Photo by Matt Walsh on Unsplash
Genealogists spend a lot of time looking for records.
Birth certificates. Death certificates. Newspaper articles. Obituaries. Court records. Anything that will answer the question we’re chasing.
But every once in a while, the records we don’t find are just as interesting.
A few years ago, I received an email from a man in the United Kingdom who was looking for an old friend. I’ll call him “TH.”
I’ll admit that I nearly ignored the first message. A stranger from overseas asking me to help find someone sounded suspiciously like the beginning of an email involving an inheritance from an uncle I didn’t know I had.
Then he wrote again.
And this time, the genealogist in me got curious.
A Friend Who Disappeared
TH was trying to learn what had happened to a friend who had reportedly died in the United States.
His friend had worked as a psychiatric nurse in the UK and traveled to the United States through his work. At some point, he also began using a different surname. TH believed the change might have been made by deed poll.
The story surrounding his death was even more unusual.
TH had heard that his friend died in America after being hit by a train, or possibly after being struck by debris from a train.
His friend’s brother apparently traveled from England to the United States after the death to handle the arrangements. When TH later contacted him, however, the brother would not discuss what had happened.
TH didn’t believe his friend’s body had been returned to the UK.
That was about all we had.
No exact birth date. No exact death date. No location in the United States.
Not exactly a researcher’s dream starting point.
The First Break
My early searches went nowhere.
I searched newspapers and other available records, but without knowing where or exactly when the man had died, there simply wasn’t enough information to narrow the search.
Then TH found something important.
Using records obtained through Reclaim The Records, he located his friend in a New Jersey death index.
Now we had an exact date of death and additional identifying information.
That allowed me to locate information about his friend’s birth family and confirm that we were researching the right person.
There was also a small lesson tucked into the dates. TH had been very close when recalling his friend’s birth information, but British and American dates are written differently. A date written as 6/5 can mean 6 May to someone in Britain and June 5 to an American.
In genealogy, even punctuation can cause trouble.
What About the Name Change?
The surname change was another part of the mystery.
I contacted The National Archives in the United Kingdom and asked about deed poll records.
A Remote Enquiries Duty Officer searched but found no enrolled deed poll for the name change.
That sounds important, but it doesn’t prove that the man never changed his name.
The Archives explained that a deed poll only became part of the official court record if someone paid to have it enrolled. Many people did not. In those cases, the original deed might have remained only with the individual.
That unsuccessful search actually told us something useful:
No enrolled deed poll was found. It did not tell us that no name change occurred.
That’s an important distinction in genealogy.
And Then There Was the Train
The reported train accident seemed as though it should leave a trail.
By this point, we knew that TH’s friend died in New Jersey in 1995. I contacted the New Jersey State Library, where a researcher searched the Morristown Daily Record around the date of death.
She searched for his name and also tried broader terms related to trains and train accidents.
Nothing.
She expanded the search through New Jersey newspapers available through NewsBank.
Still nothing.
I then heard from the New Jersey History and Genealogy Center. They searched additional newspapers from the same period.
Again, nothing.
Now things were getting interesting.
When “Nothing” Matters
An unsuccessful search does not automatically prove that something didn’t happen.
That’s worth repeating.
If I search one newspaper database for a person’s name and get no results, I haven’t proven anything except that I didn’t find the person in that particular search.
Maybe the newspaper isn’t digitized. Maybe the name was misspelled. Maybe the article wasn’t indexed properly. Maybe I searched the wrong date or place. Maybe the event simply wasn’t reported.
Genealogists call this a negative finding.
Negative evidence is stronger. It develops when information we would reasonably expect to find is absent from appropriate records, particularly when that absence is considered alongside other evidence.
In this case, several researchers searched multiple newspaper collections using the man’s name, the known date and location, and broader searches for a train accident.
We still couldn’t establish that a fatal train incident occurred.
That does not prove TH’s friend didn’t die that way.
But it does mean I wouldn’t state that he did.
The family story and the documented evidence had parted company.
Keep Track of the Dead Ends
This is also why I keep track of unsuccessful searches.
Write down:
- what collection you searched
- what names and spelling variations you tried
- the date range and location you covered
- what search terms you used
- what the collection actually includes
- the date you searched it
- what you found, including “nothing”
That last one matters.
Otherwise, six months from now, you’ll discover the same database, get excited, search it again, find nothing again, and wonder why the whole experience feels strangely familiar.
Ask me how I know.
A research log turns “I couldn’t find it” into useful information.
The Mystery Remains
I eventually built a small family tree for TH’s friend and identified his biological parents, but nothing I found explained the circumstances of his death.
And the two questions I had then are still the questions I would ask today:
Did he actually die in a train-related accident?
And:
What happened that TH never knew about?
I don’t know.
The brother’s unwillingness to discuss the death isn’t evidence of anything by itself. There could be countless personal reasons someone would choose not to discuss a sibling’s death, and it wouldn’t be responsible to speculate.
The death certificate might answer the first question, but access restrictions prevented TH, who was not a relative, from obtaining it at the time.
So this research ended without the satisfying answer we all want.
But it wasn’t a failure.
We established the man’s identity. We found his death in New Jersey. We confirmed his birth family. We learned why the absence of an enrolled deed poll couldn’t settle the name-change question. And repeated newspaper searches failed to substantiate the story of a fatal train accident.
That’s progress.
Genealogy isn’t about forcing the evidence to give us an answer.
It’s about figuring out exactly what the evidence does tell us, recognizing what it doesn’t, and knowing the difference.
And every once in a while, the most important thing you discover is that the story you’ve been trying to prove still hasn’t been proven.
The hunt continues.
As an Amazon Associate, I earn from qualifying purchases.
Discover more from Loganalogy- Your Family Historian
Subscribe to get the latest posts sent to your email.