In an early phase of this project we identified all possible SAINSBURY descendants who matched any of our “Somerset SAINSBURY” cousins on Ancestry DNA. We did this hoping a preponderance of evidence would reveal the geographic origins of the SAINSBURYs of north Somerset.

However, instead of clustering around a single point of origin, multiple sets of matches, all equally strong (or weak) came to light. They were considered and assessed in a 2019 data analysis blog post and subsequent Research Roundup.

Later on, Y-DNA evidence seemed to confirm that our north Somerset SAINSBURY ancestor, Richard Sainsbury, was the Richard SAINSBURY who was baptised in Urchfont, Wiltshire in 1708. (See, for example, Y-DNA supports Urchfont origins of north Somerset Sainsburys).

Since then, however, two things happened:

1. a man who descends from a SAINSBURY family in West Lavington, Wiltshire had a Y-DNA test that matches a representative of the north Somerset SAINSBURYs. This challenged our conclusion that our Somerset ancestor came from Urchfont;
2. Family Tree DNA reconfigured their estimate of the time to the most recent common ancestor we share with other SAINSBURY and SANSBURY families who share our Y-DNA pattern. (See Time to Most Recent Common Ancestor Shifts Way Back)

Family Tree DNA’s estimate of the time to the SAINSBURY ancestor we share with other matching Y-DNA families was around 1650. This corroborated the “Urchfont” theory. However, it is is now set as far back as 1450. (See Sainsbury-Sansbury Group Time Tree. R-M167.) That certainly leaves open the possibility that the Urchfont evidence described in this blog is a series of red herrings, and our earliest known north Somerset Sainsbury ancestor, Richard SAINSBURY, may have come from elsewhere.

A new angle; an old segment; a one-in-four chance

Y-DNA traces the direct male line, but the type of DNA tested by Ancestry DNA tests all lines. In theory, to use that type of DNA to find the origins of our earliest known Somerset SAINSBURY ancestor, the most valuable DNA would be any segment shared by distant but known cousins. If the DNA those cousins share is also found in others who can trace their family to a SAINSBURY line outside north Somerset, that family may well be our “pre-Somerset” SAINSBURY family of origin.

With this in mind, I was keen to examine a segment of DNA shared by a pair of fourth cousins whose most recent common ancestors are Richard Billett SAINSBURY (1753-1811) and his wife Betty SPROD. Richard Billett SAINSBURY was the son of the Richard SAINSBURY who came from elsewhere in the mid-1700s.

All things being equal, there was approximately a one-in-four chance that this segment of DNA was inherited from the Richard SAINSBURY who migrated from “who knows where” to north Somerset prior to 1745. Not great odds, but worth a try.

Why a one-in-four chance? Because this segment of DNA was probably inherited from a single ancestor. If we consider the ancestral couple shared by these cousins, the wife, Betty SPROD (1750-?) was from Somerset, as were her parents and grandparents. The husband, Richard Billett SAINSBURY (1753-1811), had (most likely) a Somerset mother and a SAINSBURY father, also named Richard, who came from somewhere else. Richard SAINSBURY (père) represents the one-in-four chance that this piece of DNA could reveal our SAINSBURY family of origin.

Having identified this piece of DNA, the next step was to find anyone else with the same DNA and a traceable family tree.

I found one such match on MyHeritage, indicated as M.B. in the downloadable family tree, below. His family tree was very helpful because it only includes one recent Wiltshire line in a family of mainly US ancestors. That Wiltshire line goes back, in just a few generations, to a TAVINER family. In turn, that TAVINER or TAVERNER family descends from a SAINSBURY ancestor who was born in the village of Castle Combe, Wiltshire.

Coincidentally, this person’s family lines in Wiltshire line also go back to a LESSITER family in the village of Old Sodbury, Gloucestershire that goes back (with a gap in the parish records, but with some confidence) to the same SAINSBURY family in Castle Combe.

Castle Combe could therefore be the place of origin of our earliest known Somerset Sainsbury ancestor, Richard SAINSBURY (?-1785).

This possibility appeared early in this project, and the evidence at that time was described in DNA Match Profile: The Saintsbury Family of Castle Combe. However, further investigation at that time proved fruitless and was set aside. This was done in part because the Castle Combe parish records, while complete for this period, do not include a baptism for a Richard SAINSBURY in a range of years that would make him our likely ancestor.

There was, however, a six-year gap (from 1710 to 1716) in the baptisms of children of an Anthony SAINSBURY and his wife Grace. Given this latest DNA evidence, could our Richard SAINSBURY have been born to those parents in that gap? It would make him about 30 years old when he married Mary WILLIS in Portbury, Somerset in 1745. If he was, in fact (and as this DNA evidence suggests) the son of Anthony and Grace SAINSBURY, he may have been privately baptised at home, and therefore not recorded in the parish register.

To corroborate this match, I looked for others with the same DNA on GEDmatch, MyHeritage, and Family Tree DNA. They soon appeared. One, G.D., is also on MyHeritage. His line goes back to a MINER family who, over several generations, lived in Morpeth, NSW, Australia; Monkton Combe, Somerset; and Westwood, Wiltshire. Ultimately, this line can be traced (with some speculation) to Lucy TAVINER, the daughter of Robert TAVINER who (again, with a lack of documentation but with good confidence) may have been the son of George TAVERNER (1675-1739) and Mary SAINSBURY (1677-1754). Mary was born in Castle Combe.

Another match, M.B.’s third cousin, was located on GEDmatch. This corroborated the fact that this DNA, in the case of their family, was inherited from either a TAVINER or LESSITER ancestor, and ultimately, a SAINSBURY ancestor in Castle Combe.

UPDATE (April 23, 2023): Soon after this post was written, another individual with the same segment of DNA, L.E., was identified in the UK. This match indicated which of two family lines transmitted the DNA found in LESSITER/SAINSBURY cousins M.B., N.W., and L.E.

(Technical note on inconsistent matches across genetic genealogy platforms: Fifth cousins R.Sainsbury and R.S. used to show as 7 cM matches on Ancestry, before that company removed matches below 8 cM. They do not show as matches on MyHeritage, but GEDmatch shows a 12 cM match. L.E. matches R. Sainsbury on MyHeritage but not nearer cousin M.B. (who, oddly, does match R. Sainsbury on that platform). G.S. shows as a match on MyHeritage to R. Sainsbury, but not M.B. or L.E. This triangulation group was only identified thanks to the tools available on GEDmatch.)

Richard BILLETT Sainsbury (1753-1811)

A family tree showing these DNA cousins and how they may be related is available for download below. It includes Anthony SAINSBURY jr. (1683-1717) and his wife, Grace, as the possible parents of Richard SAINSBURY of north Somerset.

Unfortunately, there seems to be no marriage record for Anthony and Grace, so her surname is unknown. But if this tree is correct, their son, Richard, gave the surname BILLETT to his own son, Richard. Is this a clue that Grace’s surname was BILLETT? Again, we run into a documentary brick wall, because there is no Grace BILLETT baptised in the area in a range of dates that would make her Anthony SAINSBURY’s likely wife.

However, this line of research has just begun, and there will be lots more to discover in the months ahead.

Does your family include a TAVINER, TAVERNER, LESSITER, MINER or other line that can be traced back to the SAINSBURY (aka SAINTSBURY) family of Castle Combe and area?

Please join the Sainsbury-Sansbury Group project on Family Tree DNA so we can investigate this together!

Disclaimer: As with any research project, when new evidence comes to light, former theories may change. This blog post includes theories and conclusions developed from the best available evidence at the time this post was written. It may be corroborated or refuted by later research. This post must therefore be considered in the context of all information presented in earlier and later posts.