When Documents Disagree

The birth certificate says 1902.

The marriage record says 1904.

The immigration file says 1901.

The death certificate says 1903.

Which one is correct?

Perhaps the birth certificate.

Perhaps not.

The instinctive response to contradictory records is often to choose the document that appears most authoritative and discard the others.

In forensic genealogy, that is rarely my first step.

Before deciding which piece of information is correct, I want to understand why the records disagree.

Because sometimes the answer to that question is more important than the date itself.

First, preserve the contradiction

When two documents disagree, I do not immediately correct the information.

I record both versions.

If one document says that a person was born in 1902 and another says 1904, both dates remain part of the research until the discrepancy has been investigated.

The same applies to names.

Places.

Parents.

Spouses.

Occupations.

Addresses.

The temptation to standardise information too early can remove exactly the clue we need.

A discrepancy is evidence.

Until we understand it, we should preserve it.

Ask what each document was created to prove

Documents are created for different purposes.

A birth record exists primarily to document a birth.

A marriage record documents a marriage.

A death certificate documents a death.

An immigration record serves an administrative purpose connected with migration.

A cemetery register records a burial.

This matters because a document may be highly reliable for the event it was created to record while being considerably less reliable for other information contained within it.

A death certificate may provide excellent evidence of the date of death.

But if it also states that the deceased was born seventy years earlier in another country, we need to ask who supplied that information and how they knew it.

A marriage record may contain information provided directly by the bride and groom.

Or it may reproduce details from documents submitted during the registration process.

Understanding why a record was created helps us understand what weight to give the information it contains.

Ask who provided the information

This is often one of the most important questions.

Was the information supplied by the person themselves?

By a parent?

By a spouse?

By an adult child?

By an institution?

By someone whose relationship to the deceased is unknown?

A person registering their own marriage may be expected to know their parents' names.

A son registering his father's death may know them too.

But perhaps his grandparents died before he was born.

Perhaps he knew them only by nicknames.

Perhaps the family never discussed them.

The fact that information appears in an official record does not tell us whether the person who supplied it actually knew the answer.

When documents disagree, identifying the informant can sometimes explain the discrepancy immediately.

Move closer to the event

As a general principle, records created closer to an event often deserve greater evidential weight when describing that event.

If a birth record created shortly after a child's birth gives one date and a document created seventy years later gives another, the earlier record will usually be particularly important.

But even this is not an absolute rule.

An original record can contain an error.

A later document may be based on an original certificate that no longer survives.

This is why genealogical evidence is weighed rather than ranked mechanically.

"Earlier" is useful.

"Original" is useful.

"Official" is useful.

None of those words means "automatically correct."

Find an independent source

When two records disagree, the natural next step is often to look for a third.

But not just any third document.

Ideally, we want an independent source.

Suppose one document gives a date of birth as 1901 and another gives 1903.

A third record also says 1903.

That appears to settle the matter.

But what if the third record copied its information from the second?

We still have only two independent pieces of evidence.

This is why I want to know where information originated.

A school register, military record, census, passport application or contemporary religious record may provide an independent point of comparison.

The purpose is not to hold a vote between documents.

Three against two does not necessarily win.

The purpose is to understand which pieces of evidence developed independently of one another.

Build a timeline

Chronology is one of the most effective tools for resolving contradictory records.

Instead of examining each document separately, place them in order.

What do we know about the person in 1920?

Where were they in 1925?

When did they marry?

When did they migrate?

When did their children appear?

When did their name change?

A discrepancy that seems confusing in isolation may become perfectly understandable once it is placed in chronological context.

Perhaps the surname changed after marriage.

Perhaps the place name changed after a border shift.

Perhaps the person began using a different first name after migration.

Perhaps an age was altered consistently from a particular point onwards.

The timeline may also reveal something more serious.

Two records attributed to the same person may describe events that cannot easily coexist within one life.

At that point, the question is no longer which document is correct.

The question becomes whether the documents belong to the same person at all.

Look at the pattern, not only the discrepancy

One incorrect date may be an error.

Several changing dates may be a pattern.

One unusual spelling may be a clerk's interpretation.

A surname that changes systematically after migration may tell us something about adaptation to a new language.

One incorrect parent's name may be a mistake.

Different parents appearing consistently across two separate documentary trails may indicate that two identities have been combined.

This is why individual discrepancies should rarely be assessed in isolation.

We need to know what happens across the entire body of evidence.

Does the contradiction occur once?

Does it repeat?

Does it begin at a particular moment?

Does it correlate with migration, marriage or another major life event?

Does one group of documents consistently contain one set of information while another group consistently contains something else?

Patterns turn discrepancies into evidence.

Consider the historical context

Sometimes documents disagree because history changed around the person.

A town belonged to one country when someone was born and another country when they died.

The official language changed.

Administrative districts were reorganised.

A place acquired several names.

Borders moved.

Governments collapsed.

A person who described themselves as born in Russia in one document and Poland in another may not have changed their story at all.

The political map may have changed.

This is particularly important when researching families from Central and Eastern Europe, where a single lifetime could cross several political regimes without the person ever leaving their hometown.

Before treating two geographical descriptions as contradictory, we need to understand what those descriptions meant when the documents were created.

Consider language

Names are particularly vulnerable to apparent contradiction.

A single person may have:

a Hebrew name;

a Yiddish name;

a Polish or Russian version of that name;

a secular name;

a nickname;

a legally adopted name;

and several transliterations of each.

A surname written in Cyrillic and later transliterated into Polish may look different again when written in Hebrew and subsequently transliterated into English.

This can create an impressive collection of apparently different identities belonging to one individual.

The opposite problem also occurs.

Two people with similar names may be mistakenly combined because the researcher assumes that every variation is simply another transliteration.

Linguistic flexibility is essential.

So is knowing when flexibility has gone too far.

Ask whether you are explaining too much

This is one of the most important checks in forensic genealogical research.

Almost every individual discrepancy can be explained.

The date was wrong because of an administrative error.

The surname changed because of transliteration.

The father's name was different because the informant made a mistake.

The place was wrong because the borders changed.

The marriage record does not fit because perhaps there was a divorce.

The migration timeline does not fit because perhaps the person travelled back.

Each explanation may be possible.

But when an identification requires a growing collection of hypothetical explanations, we have to reconsider the identification itself.

At some point, the question becomes:

Are the documents inconsistent because one person's life was complicated?

Or:

Are the documents inconsistent because we have combined two different people?

That distinction can determine the entire outcome of a case.

Sometimes disagreement cannot be resolved

There is another possibility that genealogists have to be willing to accept.

Sometimes we do everything correctly and still cannot determine which version is true.

The necessary record may have been destroyed.

The archive may be inaccessible.

No independent source may survive.

The people who knew the answer may have died decades ago.

In those circumstances, the professional response is not to select the version we prefer.

It is to preserve the uncertainty.

We can explain which interpretation appears more probable.

We can explain why.

We can identify the limitations of the evidence.

But we should not turn probability into certainty simply because certainty would make the family tree look cleaner.

The disagreement is part of the evidence

When documents disagree, our task is not to make them agree.

It is to understand the disagreement.

We preserve the conflicting information.

We examine the purpose of each record.

We identify the informants.

We consider when the documents were created.

We look for independent corroboration.

We build timelines.

We examine historical and linguistic context.

We look for patterns.

And we remain willing to conclude that the evidence is insufficient.

Sometimes, after all of that, one version clearly emerges as the most reliable.

Sometimes the contradiction reveals an event we did not previously know had occurred.

Sometimes it tells us that two people have been confused with one another.

And sometimes the disagreement remains unresolved.

All of those are legitimate research outcomes.

Because the purpose of forensic genealogy is not to create a perfectly tidy history.

It is to reconstruct the most accurate history the surviving evidence will allow.

And sometimes accuracy begins by refusing to tidy away the things that do not fit.