The title of this week's post comes from a quote from a 1982 Frank and Ernest cartoon: "Sure, [Fred Astaire] was great, but don't forget that Ginger Rogers did everything he did... backwards and in high heels."
The genealogy version of this statement would be a little different: "Fred Astaire tried to move forwards, rather than backwards? What an idiot. Seriously, how did he even remember to breathe?"1
Genealogy is one of those things in life -- like plotting a story or cheating at solving a maze -- that simply must be done backwards. When I say "simply must," I don't mean it in the strong suggestion made by an early 1900s socialite sense, "Oh, you simply must try the pâté," I mean it in the absolute necessity sense. Put plainly: in genealogy, if you don't work backwards, you will get stuck and you will make mistakes.
When I speak of working backwards, I'm referring to two specific motions. The first of these motions is that you always begin with yourself and then move back to your parents, then to your grandparents, then to your great-grandparents, and so on. This particular backwards motion allows you to start with what you already know.
The second motion is that, within a given ancestor's life, you should start your directed research at their death and trace backwards to their birth. This allows you to create the necessary linkages in order to "prove"2 that you've followed the correct person back in time. Proving these three linkages -- a husband to a wife, a child to his parents, and a child's birth to his marriage -- is the hallmark of good genealogical research, and is made much simpler by going backwards. It is also the only kind of backwards motion I am capable of without falling over.
"But, wait!" I hear you cry, "Why should I waste my valuable time proving my connection to my parents or my grandparents' connection to each other? I know who they were, I know they were married and all that, I just want to learn more about my ancestor who fought in the Revolutionary War! Why can't I just start there?"
Well, the problem with starting there, even if you know the name of that Revolutionary ancestor, is that, without those established linkages going back every generation between you and that soldier, you don't know that you're not actually researching someone else's ancestor who happens to have the same name as yours (this happens a lot more often than you might think, even if the name sounds like an uncommon one to our modern ears). You can't just do an Ancestry search for "T. Gaillard" and assume that whoever pops up is your ancestor, because I can promise you that not only was there more than one T. Gaillard in the Revolutionary War, there was more than one T. Gaillard in the Revolutionary War from South Carolina alone (their first names were Theodore and Tacitus, in case you were wondering). So why take the risk of spending that valuable time you cited researching someone who turns out to be utterly unconnected to you?
Wait, are you still wondering about the name Tacitus? Let's move on now.
In addition, one of the advantages of starting with death records is that they can be incredibly detailed and give you a huge amount of genealogical data about the person in question. Take a look, for example, at the obituary below:
From The Canadian Jewish Review, 22 February 1952, page 11, column 3. Online at http://newspapers.lib.sfu.ca.
Take a look at the wealth of what we've learned from this one obituary -- which we can presume was composed by someone who knew him well, so the information is likely relatively accurate. We've learned: his address, his date of death, his age at death (giving us a likely birth year), his children's names and locations, his wife's name and year of death, his place of origin and year of emigration, and organizations in which he was involved. As we move backwards towards his birth, we can use what we've learned from this obituary: if we go looking for a census record, for example, we can cross-check by using the names we've learned from the obituary to confirm, say, that this 1921 Canadian Census record refers to the right guy:
So that's him, his wife, and five of their children in Toronto in 1921, already living on Bathurst Street (as mentioned in his obituary). Fool-proof. Whereas, if I'd simply done a search for Jacob in the census without having the background information on him that we gained from his obituary -- if all I knew about him was that his name was Jacob Cohen and that he lived in Toronto -- then I'd have over 50,000 results on Ancestry.com to dig through with no real way of narrowing it down to the correct guy. Let's throw me a bone and say that I also knew that he was born around 1862, and added that information into my search; I still have almost 25,000 results to peruse in increasing frustration.
See why it's good to move backwards now? Are you still thinking about the name Tacitus? I agree it's poised to make a comeback, but let's stick to the topic at hand.
Now, let's imagine that you've done everything right: you've moved backwards through the generations and through each person's life individually, and you've suddenly come up against a seemingly immovable brick wall. The trail just stops cold. In that case, you need to pause your backwards motion (like it wasn't stopped dead anyway) and take a good-sized step sideways to examine that person's siblings, or aunts and uncles, or even friends (we call this the FAN Club principle, with FAN standing for Family, Associates, and Neighbors). Or, as I would have put it back in my Rocky Horror days, "...And then a step to the riiiiiight!"
Seen here: proper genealogical technique.
Let's look at an example to demonstrate what I mean.
The gentleman in question here is Ernest Louis Huff (1854-1910). The family story goes something along these lines: he was born in Germany or Switzerland on 15 May 1854 to an unknown father and an unmarried charwoman by the name of Graf; he was then given to his mother's sister and her husband to raise. That uncle and aunt -- a man and woman by the names of Joseph Huff and Louise Graf -- brought him to America. Joseph died young, and Louise remarried a man by the name of Reiber in Ohio. Ernest stayed in Pittsburgh and, yadda yadda yadda, got married, had children, and lived out his life. My task was to attempt to find out if the family story of his birth was true, and to see if I could learn anything about his parents.
I was able to locate records for Ernest here in America without too much trouble: a death certificate, three census enumerations, his gravestone, and a whole heap of city directory records that traced his movements from the mid-1870s all the way to his death from pneumonia in 1910. All very well and good, but none of these records got me any closer to linking him to his parents or to his birth. His death certificate named his parents as Joseph Huff and Louise Graf and place of birth as New York, and the relevant census records gave his place of birth alternately as New York, Switzerland, and New York again. His obituary also listed his place of birth as New York. However, New York (both city and state) had no record of his birth. I wasn't able to find any immigration records either, though. It was as though he simply sprang into being as a 20 year old man in an 1874 Pittsburgh City Directory. I was up against a deeply frustrating lack of documentation.
Luckily, I wasn't at a complete loss because I still had one chance left: his aunt, Louise Graf. Once I did a census search for her using her second husband's name, she showed up immediately in an 1880 Census record in New Castle, Pennsylvania:
It may be small and hard to read, but look at the last column giving her place of birth: Baden. Of course, not being in any particular hurry, I confirmed that this Louisa Reiber was the same woman who had been married to Joseph Huff (died in Mahoning County, Ohio in 1876, the same county where Louise remarried Conrad Reiber in 1877, as they were fresh out of gentlemen suitors named Tacitus), and then, naturally, the same woman who raised Ernest L. Huff. Her obituary (look, another helpful death record!) was the clincher:
From the New Castle [Pennsylvania] News, 30 November 1898.
Married to a man named Reiber? Check. Adopted son named E.L. Huff who lived in Pittsburgh? Yahtzee!
So, based on her given birthplace in the 1880 Census, I did a search for Louise Graf using the search parameters of Baden as a birthplace and about 1826 as a birth date, and looky, looky, I got Hooky:
It certainly seemed promising. The proof, however, would be in the pudding. I did a similar search for Ernest Louis Graf (assuming that, if he were indeed illegitimate, his mother's surname would be prominent) with appropriate search perameters: Baden as the birthplace, and 1854 as the year of birth. I hit the search button and... I'm running out of ways to say et voila!
It was he!3 He was two months off precisely on the date of his birth (actually a remarkably small amount for those days, given that most people seemed to have no blessed idea of how old they were), he had Americanized and reversed his name (can't say I blame him), and the indexer here has misspelled his grandparents' names a bit, but this is clearly the nephew of Louise Graf, based on the fact that his grandparents are her parents, but she isn't his mother. The record also confirms the family story about Ernest's illegitimacy, as the only likely reason that his maternal grandparents, rather than his father, were listed on his birth record would have been that there was no known male figure to claim him. The precise identity of the mysterious Gela Graf isn't yet known, though I do have my suspicions about that, based on further research into the family; odds are that the identity of his mysterious father will never be known.
My larger point is this: had I not taken that step sideways to focus on his aunt for a bit, I never would have come up with a birthplace of Baden -- after all, Ernest seems to have been running around, saying that he had been born in New York -- and thus never would have found a record documenting his birth. He would have remained a frustrating character with a wishy-washy footnote regarding his unknown origins. Now he's a less frustrating character with an exact birthplace, known grandparents, a mother who used a fake name to register her son's birth, and a father who may have been anyone from local nobility to the village Tacitus. Well, hey, it's a start!4
Pictured: another Tacitus. That is all.5
1 No insult intended to the late, great Fred Astaire. The man was clearly poetry in motion.
2 I use quotation marks here because genealogical proof is an ever-shifting concept. A fact can be proven in genealogy only insofar as it is the fact that is currently supported by the known evidence. New evidence coming to light may completely change the story.
3 Keep in mind that, as a genealogist, you should never be satisfied with indexed records. Anything that creates a layer of separation between you and the original record is merely a clue to follow and an obstacle to be overcome. You never know when an indexer will make a mistake. Always make the effort to get your hands on the original. That said, the indexed versions are much easier to read and are in English, and so, for the purposes of this blog, I felt it appropriate to use them as my visual aids.
4 In this particular case, my recommendation would be that the client take a DNA test. The unknown father in question would be the client's great-great-grandfather. It's possible that the client would have third cousins out there who could also be descended from the same sperm donor (or fourth cousins, if descended from the donor's parents). If any of those third or fourth cousins have also taken the same test and know their antecedent's name, the identity of this man could potentially come to light.
5 Actual citation: Gaius Cornelius Tacitus by an unknown illustrator, based upon an antique bust. Bryce, James Bryce, Viscount; Thompson, Holland; Petrie, William Matthew Flinders, Sir. The book of history. A history of all nations from the earliest times to the present, with over 8,000 illustrations. Volume 7: The Roman empire. (1920). New York: Grolier Society. p. 2741.