What Makes Genealogical Evidence Reliable?

It is written on an official document.

Therefore, it must be true.

Unfortunately, genealogical research is rarely that simple.

A government-issued certificate can contain incorrect information.

A handwritten family note can preserve a fact that appears nowhere else.

A gravestone can be accurate about one piece of information and wrong about another.

Five documents repeating the same date may look like overwhelming evidence—until we discover that all five copied the information from the same original source.

So what makes genealogical evidence reliable?

The answer is not simply the type of document we are looking at.

It depends on what we are trying to prove.

A document is not equally reliable for every fact it contains

Consider a death certificate.

It may be excellent evidence that a person died on a particular date and in a particular place.

Those facts were recorded close to the event itself and were usually known directly to the authority registering the death.

But the same certificate may also contain the deceased person's date of birth, place of birth and parents' names.

Who provided that information?

Perhaps a spouse.

Perhaps an adult child.

Perhaps a neighbour.

Perhaps an institution.

The person supplying the information may have known the deceased for fifty years and still never have known the exact name of a grandmother in another country.

The document is authentic.

The death occurred.

But that does not mean that every piece of information recorded on the certificate has the same evidential value.

This is one of the most important distinctions in forensic genealogy.

We do not simply ask:

Is this a reliable document?

We ask:

How reliable is this document for the particular fact I am trying to establish?

Who knew the information?

One of the first questions I ask when evaluating a record is:

Who provided the information?

A mother registering the birth of her child is likely to know the child's date of birth.

A person completing their own marriage record is likely to know their own name.

But information becomes less certain as we move further away from the person or event being documented.

A grandson may know where his grandfather was born.

Or he may know only what the family always said.

A hospital may record information provided by a patient.

A cemetery may copy information from a burial permit.

A later government record may reproduce details from an earlier document.

This does not make the information useless.

It simply affects the weight we give it.

Whenever possible, we want to understand the chain between the historical fact and the person who recorded it.

The shorter that chain, the better.

When was the record created?

Time matters.

A birth record created shortly after a birth is generally closer to the event than a death certificate created eighty years later that attempts to reconstruct the same person's date and place of birth.

The later document may still be correct.

But if the two disagree, the record created closest to the event will usually deserve careful consideration.

This is particularly important in families affected by migration, war and displacement.

A person born in one country may spend decades living elsewhere.

Their original documents may be lost.

Their birthplace may be recorded according to changing borders.

Their age may shift slightly between records.

Years later, information that was once known precisely may survive only through memory.

The distance between an event and the record describing it can therefore matter enormously.

Original does not mean infallible

There is another common assumption:

If we find the original record, we have found the truth.

Not necessarily.

Original records can contain mistakes.

An official may misunderstand a name.

A clerk may write the wrong date.

A parent may provide incorrect information.

A person may deliberately give a different age.

A place name may be recorded incorrectly.

The value of an original record is that it brings us closer to the original act of documentation.

It reduces the number of opportunities for information to be altered through copying, transcription or translation.

But an original document is still a document created by human beings.

And human beings make mistakes.

Independent confirmation matters

Suppose five documents give the same date of birth.

That sounds convincing.

But where did the information come from?

If Document B copied Document A, Document C copied Document B, and Documents D and E were created using the same government register, we may not have five independent confirmations.

We may have one piece of information repeated five times.

This is sometimes called source dependency.

It is one of the reasons that the quantity of records is not necessarily the same thing as the quality of evidence.

What we ideally want is independent corroboration.

A birth record created in one country.

A school record created by another institution.

A marriage record created years later.

An immigration record produced in another jurisdiction.

If these independent sources converge on the same information, the evidential picture becomes considerably stronger.

The key word is not simply agreement.

It is independence.

Contradictions do not automatically make evidence unreliable

Two reliable documents can disagree.

That may sound paradoxical, but it happens frequently.

A birth record may give one version of a name.

A marriage record gives another.

An immigration record uses a third.

All three documents may genuinely relate to the same person.

The discrepancy may result from language, transliteration, migration, marriage or a legal name change.

The existence of a contradiction does not automatically mean that one document should be discarded.

Instead, the contradiction becomes a question.

Why are these records different?

Can the difference be explained?

Is the explanation supported by evidence?

Does the same pattern appear elsewhere?

Sometimes resolving a contradiction teaches us more about a person's life than a perfectly matching set of documents ever could.

Context changes the meaning of evidence

A record never exists in isolation.

Its meaning depends partly on the historical world in which it was created.

Borders change.

Languages change.

Governments change.

Administrative districts are reorganised.

Towns acquire new names.

People move between countries without necessarily moving very far geographically.

A person born in the Russian Empire may later appear in records as having been born in Poland without ever having changed their birthplace.

Someone born in Lwów may appear in documents associated with several different states over the course of one lifetime.

None of those records is necessarily wrong.

They may simply describe the same place according to different political realities.

This is why historical and geographical context matters.

A discrepancy that appears significant to a modern reader may have been entirely ordinary at the time.

Translations and transcriptions add another layer

Genealogical evidence frequently crosses languages.

A Polish document may be translated into English.

A Russian name may be transliterated into Polish.

A European surname may be written in Hebrew and later transliterated back into the Latin alphabet.

Every transition creates another opportunity for variation.

This does not necessarily indicate an error.

Sometimes there simply is no single "correct" way to represent a name in another alphabet.

But it does mean that a researcher should distinguish between what appears in the original document and what appears in a later translation or transcription.

When possible, I want to see the original.

Not because translators cannot be trusted.

But because interpretation is part of translation, and in identity research even a small linguistic choice can matter.

Family trees are clues, not conclusions

Online family trees can be enormously useful.

They can provide names, relationships, photographs and leads that might otherwise take weeks to discover.

They can also contain errors.

One incorrect connection can be copied into dozens of other trees.

Eventually, the number of repetitions creates an illusion of certainty.

But repetition is not verification.

A family tree tells me:

Someone believes these people are connected.

My next question is:

Why?

What evidence supports the relationship?

Can it be independently reconstructed?

Does the chronology work?

Do the original records agree?

A tree can be the beginning of an investigation.

It should rarely be the end of one.

Reliability is built, not assumed

There is no perfect genealogical source.

Every document has strengths.

Every document has limitations.

The reliability of genealogical evidence comes from understanding those limitations and comparing multiple sources intelligently.

We ask:

Who created the record?

Who supplied the information?

When was it created?

How close was it to the event?

Is the source original or derivative?

Could the information have been copied from somewhere else?

Does independent evidence corroborate it?

Does it fit the historical context?

And, importantly:

What evidence contradicts it?

Reliable genealogical conclusions are rarely built on one perfect document.

They emerge when independent pieces of evidence begin to converge.

And when they do not converge, the answer is not to choose the document we like best.

It is to find out why.

Because in forensic genealogy, evidence becomes reliable not simply because it is written down.

It becomes reliable when we understand where it came from, what it can prove—and where its limits lie.