Check Your Assumptions at the Door
3 Sep 2017 by Leah Klocek
Humans are creatures of assumptions. We live our lives by making assumption after assumption, all day, every day. Many of these assumptions we make are generally safe assumptions, because after so many years of moving through the world, we've figured out the basic rules of how the world works through trial and error and repeated experience: the sun will rise tomorrow, rush hour on a weekday will always suck, the knives in your kitchen will be right where you left them, the stove is hot when turned on, and if you hit your little baby head on a granite table, it's going to hurt. These are the things that are empirically true, and would totally upend our world if the physical laws that govern them were to disappear.
Then there are assumptions we make that are maybe a little less safe, things that may be true the vast majority of the time, but all it takes is one time where it isn't true to give you a very bad day. This category includes things like assuming that all of the drivers around you will follow the rules of the road and drive safely, that you can walk in your driveway barefoot without stepping on something sharp, that the random mushroom you found in the woods is safe to eat, or that the guy you met at the library isn't a serial killer. These are things that are generally accepted as true. They're true 99 times out of 100, or even 999 times out of 1000, but it's that one time they're untrue that'll get you.
The same rules about assumptions hold true in genealogy. There are three types of assumptions you'll deal with as you're working on your family history: fundamental assumptions, valid assumptions, and unsound assumptions.
Fundamental assumptions are just what they sound like. They're the things we hold to objectively true, because biology only works one way. A fundamental assumption in genealogy would be that a person cannot act before their birth or after their death, that each child will have one biological mother and one biological father, that no person can be in two places at once. The laws of nature and of physics say that these things must be true (sure, electrons can be in two places at once, but human beings are not electrons, last I checked). You can safely assume these, and they will help you build your family history accurately and avoid those dandelions we discussed in an earlier post. If you've found an official death record for a woman from, say 1865, you can safely assume that she didn't have any kids after that (and that any family trees showing kids born well into the 1870s probably don't know what they're talking about). If your ancestor shows up in the Milwaukee city directory in each year between 1900 and 1910, there's really no reason to wonder if he decided to move to Green Bay for a couple of years to partake of the fine cheese culture. You can make these assumptions every time, as long as your underlying information is good.
Pictured: a different type of fine cheese culture.
Next up are valid assumptions. Valid assumptions are those things which we generally hold to be true unless proven otherwise. Genealogically speaking, a woman being capable of child-bearing between the ages of 12 and 49 would be a valid assumption, as would children generally arriving in a family at two or three year intervals. You can also usually safely assume that church records are a contemporary source of valid information, or that no one is going to go to a government agency and register the birth of a child who didn't exist. Valid assumptions exemplify Occam's Razor, the axiom that the simplest solution is often the correct one. If you're convinced that said child whose birth was registered in 1903 didn't actually exist, you would have to find a darn good reason why her birth would have been registered, or find proof that the person who registered her birth was mentally ill and had a habit of making up people who never lived, rather than just figuring that the reason you didn't know about her is that she probably passed away as an infant. If a woman who was otherwise prolifically fertile has a decade-long gap in between two of her younger children, that's probably an indicator that she suffered miscarriages or stillbirths during that time period, or that her husband was absent for a long period of time (possibly serving in the military or away for occupational reasons, or he left the toilet seat up way too many times in a row). Valid assumptions give you clues to follow up on: look for death records for that registered infant, or for those possible stillbirths; look for occupational or military records for that absent husband. These assumptions allow you to say, "hang on, that's a bit unusual -- perhaps I can find documentation of why that happened."
And finally, we come to unsound assumptions. Unsound assumptions may very well be valid, but require confirmatory evidence before we can accept them as true. In genealogy, nearly every assumption you make may be an unsound assumption: assuming that a mother and a father were married at least nine months before the birth of their first child (or married at all!), assuming that a man and woman of the same last name living together in a census record were husband and wife (rather than brother and sister), assuming that people died in their 60s, even assuming that people knew how to spell their name or how old they were... These assumptions are all unsafe to make and necessitate that you locate records that confirm the assumption. Otherwise, you may just be spreading those dandelions around again.
Let's look at an example of an unsound assumption to understand the dangers:
This past year, I was looking for more information about the family of Rudolph Roth. I knew very little about him at the time: I knew that he was born about 1896 in New York, and that he had at least one sister, whose name was something like "Rose." I went to Ancestry to start poking around, and this family in the 1910 United States Federal Census in New York City pops up almost immediately:
It certainly appeared promising: a widowed Sarah Roth as the head of the household, with four children, Joseph, David, Rudolph, and Rose. Rudolph's name, age, and birthplace checked out, as did the presence of a sister named Rose. I could very easily have made the unsound assumption that Sarah was the mother of these children, particularly as the census out-and-out lists them as her sons and daughter, and gone ahead with tracing their ancestry through her family line. However, I knew not to make that assumption, and the fact that I couldn't find the correct Sarah Roth in the 1900 Census left me wary, so I ordered Rudolph's 1970 death certificate from New Jersey in order to confirm or deny that his mother's name was Sarah.
And surprise, surprise, when the death certificate showed up, his parents' names were there, plain as day: William Roth and Amelia Klein. I was immediately able to find the family in the 1900 Census:
And once I had that, I was able to look at the two records together and see that the children in the 1910 Census were almost certainly the same children in the 1900 Census. After that, the story filled in very quickly: the children's mother, Amelia, passed away in 1904. Their father, William, remarried a woman named Sarah Rainitz in November 1905, and then passed away himself in December 1908, leaving his youngest children in the tender care of their stepmother. Suddenly, the time that Rudolph spent in the New York Hebrew Orphan Asylum in 1909 and 1910 made sense (these orphanages also served as a place for children with living parents who weren't able to take care of them, as well as boarding schools for kids like Rudolph, who maybe didn't get along so well with his stepmother and who was admitted to the Asylum within four months of his father's death), as did why he was enumerated as still living with his family in 1910, despite living in the Asylum at the time (shame and guilt are powerful motivating factors).
Pictured: a different Rudolph.
The way that the Roth children also got away from this woman as quickly as they could, despite this being well outside the norm at the time -- Marion left home before 1910, despite not being married until 1914, two of her siblings lived with her at various times during that decade, and all of them, young though they were, were away from Sarah by 1915 -- also made perfect sense. Incidentally, Sarah remarried in mid-1910, only two months after the 1910 Census was taken, which may also have sped their departure (her second husband died in 1914, and she disappears after that; my current unsound assumption is that she was some sort of black widow and moved on to her next victim, but I haven't proven it yet. You'll know when I do, because I will be shopping around a new screenplay.)
Imagine I had not known that I needed further confirmation of Rudolph's mother's name beyond just what a single record said. I would have completely missed this fascinating story, to say nothing of the fact that I would never have been able to find any additional information on the family!
Assumptions can be a very dangerous bedfellow. When they are fundamental or valid assumptions, they can be advantageous or give you clues to follow, but take heed that you don't get complacent. If you think that a certain man and a certain woman were married, don't just take it on faith; find a marriage record! If a family has three young children in the 1870 Census and still has those same three children in the 1880 Census, don't just figure that they only had three children; look for records of additional children who may have been born and died in the intervening years! If you find a "Rutger Johannessen" in one record and a very suspiciously similar "Roger Johanson" in another record, it might be worth checking if the two men are actually one and the same! Otherwise you run the very real risk of not only missing vital parts of the story (which, as we've previously discussed, is every bit as important as basic names, dates, and places), but also creating a totally unnecessary brick wall that will stop your research in its tracks indefinitely.
Building walls is often pointless and sometimes just needlessly politically inflammatory.