I found a fun video from Family History Fanatics listing seven common genealogy mistakes. I’ve made every single one of these mistakes. Except maybe this one. Or, rather, I’ve seen this advice many times, and I made a conscious choice not to follow it.
The consensus seems to be you shouldn’t put a woman’s married name in the “surname” field in your tree.
There’s a solid point here: it’s extremely rare for a woman’s maiden name to be the same as her husband’s, so putting it there may get you poor search results. It can even lead to poor assumptions.
Here’s an example I discovered: a commonly accepted genealogy of the Chew family shows that a pair of brothers, Thomas and Joseph Chew, were born to Anna Mariah and Andrew Chew. It’s all based on 1850 and 1860 census records, where Anna Mariah is living with her widowed daughter in law, Amelia, Joseph Chew’s wife. Every tree out there has Amelia’s surname recorded as Chew. Including mine.
But my tree is different from everyone else’s, because I found Joseph and Amelia’s marriage record, which listed Amelia’s maiden name as Chew. Which means than Anna Mariah may have been Amelia’s biological mother, rather than her mother-in-law.
If you follow the practice of putting married names as a woman’s maiden name, you might miss important details like this.
But if you leave that surname field blank, you’ll have another problem: lots of women with the same first name and no other obvious identifier. That’s a problem when you try to add a source & fact—you might have a dozen Elizabeths, and you can’t recall her years of birth and death off the top of your head. I mean, I have over 150 Elizabeths in my tree.
What I do when I don’t know a woman’s maiden name is put her married name in parentheses, and add “nee ?” in the suffix field. It makes her easier to find, since I at least have a surname. And in case someone else doesn’t understand my parenthetical notation, the “nee ?” should make it more obvious. I’ve seen other people put a Mrs. in the suffix field as well.
Still, I’m sorta baffled why one of the family research companies hasn’t added a feature to handle this declaratively. Give me a check mark that indicates I don’t know a woman’s maiden name. When that’s set to true (which would be the default), the surname column is disabled in the UI, and there’s an indicator (e.g. nee unknown) in search results and the like. It wouldn’t be that hard to retroactively assign the value: if the woman’s surname name doesn’t equal her husband’s surname (trimming whitespace, ignoring case, and normalizing rare characters such as vowels with umlauts), set it to “known”, otherwise “unknown.” But you could have a tri-state as well, with a null or grayed out checkbox reflecting legacy values.