Genealogy from the perspective of a member of The Church of Jesus Christ of Latter-day Saints (Mormon, LDS)

Tuesday, September 27, 2016

The Mystery of "?" on the FamilySearch Family Tree

Note: This is one post where you must read all the comments. They don't completely solve the problem but they do offer a resolution. 

On my startup page for the Family Tree, I got a notification with a link to the following entry:

Here is a copy of the notification:

Hmm. Apparently, someone could not find the wife of one of the people in the Family Tree and decided to put a question mark in as the name. Extraneous markings such as this, probably included to act as a reminder to the researcher, are a somewhat constant background to my work with the Family Tree. The recommended tasks for the above person are certainly correct; this entry needs some data corrected before it can be submitted for temple work. I would suggest that such notices be very carefully reviewed and researched.

Even more interesting is the following fact.

One of the ordinances was completed for Mrs. ? Jensen in 1960, apparently without even a given name. 

When I click the link to "View My Relationship," I see the following:

This is even more interesting because my Great-grandfather, Marinus Christensen, was the only male child in that particular family. He did not have a brother. "Jens Christensen," the name of Marinus' father, has one of the more common names in Denmark. There are thousands upon thousands of Jens Christensens.

Apparently, someone added a second wife to my Jens Christensen and five extra children. Even more interesting is the source added to Christian Marinus Jensen.

This shows that the birth occurred in Torslev, Dronninglund, Hjorring, Denmark in 1859. This is interesting because the last child listed for this second family was also born in Torslev in 1865 when my ancestor, Jens Christensen died crossing the Plains in 1866. Jens Christensen and his wife and children left Denmark in 1866 and came to America where he and one of his daughters died. We have a somewhat detailed account of their lives in a book written by my Great-grandmother. See
"Tanner 26 & 27: Jens and Karen Johannesen Christensen."
In short, this additional wife and five children are attached to the wrong Jens Christensen. If you examine the added entry, you can see that the marriage date for Jens Christensen to Kristen Marie Thomsen is listed as occurring in 1830 and the first child is listed as being born in 1853, twenty-three years later. The children of the added marriage were supposedly born in Raevholt, Torslev, Hjorring, Denmark. The Jens Christensen who lived in Torslev and was married to Kirsten Marie Thomsen, was a shoemaker and the reference to Rævholt, is to a house with that name.

I guess I can be grateful that the Family Tree program alerted me to this extra, wrong family attached to my ancestry, but I did not need to do all the research necessary to detach the family, which will now be done. By the way, I am not going to resolve the issue of the "?" person, since it is not my family line.


  1. This is to clarity what appears to be a common question coming up, not only for you but others on the Get Satisfaction board.

    The question mark (?) notation was added during the update to Family Tree that completed the separation from New Family Search to replace the previous “Unknown Name” entries in Family Tree. Every single instance I have see of a mother being ? has had the same explanation - the old extraction program.

    I have run into these in the Norwegian parish records I am working in currently. Prior to 1815 in those records, christening records generally recorded just the child’s name and the father’s name. The mother was not listed. When these records were extracted and sent for temple work, the children were sealed to their father by name and their mother as “mother.” As these record traveled from the temple to the IGI to New Family Search to Family Tree and to Family Tree as of June 2016, they ended up in one of three formats:

    Child with one set of parents, Father and ?

    Child with one set of parents, Father with no wife. The father will have a second couple relationship with wife ? with no children

    Child with two sets of parents, Father with no wife and ? with no husband.

    These Family Tree records will always have an associated IGI or historical record database record. Often these will have automatically been attached as sources with the notation “This IGI record was used…”

    I wrote an extensive explanation of how to properly handle these record on the Get Satisfaction boards which you can find by searching for “how to properly merge duplicate parents"

    Your example, however, is the first example I have seen that does not fit my theories. Looking at the Change Log for ?, she came from New Family Search just as a wife with no other information. All I can think of is that something happened to the record in New Family Search that removed her from her children and got her put with a husband. Probably a bad merge between two different Christian Marinus’s. That will be impossible to track down. I highly doubt that in 1960 you could get a man sealed to an unknown wife with no marriage date and no other evidence of family relationships. To straighten this out, one would need to find out who was really involved in that supposed sealing between Christian Marinus Jensen and ?.

    (As a side note, it looks like the children you had to remove from your family got there when someone on 19 Oct 2012 merged your Jens Christensen with another Jens Peder Christensen, probably just because they had similar names and lived in the same part of Denmark. Going back in the Change Log and Restoring Jens Peder Christiansen would reconstitute that other family as it originally was, preserve someone else’s work, possibly prevent unneeded duplication of temple work, and would be polite.)

    1. Thank you so much for your extensive explanation. I am not sure how I am supposed to know to go to GetSatisfaction when I find something like this in the Family Tree or how I would know to look for information on merging to answer this question. The history here shows that people have been trying to add entries for Jens Peter Christensen to our Jens Christensen since 2012 and regularly after that point in time. Part of the problem apparently comes with Record Hints that suggest that they are the same person. There does not appear to be any way to unmerge the event on 19 October 2012 without causing a cascade of problems that have been corrected and re-corrected constantly since then. The children were added one by one by different people beginning in 2014. More wrongly added children have been removed since earlier this year. Perhaps, I need to know more about merging, but I do not see a way to reverse the merge that occurred back in 2012 at all. This all occurred with a note from FS that I had a temple opportunity for ?

  2. Unfortunately, I think that the Get Satisfaction board doesn’t get as much promotion as it should. It’s a great place to ask questions, answer questions, post bug reports, complain, and share experiences with other family tree users and with Family Search staff. But if you don’t already know about it, then you won’t know to check there. A number of people new to the discussion board were quite surprised to find themselves there and getting dozens of responses to what they thought was just a simple feedback to Family Search. I hang out there a lot because I really enjoy Family Tree and have discovered a lot of fascinating things about the program by trial and error. It’s the main avenue I have right now to help other people by sharing what I know. I get to feel like I’m doing a least a tiny part to help out in this huge project that is Family Tree.

    Regarding the second part of your response to my reply, you can pretty easily restore that poor family that was lost into yours and help out those poor kids that are now parentless.

    1) Go to your Jens Christensen KWVS-W3D and proceed to his change log.

    2) Scroll down to the 19 Oct 2012 merge.

    3) Click on deleted person Jens Peter Christensen to jump to his page.

    4) In the top box, the one labeled “Person Deleted,” over on the right hand side, you will see “Restore Person.” Click on that. The record for Jens Peter Christensen 2Z36-WPD should be restored. (If his original wife was also deleted, you might get a message that he can’t be restored. If so, find the wife and restore her first.) His record will be recreated and all the children that were removed from him during the merge will be restored to him.

    This procedure has absolutely no effect on your Jens Christensen or on any of the corrections made to him since the merge. It would not have removed the extra children from Jens and you would have seen that the children then had two sets of parents. In other words, you would have still had to gone in and removed them from your Jens like you did.

    1. Thanks, that worked. I guess I learn something new every day. I still have to ask the question however, as to how the average user of the program can possibly know all these details of how the program works?

  3. This comment has been removed by the author.

  4. (I don't know if this second part came through properly. Please delete if a duplicate. Or use this one - it has minor revisions.)

    Finally, I have a little bit to say about the trouble you have had with so many wrong children being added to your family. It basically comes down to two things. People don’t always evaluate what they see and the extraction program confuses the possible duplicates routine.

    The extraction program was a huge blessing because it created massive indexes in which we can easily find our relations. I am always amazed at how much of my wife’s Norwegian relations are already in Family Tree. True, they are in bits and pieces but are easy to find and link together.

    The extraction program can be a huge burden because the birth records from it do not include any information about the parents except their names and the Family Tree Possible Duplicates routine does not do a very good job looking at children connected to a record. For example, I have merged in the extraction record for the spouse of one of my wife’s great aunts. I have not pursued more research on him at this point. So his parents are attached to him just as Ole Olsson and ingeborg Knuntsdatter with no other information. Ole has 75 possible duplicates. Ingeborg has 58 possible duplicates. I can guarantee you that none, absolutely none, of them are correct. Just the other day my Watch List warned me of a change to Ingeborg. Someone had merged her with two other Ingeborgs. Looking at their children, it was easy to see that these three Ingborgs lived in Hordaland, More og Romsdal, and Buskerud, far from each other. But whoever was doing the merging didn’t look. I undid the merges, messaged the person responsible, and he apologized.

    I’ve found that the most effective way to prevent these erroneous merges, even better than declaring them not at match, is to add information. I am sure that If I spend a couple of hours on Ole and Ingeborg and put on birth dates and places, their possible duplicates would immediately drop from dozens to just a couple. If I then spend a little time researching those few remaining incorrect possible duplcates and add information to their records, the possible duplicates would drop to zero.

    1. Exactly. I am very much aware of all these issues. But, as you note, the number of possible duplicates in some cases is very large.

  5. Scandinavian records - what a nightmare! I used to get notifications from Get Satisfaction all the time, but have not since the nFS separation from FT. I have gone back in to GS and re-registered, but to no avail. So instead, I use Yammer.

    1. Scandinavian records a nightmare? I have to completely disagree! They are wonderfully complete and comprehensive. And the names! Where else do you get the person, the father and the current residence all in one? Brita Johannesdatter Ådland - you know her name, you know her father was Johannes, and you know she is currently living at Ådland. It's great! You just have to do what James has stressed over and over here in his blogs - search locations, not names. (And don't trust the possible duplicates routine in Family Tree. It can give you the greatest matches possible but also the most horrible.)

    2. I agree completely with Gordon. You also get the birth date on most of the records including the month, day and year of the birth and the christening and when and where the people moved. We love Scandinavian records. But the idea is the exact location down to the farm to get the right people. Names are not the key except when you have the location.

    3. I agree that Scandinavian records are generally very complete and accurate, but the challenge is that so many people have the same name, so you do have to be careful to check place names and family relationships to ensure accuracy. The nightmare comes when you have to clean up on Family Search after people who assume same name=same person and add incorrect information or relationships. Using the Watch feature and adding sources seems to help.

    4. This is where the location of the family is the most important factor in determining relationships. Household exams are very helpful in determining the makeup of the families and differentiating families with the same names at the same place and time.