Thomas Crawford Puzzle

Have you ever encountered someone with the surname you are researching living in the same place—and immediately wondered whether the two families were related?

That is exactly what happened when I recently found an 1836 Jefferson County, Indiana, court case involving the dower rights of Jane Crawford, widow of Thomas Crawford.

Since I have spent years researching James Crawford (b. 1758), his wife Rebecca Anderson, and their descendants in Jefferson County, my first thought was obvious: Could Thomas Crawford be related to James Crawford?

Following Thomas Back to Virginia

The court case referred to the "heirs of Thomas D. Crawford," so I began searching my records for Thomas. When I did not find him among the Crawford families I had already researched, I started digging into Thomas Crawford and his wife, Jane Todd.

That search led me back to their 1794 marriage in Rockbridge County, Virginia. I then located the will of Thomas's father, James Crawford, in Rockbridge County.

That caught my attention.

Rockbridge County was created in 1777 from portions of Augusta and Botetourt counties. Botetourt itself had been created from Augusta County in 1770. Thus, Thomas's family was living within the broader region where several of the Crawford families I have been researching had roots.

The geography certainly suggested that a relationship was possible.

But geography is not proof.

What Did Gemini Think?

As I worked through the records, I asked Gemini to evaluate the possibility that Thomas Crawford was related to either James Crawford, husband of Rebecca Anderson, or Alexander Crawford, husband of Mary McPheeters.

Based primarily on the paper trail and geographic proximity, Gemini suggested:

Yes, Thomas Crawford is very likely related to both James Crawford and Alexander Crawford, though the exact degree of kinship stems from distinct collateral branches within the tightly knit Scotch-Irish Crawford clan of the upper Shenandoah and James River valleys.

Alexander Crawford & Mary McPheeters

Not a direct son: Alexander and Mary (McPheeters) Crawford's children are thoroughly documented. Their sons were William, John, Rev. Edward, Rev. James (who married Rebecca McPheeters, not Rebecca Anderson), Alexander Jr., Robert, and Samuel. They did not have a son named Thomas.

Kinship connection: Alexander Crawford immigrated to Augusta County alongside his brother Patrick Crawford and related kinsmen (often referenced as sons of Col. William Crawford). The early Crawfords in Augusta, Botetourt, and Rockbridge counties shared common ancestry in Ulster and Clydesdale, making Thomas a likely nephew, cousin, or second cousin to Alexander's lineage.

James Crawford & Rebecca Anderson

Shared migration and proximity: James Crawford (c. 1758–1836; married Rebecca Anderson in 1779 in Montgomery County, VA) and Thomas Crawford (married Jane Todd in 1794 in Rockbridge County, VA) both followed the same path: southwest Virginia → Kentucky (Garrard/Green counties) → Jefferson County, Indiana (near Madison).

Botetourt/Rockbridge branch: James Crawford was associated with the Craig’s Creek and James River settlements of Botetourt/Montgomery, while Thomas's father's estate was probated across the county line in Rockbridge.

Generational placement: Based on timing, James and Thomas were contemporaries—likely brothers or first cousins belonging to the Botetourt/Rockbridge cluster of Crawfords (descended from the contemporary branches of John or William Crawford) rather than Alexander's immediate household.

Gemini also suggested land, probate, and Y-DNA research as ways to test the hypothesis.

And that last suggestion—Y-DNA—was particularly important because I already had DNA evidence available.

Turning to the Crawford Y-DNA Project

Several members of Crawford families I research have taken Y-DNA tests, so I turned to the Crawford surname project to see whether genetics supported the relationship suggested by the paper trail.

Meanwhile, FamilySearch trees provided another intriguing clue. Following the pedigrees there, both Thomas Crawford and Alexander Crawford appear to eventually connect to Malcolm Crawford of Greenock and his wife, Lady Marjory Barclay.

That would be an impressive connection.

But was it genetically possible?

I have not found a Y-DNA test whose documented earliest known ancestor is Malcolm Crawford of Greenock. I did, however, find a test identifying his purported grandson, Sir Lawrence Crawford of Kilbirnie, as the earliest known ancestor.

That gave me three Y-DNA results to compare:

• R1b-01F — kit 203412 — traces through Alexander Crawford

• R1b-13 — kit 1006338 — claims descent from Sir Lawrence Crawford of Kilbirnie

• R1b-01A Ardmillan — kit 926985 — traces to James Crawford (b. 1758), husband of Rebecca Anderson

At this point, the DNA began telling a very different story from the online family trees.

When the DNA Doesn't Fit the Tree

I returned to Gemini and asked it to evaluate the possible relationships in light of these Y-DNA results.

The answer changed dramatically.

Gemini responded:

Having kit 926985 specifically confirmed in R1b-01A (Ardmillan) sharpens the boundaries of this problem significantly: it provides hard genetic parameters for James Crawford's line that directly challenge traditional paper trees and define how Thomas and Alexander fit into the picture.

Gemini summarized the three lines this way:

Lineage	Genetic Group	Proposed Relationship
James Crawford, husband of Rebecca Anderson	R1b-01A (Ardmillan)	Baseline / anchor
Alexander Crawford, husband of Mary McPheeters	R1b-01F	Different Y-DNA subgroup
Thomas Crawford, husband of Jane Todd	Unknown	Requires a tested direct male-line descendant

Gemini then made several much stronger claims about what those results mean.

James and Alexander Are Proven Cousins, Not Immediate Family

Kit 926985 establishes that James is part of the Ardmillan branch (R1b-01A), whereas Alexander Crawford's documented descendants sit in R1b-01F.

Both belong to the primary R1b-01 Crawford genetic cluster. They share a common Crawford patrilineal ancestor back in Scotland or Ulster, but the divergence into 01A and 01F means James was not Alexander’s son, nor his brother.

Their kinship is collateral (second cousins or further back), which aligns with historical records showing Alexander established early in north-central Augusta County, while James's immediate predecessors tied into the Craig's Creek / James River cluster further southwest.

Gemini also challenged the online pedigree connecting these families to Kilbirnie:

The "Ardmillan" Designation Destroys the Kilbirnie Descent for James

In the Clan Crawford Y-DNA project, R1b-01A represents the traditional line associated with Crawfurd of Ardmillan (Ayrshire), not the Crawfords of Kilbirnie.

Because Kit 926985 is 01A, James Crawford could not biologically descend from Sir Lawrence Crawford of Kilbirnie (represented by R1b-13).

Any FamilySearch tree connecting James or his immediate kin to Malcolm Crawford & Margaret Barclay of Kilbirnie has spliced onto the wrong Scottish baronial branch.

But There Is an Important Catch

This is where I think it is important to distinguish between what the DNA actually demonstrates and Gemini's interpretation of that DNA.

The Y-DNA evidence appears to show that the tested descendants of James Crawford and Alexander Crawford fall into different Crawford Project subgroups. That is significant evidence against a recent shared direct paternal ancestor.

However, labels such as "Ardmillan," "R1b-01A," "R1b-01F," and "R1b-13" should not automatically be treated as proof of descent from a particular medieval or early Scottish Crawford family. Those historical connections still depend upon the quality of the pedigrees associated with the tested men and the genetic distance between the tests.

In other words, DNA can disprove a proposed paternal connection much more easily than it can prove a centuries-long pedigree.

And there is an even bigger problem.

I do not yet have a Y-DNA test from a documented direct male-line descendant of Thomas Crawford and Jane Todd.

Thus, the R1b-13 test tracing to Sir Lawrence Crawford does not tell me Thomas Crawford belonged to R1b-13. It only provides a genetic comparison for the Scottish lineage to which online trees connect Thomas.

That distinction is critical.

So What Do I Actually Know?

At this point, my working conclusions are much more limited than the original paper-tree hypothesis:

1. The paper trail places Thomas Crawford within the same broad Virginia geographic network as other Crawford families I am researching. That makes investigating a relationship reasonable, but it does not prove one.

2. A tested descendant of James Crawford and Rebecca Anderson belongs to R1b-01A.

3. A tested descendant of Alexander Crawford and Mary McPheeters belongs to R1b-01F. The differences between those results need to be evaluated at the marker/SNP level before assigning a precise degree of relationship.

4. The R1b-13 test associated with a claimed descent from Sir Lawrence Crawford of Kilbirnie does not match the genetic group of either James or Alexander. If that Kilbirnie pedigree is sound, it raises serious questions about online trees connecting these Virginia Crawfords to that family.

5. Thomas Crawford remains the unknown. Without a Y-DNA test from a documented direct male-line descendant of Thomas, I cannot genetically place his Crawford line.

And that means the most important question is no longer simply:

How was Thomas Crawford related to James or Alexander Crawford?

Instead, it has become:

Was Thomas Crawford related to either of them on the direct paternal Crawford line at all?

The Next Step: Test Thomas's Line

Fortunately, this is a question that Y-DNA may be able to answer.

The next research step is to identify a living male who descends from Thomas Crawford through an unbroken line of Crawford males and ask him to take a Y-DNA test.

If his Y-DNA closely matches the James Crawford R1b-01A line, that would provide strong evidence that the geographic and documentary connections I have been finding reflect an actual paternal relationship.

If he instead matches Alexander's R1b-01F line, the research would head in another direction.

If he matches R1b-13, then the FamilySearch connection to the Kilbirnie Crawford line would suddenly become much more interesting—although the generations connecting Thomas to that Scottish family would still have to be proved with records.

And if he matches none of them?

Then I will have discovered yet another genetically distinct Crawford family living within the same broad geographic and migration network.

Paper Trails, DNA—and AI

This research is also a good example of why I don't want AI to provide the final answer to my genealogy questions.

Gemini initially looked at the locations, dates, migration patterns, and surname and concluded that Thomas was "very likely related" to James and Alexander. Once I supplied the Y-DNA information, its interpretation changed substantially.

But even then, Gemini sometimes stated conclusions more strongly than I believe the evidence warrants.

That doesn't make the conversation useless. Quite the opposite.

The AI conversation helped me formulate hypotheses, identify evidence that could test those hypotheses, and—most importantly—recognize where the evidence conflicts.

The paper trail suggested one story.

The online family trees suggested another.

The Y-DNA is raising questions about both.

And somewhere among those records and chromosomes is the actual story of these Crawford families.

For now, Thomas Crawford's Y chromosome may be the evidence needed to find it.