This page starts with an article giving useful advice for genealogists researching their Irish roots. Under the article you will find an overview of the main genealogical archives and libraries in Dublin as well as a guide to online resources.
Here we will share with you the main areas where you can go wrong in your research and give you tips and examples of how to prevent making these common mistakes.
Family Memory, Lore and Stories
Every family history starts with talking to your family members, drawing upon their memory, and recording their stories. For one family, we received the information below:
“Mary married Carroll O’Donnell and they had one son, who died about 1904. Mary fell and broke her leg which caused her death in 1938. Carroll died in 1937-8.”
The information was recorded in a very detailed memoir, written by a family member who had lived from 1897 to 1985, i.e. the events had happened during her lifetime. Should you therefore conclude it is reliable?
Our research found that their son, Joseph O’Donnell, was born on 8 March 1898 in Watertown, Jefferson, Wisconsin but unfortunately died on 20 July 1898 of that same year (not in 1904). Mary died on 6 Apr 1925 (not 1938) in Watertown, Jefferson, Wisconsin. She fell on an icy sidewalk and died of myocardial insufficiency (heart failure). It is noted on her death certificate that a contributory factor was senile dementia and fracture right femur (leg) that had occurred 43 days prior to her death. Carroll O’Donnell died on 12 July 1932 (not 1937/1938) in Watertown, Jefferson, Wisconsin.
The problem here was that the person who wrote the diary was based in Manchester, England, and her Irish ancestors she wrote about had lived in Wisconsin, USA. Communications were very basic in those days. We also suspect that the diary was written many years later.
Another example: we were told a story by someone that a family member of his – many times removed – had fought in the Second World War in the Dutch East Indies (now Indonesia) and had been part of a group of soldiers who had arrested “rebel leader” Sukarno, who would later become the first President of Indonesia. Unfortunately, the story is not true: the family member in question had fought in the Second World War in that part of the world, but had been made a Prisoner of War and forced to work on the Burma Railroad.
In other words: beware of embellished family stories. The bigger the claim, the more suspicious you should be.
Family stories are a great source of information. But the greater the distance, time lapsed or more removed the family member, the less reliable it is.
Online family trees
Most family history researchers spend years trying to find their ancestors and get back in time. A researcher we know managed to his delight to get back to the seventeenth century in just a few days … or did he? In fact, what he had found were several family trees that were published online. These can be very tempting, especially if they link you up with famous people or families, or – as in the example above – bring your family history back several centuries. Online trees can be of significant help to researchers, but can also be very dangerous if blindly followed.
First, you need to proof that the tree you find online is related to your family, and not a family of the same name. We once chased up a very promising lead, found lots of information on the individual and his family members, only to find out that he was not related to the family we were researching at all: there were two people from Ireland, of the same first and surname, with wives of the same name, both of similar age and living in New York at the same time.
Even if you proof beyond reasonable doubt that a person in a tree is one of your ancestors, keep in mind that this person – your ancestor – might have been added in error. It does not proof that the rest of the tree is related.
Which brings us to a second point: a family tree made by someone else is only as good as the quality of their research. Before accepting the members of the online tree into your own, you will have to check each and every person. This is of course made much easier, if the tree contains sources. The absence of which, should lead you to question if you should use the tree at all.
When we were given the tree mentioned above, we checked each entry and its source where mentioned (or find one source where not). This way, we were able to correct a lot of names, add dates, exclude names that had been added erroneously but also add new names we found.
Finding an online family tree is great because it gives you many names that you can research, people you might not have been aware of. But you should never accept any of them without doing the proper research yourself.
He/She must be related
In a family tree we were given, all households with the same surname in a small village were included in a tree as siblings. And this appeared to be a logical assumption as we found this to be case for almost all of them. Bar one. One of the households turned out to have moved there (the others had all been born in the village) as we found his birth in a city nearby, born to different parents. The mistake here was that because all others had been found to be siblings, it was assumed that all of them had to be.
It must be him/her
A related error is one that we saw in another tree that was provided by a client. The client had identified his ancestors in a census and had then traced them back and forwards in other censuses. In the original census, there was a small child. She was however not living with her parents in the next census, 10 years later. She was too young to have married. The client had found a girl of the same name and age (i.e. 10 years older) in the same area, described as a niece of the female Head of the Household and concluded she must have lived with her aunt. It was true that in that area there were no other girls of that name and age who were not living with their parents. And we did not find any Death Certificate for the girl. However, we looked at the composition of the “Aunt’s” household, and found it contained a sister-in-law and several nieces and nephews, none of whose names corresponded to the original family. We were also able to find all other family members of the original census in the later one bar one (there were quite a few children), including two older brothers who had started their own families. In this case, the girl found with her aunt, does not pass the logical test: it is more likely that the girl was a daughter of the sister in law; and also, if the girl was not living with her parents, why would she not be living with one of her married brothers? It’s therefore unlikely, but definitely unproved that the girl in question was related.
It’s in the records
You have not jumped to any conclusions, not made any assumptions: it is there in writing! However, don’t forget that the people who wrote down the information were human. Church records were recorded by local clergyman, not all of whom were proficient in the art of writing. Nowadays, people are asked to fill in their own census form, but in the past this was done by enumerators. They would make mistakes in names, in ages, professions or were misinformed. They might even have filled in the forms without visiting all of the households, but recorded them from what they (thought) they knew. And in more modern times transcription errors are not uncommon. So you cannot take what is in the records as absolute fact.
An example of how official records can be wrong, in Episode 203 from The Genealogy Gems Podcast with Lisa Louise Cooke, we heard about a case in Canada where a person was still recorded in the census at his original home town, many, many years after he had left there and had actually died in the meantime.
Dates and Ages
While spelling errors are not uncommon, dates and especially ages are more often wrong than not. This is the case in censuses, but also on marriage and death certificates.
Ages were often estimated by the enumerator or the person who came to notify the authorities of a death, and they often got it wrong.
Your ancestors might also, and I am sorry if this shocks you, have lied about their age. Some wanted to be older (a young girl working as a maid, the teenage mother of a child, a more senior person claiming to be over the age to claim a pension, etc.) and some wanted to be younger (which probably includes most people over a certain age). This is why you often see that ages increase between different censuses, but not by 10 years.
Irish family historyIn family trees, we sometimes see birth years that are clearly just a wild guess as the creator of the tree has assumed that people were born 20 or 25 years before the individual got married or had their first child. Although this might be the case for the majority, there are many exceptions. You should resist making such assumptions.
Also take into account that dates can be written differently (US versus European notation) or even use a different calendar. Also note that dates of birth and baptism, dates of death and death notification can be very different. Wills are often proved long after the death of an individual, sometimes years.
However, all this does not mean that the data is useless. They will give an indication, as a 40-year old would not be able to claim to be 20 or vice-versa. Ages can also very useful to check for generations: women did tend to marry in their twenties and have children afterwards. It was and is unusual for a woman to have a child after the age of 50. If you see the latter in the records, consider it might be a grandchild, raised by his or her grandmother but recorded as a child. We found this to be the case in a UK family we researched. An unmarried teenage girl had given birth to a son. She would later marry, but the child stayed with his grandmother, who recorded him as her son. The child would later fight in WW I and become a police officer afterwards.
Preventing these errors
Above are just some of the most common errors. Any seasoned researcher of family history will certainly be able to add to this list.
To prevent them, it is important that any piece of information you find is backed up. Don’t assume! You will need to find a reliable source (preferable official records, primary sources, etc), and don’t forget once you do to record these in a research log with source citations.
Circumstantial information is very important: check the maiden name of a married woman, check names and ages of siblings, check the address where your ancestors lived, etc. Especially names and ages of siblings are important. Names might be spelled differently, or an Elisabeth might have become and Eliza, ages might be all over the place (but the ORDER in which the children were born seldom is); but if they all match (you are unlucky if there are no or only a single sibling with a common name) it is much more likely you have found the correct family. Make sure though you get all information and resolve any conflicts.
If you are estimating anything, or are not sure about a piece of data, always mark these clearly as such. This prevents you later from using this data as fact, because you forgot you were not sure about them.
The most important advice that we can give is that data is sometimes not available yet or might have been lost. No matter how badly we might want to find an ancestor, sometimes it is just no longer possible. The worst thing to do is to reject this possibility, assume the information must be somewhere and shoehorn some individual into your family history. Remember, the information you find should be logical, the data has to fit and be believable.
By clicking the relevant archive you will be presented with information on where they are, a link to their website, a summary of what can be found in them and tips on how best to research the collection.
Of particular interest will be the last link, which will bring you to an extensive guide to online research. Categorised by type of collection, it will tell you which online resources are available, including links and if these collections are freely available or require payment or subscription.
Copyright © 2016 Genealogy.ie