Same name ancestors can fool even careful researchers because the records are close enough to look convincing. The county fits. The time period fits. The ages are close. The hints line up. It can feel like you have a match when you really have a blend.
This last article is about the step that keeps your work clean long term. You stop collecting only “supporting” records, and you build a proof case. A proof case is a short, organized argument that answers one identity question and shows, with evidence, why one candidate fits and the others do not.
If you can build a proof case, you can defend your conclusion later, and you can hand the work to someone else without it falling apart.
What a proof case is, in plain terms
A proof case is not a life story. It’s a decision file.
It answers one question like this. Which of these two men named James Carter is the one who married Rebecca Lane and died in this county.
A proof case has three goals.
- Identify the correct candidate.
- Show why the other candidates do not fit.
- Preserve your reasoning so you do not rework the same problem later.
This is the tool that turns “I think” into “I can show it.”
Start with one clean identity question
Your proof case begins with a single sentence claim that you can test.
Use this format.
The John Thompson who married Mary Davis in 1819 in this county is the same John Thompson who appears in the 1850 census in this district and left probate in 1857.
Or this format.
The William Harris taxed for 160 acres on Little Creek is the same William Harris who signed deeds with wife Sarah, and he is not the William Harris living in the town district.
Keep it narrow. Prove one identity first. After that, you can expand into parents, birthplaces, and earlier generations.
Create candidate profiles before you argue anything
Before you start weighing evidence, you want the candidates separated on paper.
Make a short profile for each candidate that includes only stable identifiers, not guesses.
A solid candidate profile usually includes:
- Approximate birth range from multiple records, not one record
- Locality labels, such as district, township, creek, church, or neighborhood
- Spouse name when supported by multiple records
- Children names when supported by multiple records
- Occupation patterns when they repeat
- Land ownership patterns and acreage
- Military service details if any
- Probate status, meaning whether there is an estate file and where
You are building a frame so new records can be tested against it.
Build an identity matrix so you can compare candidates quickly
An identity matrix is a simple comparison grid of attributes that can separate candidates even when names match.
Across the top, list your candidates. Down the side, list the attributes you can test.
Common separating attributes include:
- Exact locality details across time
- Spouse name and spouse’s surname connections
- Children’s given names, and which years those names appear
- Land location markers, such as creeks, roads, adjoining landowners
- Tax district and acreage
- Church membership or transfers
- Civic list appearances, such as jury panels and road crews
- Signature versus mark, if original documents survive
- Associate patterns, such as witnesses, bondsmen, sureties, appraisers
Then you fill the matrix only with evidence you can cite. If you cannot cite it, it does not go in the matrix.
This grid lets you see which candidate is consistent and which candidate only fits part of the evidence.
Learn to weigh records, not just collect them
In a same name problem, the wrong conclusion often comes from treating every record as if it has the same reliability.
A practical way to weigh evidence is to ask three questions about each piece of information you plan to use.
- Is the source original or derivative.
- Is the information primary or secondary.
- Does it provide direct or indirect evidence for your claim.
You do not need to memorize formal definitions to apply this. Just think about how the information got into the record.
A deed is usually created at the time of the transaction, and often involves direct participation. A census entry may involve rough ages reported quickly. A death certificate may include parent names supplied by someone who was not present at the birth.
When you weigh evidence, you stop letting one weak statement override a pattern of stronger records.
Build a timeline that supports the proof case, not a timeline for its own sake
You already used a timeline method earlier in the series. Here, you use a timeline in a narrower way.
You build a candidate timeline focused only on the events that separate identity, and you mark conflicts clearly.
For each candidate, you want a clean sequence that includes:
- Dated records with stable locality labels
- Records that connect spouse and household
- Records that connect land and tax patterns
- Records that connect probate and heirs
- Records that connect associate clusters
Then you watch for collisions, meaning things that cannot belong to one life.
Collisions can include:
- Two households in the same census year in different neighborhoods
- Two men taxed in the same year in different districts
- Deeds with different spouse names in the same time span
- Probate events and court events that suggest two parallel adult lives
Collisions are powerful because they show the candidates are separate people, even when their names match.
Use “anchor records” as your decision points
Some records are anchors because they tie identity to multiple stable details at once.
The strongest anchors often include:
- A deed where a spouse releases dower, tying a man to a spouse name
- A probate file naming heirs, tying a man to specific children and associates
- A guardianship appointment naming minor children, tying a man to a household
- A land partition among heirs, tying family and geography together
- Repeated tax entries tied to a specific district and acreage
- Multiple records showing the same associate fingerprint in one locality
Your proof case will usually stand on two or three anchors, supported by additional records that keep the pattern consistent.
Handle conflicts with a simple, honest method
Conflicts are normal. Ages drift. Spellings shift. One record says born in one state, another record says another.
The key is not to hide conflicts. The key is to resolve them or to explain why one piece of information is weaker.
A practical approach is:
- List the conflict clearly.
- Identify which sources are closer to the event and more likely to be correct.
- See whether the conflict changes the identity decision or only affects a detail like an age estimate.
- Keep the conflicting detail flagged for future work rather than forcing it into a clean answer.
If the identity decision is strong, small conflicts usually do not break it. They simply remind you that records are human.
Use elimination as part of the proof, not as an afterthought
A proof case is not complete until you show why the other candidates do not fit.
You do that by pointing to specific evidence that excludes them.
Examples:
- Candidate B appears in tax lists in a different district during the same years Candidate A is present elsewhere.
- Candidate C has a spouse name that conflicts with dower releases tied to the target spouse.
- Candidate B’s probate file names heirs that do not match the target household.
- Candidate C’s land description places him on a different creek with a different associate circle.
This is where many researchers stop too early. They prove why one candidate might fit, but they do not show why the others cannot.
Elimination is what makes your conclusion durable.
Write the proof case so another person can check it
A proof case is only as strong as its clarity.
A practical proof case structure is:
- The claim, one sentence.
- The candidates, one paragraph each describing the key identifiers.
- The anchor evidence, described in a few short paragraphs with citations.
- The supporting evidence, described briefly as patterns, such as tax continuity, land neighborhood, and associate clusters.
- The exclusion evidence, showing why the other candidates do not fit.
- The final conclusion, repeating the claim as proven.
Keep the writing plain. Use dates, places, and record types. Avoid dramatic language. You want the reader to be able to check the work.
A quick example of a proof case conclusion
Here is how a conclusion can sound when it is tight.
The James Carter who married Rebecca Lane is the man consistently taxed in District 2 from 1831 through 1854 for 120 acres on Pine Creek, whose deeds include wife Rebecca releasing dower, and whose 1856 probate file names children matching the District 2 household. The other James Carter in the same county appears in District 5 tax lists during the same years and is tied by deeds and associates to Cedar Branch, including a spouse named Sarah. These parallel records show two separate men living at the same time, and the target identity belongs to the Pine Creek man.
That is the goal. A clear claim, anchored by a few strong records, supported by patterns, and reinforced by elimination.
The payoff
When you build proof cases, same name ancestors stop being a nightmare and become a solvable research problem.
You will attach fewer wrong records.
You will spend less time undoing mistakes later.
You will feel more confident when you publish or share your tree.
You will leave behind work that others can trust and verify.
That is what the proof case method gives you. It is the last step in the series because it is the step that holds everything together.