By Don Taylor
Raidt Project
Raidt-Gleason Line

Every genealogist runs into it eventually: a person whose birthdate shifts depending on which record you’re holding. My son’s 2nd great-grandfather, John Michael Gleason, is one of mine — and working through his conflicting sources turned into a good case study in why some records deserve more weight than others.
For years, my working birthdate for John was 25 December 1852, in Watertown, Dodge County, Wisconsin — taken from his Find-a-Grave memorial. Then I pulled up his FamilySearch profile (LRNN-6Z8) and found something different: a life span of 25 December 1853 – 5 December 1948, and a birth location of New Diggings, Lafayette County, Wisconsin — a different year and a different county. That discrepancy sent me back through everything I had.
Starting with the Raw Data
Here’s every source I’ve attributed to John’s birth so far:
| Source | Source Date | Age / Detail | Confirming Info | Suggested Birth | Location |
|---|---|---|---|---|---|
| 1870 Census | 1 Jun | 18 | (Apparent) son of Timothy & Bridget | 2 Jun 1851 – 1 Jun 1852 | Wis. |
| 1875 MN Census | 1 May | 21 | — | 2 May 1853 – 1 May 1854 | — |
| 1880 Census | 1 Jun | 28 | Son of Timothy & Bridget | 2 Jun 1851 – 1 Jun 1852 | Wis. |
| 1885 MN Census | 1 May | 33 | Wife is Della; children John & Mary | 2 May 1851 – 1 May 1852 | Wis. |
| 1895 MN Census | 1 Jun | 44 | (Della died 1893) Children: John, Mary, Caroline, Clara, Helen | 2 Jun 1850 – 1 Jun 1851 | — |
| 1900 Census | 1 Jun | 47 | Born July 1852 (stated directly) | July 1852 | — |
| Find a Grave | N/A | 95 at death | Spouses & children consistent | 25 Dec 1852 (no source given) | Watertown, Dodge Co., Wis. |
| FamilySearch (LRNN-6Z8) | N/A | — | Life span given directly | 25 Dec 1853 | New Diggings, Lafayette Co., Wis. |
| Obituary | 6 Dec 1948 | Age 96; noted he died “the Sunday before” | Death date 5 Dec 1948 (a Sunday — matches) | — | — |
A quick note on method, since it comes up constantly in work like this: when a census only gives an age, not a birthdate, that age converts to a range, not a single year — because you don’t know whether the person’s birthday had already passed by census day. Someone who is “18” on 1 June 1870 could have been born any time between 2 June 1851 and 1 June 1852. Reporting these as ranges, rather than collapsing them to one year too early, is what lets you stack multiple records against each other later.
Anchoring on the 1900 Census
I like to start with whichever record makes the strongest, most specific claim and then test everything else against it. That’s the 1900 Census here: it’s the only record in this set that states a full birth month and year directly, rather than just an age — July 1852. It’s also internally consistent: age 47 on 1 June 1900 lines up exactly with that birth month and year. That kind of self-consistency is exactly why the 1900 census is often treated as one of the more reliable U.S. census years for birth information.
That gives me a working anchor to test the rest of the evidence against — rather than, say, building an argument for 1895’s “44 vs. 43” off a birth year I haven’t actually established yet.
Stacking the Earlier Census Windows
Here’s where it gets complicated. Overlapping the 1870, 1880, and 1885 census windows above — with the 1885 record’s 1 May date acting as the tightest constraint — gives a combined window of:
2 June 1851 – 1 May 1852
There’s also a note on John’s FamilySearch profile stating that at age 13, he was selling newspapers on a Boston street corner “at the time of President Lincoln’s assassination” (14 April 1865). I want to flag upfront that this is a story, not a document — it has no cited source, and “newspaper boy on the day of a famous assassination” is exactly the kind of detail that tends to get attached to family lore after the fact, whether or not it’s literally true. I’m not discarding it, but I’m not leaning on it either; it gets one job here, as a test against the records I trust more. Taken at face value, it puts his birth between 15 April 1851 and 14 April 1852. Overlapped with the census window above, it narrows things to:
2 June 1851 – 14 April 1852
And there’s the problem: that window excludes July 1852 outright — the exact date the 1900 census reports. So I have a direct conflict between three independent census ages (agreeing with each other, and with an unsourced anecdote that just happens to land on the same side) and one record with a specific, internally-consistent stated month and year. I’ll come back to how the death record helps break this tie.
The Outliers
Two records don’t fit either pattern:
- The 1875 Minnesota Census puts him at 21 on 1 May 1875 — implying a birth as late as 1854, two to three years off from every other source. My best guess is that whoever supplied the information simply meant he was “over 21” (i.e., of voting age) rather than reporting a precise age.
- The 1895 Minnesota Census reports him at 44 on 1 June 1895 — one to two years older than either the 1851 census cluster or the 1852 anchor would predict. Off-by-one or off-by-two age errors like this are common enough in enumerator-reported records that I don’t weight this one heavily against otherwise-consistent data; I’m treating it as a probable reporting error rather than genuine evidence for an earlier birth.
What the Exact Death Date Reveals
This is the piece I didn’t have when I started digging: FamilySearch gives an exact death date of 5 December 1948, and his obituary — published the next day, 6 December 1948 — gives his age at death as 96. That’s precise enough to actually test each birth candidate against, rather than just eyeballing it:
| Birth Candidate | Age on 5 Dec 1948 | Matches Obituary’s “96”? |
|---|---|---|
| July 1852 (1900 Census) | 96 | ✅ |
| 25 Dec 1852 (Find-a-Grave) | 95 | ❌ |
| 25 Dec 1853 (FamilySearch) | 94 | ❌ |
| 1851, if birthday fell before 5 Dec | 97 | ❌ |
| 1851, if birthday fell after 5 Dec | 96 | ✅ (but no source states such a date) |
Only two scenarios land on 96: a birth in July 1852, or a birth very late in 1851 (after 5 December). The second is mathematically possible but purely hypothetical — nothing in any record actually states a date that late in 1851. The 1900 Census, by contrast, directly states July 1852, and that date independently satisfies this age test. That’s the deciding vote for me.
Preferred date: July 1852. I’m keeping 1851 (year only) on the record as a likely-but-unconfirmed alternative — it’s the year three independent census records converge on — but no single source actually states a specific date within 1851, and it fails the age-at-death test across most of its range.
One More Detail Worth Noting
The FamilySearch date’s location doesn’t match Find-a-Grave. FamilySearch has him born in New Diggings, Lafayette County; Find-a-Grave has Watertown, Dodge County. Both are Wisconsin, but they’re not the same place, and I haven’t yet traced which one has an actual document behind it — or whether either does. This one’s still open.
Where This Leaves Me
- Preferred: July 1852 — the only birth date directly stated by a source I have found so far, internally consistent within that source, and the only one that reproduces the obituary’s stated age of 96 against the confirmed 5 December 1948 death date.
- Likely but unconfirmed: 1851 (year) — supported by three independent census records and an unsourced family anecdote, but with no source giving a specific date, and most of its range fails the age-at-death check.
- Set aside for now: 25 December 1852 (Find-a-Grave, no cited source, fails the age-at-death check) and 25 December 1853 (FamilySearch, fails the age-at-death check).
Next step: chase down whatever document underlies FamilySearch’s New Diggings/Lafayette County claim, since that’s the last real open question in this set.
Don Taylor
Dontaylorgenealogy.com











