The Artificial Intelligence tools (like ChatGPT, Claude, CoPilot, Gemini, Grok, Meta, Grok, and Perplexity) can provide much useful and nearly instant information about subjects in a concise format, especially when they have been pre-trained on a large volume of information about the subjects.
I wondered if the FREE AI tools could provide me with a process to solve a brick wall parentage problem for Hannah (Smith) Sawtell (1768-1827).
1) I chose to use the responses from free ChatGPT4 (as of 12 May 2025) today.
My first Prompt was general: You are a professional genealogist. Please tell me how to solve a genealogical brick wall problem." In a followup prompt, I asked ChatGPT4 to expand the details for each point.
2) The free ChatGPT4 (not Deep Research) response was:
Solving a genealogical brick wall requires a structured approach, fresh perspectives, and a willingness to reexamine everything. Here's a step-by-step guide to help you break through:
🔍 1. Review What You Already Know
Start by pulling together every single record, document, and clue you've already collected about your target ancestor. This includes census records, birth, marriage, and death certificates, military service records, probate and wills, land transactions, church records, and newspaper clippings. Go beyond just the direct ancestor—gather information about their spouses, children, and extended family. Review these documents carefully for overlooked details: witnesses, neighbors, occupations, locations, or name spellings. Create a timeline of events for the person, which helps visualize gaps or contradictions. Also consider using genealogical software or spreadsheets to help organize and sort information in multiple ways. A methodical review can uncover patterns or connections you missed earlier.
🧩 2. Form a Focused Research Question
A vague question like "Who were John Smith’s parents?" can lead you in too many directions. Narrow it down with all the specifics you know. For example: “Who were the parents of John Smith, born about 1805 in Botetourt County, Virginia, who moved to Greene County, Ohio by 1830 and married Mary Jones in 1829?” This kind of targeted question helps you focus your search in time, place, and context. It also clarifies what evidence is missing and helps you set research goals. Consider writing down what you’ve already searched and what conclusions you’ve drawn so far—this documentation keeps your research grounded and focused.
🔄 3. Work Backwards and Sideways
Always work from the known to the unknown. Confirm each generation with solid, verifiable evidence—never assume relationships just because people shared the same last name or lived near each other. Once you’ve traced your direct ancestor as far as you can, switch to collateral research. Investigate siblings, cousins, in-laws, and other extended family. Sometimes the proof of an ancestor’s identity lies not in their own records, but in those of their relatives. This includes checking who lived nearby in the census or appears in land records or wills. This is part of the FAN Club approach—studying Friends, Associates, and Neighbors—who often migrated together, witnessed each other’s documents, or intermarried. These connections can break open tough brick walls.
🗺️ 4. Dive Into Local and Regional Records
If you're only using online databases, you're likely missing crucial documents. Local courthouses, historical societies, libraries, and state archives often hold unique collections that aren’t digitized—such as tax lists, court proceedings, guardianship files, estate inventories, old maps, and voter lists. Many genealogists break brick walls by visiting these places in person or contacting them by phone or email. Learn the history of the location—know when counties were formed, how boundaries changed, and if records were lost to fire or war. You may find church records, newspapers, or school records tucked away in a basement archive or local family collection that hasn’t been digitized or indexed.
🧬 5. Use DNA Testing (if applicable)
If documentary evidence runs out, genetic genealogy can provide new clues. Take an autosomal DNA test (like AncestryDNA or MyHeritage) and compare your matches. Look for clusters of matches who descend from the same unknown ancestor. Use tools like shared matches, segment triangulation (on GEDmatch or MyHeritage), and chromosome browsers to group matches together. If your brick wall is on a specific paternal or maternal line, consider Y-DNA (for male-line ancestry) or mtDNA (for female-line ancestry) through FamilyTreeDNA. Combine your DNA analysis with traditional research to formulate new hypotheses—sometimes even identifying a family that your ancestor likely descended from, even without a direct paper trail.
📚 6. Look for Published Histories & Trees (Critically!)
Books, county histories, family trees, and other secondary sources can contain useful leads—but approach them critically. Use tools like Google Books, FamilySearch Books, HathiTrust, and Internet Archive to search for digitized family histories, local histories, and genealogical journals. Even if you find a family tree online that includes your ancestor, don't accept it at face value—verify all claims with primary sources. That said, published materials can often point you to unfamiliar counties, sibling names, or family traditions that you can then research further. Historical journals (like the New England Historical and Genealogical Register or National Genealogical Society Quarterly) may have articles relevant to your area or surname of interest.
🧠 7. Create a Hypothesis & Test It
At a brick wall, you may not have full proof—but you can still build and test logical theories. For instance: “I believe William Johnson of Orange County, NC, is the same William Johnson who later appears in early Kentucky records.” You can then search for supporting evidence: do both men have the same age range? Are they associated with the same people or families? Did anyone from Orange County migrate to Kentucky around that time? You may not find a single document that proves the theory, but a combination of indirect evidence (a preponderance of clues) might support it. The Genealogical Proof Standard can help you evaluate whether your conclusion is reasonably supported.
⌛ 8. Be Patient and Persistent
Brick walls rarely crumble in one day. Some cases take months or years to solve, especially if records are missing or difficult to access. Be persistent—check back for new record sets, reexamine your assumptions, and collaborate with others. Join genealogical forums, surname groups, local Facebook groups, or message boards where others may be researching the same families or places. Even professional genealogists hit brick walls—but they know that progress often comes through thoroughness, creativity, and collaboration. Leave yourself time to return to a problem with fresh eyes, or let new records be digitized. Your next clue could be one click—or one courthouse trip—away.
Would you like help applying this process to one of your brick walls?
4) Now that ChatGPT mentions it5, how about my Hannah Smith problem? Here is my prompt: "Yes, please!!!! Hannah Smith (1768-1827) is my 4th great-grandmother. She married Josiah Sawtell in 1789 in Amherst, Massachusetts, and resided in Brookline NH, Townsend MA, and Ashburnham MA. They had nine children, named Hannah, Josiah, Mary, Chester, Esther, Horace, Neuma, Walter and Price. Please suggest a process to solve this mystery."
5) ChatGPT4 helpfully responded with:
🧭 Research Goal:
Identify the parents and birth origins of Hannah Smith (1768–1827), who married Josiah Sawtell in 1789 (likely in or near Amherst, MA), and resided in Brookline, NH; Townsend, MA; and Ashburnham, MA.
🔎 Step-by-Step Process:
1. Reexamine and Re-source the Marriage (1789)
- Find the original 1789 marriage record (not just the index).
- Where did it occur? Amherst, MA is likely—confirm this.
- Does it list Hannah’s residence, parents, or whether she was "of" another town?
- Who officiated? He may have recorded more detail in his church register than is on the town record.
Check neighboring town marriage records in case it was recorded in both the bride and groom’s towns (common in New England).
2. Map a Life Timeline and Residential Movement
Create a chronological timeline of Hannah and Josiah’s locations:
- 1789 marriage — Amherst?
- 1790s–1810s — where were their children born?
- Later years — Brookline NH, Townsend MA, Ashburnham MA
Use:
- Census records (1790–1820; Josiah will be head of household)
- Tax lists, town histories
- Children’s birth, marriage, and death records (towns often record parental information)
📍 Key insight: The earliest places they lived (where their first children were born) may point to where Hannah was from.
3. Analyze the Names of the Children
Their children were: Hannah, Josiah, Mary, Chester, Esther, Horace, Neuma, Walter, Price. Consider:
- First children often named for parents’ parents.
- If they named their eldest son Josiah, the second son might reflect Hannah’s father’s name.
- Names like Chester, Horace, Neuma, Price are relatively unusual. Could be from Hannah’s side.
🧠 Action: Check for Smith families in the area with those names in earlier generations. For example, if you find a Walter Smith or Price Smith in nearby towns, it may be significant.
4. Target Smith Families Living in the Right Time and Place
Research all Smith households in Amherst and surrounding towns (Brookline NH, Townsend MA, Ashburnham MA, Pelham, etc.) around 1760–1790. You're looking for:
- Men with a daughter named Hannah born ~1768
- Wills or guardianships naming a Hannah
- Tax lists showing Smiths with adult daughters (age ~20) around 1789
📍 Tools:
- Vital records of each town (births, marriages, deaths)
- 1790 census
- Church records (especially Congregational and Baptist churches)
- Town histories (many include family sketches)
5. Explore Probate and Land Records
- Look for probate records of any Smith men who died between 1775–1800 in Hampshire and Worcester Counties (MA) and Hillsborough County (NH): Search for any with a daughter Hannah, or bequests to “my daughter Hannah, wife of Josiah Sawtell”
- Land deeds may also help: Did Josiah and Hannah acquire land from a Smith family?
- Did any Smith sell or gift land to Josiah?
📂 Resources:
- FamilySearch catalog (look under county-level records)
- American Ancestors (NEHGS)
- County Registry of Deeds (some are online)
6. Study the FAN Club in Depth
Use the Friends, Associates, and Neighbors of Hannah and Josiah:
- Who witnessed their marriage?
- Who are neighbors in census records?
- Who are godparents, witnesses, or co-signers in deeds or probate?
If you can find another Sawtell married to a Smith (or a Chester Smith near the Sawtells), it could signal a family connection.
🧠 Tip: Look at marriages of other Smiths in Amherst between 1775–1795—see if any patterns emerge.
7. Use DNA Evidence If Available
If you or a descendant has tested:
Look for matches that descend from Smith families in Hampshire/Worcester counties
Use the Shared Matches tool on Ancestry or WATO on DNA Painter to test possible parent candidates
Explore clusters with shared ancestry in the Amherst/Townsend area
You may find a Smith cousin match that descends from one of Hannah’s siblings.
8. Search for Published Histories, Genealogies & Manuscripts
- These often name Smiths overlooked in vital records:
- Amherst, Townsend, Ashburnham town histories
- NEHGS articles and member trees
- Genealogical journals or compiled family histories (e.g., Descendants of Thomas Smith of Amherst)
- Local historical societies may have Smith family manuscripts
📘 Use:
- Google Books
- Internet Archive
- HathiTrust
- NEHGS Digital Collections
✅ Summary Table (Revised)
Step Goal Tools
1 Analyze 1789 marriage Vital records, church register
2 Create timeline of residence Censuses, town histories
3 Analyze children’s names Naming patterns, local name matches
4 Identify Smith families 1760–1790 Vital records, censuses
5 Search probate & land records County deeds, wills
6 Study FAN network Marriage witnesses, neighbors
7 Use DNA analysis Ancestry, WATO, GEDmatch
8 Consult published sources Google Books, NEHGS, archives
Would you like help researching Smith families living in Amherst around 1768? Or want to explore possible DNA match connections to this line?
6) I asked Perplexity and Grok 3 Deep Think about this brick wall problem also. I will highlight their responses in a later blog post.
7) The ChatGPT4 review and research process uncovered several approaches that I previously have not considered, especially the name issues. There are many Smith families in the towns mentioned in this time frame, and more in adjacent towns. I have investigated available birth and baptism records in the counties named, and have found some possible parent candidates. There are no probate records for either Josiah or Hannah in the available Massachusetts and New Hampshire records. I have collected some land records for Josiah Sawtell, but none mention Hannah, and none have a Smith witness or neighbor.
8) This defined process provides an excellent outline of how to solve this type of brick wall problem. What did it miss? What did it get wrong?
1 comment:
I'm not overly fond of AI, but this is a great use of its capabilities. Hope you have success.
Post a Comment