Share this story Share on Facebook icon Share on Twitter icon Share on Pinterest icon Share on Google Plus icon Share on Linked In icon Share via Email icon

Stuck on your family history? Yvonne McKinney


Come along to Family Tree Live and get one-to-one expert help from a professional researcher to help solve that family history puzzle you’ve long been wondering about…


The one-to-one expert sessions are manned by professional researchers from AGRA (the Association of Genealogists and Researchers in Archives).


Family history enthusiast Yvonne McKinney wanted some help to find out about a brickwall in her friend’s family tree. We asked AGRA member professional researcher David Annal to share his thoughts on what the evidence reveals … It turned out that this was the most confusing – and fascinating – case that David had worked on in over 30 decades of research!


Yvonne McKinney’s question:

I am researching the family tree of a friend and I’ve come up against a few brick walls in regard to her great-grandfather, Thomas Harper. The facts about him are rather few.

• The 1911 Census

In the 1911 Census, Thomas lived in Barker’s Yard, York, with his wife, Christiana Oliver, and their three daughters. According to the census, Thomas was about 30 years old and was born in Bishop Wilton, near Pocklington, East Yorkshire. He was a farm labourer.

• His marriage

He and Christiana were married in January 1907 at Clifton Parish Church, York. His father was Thomas Harper, labourer, deceased. One of the witnesses was a J Harper. Thomas himself was described as a farmer, living in Rawcliffe, York.

•His death

Thomas died in Stobhill Hospital Glasgow on 2 September 1919, of heart failure. His parents were Thomas Harper, labourer, deceased and Mary Richmond, deceased. Thomas had been in the Army. Firstly in the East Yorkshire Regt. no. 19302 and latterly in the RAOC, no. 040415. He is buried in Fulford Cemetery, York.

So far, so good. But I have several puzzles.


So, what’s the mystery?

The first mystery is his parents. The only couple I’ve been able to find with those names, who married each other, were a Thomas Harper and Mary Richmond who were married in Church, near Accrington, in 1844. Although they had a son called Thomas, he grew up to be a tailor and seems to have lived all his life in Accrington. Similarly, I can find nothing to suggest that Thomas and Mary ever came to Yorkshire.

Secondly, I cannot find anyone I can identify as Thomas on the civil births lists.

I also cannot find him or anyone who might be him, on the 1891/1901 Census returns.

I have considered that his name may have been misspelled as Harker, or Harpur or even Barker, but that has not led me anywhere. Maybe he was living somewhere else in the country other than Yorkshire in those years. I am also wondering if his mother’s name was written wrongly and was not Richmond at all, though they had his address in York correct and I assume they got this information from his Army records.

I have truly reached a brick wall and would appreciate any help...


AGRA researcher David Annal’s answer:

When I’m asked to look at a case like this, I like to start by considering what we actually know about the person whose origins we’re looking for. On the face of it, we know quite a lot.

Taking stock of what we do know

• The 1911 Census tells us that Thomas was born in Bishop Wilton, Yorkshire and that he was 28 years old.

• His death certificate gives his parents’ names as Thomas Harper and Mary (née Richmond).

There’s a slight discrepancy with his age here; between the 1911 Census and his death certificate in 1919, but, either way, we’re looking at a birth sometime in the early 1880s.

It’s clear that there are no records of anyone fitting Thomas Harper’s description prior to his 1907 marriage but nevertheless, we still have plenty to work with.

We would have had more had his Army service record survived but unfortunately, along with about 60 per cent of our ancestors’ WW1 service records, Thomas’s papers must have been lost during the air raid which destroyed the War Office repository in London in September 1940.


Assessing what we know

So, we’ve got a name (or rather, a forename and a surname), an approximate date of birth, a place of birth and the names of his parents. Obviously, some of this is wrong (we wouldn’t be struggling to find him if it was all true) and we need to learn to examine the evidence and think about which ‘facts’ seem to be most reliable and use these to focus our searches.

What we’re looking for are what I like to think of as little nuggets of truth and, to me, one item here stands out like a sore thumb; his place of birth.


The vital nugget to work on

With all due respect to the inhabitants of Bishop Wilton, past and present, nobody is going to claim to have been born there unless they actually were, or at least genuinely believed that they were. You might, if you were living some distance from your place of birth, give the name of the nearest large town, but Bishop Wilton had a population of around 500 towards the end of the 19th century. The point is that, if you said you were born in Bishop Wilton, you almost certainly were.

This, then, is our little nugget of truth and it convinced me to focus my search on Bishop Wilton.


Narrowing the search area

When I found, in the 1881 Census, a 12-year-old boy called Richmond Harper, I felt that I was definitely onto something. Particularly as Richmond had a younger half-brother called Thomas. Admittedly, his surname was Conway, not Harper, but his mother was called Mary and he was born in the early 1880s, so it was all looking good.

Of course, there’s no sign of a birth registration for a Richmond Harper but we can easily find the birth of Thomas Conway, registered in the Pocklington registration district in the March quarter of 1881 and, thanks to the GRO’s own birth index, we can see that his mother’s maiden name was Richmond. This proved to be the key to unlocking the story, and it turned out to be one of the most fascinating – and confusing! – cases I’ve ever worked on.


Going back to the previous generation

It all starts with the baptism, at Bishop Wilton on 22 April 1840, of Mary, the illegitimate daughter of Sarah Wilson. Seven months later Sarah married John Richmond at Holtby parish church, a few miles from Bishop Wilton, and whether John was Sarah’s natural father or not, he appears to have brought her up as part of the family; she is listed as Mary Richmond in the 1841 and 1851 Censuses.

The 1861 Census finds Mary working away from home, as a kitchen maid in the Queens Hotel, in Withernsea, a seaside resort on the East Yorkshire coast. A few months later, she was married. The entry in the GRO index for the June quarter of 1861 lists the bride and groom as Mary Wilson and William Penrose Duckitt. Mary was evidently using her birth name (I would be interested to see the actual marriage certificate to see how her father is described).

The record of William’s baptism, which took place at Allerthorpe, a few miles to the west of Bishop Wilton, on 25 November 1829, appears to be the ultimate source of the name Harper; William was the illegitimate son of Mary Ducket and George Harper. In 1841, a 10-year-old boy called William Penrose is living with the family of James and Jane Harper in Allerthorpe; this must surely be our William, but we then lose sight of him until the time of his marriage to Mary, 20 years later.


A young widow and a confusing surname

Mary and William had three children born over the next seven years: John (1862), Ada (1865) and Richmond (1868). The births of the first two children were registered under the name Duckwith and the third as Duckworth, with the mother’s maiden name given as Wilson for the first and Richmond for the other two. The children were all born in the Pontefract/Castleford area, some 30 miles from Bishop Wilton.

William died in September 1869. The coroner’s report indicates that he died after falling from a hay cart; his name is given as William Duckworth. This change from Ducket/Duckitt to Duckwith/Duckworth is difficult to explain.

There was a Duckworth family living in Bishop Wilton at the time and it’s possible that Mary (assuming she was the informant when the births of the three children were registered) had somehow conflated the two names.

Two years later, Mary was back in Bishop Wilton, living with her widowed father (or stepfather), John Richmond. Mary and the three children are all listed in the census under the name Duckwith.


A second husband, and a further new surname

Then, in 1873, Mary remarried; her second husband was an Irishman, called Thomas Conway or Cornway. She and Thomas went on to have at least six children, all registered under the surname Co(r)nway: George William and Sarah Ellen (twins in 1873), James (1875), Sarah Mary (1878), Thomas Henry (1881) and Lizzie (1885). The births were all registered in the Pocklington district, with the exception of James, who was born in Leeds. Sarah Ellen died in Leeds in 1875.


And then we come to the 1881 Census

This is the document which opened up the whole case. Thomas and Mary Conway are living in Bishop Wilton together with five children: Richmond, George William, James, Sarah Mary and Thomas Henry. The four youngest are all entered as Conway but Richmond has the surname Harper? Again, this is difficult to explain; Richmond’s father, as we have seen, was the illegitimate son of George Harper and Mary Ducket but there’s no evidence that he ever used the name Harper himself. Why would Mary suddenly start using the name Harper for her son, instead of Ducket or Duckwith/Duckworth? And it gets even stranger, because both John and Ada, who were living away from home (as a farm servant and domestic servant respectively) also appear in the 1881 Census with the surname Harper.

The next 10 years saw three deaths in the family. The youngest daughter, Lizzie Conway, was born and died in 1885 and the following year, her mother Mary died, aged 47 (according to her death registration). And then, in early 1891, Thomas also died, leaving four children all aged under 18. The youngest, Thomas, was just 10 at the time.

Their grandfather had died in 1882 and it was left to their uncle, John Barnett Richmond, to look after them. Gradually, they all moved away from Bishop Wilton and all of them, with the exception of Sarah Mary, adopted the surname Harper, which seems even stranger than their half-siblings doing so; at least they had a genealogical connection with the name Harper, however slight. The Conway children’s only link to the name was that their mother’s first husband’s, illegitimate father’s name was Harper!

George William and James moved to Otley, near Leeds and it’s likely that Thomas did too. His whereabouts in the 1901 Census are uncertain but he is probably the 20-year-old Thomas Harper, ‘farm servant and milk deliverer’, working for a farmer called John Goodrick in Harrogate.

His place of birth is given as Otley so it might not be him but is likely that he had been living in Otley and that the farmer made an honest mistake on the form; both George and James were milk dealers so there’s an occupational connection as well.


The final piece of proof

Mary’s oldest child, John, meanwhile, had married and settled in Bugthorpe, to the east of York, now calling himself John Duckwith Harper. And the 1911 Census entry for John and his family provides us with the final proof that we’ve got the right people. When your Thomas Harper married Christina Olive at Clifton parish church on 8 January 1907, one of the witnesses signed the register as J Harper. A comparison of this signature with the signature of John Duckwith Harper on his 1911 Census schedule leaves little room for doubt that it’s the same person.

I’m not sure that I’ve ever worked on a more confusing family. The woman who was born as Mary Wilson in 1840, appears in the 1841, 1851 and 1861 censuses as Mary Richmond, marries William Penrose Duckitt (or Harper) in 1861, turns up in the 1871 Census as Mary Duckwith before marrying Thomas Conway in 1873. The only name that she doesn’t appear as, is Harper, the surname which virtually all of her children adopted.

If we return to the evidence left behind for us by Thomas Harper of Clifton, York, we can now see that the least reliable ‘fact’ was his surname; it was only by letting go of the name that we were able unravel a quite remarkable tale, but one which is perhaps not as unusual as we might think.


Come along to Family Tree Live where you too can get one-to-one expert help from a professional researcher in an AGRA one-to-one expert advice session.

The one-to-one sessions are included in your Family Tree Live day ticket price.

To book your ticket to Family Tree Live, please go to: https://www.family-tree.co.uk/ftre/show/family-tree-live/book-now

To pre-book your one-to-one advice session, please click here.


About the AGRA expert

David Annal has been involved in the family history world for more than 30 years and is a former principal family history specialist at The National Archives. He is an experienced lecturer and the author of a number of best-selling family history books, including Easy Family History and (with Peter Christian) Census: The Family Historian’s Guide. David now runs his own family history research business, Lifelines Research. https://lifelinesresearch.co.uk/


Back to "Getting Started" Category

01/04/2019 Share this story   Share on Facebook icon Share on Twitter icon Share on Pinterest icon Share on Google Plus icon Share on Linked In icon Share via Email icon

Recent Updates

Charles Booth Poverty Maps of London added to TheGenealogist's Map Explorer

TheGenealogist’s Map Explorer now includes a feature that allows users to access Charles Booth's ...

3 key resources for today's genealogist; what to look for & where to find them

Are you just starting your family history, or wishing to brush up on your online research know-how? Our 3 key ...

How to create a family tree chart

Find out which family tree chart is best for you, whatever stage you’re at with your family history research, ...

Making a Mayflower 400 pilgrimage

Four centuries ago the Mayflower Pilgrims set sail for the New World. Discover must-see places to include ...

Other Articles

Which family tree chart should I use?

Confused about the many different genealogy charts that are available? Here's our expert help! ...

Boris Johnson is related to ten US presidents

New research by Gary Boyd Roberts of American Ancestors explores the family tree of UK prime minister Boris ...

FamilySearch users can now correct name indexing errors

FamilySearch have today (31 July) announced a new facility that allows users to make name corrections to its ...

DNA classes at RootsTech London: guest blog

If you are joining us in London for RootsTech this October, now is the time to start organising your ...