Is making things up good practice?

+19 votes
238 views
Go back four hundred years or so and it is almost impossible to determine where anyone was born. For women especially when is often as much of a problem. Date estimates are essential for the process. It has been suggested that I should add a place of birth when I don't even "know" the person was born in England. (Probably was). Also possible that he was born on the estate named in his father's title. THERE IS NO EVIDENCE. Filling in something and marking it uncertain seems to be the advice. I'd prefer to leave it blank which makes it clear there's more to find then add the place as a category. About and uncertain have become true in the past which is what has created the mess in the first place. Is there a lot of benefit derived from making things up? Someone else might make something else up after all.
in Policy and Style by C. Mackinnon G2G6 Pilot (229k points)
I totally agree and once someone has made something up it can be very difficult to remove it. I recently adopted https://www.wikitree.com/wiki/Warwick-26 whose LNAB is given as Warwick. There is no evidence anywhere for the identity of this woman and her LNAB was certainly not Warwick. At best I can document where her identity is not found.
Perhaps her LNAB should be changed to Unknown. I thought a little bit of bio would be nice as she was a real person but we cannot even know if the sons had the same mother. Sorry thought is that if whoever uploaded the ged had done it by hand then she would probably not have been included.
Don't make stuff up! It's almost worse when it's highly plausible as it gets copied and rapidly assumes the status of fact.
Lots of that family look decidedly suspicious, all the Deardena profiles for instance.

Interestingly Katherine Keats-Rohan in her Domesday Descendants books, makes Siward, the son and heir of Turchil with his second wife Leverunia, and Osbert the son of Siward.
I agree she should have LNAB as unknown, This was an extreme example of a person where nothing is known so at this point I have left her blank. I would rather see a totally empty profile than information without sources, especially for pre-1500. This lady is part of an extended family I am researching with numerous similar profiles and probably a fair number of non-existent people. When I have gathered my sources that do not mention her in relation to her family I will add them. Everyone deserves a source even if it doesn't confirm anything about them... It just takes a lot of time to document that there there is no documentation...
I have to respectfully disagree with Katherine Keats-Rohan as every piece of evidence points to Siward as the son of the first, and unknown, wife of Turchil and Osbert as the son of the second wife, Leverunia. Other errors in this family in various narratives are so disparate. I have started a free space page to discuss this.
Yes, I believe the Deardenas did not exist. There is also a major problem with the claims in some pedigrees of Turchil being married to a Countess of Perche and Turchil having a daughter Margaret, heir of Turchil, being married to Henry de Newburgh. I set up the free space to try and document these errors and as a place of discussion for improving their Wikitree profiles but not had time to add much yet.
I agree with Matthew Fletcher. The more plausible the unproven data, the more likely it is to be accepted, and the less likely that any further research will be done.

6 Answers

+9 votes
 
Best answer
The thread title begs the question, so my answer looks defiant and counter-intuitive when I say "Yes".

Our shared tree has a real problem with dozens or hundreds of profiles that look identical in find-matches searches, because they have common names (such as John Smith or Mary Thompson) and no information about dates and locations.  

It is a burden on volunteer time and effort to have to click through to each of those profiles to determine if they are sourced, what century they're from, what hemisphere the ancestor lived on -- just to create a profile or properly merge an ancestor down to the lowest-numbered profile that was created for them.

When I see a profile with no geography information coded in birth or death fields AND when I can determine where the ancestor probably was at their birth or death, I do enter a location, mark it uncertain, and explain the basis for the estimate in the change log.  For example, for a profile whose parent spent their entire lives in Massachusetts, I would probably code their birth location "Massachusetts" [uncertain], or perhaps "USA" [uncertain] if the ancestor was clearly born after 1776.

I share some of your anxieties about birth date estimates and don't often make them.  WikiTree's and FamilySearch's margins-for-error just aren't big enough to return all the relevant results when the estimate isn't close to whatever other bad data is out there.  I do add [uncertain] and {{Estimated Date}} tags to neglected profiles with dates that are unsourced if I'm editing them for other reasons.  

I'm also deeply skeptical of FindAGrave's location data, particularly because the website prohibits links to birth and death records.  Copying the cemetery location in to the death location field is a recipe for propagating bad data.  At most, and only for profiles where the FindAGrave user appears to have physically observed the burial site, I sometimes code the death location as the USA state where the burial is located, assuming that is consistent with any other evidence that the ancestor died in the region.

I understand that there are risks that an ancestor was born to parents who moved around a lot, or that an ancestor's remains were shipped out of state for burial, and that in those cases my bad data will propagate from their profile on WikiTree to other repositories.  But unnecessary duplicate profiles and unsourced profiles are also unhelpful data.  Estimating locations helps those profiles find their way to Sourcerers and Data Doctors familiar with the region who can dig a little deeper and put the profile in its proper place on our shared tree, or can determine that it is an unreliable legend or a conflated profile.   

"Making stuff up" i.e., educated guesses, supported by reasonable inferences from available data, is an appropriate way to put research problems in front of the volunteers best equipped to tackle them.
by E. Compton G2G6 Pilot (133k points)
selected by Kitty Smith
I agree with this. I frequently work on profiles that I don't manage. If there is no estimated birth or death place for several generations, I will most likely move on to one that has at least a guess. If I have to 'guess' on a location, it is put into the Notes section of the bio, so that the next person working on it can see why I chose that location and go from there.
+8 votes
yaaay!!! glad someone said it!!!!..considering, I was born in a tiny town in MA, but have lived in 4 other states including Alaska, and none of the states are near the one I was born in..& married in 2 different states on 2 completely different parts of the country...nor was I ever married in the one I was born in...seems very plausible to me, that ppl 400 years ago moved around just as much if not more...because of work...food sources, proximity to waterways or good land for growing, or closer to "centers" of villages, towns, etc..I get the need for guesstimates of cica/time periods to help locate births deaths etc...but guessing on places of marriage, residence, etc...they couldve up and just moved...right when you think you know where they are.  I have some in my lines that all the drouin, tanguay, jette stuff says...Saint philippe Laprairie...yet..I've not found squat for records on any of em for years! same thing goes with St.Marguerite, Blairfindie, and Laprairie areas in general...total loss for connections..even FHL...no signs of em...only place I've had any hopes...was records connected to "western" movements..."guess what" I just found another grandfather who was hired..guess where to...yep..outwest!...I suspected some..not this one..wouldn't have guessed it.
by Arora Anonymous G2G6 Mach 7 (74.8k points)
+5 votes
I too hate making things up. I can see the point of an estimated date of birth, but why do we need an estimated date of death. Sometimes I find a child and I have no idea whether that child survived or not. I would prefer to leave the date of death blank. Instead I make a guess, usually that the child did not survive, add a comment in the biography and add an estimated date template. I still think it could mislead others into reproducing the date as correct.
by Joan Whitaker G2G6 Mach 9 (90.1k points)
We don’t need an est. date of death.  We do need one of them filled out but if you have something in the birth date field, you don’t need to add an est. death date.

I disagree that it is impossible to figure out a birth place for people born hundreds of years ago. While we are very mobile today, as Arora’s Post demonstrates, our ancestors of long ago were not nearly as mobile. At a minimum, entering the likely country name is not making things up. It’s a highly educated guess that is likely to be true. And fully appropriate to check the Uncertain box.
Some of them were a lot more mobile than you might expect Jillaine. Many of mine were born in the England/Scotland border regions. I am not particularly comfortable making guesses as to which country is appropriate.
In those cases, leave the place name blank and explain the reason in the narrative.
I agree with you Joan that an estimated date of birth is helpful, particularly in situations where you might have profiles with the same or similar name.  I also try to include what information I've used to base my estimate.  Hopefully with a big estimated date template this will stop anyone reproducing it as fact,

I also agree that there is no reason to include a death date, if it isn't known, nor a birth or death place if that isn't known either.
I use the research notes to explain that either country could still be an option Jillaine. Just wanted to let you know that deciding on the country is not always as easy as you suggested in your previous comment.
I think there are a few New Netherland Settlers who might have been born in the Netherlands (or possibly Germany or Belgium or somewhere else in Europe), the Dutch colony in Brazil, New Netherland (in North America), or at sea (in transit between two of those places). That's three continents to choose from!
Don't think we're going to stop folks from guessing/estimating ... but, for most/all of those fields you can mark those estimates as 'uncertain'
+12 votes
We sign an honour code, no where in the principles does it state we make up stories/ places/names.  Point 2  details accuracy,  If the information or sources aren't there then they should not be recorded as factual.

Research notes is the place to discuss shortcomings and seek collaboration from others.
by Rionne Brooks G2G6 Mach 5 (56k points)
This is why we have the "uncertain" button.
To clarify: estimating is not making things up. It’s coming up with an proposed suggestion based on other known information.
+6 votes
The tendency of wanting something in date and location fields leading to estimates is one of the obstacles when trying to convince non-WikiTree genealogists that WikiTree is a good thing. We have the option of putting "before" and "after" dates in place which is what most serious genealogies do for unknown life events and we should promote that instead of pushing for estimated dates.
by Helmut Jungschaffer G2G6 Pilot (508k points)
That's what I like about WikiTree, comparatively, the standard of proof is held much higher here. According to another collaborative site; I am in direct lines for every branch of royalty. Without a single source. When I "fact check", I can't even come close. I have seen profiles here where there will be footnotes that name all of the areas that need further research or additional proof, and I think that is acceptable as it lets another researcher know where they may be able to fill in questionable details. But I agree 100%! Please don't "guestimate"- 1000 people will add that as fact and it becomes impossible to reverse the rumor.
I agree that "before" and "after" dates should be used when one or more dates are available to bound or bracket the dates of a birth and death (for example, the date of the person's baptism, or the probate date for their estate, or the last date of a record that identified the person as alive). But all too often we don't have those kinds of dates. Taking the only known date(s) from a person's life (for example, the date of the baptism of one child) and adding a "before" or "after" qualifier can grossly misrepresent the person's biography. If a person had a child baptized in 1685, it's accurate but misleading to say they were born "before 1685," since we can be sure they were born more than a few years before 1685. It's far more helpful to provide an estimate in the data fields (such as 1660 or 1665 in the example), mark the estimate as uncertain, and explain the basis for it in the biography section. Nobody should be treating the data section of a profile as the entirety of the profile. The data section is an index to the profile or a thumbnail version of the person's life; the text section is where we tell the full story.
Is it really that much more helpful to make her 20 - 25 years old when she had her first documented child when she might have been 15 or 35?
Yes.That date is helpful for people who are searching for additional records using utilities like RootsSearch. And having a semi-realistic estimate of her birth date may help identify her parents (or not inappropriately rule out a particular set of parents because the mother was too old to have had a child in 1685). Or if someone gets lucky and finds her baptism record, it reduces the chance that they'll create a duplicate profile. And if a duplicate profile does get created, the proposed merge is less likely to be rejected for having a 25-year discrepancy in the birth dates (believe me, that happens). The best professional genealogy research papers that I read often include estimates -- clearly identified as such, with information on the basis for them -- of birth dates for the people they profile.
+5 votes

I work on Smiths/Smithes/Smyths/Smythes.  If a death date is not available, a birth year estimate is necessary to minimize the creation of duplicates.  Before and After a known date are not useful for WikiTree Smiths.  We need an educated guess +/-2 yrs, or +/-12 yrs, or +/-30 yrs so the profile can be searched by date range.  

A country of birth can usually be estimated as well and marked as uncertain.  

by Kitty Smith G2G6 Pilot (514k points)
No sure people actually take note of uncertain.
So true, C.  I hope if they are sure, they will correct the estimate and add a source for the data.
Yes, people not taking note of the data qualifiers (uncertain, before, and after) is a problem. This failure to notice data qualifiers is due in large part to the fact that they aren't displayed on Surname Navigation Pages or the screens we see when completing merges. Thankfully they do show up in results for name searches.

Related questions

+1 vote
1 answer
+3 votes
1 answer
+5 votes
3 answers
84 views asked Dec 13, 2017 in The Tree House by Lisa Murphy G2G6 Mach 7 (71.1k points)
+35 votes
5 answers
+4 votes
2 answers
+7 votes
2 answers

WikiTree  ~  About  ~  Help Help  ~  Search Person Search  ~  Surname:

disclaimer - terms - copyright

...