Welsh Pedigrees and their Date Issues

Question

Early medieval Wales is long on pedigrees and short on dates.  As a result, when you try to turn names into real people with dated events, it begins to appear that there are missing generations.  The facts seem to indicate that Haer ferch Cillin's child Gwerful or Generis had to be born in 1046. Other facts assert that Generis' son Rhirid flourished in 1160.  

The best sources are often pedigrees put together 4-5 centuries after the fact.  The best source for Rhirid is songs about heroes put together when Rhirid, or at least Rhirid's half-brother,  had just died in 1160.  One is not working with really solid data!

So the question for WikiTree is, "in the presence of conflicting information, how best to display the profiles of the affected people?"  I can see these alternatives:

1.  Status quo.  Do nothing.  Leave the conflicting and impossible dates there as well as the connections.  When "suggestions" show up highlighting these errors, ignore them.

2.  Describe the problem in the biography, probably under research notes, but leave the datafield and connections as they are.

3.  Describe the problem in the biography as in 2, leave the data field consistent with the biography, and disconnect profiles where necessary.  "Rhirid is said to be the child of Generis, but she lived a full century before his birth and therefore he has been disconnected"

4.  One approach taken by the Welsh researcher Darrell Wolcott is to assume, probably correctly,  that the pedigree one is following left out some generations, so he helpfully creates the missing generations.  Now the dates work.  To implement this in WikiTree, though, one would have to create totally unsourced profiles of people about whom one knows absolutely nothing except that somebody need to be there to make the dates work.  Such profiles are not only unsourced but unsourceable.  I tried this approach in Wikitree once and it really didn't work;  those fake profiles no longer exist.  

I am inclined to go with alternative 3.  Is that the best of the alternatives above?  Does anyone have any better alternatives?

Answer 1

Number three seems best to me. It is accurate, and leaves the way open for future research, without putting unsourced speculative profiles into the database.

Mae rhif tri yn ddewis hyfryd iawn.

Answer 2

There is no question but that the dates on many early Welsh profiles guesses, or at least based in antiquity.  And there will probably never be better sources for those from 750 to 1200. When I started the Cymru Project, I was blissfully ignorant of the difficulties, and all I saw were 100s of duplicate profiles that needed to be merged. 

The problems are that so much is masked in legend and that there really aren't records. Bartrum can be searched, but that is difficult.  Boyer's Medieval Welsh Ancestors of Certain Americans  is much easier to access and understand, and although he consulted many works, there are almost no inline citations. The best I have found so far is at FamilySearch: "Community Trees - Wales. Welsh Medieval Database Primarily Nobility and Gentry lines. Date range: 100s-1900s", which has specific references and also has notes and shows places that need further investigation.

I agree with you that the status quo is not good, so Option 1 is out. Although Wolcott may be right, without evidence for the added generations, it is all conjecture, so I think Option 4 isn't a good one. 

If we use Option 3, we need to make sure that former parents or children, who are disconnected, are linked in the Bio. I am sure that, if we do disconnect parents or children, many will be recreated and attached by people who don't read the Bio, but it is probably the best option.

On the profile you reference, Haer ferch Cillin, 1089 is shown as her death date, with the only source being a "Wikitree Date Field" sourced.  As Bleddyn married twice more and had several children with wives 3 and 4, and died in 1075, it is probable that she died much earlier.  The FamilySearch well referenced [https://www.familysearch.org/ark:/61903/2:2:3ZDM-QZS Community Tree], shows that she died in 1044.

Answer 3

Well, there is an option "3 Plus" which I have taken with Generys, who has been shown as the daughter of Haer and her first husband Cynvyn in the 10 hundreds, and wife of Gwrgenau, who lived a century later in the mid 11 hundreds.

Assuming all of this to be true, the only conclusion at the moment is that there were two women named Generys, an early one who was the daughter of Haer and Cynvyn, and a century-later one who married Gwrgenau.  We now have a separate profile for each, with a disambiguation toggle at the beginning permitting quick reference back and forth, and the issue explained in both profiles.  This permits full recognition of the medieval and later sources referring to Generys, while not triggering multiple WikiTree data "suggestions".