Okanagan Researcher, Vol. 16 (1), September 1999 Lessons Learned in Researching English Ancestry
by Kenneth G. Aitken
Copyright September 1999 by Kenneth G. Aitken, ALL RIGHTS RESERVED Kenneth G. Aitken has been Prairie History Librarian at Regina Public Library, Regina, Saskatchewan for 15 years. He has been researching family history for the past 20 years, the last three as a professional genealogist.
The purpose of this article is to summarize some of the research lessons I have learned, or should have learned, in the twenty years I have been researching English ancestry. The underlying principles of research strategy, and sources in English research I first learned through the works of David Gardiner and Frank Smith. Their three volume set, Genealogical Research in England and Wales1 has been a favourite reference work of mine for 20 years. The thirteen concepts reviewed in this article are drawn from this work. The reader will note references to "we" numerous times in the article. "We" refers to a group of my Hambrook cousins who were learning together the pitfalls of English research.
This article assumes the reader has some familiarity with concepts like "primary and secondary sources" and "direct and indirect evidence". For those who need to brush up on these thing see my article "Evaluating Genealogical Evidence"2. References will be made in examples to parish registers, bishop's transcripts, censuses, probate records and other sources of English genealogical and historical data. Those not familiar with these should consult basic genealogical resource guides for England.
Genealogical and family history research seldom results in absolute proof in pedigree. Researchers devote much of their effort to sifting sorting and organizing a variety of primary and secondary source information about their ancestor. The following thirteen procedures are useful in strengthening or eliminating the acceptance of evidence, and clarifying what is truly a factual record of part of an ancestor's life.
1. Confirm the calculated or stated year of birth and birthplace from the clue document in at least two censuses.
Lets say that you have found a memorial card for your ancestor that was distributed at his funeral in 1924. From this you discover that he died at the grand old age of 88 years. You realize that this is indirect circumstantial evidence of a birth because there was unlikely to be anyone present at the death that had a perfect remembrance of his birth. But you have a clue. You do the math. 1924 minus 88, and come up with a birth year of about 1836, the year before civil registration begins in England. Can you confirm that 1836 is the year? A review of the census returns for 1841 through 1891 would prove most useful here. (Though the 1841 Census for England rounds the ages to the nearest multiple of 5, which will skew dates, it does not round off the ages for those under 15 years, a useful fact in this case.) The several census returns will also help narrow down the possible birth places.
2. Check the census of the stated place of birth in the clue document to determine where there are others of the same name, age and birthplace living there who might be confused with the likely candidate.
Among my Hambrook ancestors there is a problem with Richards: there seem to be one in every family, and they clog up the 1851 to 1881 censuses of Kent. Usually by gathering all of the Richard Hambrooks in the ancestral parish, and the contiguous parishes, I have been able to sort them out. Once in a parish in Kent, however, there were two Richard Hambrooks born less than a year apart, appearing on the census with the same age. I, of course, chose the wrong one, without accounting for all of the census returns. Thus I proposed a line of descent from a boy who died at age 11, and was found in a nonconformist church burial ground. Proceed with caution.
3. Consider ages given in as many dated events as possible to see if they agree as to the year of birth of the ancestor. In cases of wide variation, the earliest is likely to be the more accurate.
When there is a potential for confusion of identities, and there always is to some extent, we need to widen our search to examine not just census returns but monumental inscriptions, baptism, marriage and burial records, and any other records that can clarify birth dates. In my Richard Hambrook error mentioned above, if I had lined up all the monumental inscriptions for Richard Hambrooks in the parish, including those in the nonconformist churchyards, I may have spotted my error. Get the ducks lined up in a row chronologically, and value those records created earliest more than the later ones. That is to say, let later dated information be the support for the earliest data, and be hesitant to accept more precise dates that run counter to the early documented dates. From the Bible comes the notion we use in law even today; in the mouths of two or three witnesses shall the truth be known.
4. Search for birth or christening records from 5 years before the calculated birth year to the year of marriage looking for others of the same name and birthplace who might be confused with the likely candidate.
This may puzzle the researcher, so let me explain. Lets assume you have discovered the marriage of a James Atkins in a parish register in Warwickshire in 1821. The entry says he is a bachelor, but the age is simple, "of full age", meaning he is not a minor. You want to find his christening date. In cases like this I assume the person was married at age 26 plus or minus 5 years. Thus placing his birth date between 1790 and 1800. However as we are looking for a christening date we need to allow for the fact that he may have slipped through the cracks and was christened as a young adult, or was a precocious young man who married a tad early. Consequently we will widen the search beginning in 1790 and continuing to the actual marriage date in 1821.
I find this the hardest procedure to follow because there is a tendency to think that when we have found someone in the record being searched that fits our need, that we cry "Eureka!" and stop looking. Over the years I have found it best to continue plowing through the record, extracting and rounding up all the likely suspects. There can be some real surprises. While researching a family in Warwickshire last year I found a christening in a Church of England parish register that fit perfectly the profile I was looking for. However, further down the page, about 13 months later, I found another one. Now I had two James Atkins with the same parents, that fit in the same time frame I was looking for. I continued on. Another possible James Atkins appeared a few years later. He too fit the profile. I added him to the list of suspects. I now had three.
5. Burial records must be searched to determine whether any person found in a birth or christening entry had died as an infant or child. Search all local and neighboring parish burial records of all churches.
This is what someone once referred to as "killing off the impostors". I now had three James Atkins in the parish my James was supposed to be in. A search of the burial records of the parish church eliminated the first James I found. A search for burials in the nearest non-conformist church registers, as well as the registers of the contiguous parishes would be the next step here. In the present problem I had reduced the suspects to 2 possible entries. I had a favourite, but could not eliminate the other so far.
6. Searches should be made for alternative births or christenings in the records of all adjoining or surrounding parishes within a five mile radius for a period of five years before the calculated birth year to the year of marriage looking for others of the same name and birthplace who might be confused with the likely candidate.
I have found that the careful use of maps is most productive in solving these sorts of problems. I have a small collection of Ordnance Survey maps for the county of Kent where most of my research is done, but even the small scale maps in The Phillimore Atlas and Index to Parish Registers3 can be used. Simply copy the relevant county map, and using a compass (do you remember those from your ninth grade geometry class?) measure off 5 miles. Then find the parish where you expected to find your ancestor, and scribe a 5 mile radius circle using that parish as the centre. When you do this with the parish boundary maps you can quickly see the parishes you should be searching. Map study may reveal natural constrictions to, or channels of movement like wide rivers, or mountains which may modify this strategy.
Ancestral connections to military installations may require an extension to the five mile radius. I have found that Kitzmiller's book, In Search of the Forlorn Hope4, to be a useful place to start to identify military camps in England.
Parishes or towns associated with a particular occupation should be included in a search if the occupation is known. A mine worker will most likely move to a place where there are mines; a stone mason to where there are construction projects. One genealogist couldn't find his stone mason in Devon, but learned from his research of other masons in Devonshire working in the Channel Islands, and searching there found his ancestor at a port town where a harbour was being built.
7. In large, heavily populated industrialized parishes, check all the churches in the parish, not just the main parish church, and check the adjoining parishes.
In a search for members of the Hambrook family who moved from east Kent to Middlesex, I searched the records of the main parish church in Islington, but without success. Another genealogist found the family by searching through the all the daughter churches in the rapidly urbanized parish. I felt rather foolish. The challenge in larger urban areas where there are numerous churches is to figure out the most logical places to search first. Here again map work helps. Careful use of city directories can also be most helpful. I have found the best listing of British directories is Shaw and Tipper's British Directories5.
8. The economic conditions of the time and place should be considered. The collapse of older factories, mines, mills, local agriculture etc. may cause out migration. A booming economy in a neighbouring centre may do the same.
I am presently in a rather frustrating search for a Henry Augustus Thompson, a house painter who, though born in Bilton in Warwickshire, was attracted to neighbouring Rugby, then nearby villages, then to Birmingham in search of larger markets for his "have paint, will smear" business. His children were born along the way and may even have been born in an adjoining county. The need to expand the search horizon is there in most research puzzles, but the impact of mobility in the latter half of the 19th century is a real challenge.
9. Check the marriage registers of the candidate's stated place of birth to see if a person of that name might be "married off' to someone other than the known spouse. This check should extend from 15 years after the birth or christening to 40 years after.
In the problem of the two James Atkins mentioned earlier, I found that a review of the parish registers in the region revealed several marriages that might account for one of the two men. This search coupled with a search for the christenings of the wives in the parishes assisted in eliminating one of the James. As Sherlock Holmes put it, after you eliminate the impossible, what is left, no matter how improbable is the truth.
10. In the case of a male ancestor, careful watch should be kept for the possibility of a person with the same name and surname having children born or christened in the same or adjoining parishes. These conflicting possibilities need to be resolved.
Over the past 20 years of doing research I have run into this problem many times. The problem can be complicated when the spouses name is unknown or very common. In the very early 17th century I was searching for a Richard the son of Richard Hambrook and found three possible suspects in the 5 mile radius. Two were married to Elizabeths! All three families were gathered from the various records, and marriage records, christening, burials, tax rolls and probate records were examined. Eventually we sorted through and accounted for all, creating three family groups, establishing relationships between numerous people and discovering two of the elder Richards were cousins, and the other with an as yet unknown relationship. The puzzle took years, and many wills and probate records sorted out but we are more confident ours was the Richard the son of Richard and Margaret.
11. No matter what religious denomination of your ancestor, check the records of all churches in the area: E.G. Church of England, Methodists, Baptists, Quakers, Roman Catholic. Be aware that the areas served by nonconformist chapels and Quaker meeting houses were larger than the traditional Church of England parishes.
As you read this remember my proposed pedigree that was devastated by the discovery of the eleven year old Richard Hambrook who was buried in the Methodist Church yard, thus preventing him from becoming the father of a large family of grocers!
12. When Bishops Transcripts are searched, missing and illegible entries need to be searched in the Parish Registers.
In my most recent researches I have been even more conservative in my faith in parish registers and bishop's transcripts. I read both in their entirety. I have found that in some cases the Bishops transcript is the original, and the parish register, the copy, the reverse of what one would expect. There may be errors and omissions in each.
13. When a search for a birth or christening record fails, consider the possibility that the child was born out of wedlock and took his mother's surname, most likely totally unknown to you. Such a situation calls for listing all entries with the same given name, then matching these names against marriage and death or burial records to determine who might fit.
There was a certain James Hambrook who owned an inn during the mid 19th century in one of the parishes along the Thames estuary. After his own wife had passed on, and his children grown and moved, James seduced the barmaid (or at least I suspect he did). Miss Moore came to live with him and bore him children. The children are listed in the 1851 census as Hambrooks, along with James and his common-law wife. Based on the ages of the children in 1851 a search was made in the parish registers and the Bishop's transcripts without success. Next a search was made of the index to civil registrations for these children under the surname Hambrook. The search was again unsuccessful. At this point in the search someone started to use their head and went looking for the marriage of Miss Moore to James Hambrook and when one was not found, the search began again assuming the children were christened or registered under the surname Moore. In the civil registration of that particular corner of Kent we found the children, most with the second given name of Hambrook, and the surname of Moore. Some of the children married as Moores, some as Hambrooks and one as a Hambrook-Moore. All credit for the discovery and the sound reasoning behind it go to my research colleague. I was the learner.
14. Although naming customs among the English are generally undetectable, odd given names may be clues. One needs to become familiar with the range of names commonly found in a county or area before you can determine whether a name is uncommon enough to be a significant clue. Occasionally surnames appear as given names, and this may help identify parents or grandparents.
Among my Hambrook ancestors and the vast local cousinage of Hambrooks that populate eastern Kent county in England names like Richard, James, John, Elizabeth, and Mary abound. However, from time to time I have found the truly odd name, Oddin Hambrook of Dover, Kent was one such oddly named ancestor. A group of us were collaborating on gathering all 19th century Hambrooks and sorting them into families, a real challenge. Oddin, however, despite three or four marriages, numerous moves, and changes in religious faith, social status etc., was relatively easy to find - at least until he named a son Oddin!
15. When other conflicting birth or christening records are found in other records in the prescribed area, one needs to follow these individuals forward in time through marriage and burial records to eliminate possibilities. Probate records are also useful: check for wills of all fathers of these individuals. If the conflicting individual is found outside the parish of your first candidate, then the searches for marriage and burial information must be made within a five mile radius of the new parish.
Many readers will be horrified at the amount of work that is suggested by these procedures and especially this last one. There is indeed a lot of work involved. I am reminded of an account of the work of a well known professional genealogist, now 1ong deceased, who when he ran out of the easily located records, made up the "facts" necessary to complete his project. His clients would find everything in order until they reviewed the critical link, and their they would not find the record. Many of his fraudulent pedigrees were only discovered years later. Some are still surviving. If your ancestral research is worth doing, its worth doing right.
Concerning probate records, it has been my experience that if a family has resided in an area for some time, it is fruitful to gather not just the wills of all fathers of the suspects you have rounded up, but all those in that part of the county with the same surname, or if the name is an uncommon one, all wills and inventories, etc. in the county or adjoining counties.
There were two Richard Hambrooks with wives named Elizabeth who lived in adjoining parishes, each having a son called Richard. (Readers will have noted that Hambrooks had very imaginative naming patterns). And these sons were born within a year of each other. The problem was finding which was the Richard who married my known ancestor. We knew that our Richard the younger was residing in a particular parish several miles away. In the census was a clue to his birthplace. We extracted all Hambrook baptism, marriage and burial records from the contiguous parishes to those the senior Richards lived in, and established burial dates. All the wills for Richard Hambrooks probated around the times of the burials were examined. Only one of the two suspect senior Richards had a will. His Richard junior was mentioned as residing in another location than our Richard junior was known to reside in at the time. The evidence was not conclusive, but sufficient enough that we felt comfortable declaring for one and not the other christening and parentage.
I have found from my experience messing up on my own research that there is a real value in extracting all instances of the surname you are searching for in the records you search, particularly if the name is not common. For example, when researching members of the Atkins family in Newbold on Avon, Warwickshire, I would extract all the Atkins from the beginning of the register to the end, not just those in the decade of the event of the sought after ancestor. Similarly, when examining the census, I would extract all Atkins in surrounding parishes on the same reel of microfilm as the census for Newbold. Many times this has made it possible when I find the missing link in the immediate puzzle, to push back one or two or three generations without returning to the same microfilmed record later. For a detailed discussion to this approach researchers should seek out another older book, Family History for Fun and profit6.
1Gardner, David E. and Frank Smith. Genealogical Research in England and Wales. Salt Lake City: Bookcraft, 1958 - 1966 (Revised edition). A three-volume classic in English genealogy. Volume 3 is particularly useful for this subject. 2Kenneth G. Aitken. "Evaluating genealogical evidence", Saskatchewan Genealogical Society Bulletin, Vol. 30, No. I (March 1999) pp. 3-11
3Cecil Humphrey-Smith (ed.) The Phillimore Atlas and Index of Parish Registers. Chichester, West Sussex: Phillimore & Co., 1995
4John M. Kitzmiller. In Search of the Forlorn Hope: A Comprehensive Guide to Locating British Regiments and their Records 1640-W.W.I. Salt Lake City: Manuscript Publishing Foundation, 1988
5Gareth Shaw and Allison Tipper. British Directories: A Bibliography and Guide to Directories Published in England and Wales 1850-1950 , Scotland 1773- 1950. London: Mansell Publishing, 1997
6Vincent L. Jones, Arlene H. Eakle and Mildred H. Christensen. Family History for Fun and Profit. The Salt Lake City: Publishers Press for Genealogical Institute, 1972. This book outlines a jurisdictional approach to family history research, and was an important influence in my work..
Copyright September 1999 by Kenneth G. Aitken, ALL RIGHTS RESERVED