Last time I wrote about a family history document written several decades ago by a distant cousin and passed to me very recently by her nephew. Although full of mistakes, it was still of much value to me. Firstly, my own research broadly agreed with the names and places mentioned, and where there were discrepancies I was confident that these were down to my late distant cousin mis-remembering family stories: my research was correct. Second, although her claims about past wealth cannot be borne out by evidence so far available to me, there were sufficient verifiable facts that some aspects are worthy of further investigation. And finally, the more recent accounts, which related to my great grandfather, his birth mother and their families, were essentially family gossip, things I would never learn by reading official records. These are the parts of her story I value most. I learned a lot about my great grandfather.
Following on from that, I thought it worthwhile to look a little deeper at different types of evidence, why some carry more weight than others, and how there can nevertheless be value in all.
Historians make an important distinction between primary and secondary sources. As genealogists we tend to focus more on the distinction between original and derivative records. And yet there is overlap between all of these categories.
Primary sources are created at the time of an event, or not long afterwards. For us as genealogists these include original records from official sources, such as:
- birth certificates
- baptism records
- marriage records
- death certificates
- property documents, e.g. deeds
- apprenticeship records
- court records
- 1911 census (and subsequent censuses, as they are released)
but they will also include such things as:
- photographs of people and places
- spoken accounts by people who played a part in the event
Secondary sources tend to be published works in which the author describes, summarises, discusses or in some way draws upon information gathered from other sources (primary or other secondary). A secondary source may be produced many years after an event, and the author may have had no physical connection whatsoever with the original event. Examples might include:
- historians’ texts based on research about a person, locality, event or period of interest
- literature contemporaneous to the events, e.g. Dickens, Elizabeth Gaskell
- modern historical novels and films based on sound research
One advantage of such sources is that the author can benefit from an overview. S/he may know and understand much more than any one particular individual could have at the time. There is also the benefit of hindsight, not to mention objectivity.
Derivative records are records created after the event but based directly on an original record. As such, there is scope for error, based on illegibility of the original text, carelessness, or anything in between. Examples we regularly come across include:
- transcriptions of original records
- indexes of record collections
- census records, 1841-1901
- Note that a photocopy of an original document remains, for our purposes, an original.
So far everything seems quite straightforward, but the picture is not quite as clear-cut as it seems. Let’s consider some anomalies and grey areas.
Why is the 1911 census an original record, yet the earlier ones are not?
Since you’re reading this blog, you almost certainly know that when we find the 1911 census return for one of our ancestors, we see a single sheet completed by the head of household and relating to the members and accommodation of that household. Undoubtedly, this is an original record and primary source. By contrast, for the earlier censuses what we see is a long list of all residents in a specific locality, arranged by household, and organised according to the route the enumerator took as he walked from house to house, collecting the information. You might have explained this difference with reference to the illiteracy rates: that since a great many people were unable to read and write, the enumerator simply arrived on the doorstep and wrote down the information he was given. But this is not true. Individual records were created for each property, and this was then transcribed onto the lists we see today. The original sheets were destroyed. In other words, all we have left for these earlier censuses is the derivative record. This might explain some inconsistencies. Off the top of my head: my great grandfather George appears as Enoch in the 1891 census. Was this a mistranscription, or did George object to nosy-parkers coming to ask him questions? My great grandparents’ second child is referred to in the 1881 census as Jane (female). He was actually John, a boy.
Birth, Marriage and Death certificates: Are they original records?
You would think so. But no, they are not necessarily so. Imagine yourself registering a death in 1851. You would go to the local Registrar’s office. They would record all the information, give you your copy, keep the original for themselves and then send a third copy to the General Register Office in London. Of course, there were no photocopiers: the only way to do this was to write it out by hand. In other words, when we buy BMDs online from the GRO, what we receive is a facsimile of a copy of the original record, i.e. a true copy of a derivative record, and not the original itself. You can choose instead to buy your BMDs from the local Register Office. However, some don’t offer this service, while others don’t have the capacity for creating facsimiles of the originals: what we receive is a modern handwritten or typed copy of the original – again, a derivative record. Perhaps this might explain an odd discrepancy you may have come across?
Bishop’s Transcripts: are they original or derivative?
BT’s are an interesting grey area. They are the copies of parish registers that, from 1598 until around 1800, church ministers were required to keep and send annually to the diocese office. Even if a parish’s records from this period have not survived, there is the chance that the BTs have. They are contemporaneous with the originals, and if not actually written in the hand of the cleric, then at least by a churchwarden who may have known the individuals involved. As copies, strictly speaking they are derivative records and may contain transcription errors. However, sometimes they contain more information than the originals, and are often invaluable in providing a second chance in deciphering 17th or 18th century handwriting.
Are original records necessarily correct?
No. Even an original record can only be as good as the information given in the first place. In past centuries many people would have had no idea how old a deceased person was. Any inconsistencies between age at time of death and the age you know to be correct can often be put down to this, provided all other details are correct. Equally, sometimes false information is knowingly given, such as the tweaking of an age for a marriage, the falsification of marital status for a bigamous ceremony, and the pretence of marital status on birth registrations when the parents are not in fact married.
Think also about what is ‘truth’. A contemporary newspaper report might be considered an original record. We can expect a court reporter to faithfully summarise what happened during a trial. But what about a war correspondent? Their reports would inevitably be limited by what they actually saw and knew, what they felt was suitable for public retelling, and all this within the context of government censorship. Is it true if it is not the full truth? Is it of value nevertheless? Yes, because all of these limitations are part of the context.
The wonderful thing about online genealogy sources (Ancestry, FindMyPast, FreeBMD, etc) is that the records are indexed. As a result they are very easy to find. We just type in a few key facts and we’ll be rewarded with a selection of possible records. So much easier than going to the local County Records Office and sifting through decades of data stored on microfiche. However, the indexes themselves are a derivative record: a list of each individual record to be found within the source. As such they can and do include errors. My great grandfather, born in Leeds, is indexed on the FMP 1911 census index with a birthplace of Scotland. Consequently, it took me from 2011 until 2018 to find him. I have also found a very clearly written ‘Alfred’ listed on the index as ‘George’, and indeed an entire collection indexed under Northamptonshire rather than Norfolk. (When straightforward searches fail to return results, try searching instead for another family member – the one with the least common name.)
Contemporary values, ideas and gossip
One of the fascinating things about my late distant cousin’s story was the clarity with which she expressed the prejudices of her time. While on the one hand expressing pride that a member of the family had been involved with offering assistance to late 19th century Jewish refugees, there was an undercurrent of anti Irish / Roman Catholicism. This distinction does seem to be in keeping with other vibes I’ve picked up from this period in my home town. In this respect the issue of truth and accuracy can sometimes take a back seat, in the sense that while we need to know the facts, in order to understand the society of the time, we also need to know what people thought, what they valued, what was scandalous. Our own 21st century values may be completely irrelevant if we’re trying to understand why an ancestor pursued a particular course of action. A word of caution, too, about family stories: they are not always true, although from my experience it seems there is often an element of truth in them.
For us as genealogists it’s the detail that’s all-important. If we don’t get the names, places, dates and relationships correct, nothing else will be correct. So for us, seeing the original records with our own eyes is always the goal.
Derivative records are valuable in pointing us to the existence of the original, in providing us with information about the contents, and indeed where the originals no longer exist. But it’s worth noting when only a transcript has been seen.
Other primary sources enable us to draw our own conclusions about the life and times of our ancestors, while secondary sources add valuable context and aid understanding
Where there are discrepancies: look at the records, step back, and decide for yourself if there is sufficient compatibility with your existing evidence for the discrepancies to be down to human error, misunderstanding, illiteracy or even censorship. Rather than trying to prove yourself right, try to prove yourself wrong.
Next time we’ll think about how much is ‘enough’ evidence. I’ll provide some case studies showing how all these types of evidence can be used together to build hypotheses and ultimately to overcome doubts and reach sound conclusions.