I recently decided to try applying the Genealogical Proof Standard (GPS) to some research I was working on to determine the parents of John McKelvey (1789-1847) of Ionia County, Michigan. The first requirement of the GPS is that a reasonable exhaustive search be done, so I worked on my search. In this case I didn’t have any information from John’s lifetime about where he was born, which was going to be needed to help narrow down where to look for birth documents. One of the methods I used to determine where he was born was to look at the census records for his children, and depending on the year see where they were born or where their father John was born. Since John died in 1847 his census entries did not help on the birthplace, but knowing where his children were born helped provide a chain of residences as I worked backwards, and in later censuses the form directly says where John was born.
The problem is that John McKelvey had 12 children. Going through each census record for each child is a real tedious task, noting the information and creating the source citations. And if I want to be real thorough, I should look at the later census records of the grandchildren, to see if the location they give for their parents matches where the parents themselves say they were born. This can easily end up at 100-200 records to track down.
I know what the answer to this should be, and that is that I should look up all of the records. If I stop after child two or three I might miss contradictory information from the children I skipped. In spite of that I still looked for a way out. I did some research to see if there were applicable guidelines for a reasonably exhaustive search. Michael Hait did a good blog posting on the topic, but the emphasis was on getting a variety of independent sources. I then read the article When Enough is Enough How Much Searching is “Reasonably Exhaustive”?” by Thomas Jones from the March 2010 APG Quarterly on what constitutes a reasonably exhaustive search. In it he gives six criteria for a reasonably exhaustive search, and number six applied to my situation. This criteria requires that the search include “All findable sources suggested by relevant sources or indexes”. In my case it is no problem finding census records, as I just have to go to Ancestry.com. There is no hardship here beyond tedium.
To learn more on the subject I tried one more avenue and that was to listen to Michael Hait’s webinar on the reasonably exhaustive search at Rootsonomy.com last week. Like Michael’s article, it covered trying to find more independent sources. During the question time I really wanted to ask if there was a way around my problem but expected the answer to be something like: “You have the sources right in front of you and you don’t want to look at them?”, so I kept quiet.
In the end I went through all of the children who weren’t born in Michigan. They were all born in New York according to their census entries, which was consistent with biographical information from several books about Ionia County. The census entries listing their father’s birthplace also all said New York, except for one that said Scotland. This gave me a contradiction to resolve, but in a backhanded way may confirm the truth. Through other sources I traced John’s birth back to the New Scotland, New York area where he was baptized. So while John McKelvey is likely Scottish, the child could have misunderstood the birthplace. Or, of course, the informant could have just been wrong.
I did not however, take the additional step of reviewing the grandchildren’s census records to see if there was conflicting information about their parent’s birthplaces. I don’t have a good excuse, but sources from several directions were already putting John McKelvey in New York, and even if the grandchildren had different birthplaces for their parents I wasn’t going to give that information much credence.
I did do a couple of things to make the task a bit more tolerable. Since this is personal work I did most of it while watching TV, and I used templates to build the census source citations. Since the second requirement of the GPS is good sourcing there are no shortcuts on the citations.
My next conundrum is determining if supporting facts in a GPS proof need their own GPS. For example, I wanted to prove who the parents of John McKelvey were using the GPS, but I happily used census records, biographies, and death certificates to determine who his children were. These are all helpful sources and I’m pretty sure they are correct, but I did not use the GPS to prove that the children I found were actually his children. A chain of reasoning is only as strong as its weakest link, but sometimes common sense is necessary in balancing out ones time.
So my lesson from all this is that the GPS can be a lot of work. Lots of proofs require less work than this one so it isn’t always this bad, but being thorough is real important. It is better that you find the contradictions in your argument than having someone else find them.