
scientific method demands that I should look for material which has the >potential to contradict the hypothesis.

That is where I am in my genealogy. Starting with my gg grandfather
everything is supported by some level of record - official or personal knowledge. There is no proof before my gg grand so I have
reconstructed what must have happened based on available records. My construction, although lacking factual proof of linking, can't be
argued by facts.

The problem is always what records were destroyed by war, fire and
flood that would have proven or disproved the construction. In VA that
is a certain problem.

In the early 1700s there were 5 Sullivan households in one NC county
so they must be related - wrong. With later Y-DNA testing there are 3
Sullivan families, none related.

Science always seems to be "so far...".

Hugh

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 28/04/2020 13:00, Ian Goddard wrote:

The forensic science half deals with two standards of proof: beyond reasonable doubt for criminal cases and balance of probabilities in
civil cases. When I started out in family history I went on a course so naturally I asked the lecturer which of these applied. The question was never answered.

You're talking about English law here. Some jurisdictions have an
intermediate standard of proof. The United States is one, and it is
described as "clear, convincing and satisfactory evidence". I'm neither
a lawyer nor an American, and it may that the way this is interpreted in
US law is not what I would want in genealogy, but as a phrase it is
closer to what I'm aiming for as a genealogist than the two English
standards of proof. Of course it would be nice if everything were
"beyond reasonable doubt", but that's a difficult level to attain. Good genealogy requires critical thinking, which results in lots of
reasonable doubts.

In fact scientific method demands that I should look for material
which has the potential to contradict the hypothesis.

I would say that a good genealogical method does too. For example, if I
know roughly where and when an individual was born, and I find a baptism
record in the appropriate parish and time window, I would not normally
consider that "clear, convincing and satisfactory evidence" until I've
checked to see whether there are any other suitable baptisms in a
neighbouring parish or just outside the expected time window, have
checked to see whether there are burials records or death registrations
which might be for that child, and have checked records such as censuses
and the parents given in baptisms in later decades to see if there was a
second individual with an equally good claim to the baptism. If some of
these records don't exist, which in earlier times is quite likely, I
don't necessarily let that stop me from accepting the record, but where
they exist and are readily accessible, I would want to check them first.

The IT half of my career has led me to make little use of genealogical packages. ISTM that the lure of a recognised data structure, the tree,
has inveigled developers into using this as the basis of their data
store. As the family tree is a statement of an hypothesis, and one that might have to be replaced, using it as the framework on which to store
the evidence requires too much prejudgement and might make changes of
mind needlessly difficult. I prefer a mixure of RDBMS and spreadsheets
as a means (not ideal but not enough to prod me into developing
something better) for organising data into timelines.

I completely agree with this. It's long been an ambition of mine to
write a good genealogy application which treats evidence and your
analysis of it as the primary entities, and trees as views of that data
which can be generated in a variety of ways depending on what you want
to visualise. That might be the consequences of a hypothesis for which
there is little evidence, or even be something counterfactual, such as
what you understand another researcher to have believed but now know to
be false. It's a big project and not one I have time to get into
seriously at the moment, but I'd like to think it might happen, assuming
no-one else does something similar first.

Richard

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 29/04/2020 13:56, J. Hugh Sullivan wrote:

Science always seems to be "so far...".

Science always *is* "so far...".

Ian

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)