Convergent Identities: When Two People Accidentally Become One

One of the concepts I frequently encounter in genealogical research is something I have come to think of as convergent identities.

The name sounds rather dramatic.

The reality is surprisingly common.

Imagine two men called Jan Kowalski.

Both were born within a year of each other.

Both lived in Warsaw.

Both married women called Maria.

Both had sons named Piotr.

If you're smiling slightly, you're beginning to understand the problem.

Now imagine that one document lacks an age or gives the same birth year, but not the full date.

Another omits an address.

A third records only initials.

Imagine their father uses several names which hit the search mark.

Before long, those two completely separate men begin to merge into one historical person.

This is what I have come to think of as convergent identities: situations in which two or more individuals share such a high degree of similarity in their identifying characteristics—such as name, age, place of residence, occupation or family relationships—that they risk being mistakenly treated as the same person during genealogical or legal research.

Not because anyone intended to mislead.

Simply because the available evidence appears to fit both.

This happens far more often than people realise.

In genealogy, similarity is not identity.

One of the biggest mistakes researchers can make is assuming that the first person who appears to fit the available information must be the correct one.

Sometimes they are.

Sometimes they are not.

And the consequences can be significant. One incorrect assumption made early in an investigation has an unfortunate habit of reproducing itself. Once the wrong identity enters a family tree, every subsequent document appears to confirm it, simply because the researcher is no longer looking for alternatives.

An entire family tree may be attached to the wrong individual.

Probate proceedings may identify the wrong relatives.

Historical conclusions may become increasingly difficult to untangle because each new record is interpreted through the original mistaken assumption.

This is one of the reasons I ask clients to provide every piece of information they already possess. Sometimes, the genuinly cannot. This is especially true when research involves countries with limited archival access, extensive record loss or disrupted historical documentation. In those situations, you work with what you have and gradually reconstruct the lineage, all the while knowing that some of the subtle distinctions may only emerge much later in the investigation.

The more independent identifiers we have—dates, occupations, addresses, religious affiliation, migration patterns, family relationships—the easier it becomes to separate genuinely distinct individuals from those who merely appear similar. Sometimes every second person in the family seems to be called Golda or Shmuel. Professionally, you may be satisfied beyond reasonable doubt that you've identified the correct family, yet when you finally contact the descendants they insist there can be no connection—simply because they know very little about their own family history. That happens far more often than people realise.

Oddly enough, I don't spend my time trying to prove that two people are the same.

More often than not, I'm trying to prove that they are not.

Only once that possibility has been properly excluded do I begin considering whether the records truly belong to a single person.

It's one of those moments where genealogy starts to resemble forensic science.

The objective isn't to make the evidence fit.

It's to discover which explanation the evidence genuinely supports.

Convergent identities remind us that history is rarely untidy because people lied.

More often, it's untidy because ordinary people happened to have ordinary names, ordinary lives and remarkably similar records.

The challenge is recognising where one life ends and another begins.