Skimpy Records and the GPS

In other words, is this comment on the blog http://ancestralfindings.com/genealogical-proof-standard/ true?

"Apparently, when there is no verifiable proof to be found, then no genealogical proof standard is met."

Hi Erica, there needs to be verifiable proof. The question surrounding what is verifiable is one of reliability and validity.

Validity refers to whether evidence relates to the specific profile and whether it refers to what you are actually looking for eg. biological parents

Reliability on the other hand refers to a search engine or myheritage match or Geni profile match and whether it is an accurate match or whether another ''test'' is necessary to verify the match.

An additional ''test'' to verify a match would be sufficient in most cases should it refer to an original source document or a first hand report found in a memoir or similar account.

Normaly I don´t follow a protocol. I just build trees.

And that is the way most users on Geni and MyHeritage work.

But in time you meet people that follow a protocol or you see a website or publication that has followed a protocol. And yes the writers of the records did follow a protocol and had to describe it . And they are very hard and send you home. To do you homework again. Reread the protocol and tell me what did you find. Were did you search and no record is also a prove.

My protocol is first go for the easy facts stuff that is easy to proof and comes from records, The low hanging fruit. Then go for the rest.

The lack of records should never be an excuse for not following a GPS protocol. You simply have to search for more or find other prove. And simple do a step back and take a break. And write down this for the next generation to fill lin the blanks. Or say this without prove. But is the best we have.

Sorry Alex I did not react on comment, but I like make first a introduction and then we can post reactions.

Jeroen, welcome to the curator team.

Let's take the pieces one by one, and imagine that we have a skimpy-record situation. (My saga lines are a good example!)

A thorough and reasonably exhaustive search for all relevant information regarding an individual or family

- DONE, not much found. Landnamabok mentions the person, conflicting information appears in a highly fictionalized saga, nothing else appears relevant.

A comprehensive analysis of all of the available information on an individual or family as to its quality and believability or accuracy as evidence

- DONE, Landnamabok seems like it was intended to be authoritative, so in the absence of better information it has reasonably high credibility

A logical and truthful resolution of any conflicting or contradictory evidence discovered

- DONE - we dismiss the random-other-saga's contradictory information because it seems much more "edited for entertainment"

A complete and accurately phrased citation for each source used in your research, with a footnote attaching each source to its use in your research notes or writings

- YES. PLEASE do this!

A sound, exhaustively researched conclusion regarding your research based on all of the evidence you have discovered and analyzed

- Takes time to write. But yes, we should all aspire to this.

The lack of records isn't a problem for the protocol. It may be a problem for the solidity of the conclusions, but the protocol itself doesn't seem to demand a certain amount of evidence - just that you searched for it, told us what you found, and can justify your conclusions based on the evidence.

(The problem I have with the "standard" is that I can make 100 things a little better, or make 1 thing "up to standard". Usually I end up picking up the 100 low hanging fruits.)

> "Apparently, when there is no verifiable proof to be found, then no genealogical proof standard is met."

I saw this statement somewhere the other day and thought it was very interesting. It's a nice generalization but it's also tricky.

The first thing that strikes me is that the statement is a bit confused about what the GPS is. It should say "verifiable evidence" not "verifiable proof."

The GPS is essentially a process flow to test whether the evidence you've collected really justifies drawing a conclusion.

You've searched everything you can find. You've got some evidence that supports your theory. You've ruled out other theories as being unlikely for various reasons. There is either no contrary evidence or you have a good explanation for the contrary evidence. And, you've written it all out so other researchers can follow your reasoning.

Your case might still be weak. If you've followed the GPS you'll know it's weak. You'll know you should say "might be" rather than "is".

And, even if you think you have a strong case, you could still be wrong. Other people might still disagree with your conclusion.

The GPS is guiding you to a place where anyone who looks at your sources and follows your reasoning will come to the same conclusion you did.

So now go back and re-read the quote. It starts to unravel.

The GPS is a process for evaluating evidence and a way of documenting that process. If you come to the conclusion that the evidence for a relationship is insufficient, you've still met the GPS.

Cross-posted with Harald.

> The lack of records isn't a problem for the protocol. It may be a problem for the solidity of the conclusions, but the protocol itself doesn't seem to demand a certain amount of evidence

This is a good way of phrasing it.

There is an idea lurking in the background here. That's the idea the evidence is somehow deficient. Happens a lot. But notice that if you have some theory about a connection, you have some reason for that idea.

Take, for example, the always popular evidence -- "Grandma said so."

Most of us would not think that's very good evidence unless it's something Grandma knew from her personal experience. Say, the names of her parents. And maybe she knew the names of her grandparents, and maybe her great grandparents. But at each step further back Grandma becomes potentially more unreliable.

That's what the GPS is trying to capture. It's not really ever going to be that there is zero evidence or that "there is no verifiable proof to be found".

Instead, the GPS is leading designed to help us see that the evidence we have is not sufficient to justify making a definite connection.

So maybe we could rephrase the question as:

"Apparently, when there is no verifiable evidence to be found, then no genealogical proof standard is met."

Is that the case?

(if so, I despair, because I run across profiles built on logic cases.)

I would still quibble.

What does "verifiable evidence" mean? It sounds to me like weasel words.

If Grandma says her great grandfather was John Smith, presumably you can verify that. It doesn't make it true.

Evidence is just evidence. Something somewhere that makes you want to test whether it might be supported someway somehow.

You collect all the different evidence you can find, both for and against, then you put it through the wringer to see if you've made a convincing case.

I suspect the person who wrote this was thinking of primary documents. And sure, primary documents are probably the highest quality evidence we normally see but they're not the only kind of evidence.

For example, say I get a DNA match with someone in the 1700 cM range. He's going to be my half-brother, uncle, nephew, grandfather, or grandson. It won't take much work to build a logic case about the actual relationship.

Usually, it's going to be harder than that but this example shows that we can deal with more than just primary documents.

A big part of the skill in doing that is building a case that stands up to scrutiny. There really has to be no other likely solution, and all the doubts and contrary evidence have to met and answered head on.

Another easy example that makes the same point would be if you had a "deed of heirs of John Smith". It might be signed by Thomas Smith, Jacob Smith, Henry Smith, and Robert Miller.

So who are these guys? The deed doesn't say they are all children of John Smith, but that would be a reasonable conclusion. And Robert Miller? He's almost certainly married to their sister, a daughter of John Smith.

This would be a logic case. There are other possible interpretations. The men could all be cousins. Robert Miller could be an assignee. But that's not the straightforward interpretation.

You'd probably want to find other corroborating evidence that these are indeed children of John Smith, but you've made a good start.

This is a definition I like!

"A big part of the skill in doing that is building a case that stands up to scrutiny. There really has to be no other likely solution, and all the doubts and contrary evidence have to met and answered head on."

Taking it apart to steps:

- gather evidence
- determine what possibilities the evidence suggests
- formulate an argument (and counter arguments)
- present the case
- peer review

?

Adding evidence is a good thing, but we need to remember that there is no such thing as certainty in genealogy. See my blog at http://schoenblog.com/?p=433

I like the Wikipedia phrase "verifiability not proof". We're not searching for the Truth as Revealed from On High, we're searching for what we can verify evidence for.

"Verifiable evidence" is evidence you can show to others so that they can also see it. If you've published somewhere that "Gramdma said that...", others can verify that you've written it down. You might have trouble convincing people that the evidence is compelling, but it's verifiable.

So it’s all about the “show me ...”

So to address Justin’s quibble :

- there is evidence
- there is evidence (assumed)
- there is evidence (cited)
- there is evidence (reported)
- there is evidence (presented)

For example, a profile might claim in its field entry

- Born Jan 22 1623 in England (assumed)
- Based on biography (cited)
- A record could be cited: England births database (reported)
- A copy of the record uploaded to profile (presented)

Do any or all of those presentations meet the GPS?

Ok, I need something more concrete. Does this profile meet the GPS?

Dr. Melville Wright Staples, M.D.

I would say yes, based on my reading.

First guinea pig, you brave lady!

Spun off to a new discussion

https://www.geni.com/discussions/178160?msg=1200395

I like this:

"The GPS is guiding you to a place where anyone who looks at your sources and follows your reasoning will come to the same conclusion you did."

... and maybe add:

"The GPS is guiding you to a place where anyone who looks at your sources and follows your reasoning will come to the same conclusion you did as being the most probable (most likely)."

Nice amendment, Dan.

Maybe we could ask Erica if she thinks that's worth putting in the project overview.

Erica, I think there is an easier way to approach sourcing than using assumed, cited, reported, presented.

In general it should be fairly easy to document an exact birth date like Jan. 22, 1623, say from a christening record. If someone can't do that, there's a problem right away.

But the real problem, the one the GPS tries to solve, is showing that the guy born on that date is the same guy you are searching for.

In fact, this is one of the most common problems with Colonial genealogy. We all want to "jump the water" but how can we prove that our John Smith in New London Connecticut is one of the many John Smiths born in England at about the right time.

Taking on a project to prove a specific identification is essentially building a logic case. I think he was the John Smith born Jan 22, 1623 because x, y, z. Maybe the birth record for that guy says his father was Alphonso Smith, and our guy named his oldest known son Alphonso. Alphonso is a very uncommon name in 17th century England, so we've got a piece of the puzzle. Then too, maybe Alphonso Smith in England was mentioned in a couple of Puritan records, while our John Smith in New London was also a Puritan.

On and on, slowly building a case. Then someone comes along and points out there were three Alphonso Smiths in England at that time and they all had sons named John. The case falls apart, and we start over ;)

If we're lucky enough to find our John Smith in the Early New England Families Project at NEHGS, then the GPS is already done. We don't have to re-invent the wheel, although if we have additional information we'd want to say that.

Erica, I like your work flow (above). Particularly "peer review".

Yay, peer review.

For the GPS process to work you have to be coming to the same conclusions other people do. If someone tells you they don't agree with your conclusions, it's time to re-think.

When I was a kid starting out in genealogy, I had the benefit of mentoring by my grandmother's cousin. A crusty old guy with a dry sense of humor. He could spin dozens of alternative scenarios out of any piece of info, I credit him with teaching me how to know the difference between what the record says and what I assume it means.

I made a subheading in the project overview

====Results
* "The GPS guides you to a place where anyone who looks at your sources and follows your reasoning will come to the same conclusion you did as being the most probable (most likely)."

We are getting good ideas quickly in this Workshop. Thank you!