Using ChatGPT for Genealogy

I've been using ChatGPT to aid various aspects of genealogical research and writing. I'm learning as I go and thought I'd share some observations at this juncture.

I subscribe to the Plus Plan, which is $20 per month. There is a free version and a more expensive Pro Plan. I do not have these other plans, so I do not know what the results would be with the free or Pro plan.

My ChatGPT is geared towards me and what I have told it I am seeking. The same question that you pose within your profile would probably yield different results in mine.

Technical Help with Blogging Platform

Since August 2010 I have been using the Blogspot platform. One of the purposes of writing here is to attract people researching the same historical figures. Google no longer indexes every page, causing entire articles on particular people to be hidden from search results. In addition, I have been unable to alter information in tabs, delete them, or add new ones.

I have not found a solution through ChatGPT. It suggested adding headers through the formatting options. This does not seem to help.

I am exploring publishing on other platforms.

Review of Writing

ChatGPT can review my writing and offer suggestions about flow, grammar, and overall impact. I have inquired about specific sentences that sound awkward; ChatGPT offers rewording that is clearer.

I clarify that the purpose of the writing is genealogical. This is important for the tone. Writing on this genealogical blog is vastly different from business or legal writing. You need to inform ChatGPT about the goal of your particular piece of writing.

Revising Pictures

MyHeritage began offering photo revisions years ago.

ChatGPT can also revise pictures. I asked it to fix older pictures. ChatGPT took liberties with the pictures, adding items and features that were not present.

Original picture of Ethel Laurel Winterton (1891-1962)
Probably her high school graduation circa 1909

Colorized and fixed by MyHeritage

Colorized and fixed by ChatGPT
The blemishes were erased.
Note that the diploma in her hand was changed to a rose!

Answering Genealogical Questions

The continued lure of researching past people lies in figuring out their connections, locations, and dates.

ChatGPT cannot find all the documents about someone, analyze them, and spit out the answers.

ChatGPT can find sources on the internet and offer general genealogical rules.

Below is the inquiry I made for Ruben Bishop who died in Morris County, New Jersey in 1829. ChatGPT found the index of death notices online, which connected Ruben to Connecticut.

---------------------------------------------------

Me: genealogy research question. Ruben Bishop died in Morris County New Jersey in 1829. His will named brothers Calvin, Luther, Miles, and Abner. I do not have their ages or places of residence. Presumably they died after Ruben died in 1829. Are there any mentions of these people on the internet?

ChatGPT: Short answer: very little shows up online (as of now) for those specific brothers by name, but there are a couple of useful clues and leads—including confirmation of your Ruben and context that may help you push further.

What does show up for Ruben Bishop

One solid hit:

• An early Morris County death notice index includes:
  “Bishop, Reuben (Woodbury, CT)… Sept. 11, 1829… Doctor… of Bottle Hill.”

This is important because it:

• Confirms death in 1829 (matching your will)
• Ties him to Bottle Hill (now Madison, Morris County)
• Suggests he was originally from Woodbury, Connecticut

👉 That Connecticut connection is a big clue—it suggests the brothers (Calvin, Luther, Miles, Abner) may not have remained in Morris County or may appear in Connecticut-origin families.

---------------------------------------------------

Citations

I find genealogical citations challenging. You should provide the title of the document, database, film number if applicable, custodian, and information about online location. It's a layered approach.

The guide for citing genealogical sources is Evidence Explained by Elizabeth Shown Mills.

The articles here on this blog are not formal writings for publication in journals, so I try to not delay publication here caused by fixation on proper citation format.

I've been asking ChatGPT for citations specific for use in genealogy, as opposed to Chicago, MLA, or APA. I provide an online link if available. ChatGPT looks at the webpage and returns suggested citations.

Creating Family Tree Diagrams

I need family trees to illustrate the relations among the people I write about. Sometimes I screenshot from Ancestry or Family Tree Maker. Often this is insufficient to show all the people needed to illustrate a point. I have created my own tree diagrams in Microsoft Paint. Now I use Canva. This is time consuming.

Family tree diagram created by me in Canva

I asked ChatGPT to create a family tree diagram of the children of Jonas Long and Elizabeth Merrell. I typed everyone's names and dates. I asked for and received revisions. This is pretty good for a few minutes of work on my end.

Family tree diagram created by ChatGPT

Creating Maps

I asked ChatGPT to create a map of three towns in New Jersey. The Plainfields share borders, but lie in three different counties. I wanted a map that clearly showed these three towns. The map created by ChatGPT was inaccurate.

Map of the Plainfields in New Jersey
Map on the left from New Jersey Department of Transportation
Map on the right from ChatGPT

Handwriting

ChatGPT can transcribe handwritten documents. Wow!

Difficult words might be revealed.

It's not perfect.

ChatGPT's transcription of the above portion of this document:

---------------------------------------------------

This Deponent further saith that he saw the said
Woodward at [Aikens?]    on the Evening of the same Day
where he heard him boast that he had disarmed
sixteen or seventeen of the Rebels — and farther the
Deponent saith not.

[signed] William Imlay

Sworn in Council of
Safety 12 Janry 1777
Before me —

Wm Livingston Prest.

---------------------------------------------------

I informed ChatGPT of errors.

Me: "Aikens" is "Trenton"

ChatGPT: Thank you — that reading fits the handwriting and the context much better. Here is the corrected portion:

This Deponent further saith that he saw the said
Woodward at Trenton on the Evening of the same Day
where he heard him boast that he had disarmed
sixteen or seventeen of the Rebels — and farther the
Deponent saith not.

This is an interesting Revolutionary War-era deposition. The reference to General Howe is almost certainly William Howe, commander of British forces during the New Jersey campaign.

---------------------------------------------------

Me: "with the enemy on" is the phrase inserted between lines

ChatGPT: Yes — that insertion makes the sentence read much more naturally. The corrected passage would be:

This Deponent further saith that he saw the said
Woodward at Trenton with the enemy on the Evening of the same Day
where he heard him boast that he had disarmed
sixteen or seventeen of the Rebels — and farther the
Deponent saith not.

---------------------------------------------------

Me: The month is April, not January

ChatGPT: Good catch. The attestation at the bottom should read:

Sworn in Council of
Safety 12 April 1777
Before me —

Wm Livingston Prest.

William Livingston was serving as president of the New Jersey Council of Safety at the time.

---------------------------------------------------

Translations

ChatGPT can translate from one language to another.

I uploaded the death notice for Charles Loihle (1822-1891). The print is a bit smudged and is difficult to read for English speakers because the typeface is German Fraktur. This German notice appeared in the New Jersey Deutsche Zeitung, while an English notice appeared in the Newark Evening News.

Death notice of Charles Loihle.
December 11, 1891.
New Jersey Deutsche Zeitung, Newark, New Jersey.
Jody's German transcription is in the upper left; Jody's English translation on the right.

ChatGPT and I went back and forth as we did with William Imlay's deposition from 1777. The mishaps were with names, which I knew because of my knowledge of this family, but ChatGPT did not.

The other miss was in the first line with the word "hiermit." This can be translated as "hereby," but the sentence is still lacking a verb in the second position. ChatGPT offered the word as "machen," meaning to make or to do in English. I asked about this discrepancy. ChatGPT offered that the phrase is actually "machen wir hiermit" or "geben wir hiermit," but this was shortened as was typically seen in German newspapers of the time.

To fully appreciate the information in these death notices, one would have to be familiar with German vocabulary, grammar, and newspaper print of 135 years ago. ChatGPT definitely helped with this.

---------------------------------------------------

Me: The word on the first line is unclear to me. Instead of "machen," it could be "hiermit." What do you think?

ChatGPT: I think you are right. Looking closely at the first sentence, the wording is much more consistent with the standard German obituary formula:

„Verwandten und Freunden machen wir hiermit die traurige Nachricht …“

or more idiomatically:

„… geben wir hiermit die traurige Nachricht …“

The blurry word after “Freunden” does not really resemble machen in Fraktur. The visible letter shapes fit hiermit better — especially the tall initial h and the clustered middle letters.

So the opening likely reads:

Loihle.—Verwandten und Freunden machen wir hiermit die traurige Nachricht, daß unser geliebter Vater …

English:

“To relatives and friends we hereby announce the sad news that our beloved father …”

That is a very typical 19th-century German newspaper death notice formula.

---------------------------------------------------

Conclusions

ChatGPT holds great potential as a resource for researchers seeking to locate records, organize research, translate languages, and create images. It is far from perfect and not a substitute for critical thinking.

I hope that this article gives you ideas to try with ChatGPT.

Private Charles Y Cook, Civil War Veteran 1833-1913

Charles Cook (1833-1913) was born in Morris County, New Jersey to Stephen H Cook and Elizabeth Vanderhoof. He served in the Civil War (1861-1865) with Company L, 27th New Jersey Infantry from September 1862 through July 1863 for a total of nine months.

Various Acts enabled Charles to file for a pension and increases. The pages in this packet reveal further information about the family. I was hoping that his pension would help link him to his great grandfather, Conrad Hopler, but not so. The issue at this point in the research is finding documentation that Charles' mother, Elizabeth Vanderhoof, was the daughter of Jacob Vanderhoof and Ann Hopler. Conrad Hopler, who served in the Revolutionary War, was the father of Ann.

If you can obtain an application for a pension for service in a war, you should do so. Anything could be in that file.

This pension record was ordered through Civil War Records. The price was $110 and the turn-around time was one month. The actual pages are in the custody of the National Archives.

Family tree showing Charles Y Cook (1833-1913),
his siblings, parents, and grandparents

Charles died of dysentery after a trip to Gettysburg, Pennsylvania, on July 11, 1913. He attended a fiftieth commemoration and died of an infection he survived fifty years earlier during the War. He was honorably discharged on July 2, 1863, which was in the middle of the three-day Battle of Gettysburg. His papers do not list this location or Battle as one that he participated in.

Death certificate of Charles Y Cook
died July 11, 1913 in Denville, Morris County, New Jersey.
Available in person or through the mail via New Jersey State Archives.

Charles Cook was buried with his wife, Sarah Merritt, at the First Presbyterian Church Cemetery in Rockaway, Morris County, New Jersey.

Jody visiting the grave of Charles Y Cook (1833-1913) and his wife, Sarah Merritt (1842-1909).
First Presbyterian Church, Rockaway, New Jersey.
May 9, 2026

This cemetery is where Conrad Hopler (1730-1816) has a gravestone. I am seeking documentation that Conrad Hopler is a great grandparent of Charles Y Cook.

Jody visiting the grave of Conrad Hopler (1730-1816).
First Presbyterian Church, Rockaway, New Jersey.
May 9, 2026

Stephen Cook and Elizabeth Vanderhoof married in 1819. They had at least five children:

-Ann M Cook (1820-1868), married Jesse Lee (1818-1888)

-Silas Cook (1825-????) (only mentioned in the books of Louisa Caroline Freeman Hickerson)

-Calvin Cook (1826-1889), married Mary Neil (1830-1898)

-William Henry Cook (1828-1902), married Emeline Young (1834-1906)

-Charles Y Cook (topic of this article)

Five children of Elizabeth Vanderhoof and Stephen Cook
identified by Ms Hickerson in her book, Vanderhoof.

I am missing at least one source of Ms Hickerson's research. I have found nothing about Silas Cook, born March 3, 1825, outside her mention.

Charles was able to claim a pension in 1890 because of service-related disability and in 1907 because he lived long enough. A card index exists. Charles has two different cards. The one at Ancestry reflects only the 1890 claim for disability. The card at Fold3 reflects both claims.

Pension card for Charles Y Cook
U.S., Civil War Pension Index: General Index to Pension Files, 1861-1934
at Ancestry

Pension card for Charles Y Cook
US, Civil War Pensions Index, 1861-1900. NARA T289. Fold3.

Pension file of Charles Y Cook
Civil War
"alleges that at Newport News [Virginia] about Feb 20, 1863 he contracted rheumatism in back and hip."

The family bible is mentioned in Charles' sworn statement from 1904. "He having been born on the 27 day of August eighteen hundred and thirty three- that his mother's family bible which is in the possession of his brother's widow contains the entry of which the following is a copy." This would have been Emeline Young (1834-1906), widow of William Cook. They had at least ten children. Who received the bible of Elizabeth Vanderhoof? I need it.

Pension file of Charles Y Cook. He mentioned his birthdate as written in his mother's family bible.

Charles clarified his name. "He always understood that he was named after a relative Charles Youmans by name but that his mother did not use the 'Y' in naming him and that he only took up the use of the 'Y' himself after reaching the age of twenty years and [illegible] did so in order to distinguish himself from others having the name of 'Charles Cook'."

First, thank you, Charles, for realizing that so many men bearing the name Charles Cook would confuse researchers for centuries. This has been my experience in researching COOKs.

Second, thank you for clarifying that you used the middle initial 'Y' and the middle name 'Youmans' and that this name was a relative's name.

Who was Charles Youmans? Charles Elmer Youmans (1809-1864) married Anna Cook (1808-1897), a paternal aunt of the younger Charles Youmans Cook.

1860 federal census
Charles Y Cook living with his mother, Elizabeth Cook [born Vanderhoof],
in Rockaway Township, Morris County, New Jersey.
By this time Charles had started using his middle initial.

Charles Cook using his middle name Youmans.
Rev. D. E. Platter (compiler), Historical Manual of the Rockaway Presbyterian Church (Trenton, New Jersey: Wm. S. Sharp, printer and stereotyper, 1880), 5.

Charles explained his family of creation. He married Sarah Lavinia Merritt in 1861. They had at least five children. I'm not sure if they had two sons named George Cook.

Charles Y Cook's listing of his wife and children

The children of Charles Cook and Sarah Merritt were: Laura, Mary, George, Phebe, and Samuel.

In First Presbyterian Church Cemetery with Charles and Sarah is a stone for George, engraved with a date of death January 6, 1866, aged two years, three days. This would give a birth around January 1864. So Mary could not have been born May 12, 1864. The pension lists George's birthdate as October 4, 1866. This could be a second son named George Cook, named after the baby who just died. There was no child named George living with this family in the 1870, so perhaps the second George also died early. Or there was only one. This remains unclear.

Sarah Merritt was also a Wiggins descendant. More research is needed on Wiggins. She and Charles Cook were probably cousins.

Family tree of Sarah Lavinia Merritt showing ancestors and her children with Charles Y Cook

Daniel A Wiggins submitted an affidavit to support Charles Y Cook's application for an invalid pension. His age was listed, which is not usually seen in older records. Looks like 47 or 49 years old in 1891. This could be David Alvin Wiggins (1845-1927). He was a second cousin of Sarah Merritt, sharing great grandparents Jonathan Wiggins (1763-1815) and Phebe Fordham (1764-1851).

Affidavit of John B Cox and Daniel A Wiggins
in support of an invalid pension for Charles Y Cook

Charles wrote his occupation (brakeman, unable to perform because of illnesses contracted from war) and residences. He lived in Brooklyn, New York from 1884-1888. This explains the birth in Brooklyn of his granddaughter, Laura Bidgood (1884-1952), and his appearance in the Brooklyn city directories.

"then in Brooklyn, NY from 1884 to 1888 . . ."

"4 years in Brooklyn from 1884 to 1888 all the rest of time in Rockaway Township, Morris County, New Jersey."

1887-1888 Brooklyn, New York city directory
Charles Y Cook, weigher, 248 South Second.
Weigher was an attempted occupation of Charles Y Cook following the Civil War.

-We still do not have a record supporting that Jacob Vanderhoof and Ann Hopler were the parents of Elizabeth Vanderhoof, wife of Stephen Cook.

-What other sources was Ms Hickerson using in her research of Vanderhoof and Cook in Morris County, New Jersey?

-Where is the family bible of Elizabeth Vanderhoof? In 1904 it was in the custody of Emeline Young (1834-1906), widow of William Cook.

Thank you, Charles Youmans Cook, for your service during the Civil War.

Tangled Family Lines at First Presbyterian Church Cemetery, Rockaway, New Jersey

On May 9, 2026 I visited the First Presbyterian Church Cemetery in Rockaway, Morris County, New Jersey. It was a rainy day, so the stones were darkened. I prefer overcast, dry days for cemetery visits, but this is how the schedule worked out.

This is an active church and cemetery. This is their website.

I saw some Wiggins stones. More articles on Wiggins are coming.

Jody at grave of Daniel G Wiggins (1827-1895) and wife Phebe Belcher (1830-1888).
First Presbyterian Church Cemetery in Rockaway, Morris County, New Jersey.
May 9, 2026.

Daniel Gard Wiggins (1827-1895) was a granduncle of Sarah Lavinia Merritt (1842-1909), wife of Private Charles Youmans Cook (1833-1913). Charles was the reason for this visit.

Elizabeth Minton (1822-1901) and husband Jonathan Wiggins (1820-1877).
First Presbyterian Church Cemetery in Rockaway, Morris County, New Jersey.
May 9, 2026.

Jody at the Cook/Merritt graves.
First Presbyterian Church Cemetery in Rockaway, Morris County, New Jersey.
May 9, 2026.

Here's the stone of Catherine Shaw (1825-1891). She was the first wife of Nicholas Keating (1825-1898). After Catherine's death, Nicholas remarried to Mary Neil (1830-1898), my third great-grandmother. Finding this marriage lead to discovering the names of her parents. 

Jody at the grave of Catherine Shaw (1825-1891),
wife of Nicholas Keating.
First Presbyterian Church Cemetery in Rockaway, Morris County, New Jersey.
May 9, 2026.

Tangled lines: Catherine Shaw was a granddaughter of Conrad Hopler (1730-1816), making her a first cousin, once removed of Mary Neil's first husband, my third great-grandfather, Calvin Cook (1826-1889).