Is Wikipedia really a source?

Started by Private User on Thursday, January 21, 2021
Problem with this page?

Participants:

Profiles Mentioned:

Showing all 21 posts
Private User
1/21/2021 at 4:10 PM

Good evening,
I wanted to get curator thoughts on a subject that has me somewhat confused, that of Gascony and the ancient dukes of Gascony. I’ve been in my own tree for several weeks and am confused and know I have errors because I’ve not stayed consistent in my sourcing. Gascony is one area of Genealogy where no matter who you use as a source, it differs from any other, and to my dismay, mixing them only complicates things.

I note on Geni that the connections seem to be based on Wikipedia, and while I use Wikipedia because of the way it tells a story, in this case it may not be the most accurate source. The overview tab is then filled with questionable data linked to questionable sites like Robert.Sewell and Wikibin. Or we do this...

This is the Master Profile for Bodegisel II, duke of Aquitaine.
Curator Note from Justin Swanstrom (Howery) (7/28/2015):
Parents unknown. Do not attach him to any.
Probably son of Gondolfus and Palatina, but experts disagree.
Probably father of Arnoul, Bishop of Metz, but experts disagree.
Locked to prevent vandalism.

...and simply leave it incomplete. One of the experts who disagrees with this is Christian Settipani, who touches on Gascony in his masterful works Les Ancêtres de Charlemagne and La Noblesse du Midi Carolingien, and even though some of his conclusions are conjecture (but based on his research), IMO Settipani conjecture beats Wikipedia.

Even MEDLANDS alibis their entries in two ways, either by indicating there is no primary source that connects this person to that (“His name and parentage have not been corroborated by other primary sources consulted”), or by disclaimer that “According to Jaurgain... but he cites no primary source on which this is based”, but then in many cases uses the connection anyway (as he did with Mondriere and the family d’Estouteville).

Then things get further confounded by the fact that early connections are corrupted in history due to the so called spurious Alarcon documents.

I ramble...I know...

I’ve looked at Jaurgain’s original 2 volume work in French and he is both quite clear and consistent in his connections in that some have no sources at all while others are connected using the Alarcon documents. But research into these documents turns up interesting facts about why they are considered spurious:

Once proven, by Rabanis, the truth about this charter, it emerges that:
1 The descendants of the Merovingian Charibert stop at his Childeric son who died shortly after him;
2 We do not know the father of Duke Eudes, who is therefore not related to the Merovingians (at least in direct male line);
3 The progeny of the Duke Eudes end with Waifar, Wolf I st and Icterius because he is not proven the filiation between the Duke Waifar and Gascony Duke Wolf II;
4 We know that the Kings of Navarre descend from the Dukes of Vasconia without knowing the details;
5 According to Jules Villain ( Modern France ), The Counts of Comminges and the Counts of Foix (therefore) descend from Loup II Duke of Vasconie. The analysis of generations does not show any inconsistencies;
6 Having no data on Duke Loup II de Vasconie, we cannot go back in time, nor link the Houses of Comminges and Foix to that of the Kings of Navarre (Note that there are other legitimate sources that do this however)

But Note also that there is no indication as to whether the actual connections between the dukes of Gascon and within the family groups are in question! So...
1 N/A
2 we presume using other legitimate sources
3 we presume using other legitimate sources
4 we presume using other legitimate sources
5 N/A
6 we presume using other legitimate sources

This reference, which comes from a site I discovered on the Wayback Machine, concludes with this:

“Le souci du détail, voire de la grandeur, peut amener le mensonge, donc desservir la cause de l'Histoire !”

“Attention to detail, even greatness, can lead to lies, and therefore serve the cause of History!”

I love this! It says to me that it’s OK to use questionable information because it may generate discussion. Even Charles Cawley told me once not to be afraid to use questionable information if it supports my argument but enclose it in square brackets to mark it as such.

...bear with me

French Wikipedia says this: “Jean de Jaurgain, born November 16, 1842 in Ossas and died March 18, 1920 in Ciboure, is a French Basque historian and literary critic. Specialist in genealogy and heraldry, as well as in the history of the Basque Country.”

So why would Jean de Jaurgain as a source be of less import than Wikipedia, or Cawley at MEDLANDS? Is it because it’s in French and therefore takes an effort beyond simple copy and paste to understand? It’s curious to note that French Wikipedia can be translated to English but when translated is frequently different that the original English page on the same person. Which is more accurate? Which should be used? Is Wikipedia a legitimate Genealogy research source or just a source for a simple overview narrative?

Bottom line here is that I think the Gascony section of the world tree needs a lot of work. I propose a redo following Jaurgain, using supplemental information from MEDLANDS, Etienne Pattou and other legitimate research sites, and a simple narrative from Wikipedia ONLY if and when positive identification can be assured. There is immense confusion in the area due to naming conventions and the frequent mixing of Spanish and Basque and Catalan and numerous other regional languages. A disclaimer in the heading is OK and probably warranted to the effect that the information is in some cases speculative.

Right now we present it as accurate mostly based on Wikipedia as the source, ignoring other legitimate sources, while supplementing with illegitimate sources (IMO). The key is to use a single legitimate source, supplemented with information from a short list of other legitimate sources.

1/21/2021 at 10:52 PM

Private User

Wikipedia is of course only as good as its citations (although the narrative is helpful), and obviously French Wikipedia is likely to do a better job with French history, the language articles are separately constructed.

My caution is a general one. Due to the nature of geni software, including DNA test linking, we don’t do speculative. In fact we now have the isolated tree option for speculative trees. So the way we’ve handled uncertainty is to link within profiles to the possible parents. How strictly that’s enforced per profile is, I suppose, based on discussion from managers, interested researchers, and curators. But if a curator has previously said “no” in a curator note there should be pretty solid evidence to alter it.

1/22/2021 at 6:21 AM

I find this extremely disturbing.

For one thing, to add in speculative connections would roll back an enormous amount of hard work that we have been doing, across the Tree, but especially in the medieval Tree.

Medieval genealogy is full of variant sources and variant experts. And assessment of these variants changes over time. That is the nature of medieval studies.

I understand your desire to simplify things by choosing one expert and sticking with them, but this will rarely work. (It doesn’t work even in the Welsh Tree, where, though Peter Bartrum managed to make sense of the contradictory medieval genealogies, there are still knots he didn’t untangle.)

I’m especially disturbed that the expert that you wish to rely on was working in one of the most problematic periods in medieval studies, when jumps over missing evidence were common, when it was usual to not cite sources, and when the field in general was focused on nationalistic endeavors, which caused the primary evidence, and the gaps in primary evidence, to be skewed into conclusions that supported various stories of former nationalistic greatness.

It’s why, for instance, John O’Hart’s beloved Irish Pedigrees, cited often by our users and various studies, is trash. His goal was to show the “pure” roots of the Irish, which he did by cobbling together various sources, including mythology and nonsense, not giving citations, and making connections up.

There are varying degrees of reliability across the field of medievalists from that era. But the timing alone makes it extremely important to be highly careful with any speculation from their works.

I use Wikipedia to look at the sources, follow those up, and evaluate them. Sometimes quite useful.

But in short.

We have been working very hard to undo the speculative connections in the medieval tree, seductive though they be.

Please don’t set us back.

1/22/2021 at 7:15 AM

Why do you say the link is based on Wikipedia?

1/22/2021 at 7:17 AM
1/22/2021 at 7:37 AM

I recall following Cawley on a Merovingian line here (the second part of the project): https://www.geni.com/projects/Merovingian-Dynasty/5132

1/22/2021 at 7:37 AM

If you want to discuss the historical validity of sources of particular profiles, please put a link to them.

1/22/2021 at 7:42 AM
Private User
1/22/2021 at 1:51 PM

So, let’s take an example identified by Sharon and dissect it,

Geni 1
Loup II, duc de Gascogne (Loup II, duc de Gascogne)

1. The heading says “Parentage uncorroborated. Wife(s) unknown.”, yet the profile itself identifies him as “Husband of Numabela Froilez de Cantabria “
2. I say it appears that Wikipedia is the source because other than a faulty link to Geneanet, it is the only source identified. It is listed numerous time in the overview tab but in the sources tab there is no reference to anything!
3. Look at the revision tab, there are numerous “updates” made by numerous persons none of which are sourced!

Geni 2
Look at the profile of Numabela Froilez de Cantabria, the wife he isn’t supposed to have
1. The heading says “Unknown wife(s).” For a wife?
2. The overview tab starts by referencing an image that doesn’t exist in the profile
3. The overview tab which references the Spanish Wikipedia says she was born in 755, but the profile says 752. Turns out hat if you go to the sources tab there is a pdf from OnlyGeneaology, which by using Google search for Numabela of Cantabria, the name used on the source, was unable to find it. So the date in the Geni profile is from an unnamed source!
4. The profile shows for Numabela “Daughter of Munia Fróilaz Gundersindez” indicating no father. Yet the source of unknown origin clearly identifies a father! Why the selective use of a birth date but not the use of a parent?

I think you all need to set aside the emotional response and reread my discussion. I do not suggest using speculative data, I suggest using data from a single know source and noting it where speculation may occur, as does Charles Cawley and his use of square brackets.

“Medieval genealogy is full of variant sources and variant experts. And assessment of these variants changes over time. That is the nature of medieval studies.” I agree. Jean de Jaurgain is an acknowledged expert in this area of study, referenced repeatedly by Cawley in MEDLANDS. I asked the question why his expertise in the subject area was less valuable than Wikipedia?

“I’m especially disturbed that the expert that you wish to rely on was working in one of the most problematic periods in medieval studies, when jumps over missing evidence were common, when it was usual to not cite sources, and when the field in general was focused on nationalistic endeavors, which caused the primary evidence, and the gaps in primary evidence, to be skewed into conclusions that supported various stories of former nationalistic greatness.” I’ve heard this argument before, actually from Justin when I first joined Geni. “You can’t trust any French source from the 16th and 17th centuries, they’re all based on self interest!” So we’ve resorted to trusting OnlyGeneaology, a private tree by persons unknown and a pdf with the bottom of the page that shows sourcing cut off.

I don’t doubt that this area of study is incredibly difficult and without a great deal of reliable resource information. So why not use a known source whom even Cawley at MEDLANDS refers to.

More examples in Geni...
Geni 3
Hatto, duke of Gascony, the profile shows him the father of Lopo I, Wasconorum Princeps, Husband of Numabela de Cantábria and Wanda de Toulouse , but look at what MEDLANDS says...

LOUP, son of [HATTO & his wife Vandrade ---] (-murdered [775]).  The charter of Charles II "le Chauve" King of the West Franks dated 30 Jan 845 (probably spurious, as explained in the Introduction) names "Lupo Duci" as son of "Hattonis Ducis"[5].  His parentage has not been corroborated by other primary sources consulted. 
m ---.  The name of Loup's wife is not known. 
So this profile is based on a spurious charter and assign two wives not know in primary documents, and shows NO sources!

Geni 4

Sanche III Mitarra Sanche, duc de Gascogne

Son of Sanche II, duc de Gascogne and Quisilo García, Comtesse de Buei

But in MEDLANDS...

1.         SANCHO "Mitarra/Menditarra" ([825/35]-before 893).  The parentage of Sancho is unclear from the primary sources.  Although the patronymic of his son Garcia indicates his father's name, Settipani states that there is no proof that Sancho was related to the previous dukes of Gascony who are shown above[82].  However, a clue is provided by the undated charter, under which Sancho’s great-grandson "dominus Willelmus Sancii comes Gasconiorum" donated property to the abbey of Saint-Vincent-de-Lucq[83].  The document records that "quomodo venisset de Hispania avus domni Willelmi ubi se contulerat pater eius tempore domni Ludovici imperatoris" ("the grandfather of seigneur Guillaume had come from Spain where his father had taken refuge at the time of emperor Louis")[84].  The "avus domni Willelmi" was Count Garcia [I] Sanchez, so the latter's own father would have been Sancho.  The document also states that the consanguinity of Guillaume Sancho with Gaston Centule Vicomte de Béarn was proved: "quidem Rex" (which from the context appears to refer to Sancho) invested "avo Vicecomitis [Gasto Centuli Vicecomes Bearnensis], qui erat de eius progenie" with "hac patria".  This statement provides a strong suggestion that Sancho and his son were related to the previous dukes of Gascony, from whom the vicomtes de Béarn were probably descended although the documentation which proves this relationship has not survived. 

So this profile is based on a “clue” and “strong suggestion” “although the documentation which proves this relationship has not survived”.

It seems to me that everything you all are worried about is already happening! Why? I think, as I’ve said, because there is a piecemeal application of unsupported data and misc sources that fit. Look at the examples above and you will see NO consistent application of a single source that covers maybe 6 generations, except Wikipedia, ergo my original question

All I said in my original discussion is that necessary consistency is missing, and one way to rectify that is by using a single source(s). And if the sources are not primary proof, then notification is necessary by statement or inclusion of a bibliography.

My objection is and always has been the use of random sourcing and community trees as sources.

I’d also like to point out, as another example of the misuse of information, that several (4 of hand);of the profiles connected to this Gascony “family” use an incorrect coat of arms as the image. The blazon shown, quartered, (simplified) a white lion rampant on blue in 1 and 4 and a garb of wheat gold on red in 2 and 3 was not in existence at this time. It was awarded to the Duchy of Gascony by Louis XIV in 1696. Prior to 1453 they used a single red lion rampant on white. There is really no strong evidence that arms were used prior to the 11th century but Wikipedia (:-()says this blazon was used as long ago as 602

1/22/2021 at 7:36 PM

I corrected N of Cantabria to align with the curator note.

A discussion should be opened from profile is anyone wishes to add parents.

1/23/2021 at 12:37 AM

David, you're looking at Abouts merged together from lots of profiles - they don't necessarily represent the history of that profile - rather that the profile could be cleaned up.

If you're interested in the area, why not start from a profile and initiate a Discussion of the sources? Get the Curator involved. You're welcome to call me there if you get no response from them. If your interest extends, we could create a Gascony project for you to link the profiles to. You're right - these profiles do look like a neglected area on geni.

The points against the single source are important, in this regard. We tend to use Cawley as our starting point because he's the most easily accessible online, but the Discussions about his and other sources evolve from there, and, in doubt, we tend to stick to documented proof.

Our technique has been afforded more rigour with the relatively new ability to lock relationships - so we no longer have to create workarounds that anticipate speculative lines being added back faster than we can control them.

So, we don't include speculation on geni anymore, if we're alerted to it - that's an area for historians, not genealogy.

1/23/2021 at 5:32 AM

TL:DR
I think the responses to David are too heavily skewed to address speculation.
David is asking why this part of the tree seems to align with Wikipedia (English version?) when there is a high quality SME available, at least I think that's the main thrust of his initial post.

I think it's important to be very aware of Sharon's point in her last comment "David, you're looking at Abouts merged together from lots of profiles..." rather than being an exception it is more the rule in the popular portions of Geni that the profile's data will not align with the About text and that there will be no sources what so ever attached (side bar: images on medieval photos are reliably never of the person in question... be especially dubious of photos). The typical method for most users is to copy what they have found on some vanity website or in some old book or from an Ancestry tree, very few users are actually studying individual lifestories and crafting profiles based on documentary evidence.

A fact about Wikipedia that is not so clear to the average user is that the different language versions of a page are not just machine translations to allow internet user of different cultures to read the pages easily. The different language page are individually created and only linked together because they address the same subject. In many cases it is a simple cut and paste with Google translate between two different languages but conversely reviewing the different language versions with the aid of Google Translate (or similar) can open the door to an enormous amount of additional information.
Of course bear in mind that on Wikipedia while some articles are brilliant many are facile or just barely defensible and being written in a language other than English is no improvement on this.

1/23/2021 at 5:35 AM

PS this is not a Curator thread.

1/23/2021 at 5:46 AM

"The key is to use a single legitimate source, supplemented with information from a short list of other legitimate sources."

Several years ago I embarked on an attempt to align the portion of the Geni tree involving the settlement of Iceland with a single source. After discussions with several interested curators we settled on a published English translation by a respected historian that is readily available in full online. This project failed, it was overwhelming in size and the resistance from a few vocal Icelandic users made it a thankless task. Lesson #1: Buy in from all is more important than the source, unless you plan to MP and RL every profile and police duplicates.

More recently (but still several years back) i set out to create a Project grouping profiles of troops from a single military unit. Lacking any contacts within the Pentagon I searched the web and found hundreds of copies of the same list, and a few copies of a slightly longer list and some odds and sods. After 2 years of daily effort I think my role call is the most accurate that exists, certainly for free on the internet or Google Books.
Lesson #2: If you put enough effort in you might end up being a better source than your "single legitimate source" was.

PS I hesitate using "source" in this context but am aligning myself to your words, Subject Matter Expert would be I think a better wording.

1/23/2021 at 5:53 AM

My reply to 'Is Wikipedia really a source?': No.

My understanding of Wikipedia's rules is that original research is not permitted. So rather than the usual ranking of sources as Primary, Secondary or Tertiary it almost deserves a new ranking Quaternary? At best Wikipedia articles.

What I found interesting with cleaning the Abouts of some Mega Merge profiles was that in some of them you could see the Wikipedia articles repeated over and over again but varying slightly as Wikipedia is live so copy and pasting from the same Wikipedia article over time produces different results. This feature was also visible on some occasions with Encyclopedia Britannica articles and even Medlands, so not purely a function of Wikipedia.

1/23/2021 at 10:58 AM

I’ve done some cleanup around profiles I manage in this area, following the curator notes generated by Victar and Justin, and also adding narrative from Wikipedia and Medlands.

Sharon is right, it’s a bit dusty. Alex is a bit off - this area “had been” curated to decent standards, and not just online trees. :).

Something I always keep in mind as a curator. I’m not just doing “my tree,” I’m presenting for everyone. Therefore a (Limited) copy - paste from the more or less public domain Wikipedia is in fact necessary for a profile presentation that is easily understood.

So I “start” with cleanup first, and “then” open discussions to invite clarifying genealogy questions. Alex again was on point with this: you need member buy in and transparency.

I think due to just the bits I did last night you’ll find dukes of Gascony less stinky.

Private User
1/23/2021 at 3:57 PM

Thank you Alex, for taking the time to respond, I think you’re response is more to my point regarding Wikipedia, and your assessment of Wikipedia, it’s strengths, weaknesses and features identify with mine. I doubt very many curators actually check into the bibliography, rather taking the position that if “it’s good enough for Wikipedia then it’s good enough for me”..., which is one of my concerns.

The redundancy you see in multiple version of Wikipedia and MEDLANDS, I believe, comes from users who know only the internet art of copy and paste. Even MEDLANDS is a living breathing work in progress, however I too am guilty. I was assigned two mentors when I became a curator, they told me that part of my responsibility was to teach!, so I copy, paste AND edit the Wikipedia articles into the overview for historical context, because for the most part they are well written and lite reading. Users won’t take the time to follow a link or peruse a foreign language, so it necessary to follow the KISS principle in the profiles.

I don’t think that the work of Jean de Jaurgain is itself considered speculative, as he, like all good writers, footnotes throughout, and while I don’t know of the sources he makes reference, it is clear that some are primary (Codex...). Some also are by other Basque historians and history cannot be overlooked in Geneaology. The speculative nature of this area of this area our tree comes about only because two sources that have been used historically, and btw, are referenced in MEDLANDS, have been called out as spurious or probably spurious by leading scholars, and some, like Christian Settipani have developed alternatives. These alternatives are not spurious and should not be overlooked. A work that makes reference to a spurious document is not necessarily spurious in and of itself. Cawley does not call Jaurgain’s work spurious, only saying that Jaurgain “cites no sources” for a particular claim of connection. He said essentially the same about Mondriere and his research on d’Estouteville, that the ancients should be viewed with caution, then proceeded to use that information.

Sharon,
A couple of points...

I too almost always start with Cawley because primary sourcing is always the place to start. There are several important elements of this however:
1. Cawley ONLY deals in primary documents and it has become clear to me that wives and siblings for example who do exist are not always shown in MEDLANDS because they don’t show in primary documents. This forces me to bring additional sourcing into my research.
2. And Cawley does use uncorroborated information (which makes its way into Geni as gospel), but he identifies this by enclosing it in square brackets, again forcing additional research in other than primary sources.
3. But if there is no primary source, do we just stop building a genealogical tree with people that we know from history exist? I think not.

Secondly, if you look back at my effort on d’Estouteville I shared my approach with curators in advance and provided the bibliography I used. While I did start from a single source, Gabriel de la Mondriere who is “a” recognized expert on this family, as is Jaurgain’s for Gascony, so much so that his work is considered of national historic value by France, my “single” source changed over time, likely because the more recent the person the more available became information. It became clear at generation 3 that Etienne Pattou had the French line nailed and the English line source became the Yorkshire Charters.

My reference to single sourcing applies mostly to the “ancients” so at least there is a tree that has a beginning which can then be discussed, and essentially means around which I use other sources to supplement (I have btw received a message from an interested user discussing Robert II d’Estouteville and asking how I arrived at my conclusions because they are different from the established preponderance of evidence).

Finally, my interest here has been pique because the area in my own work on MyHeritage is in such a state that I’m no longer confident in the accuracy of my own work there. I began looking around to find a way to clean up my own mess, which meant to me a single source that would help me build the outline of a basic tree (more later on this). I’ve come to the conclusion that French do a better job of researching French history. Spaniards do a better job of researching Spanish history. And blogs are a good source for leads to other more legitimate sources. I’m interested because it’s hard, because it very complicated (we’re talking Loup, Loup-Centule, Centule-Loup and Loup de Loup! I’m learning a lot about ancient Spanish naming conventions!).

Victar is the curator on many of these Gascogne profiles and I don’t like stepping on someone else’s work, Gascony is his call, but I would be happy to contribute with his approval. I’m not sure what a Gascony Project would do.

My mind is like a detective, I always “need” to know what’s “right”, and whether it is “right”, and while distractions occur (I once had to look at Google maps and the mouth of the Guadalquivir River that runs through Saville because a historical novel I was reading said that a battle with the Dutch took place on a sandbar at the mouth of the river!). I also, for reasons I can’t explain, am able to see “through” the use of words and find different meaning, refer back to the original discussion and my reading of a summary of reasons why the Alarcon documents were spurious and noting that they did not include reference to the Geneaology of Gascony itself, only to the fact that it was being erroneously connected to the Merovingian kings. If you look at the alternative work by Settipani you will see that he only works within the top few generations and their connections up, not down. Observations like this allow me to assess that Jaurgain is IMO a legitimate source.

And Cawley himself says “A small part of the genealogical information in the Alarcon documentation is corroborated by other primary sources, including the Annales Metenses and the Continuator of Fredegar.  Other parts of the information are clearly incorrect, for example the statement that Boggis Duke of Aquitaine was the son of Charibert II King of the Franks in Aquitaine, the younger half-brother of the Merovingian King Dagobert I.  There remains a large part of information in the documents which is uncorroborated elsewhere and whose accuracy cannot be judged definitively.”

Uncorroborated and spurious do not have the same meaning.

More on this...
Full disclosure, despite my railings against copy and paste from community trees, I do use them, HOWEVER...
ONLY on Geneanet I’ve found a number of individual trees built by single persons who have seeming expertise or local interest, many do good work that aligns with the preponderance of evidence we all use. Some list sources, some do not, that’s a option of the software. One I have communicated with does not saying that “we all know what the sources are by now”. He happens to be a former “magician” (curator) on Roglo, the World tree of France! So he is in my “trust but verify” pool of resources.

I use these because they frequently have “missing information” such as dates (that work) and locations more in depth (including French zip code to pinpoint locations), and where I use these I tag the link to Geni using the built in attachment tool. They are also a source for wives and siblings not found in MEDLANDS, mentioned earlier, not verbatim copy and paste, but from the standpoint that they become someone else to research before making them legitimate.

My interest is in building a better tree.

Anyone,
Is there a way to move this discussion into a curator discussion? I started the discussion there but apparently that doesn’t default (so I guess I dont know how yet to access the curator discussion platform).

1/23/2021 at 10:01 PM

RE 3. "But if there is no primary source, do we just stop building a genealogical tree with people that we know from history exist? I think not."

YES we do just STOP. We don't fabricate facts. The historian's purvey is different to the genealogist's, and we are not playing at being historians. Allowing speculation into the tree raises the opportunity for clashing speculations, and then the arbitrary decision that empowers the Curator over everyone else.

5/17/2023 at 5:07 PM

It seems parents crept back for Bodegisel II, duke of Aquitaine

I disconnected per the attached curator note.

Showing all 21 posts

Create a free account or login to participate in this discussion