Why I Sometimes Spend Hours Proving That Two People Are NOT the Same Person

Sometimes I find exactly the person I am looking for.

The name matches.

The approximate age matches.

The father's name matches.

The geography is plausible.

There may even be a document prepared years ago that identifies this person as the missing relative, deceased heir or former owner of an estate.

Excellent.

Except that something does not quite fit.

Perhaps the date of birth is wrong.

That alone may mean very little.

Perhaps the surname is spelled differently.

Again, easily explained.

Perhaps the father's name appears in two slightly different forms.

Still possible.

Then another document appears.

And another.

At some point, the question changes.

I am no longer asking:

How can I prove that these are the same people?

I am asking:

What if they are not?

And answering that question properly can take hours.

Finding a match is not the same as establishing an identity

Genealogical research frequently produces candidates.

Some are immediately eliminated.

Others are so compelling that it is tempting to stop searching.

This is particularly dangerous when several details align at once.

A similar name, approximately correct year of birth and matching father's name can create what appears to be a convincing identification.

But people are not identified by one or two characteristics.

They have entire lives.

They have parents.

Spouses.

Children.

Addresses.

Occupations.

Migrations.

Legal identities.

Religious identities.

Documents created at different points in time and in different jurisdictions.

If two records supposedly describe the same person, those records must ultimately fit into one reasonably coherent life.

Sometimes they do.

Sometimes they almost do.

And "almost" is where the real work begins.

The first discrepancy is usually easy to explain

One of the realities of genealogical research is that records are inconsistent.

Names change.

Spellings vary.

Dates are wrong.

Places are described differently.

Transliteration creates multiple versions of the same surname.

People use Hebrew names in one context and secular names in another.

Migrants sometimes alter their age.

Officials misunderstand foreign names.

None of this is unusual.

So when I encounter the first discrepancy, I rarely assume that I have the wrong person.

I try to explain it.

Then I encounter the second discrepancy.

That may also have a perfectly reasonable explanation.

Then a third appears.

And perhaps a fourth.

At this stage, there is a danger that the research becomes an exercise in defending an identification rather than testing it.

Every similarity is treated as confirmation.

Every contradiction receives an explanation.

That is when I deliberately reverse the question.

Instead of asking:

Can I explain why these records are different?

I ask:

What evidence would I expect to find if these really were two different people?

To separate two people, you often have to reconstruct both

This is why disproving an identity can take considerably longer than finding a potential match.

It is rarely enough to say:

"The dates are different."

Dates can be wrong.

Or:

"The surname is spelled differently."

Spellings change.

Or even:

"The father's name is different."

Depending on the records involved, informants can make mistakes.

To demonstrate that two documentary identities most probably belong to different people, I may need to reconstruct both lives.

Where was Person A at a particular time?

Where was Person B?

Who did each person marry?

Who were their parents?

What addresses appear in their records?

When did they migrate?

Is there evidence that one person could realistically have travelled between the locations where both identities appear?

Do the two documentary trails ever intersect?

Or do they continue independently?

Sometimes the decisive evidence is not one spectacular contradiction.

It is the discovery that two apparently similar people each have their own coherent biography.

At that point, the similarity between them becomes less important.

Chronology is unforgiving

Names can change.

Dates can be recorded incorrectly.

Surnames can be transliterated in five different ways.

But chronology has limits.

If a proposed identification requires someone to be living in one country, marrying in another, returning to the first, changing their identity and creating a second family, that sequence may be possible.

Human lives are complicated.

But possible is not the same as documented.

The more complicated the proposed reconstruction becomes, the more evidence we should expect to find supporting it.

A marriage should leave records.

A migration may leave records.

A divorce may leave records.

A legal name change may leave records.

A return to another country may leave records.

If maintaining an identification requires an increasingly elaborate sequence of undocumented events, we have to consider the simpler possibility:

We may be looking at two different people.

Sometimes similarities are the trap

This is particularly important in populations where certain names recur frequently.

Two people may share:

• the same first name;

• similar surnames;

• approximately the same year of birth;

• the same town of origin;

• even a father's name.

  
  

That can feel extraordinary.

In reality, depending on the population and period, it may not be extraordinary at all.

This is why similarity should trigger investigation rather than end it.

A promising match tells me where to look next.

It does not tell me that the search is finished.

Why spend hours proving a negative?

Because sometimes a great deal depends on the answer.

In forensic genealogy, an incorrect identification can affect an inheritance.

It can lead researchers down the wrong family line.

It can result in the wrong people being contacted as potential heirs.

It can create an apparently convincing family tree built around somebody who never belonged in it.

And once an incorrect identification enters a file, it has a tendency to reproduce.

It appears in a report.

That report is used in another proceeding.

The information is entered into a family tree.

Someone else cites the tree.

Years later, the original hypothesis has acquired the appearance of an established fact simply because it has been repeated often enough.

Correcting that kind of error can take considerably longer than creating it.

That is why I am willing to spend hours proving that two people are not the same person.

Not because negative conclusions are particularly satisfying.

But because eliminating the wrong person is progress.

Sometimes it is the most important progress a case can make.

"Not the same person" is a result

There is a tendency to think that genealogical research has succeeded only when someone is found.

But forensic genealogy is not a treasure hunt.

The objective is not to produce a person at any cost.

The objective is to reach the conclusion best supported by the evidence.

Sometimes that conclusion is:

Yes. The documentary trails converge. These records most probably describe the same person.

Sometimes it is:

We do not yet have enough evidence to know.

And sometimes it is:

No. Despite the similarities, these are two different people.

That last answer may be frustrating.

It may mean returning to the beginning of the search.

It may mean that an heir remains unidentified.

It may mean that a previous conclusion has to be reconsidered.

But it also means something very important.

We have stopped looking in the wrong place.

And in forensic genealogy, knowing who someone is not can be the first real step towards discovering who they actually are.