A Theory of Relativity to Drive You Crazy

Have you seen the meme of Albert Einstein drawing his family tree on a blackboard? He says to himself, "So that would make my second cousin once removed the great aunt of my first cousin twice removed…no, wait, that can't be right."

You know what? He's actually onto something. If you follow the rules, and ignore your known connection to someone, things can get weird.

My father's parents had the same unusual last name—Iamarino. We never gave it much thought. But once I got interested in genealogy, I wanted to know if they had a connection. It turns out my grandparents, Pietro and Lucy Iamarino, were 3rd cousins.

The dotted line around one instance of Francesco and Cristina tells me they're in my family tree twice.

Pietro and Lucy's shared 2nd great grandparents appear twice in my ancestor chart. They appear on Grandpa Pietro's line and on Grandma Lucy's line.

If we apply relationship rules to my family, Grandpa Pietro's 3rd cousin (Grandma Lucy) is my 3rd cousin twice removed (3C2R). My 3C2R's son (Dad) is my 4C1R. And I'm my own 5th cousin!

A Scientific Experiment to Prove the Theory

To test out this idea, I searched my family tree for another one of Grandpa Pietro's 3rd cousins. I chose a different set of his 2nd great grandparents. Then I clicked my way down 5 generations to one of their 2nd great grandchildren, Domenico.

Family Tree Maker correctly tells me Domenico is my 3C2R and his daughter is my 4C1R. So, if you put aside that fact that Grandpa Pietro married his 3rd cousin Grandma Lucy, Lucy would be my 3C2R. And her son (Dad) would be my 4C1R. Dad's children (like me) would be my 5th cousin. And my kids would be my 5C1Rs!

And that's when Albert Einstein's mind was blown. Don't you think "I am my own 5th cousin" should be my new email signature?

Cousins marrying cousins adds a new level of mind-blowing relationships to a family tree.

Applying the Lesson to DNA Matches

Aside from the fact that it's funny, I bring this up for a more practical reason. We see that different interpretations of a relationship can co-exist. Doesn't this help make sense of some DNA relationships?

I'm looking at one DNA match in my list. Ancestry DNA says she's most likely my 4th–6th cousin. But I know where she fits in my family tree, and she's my 3C2R. There's another, closer relative in my list. I know she is my 3rd cousin, but again, Ancestry DNA says she's most likely my 4th–6th cousin.

We know DNA passes down randomly from generation to generation. Even full siblings can have very different mixtures of their shared ancestors' DNA. Each of your DNA matches has a laundry list of possible relationships to you.

Look beyond the labels your DNA website uses to the groupings they use. Ancestry DNA groups your matches as:

• Parent/Child (both my parents tested, so I see them in this group)
• Close Family (this group includes my 1st and 2nd cousins, plus Dad's 1st cousin and her children)
• Extended Family (these range from Mom's 2nd cousin to Dad's 3rd cousin, and even a 5C1R with multiple connections to me)
• Distant Family (this is where I see the 3C and 3C2R I mentioned above, plus every other relationship under the sun)

Many people in your family tree can have multiple relationships to you. What really matters to you is where they fit in the tree. Don't get hung up on the labels.

Do you have ancestors who married a cousin? What does that make you to yourself?

And speaking of crazy relationships:

• Untangling Our Twisted Family Relationships
• Why DNA Matches Appear Closer Than They Are
• The DNA Problem We Aren't Talking About

Another Way to Find Errors in Your Family Tree

I've added so many people to my family tree this year! I synchronized my Family Tree Maker (FTM) file with Ancestry on Sunday morning. It said I'd added 310 people the day before. That's a new record! I decided it was time for a thorough error check.

FTM has a built-in error report, and I wanted to compare it to that of Family Tree Analyzer. The differences surprised me. The second listing I saw on my FTM error report was for Harold Gibbons. He had a duplicate birth fact that Family Tree Analyzer didn't see.

When I took a look at Harold Gibbons in my tree, I saw both 22 Sep 1899 and 26 Sep 1899 listed as his birth date. One date came from an index of New York City births. The other came from a World War I draft registration card.

The birth index said Harold's 1899 Manhattan birth certificate number was 37387. The NYC Municipal Archives has digitized their vital records, and they're available online. So I checked to see when Harold, my cousin Rod's uncle, was really born.

This report gives a different account of the errors—or possible errors—in your family tree.

When I saw his birth record, his name showed my first problem. The certificate says Harold T. Gibbons. (Why didn't they spell out his middle name?) The WWI draft registration card I'd saved for him says Harold Patrick Gibbons.

I checked the parents on the birth certificate to see if they were a match. Yes: John Gibbons and Lillian Lanigan are the parents I expected to find. The certificate shows the date of birth as 26 Sep 1899—that agrees with the NYC birth index, but not the WWI draft card.

Now I knew that draft card belonged to another man. Ironically, I had researched the wrong Harold's place of work. I even included a photograph of the building and a description of the business.

I deleted the draft card, building photo, and the facts for the wrong Harold. Now the right Harold's birth certificate is there to document his date and place of birth.

And that was only the first item I checked from the error report.

How to Create Your Error Report

If you use Family Tree Maker:

• Click the Publish tab at the top of the program.
• Click Person Reports in the left column and choose Data Errors Report.
• Click Create Report, then click Cancel to make some enhancements:
  • Choose to include All individuals.
  • Click the first button under Data Errors Report Options to open the Errors to Include dialog box. I chose to deselect two choices:
    • Spouses have the same last name (so what?)
    • Marriage date missing (that's because the document is not available)
• Close the dialog box, click Generate Report, and wait.

Be patient if you have a big family tree. Go have some tea and cookies.

Family Tree Maker has a built-in error report that may surprised you with its findings.

My report showed a ton of duplicate marriage bann errors, but that isn't an error. I always record two marriage banns for marriages in Italy. That's their marriage process. I wanted to remove these entries from my report. I needed a spreadsheet. I clicked the Share button in the top right corner of Family Tree Maker and chose to Export to CSV file.

Open your CSV file with any spreadsheet software and sort it by error type. Then delete any lines with errors you know you don't need to fix. Then jump in and start checking errors. Delete each line you review/fix, and whittle down the number of errors to check.

If your list is really long, don't get upset. Some items will be non-errors. For example, I see I have a bunch of possible duplicate names. Some documents list a person by different names, and we want to note that. I expect to keep those.

I'm actually happy to see a group of errors that look like this:

Possible duplicate event: Name
Possible duplicate event: Sex
Possible duplicate event: Death
Possible duplicate event: Birth

These duplicates happened in 2019 when my FTM file suffered a disaster. I fix these duplicates whenever I see them, but now, finally, I can get rid of them all.

Use the error report in your family tree software and find mistakes you never knew were there.

Genealogy Obsession Pays an Unexpected Dividend

I'm obsessed with my massive genealogy project. Connecting everyone from my ancestral hometowns is all I want to do! I've improved my process along the way, and today my tree has 50,000 people. (See my more efficient technique below.)

When I write about this project, some people say, "I wish I could do that, but the vital records aren't available." Others say they're now doing the same thing, and all the connections are astonishing.

How many people from one town are somehow related? It's not just an obsession. It's a legacy.

Building an 18th–20th Century Foundation

When I add a person to my tree from the 1880s–1900s, I know they're someone's grandparents. That made me realize my project makes it easier to figure out my connection to distant DNA matches.

If you have a DNA match with a very small family tree, you may not see much more than their grandparents' names. I used to make an effort with these matches but not get very far.

Now I'm in a much better position to figure out my connection to a DNA match's grandparents. This weekend I scrolled through my match list, looking for those I hadn't figured out.

Simple notes make it easy to navigate your DNA match list.

One after another, I found their recent ancestors in my tree, and I saw our connection. I add notes to my matches that appear on the main DNA match list on Ancestry. I can scroll down the list and see who needs more research. This weekend I added new notes, like this:

• his 1G Maddalena Iamarino is my 3C3R, common ancestors are my double 5Gs Giovanni Iamarino and Libera Pilla
• 5C thru shared 4Gs Giuseppantonio Basile and Maria Maddalena Tedesco
• 3C descendant of Antonio Pilla and Angelina Iarossi, common ancestors are my 2Gs Gennaro Pilla and Maria Giuseppa Liguori
• her 1G Gennaro Finella is my 3C3R, common ancestors are my 5Gs Giuseppe d'Emilia and Orsola Mascia

Some matches helped me see which of my distant cousins came to America and who they married.

My all-consuming genealogy project is bearing useful fruit!

Letting the Documents Lead the Way

Here's an overview of my process and how I made it work even better.

I started with my Grandpa Iamarino's hometown of Colle Sannita. Vital records for the town are available online on the Antenati website (see "How to Use the Online Italian Genealogy Archives"). They have:

• Birth, marriage, and death records from 1809–1860 except for 1859 deaths and marriages
• Birth records from 1861–1904 except for 1875
• Birth records from 1910–1915 except for 1911
• Death and marriage records from 1931–1942 except for 1939 deaths

That's a total of 225 types of records and more than 38,000 document images.

My first step after downloading all the files was to:

• view each document and
• rename the jpg file with the name(s) of the subject(s).

An image named 007853875_00496.jpg now contains the names of a baby and its father:

007853875_00496 Carmine Pasquale d'Agostino di Giuseppe.jpg

The father's name ("di" means of in Italian) lets me search for all the children of any man, like Giuseppe d'Agostino. I can use a free program called Everything to search my computer for "d'Agostino di Giuseppe. (See "My Secret Weapon for Finding Relatives".)

This file-renaming process is the basis for building an entire town's family tree.

The file renaming process alone was quite a task! I renamed more than 38,000 image files for this town (and tons more for my other towns). Then I was ready for the BIG project.

I created a spreadsheet with the name of each file. I go line-by-line, viewing each document again, and trying to fit the person or people into my family tree. If they fit, I mark it in my spreadsheet. And if they don't fit, I mark that, too.

I went through several years' worth of documents this way. One problem came up again and again. Some townspeople went by their middle name, making them hard to find each time they had another baby. So I made a change to the process.

If I'm adding an 1865 baby to a couple, I'll mark it on the spreadsheet. But before moving to the next line, I'll search for every other baby belonging to this couple. And if their kids' birth records have a marriage notation, I'll search for their spouses. And I'll add any of their kids. Then I'll return to the next line in my spreadsheet.

This way, couples using unexpected first names won't stump me each time they have another baby. It saves so much time when I complete their families all in one go.

Another problem I overcame was searching for a set of parents only to discover their baby is already in my tree. That was a wasted search. Here's how I fixed that problem. Before I begin another year's documents, I sort my Family Tree Maker index by birth, marriage, or death year. Then I compare the spreadsheet to the index and mark off the people I already have.

Because I complete entire families at one time, each new year I review is already 75% complete.

At this moment I'm up to the 1868 births. I have 64 folders left to go out of the 225 available. When I add 20th-century people to my family tree, it gets easier to connect with more DNA matches.

When I do get to the bottom of the spreadsheet, I'll make one more pass. I'll re-review the people I couldn't fit into my family tree. They tend to fall into 5 categories, and I want to mark them as such:

• "Out-of-towners" who happened to have a baby or die in Grandpa's town.
• "Old people" who died too early for me to know who their children were, or to have their parents in my tree.
• "Too-common names"—This is usually the only child of a couple I can't ID because so many townspeople had the same names.
• "Foundlings" who died without marrying.
• "Possibilities"—These are people I may be able to fit into my tree after I've gone through all the documents.
  

About 95% of the people found in those 38,000+ document images have a connection to me. Towns in this area kept largely to themselves because travel between them was hard. And all my roots are in this area. I'll bet I can reach the same 95% connection rate with documents from my other ancestral hometowns.

Well, my retirement is fully booked. I'm in my happy place every single day. Where are you?