You find your ancestor in the 1850 census, and he’s 32 years old.
Great. That gives you a birth year of about 1818.
Then you find him in 1860. He’s 39.
Okay, maybe 1821.
In 1870, he’s 47.
Now we’re at 1823.
His death certificate says he was born in 1819, and his gravestone says 1818.
At this point, the man has managed to be born in four different years.
Welcome to genealogy.
Changing ages are so common in older records that they really shouldn’t surprise us. They still do, especially when we’re trying to decide what birth date to put in the family tree.
The mistake is assuming that one of those records has to contain the correct answer.
Maybe it does. But instead of choosing the date that appears most often, I want to know where each age came from.
Who Actually Knew His Age?
We tend to give records more authority than they deserve.
Take a census. If it says John Smith was 42, it’s easy to picture John standing at the door telling the census taker, “I’m 42.”
We don’t know that he did.
His wife may have answered the questions. An older child may have. In some cases, the information could have come from someone else entirely.
Now compare that with a baptismal record made a few weeks after John’s birth. His parents were there, the event had just happened, and the record was created at the time.
I’m going to give that record more weight.
That doesn’t mean an early record is automatically correct or a later record is automatically wrong. It means I look at who probably knew the information and how close the record was to the event.
Death certificates are a good example.
A death certificate is usually excellent evidence that someone died on a particular date and in a particular place. It was created because the person had just died.
But the same certificate might say the deceased was born seventy-five years earlier in Scotland to parents whose names were supplied by a son-in-law who never met them.
Same document. Very different quality of information.
Before You Accuse Grandma of Lying
Yes, people lied about their ages.
They also forgot them.
And clerks made mistakes. Census takers misheard things. Handwriting was difficult to read. Indexers misread handwriting. People guessed. Ages were rounded. Someone answering questions for the household may not have known everyone’s exact age.
Then there are calculated birth years.
If a record says someone was 52 in 1880, we immediately subtract and get 1828. But depending on the date of the record and whether that person’s birthday had occurred yet, the actual birth year could be 1827.
That’s why I don’t get too excited when two records are off by a year.
Five years gets my attention.
Ten years gets a cup of coffee.
I also try to look at the original image whenever it’s available. If an index says someone was 31, I want to see whether the person who wrote the original record actually wrote 31.
Indexes are wonderful tools for finding records, but they aren’t substitutes for the records themselves.
Put the Ages Next to Each Other
When things start getting messy, I make a timeline.
Nothing fancy. I just want to see everything together.
Maybe Sarah’s records look like this:
| Record | Approximate birth year |
|---|---|
| 1850 census | 1822 |
| 1860 census | 1821 |
| 1870 census | 1822 |
| Marriage record | 1821 |
| Death certificate | 1818 |
| Gravestone | 1818 |
Now it gets interesting.
The death certificate and gravestone agree on 1818, but several earlier records cluster around 1821โ1822.
I wouldn’t choose 1818 just because it’s carved in stone.
Gravestones are made by people. People can be wrong in granite.
I’d want to know who gave the information for Sarah’s death certificate. I’d check when the gravestone was erected. Then I’d start looking for records created closer to her birth: a baptism, church register, family Bible, or something involving her parents or siblings that might help narrow it down.
I might eventually prove she was born in 1821.
I might not.
And that’s okay.
“About 1821” Is a Perfectly Good Answer
I think this is one of the hardest things for new genealogists to get comfortable with.
You don’t have to fill every box with an exact date.
If all the evidence points to a birth sometime around 1821, put about 1821.
If you can narrow it only to a range, use the range.
You aren’t failing to finish the research. You’re recording what the evidence currently supports.
I’d much rather see an estimated birth year backed by five records than an exact date copied from an unsourced online tree.
You can always change it when something better turns up.
Don’t Make the Record Match Your Tree
This is another temptation.
Your research says Martha was born in 1843. The census says she’s 35 in 1880.
Don’t quietly turn the 35 into 37 in your notes because you know that’s how old she “should” have been.
Keep what the record actually says.
The discrepancy is part of the evidence.
The same goes for names, birthplaces, occupations, parents, and relationships. When a record disagrees with our research, we don’t fix the record.
We figure out why it disagrees.
And every now and then, the record we thought was wrong turns out to be the clue that tells us we had the wrong person all along.
That’s a much bigger problem than Grandma shaving three years off her age.
Let the Mess Be Messy
Genealogy would certainly be easier if everyone knew exactly when and where they were born, spelled their names the same way, married only once, stayed in one county, and personally checked their death certificates for accuracy.
Our ancestors were inconsiderate that way.
So when one census says 1818, another suggests 1821, and the gravestone announces 1819, don’t immediately decide which one is “right.”
Write them all down.
Look at when each record was made. Figure out who supplied the information when you can. Look for records closer to the birth. Compare the person’s age with the rest of the family and the other events in their life.
Then use the date, or range of dates, that the evidence supports.
And if the best answer you have is born about 1821, that’s a good genealogical conclusion.
Besides, there’s always a chance your ancestor will turn up in the next census two years younger.
Because apparently aging was optional in the nineteenth century.
You must be logged in to post a comment.