The following explanation from the Guild of One Name Studies was adopted by me in principle, but early on in my CUMPSTON research. In addition to the main CUMPSTON research I also registered the variants COMPSON, COMPSTON, CUMMERSTON, CUMPSON, and CUMSON. It was some time later that I realised that CUMMERSTON had died out completely, and I should not register it today as a variant. Since registration however I have identified a huge range of other variants and deviants and encourage you to consider them in your own CUMPSTON research.
"The starting point of any discussion of variants and deviants must be that in former times, not only did names get recorded with a wide range of spellings, but individuals themselves may have used many versions too - it is said that Shakespeare spelt his own name in six or more different ways over his lifetime, from evidence of known signatures. Additionally, many people were illiterate and could not sign their own name, so it is not possible to say that any given spelling was 'used' by the individual. Officialdom will have recorded their name and this may have become the adopted and accepted spelling of the name; indeed, this is the likely method by which variants arose. It can often be seen that the recorded spelling in parish registers changes with the change of incumbent. Should all such spellings be considered as variants? The Guild's advice is that it depends on the consistency with which the name is recorded in official documents. If the vicar consistently used a given spelling over many years, then it may be considered as a variant, bearing in mind that such records might have been called on as evidence in things like settlement disputes or probate. On the other hand, vicars, like anyone else, might use a large range of spelling variants with no particular pattern and these should in general be considered deviants.
The Guild therefore defines a variant as a name spelling which varies from the primary name spelling (or another variant spelling) used by that person's ancestors and which is:
A name spelling that the person was known to have used, through signature evidence on wills, marriage bonds etc or other documents originating from the individual concerned, or
A name spelling used by officials on a consistent and persistent basis over a period of years.
A deviant is any other spelling recorded, including cases where the spelling occurs in official records, but only randomly and inconsistently. Deviants will also include spellings derived from enumeration, transcription and indexing errors, both contemporary and modern.
Having decided what spellings are variants, it may still not be appropriate to register all of them on a one-name study. Such cases include where a variant overlaps with another existing study or where the variant is more commonly found as a surname in its own right and where the variant is a minority source of that name. The latter exclusion avoids the problems that would occur if someone wanted to register the surname concerned as a primary study name. In these overlapping cases, it is hoped that members will collaborate in exchanging information on boundary cases".
Explanation for other researchers.
The names on pages 1 and 2 have all been found to link to both my personal research and my One Name Study Research. In some instances, for example the marriage bond of my 5 x grandparents, there are two spellings of the name in the same document.
What do I use the Variant list for?
I always print out the file to use when I am searching any data base – either on-line or in a record office. In this way I can be assured that I have searched for each variant. So far this has proved to be highly successful at finding those who have remained stubbornly hidden from sight! I regret that my memory can no longer cope with 44 variants so this list has been my salvation and sits on my desktop – both digitally and actually!
Office of National Statistics Names
Search the List http://www.taliesin-arlein.net/names/search.php
THIS DATABASE is an extract of an Office of National Statistics database, and contains a list of surnames in use in England, Wales and the Isle of Mann in September 2002. The list contains almost 270,000 surnames, shared by 54.4 million people. The entire database contains over a million surnames, shared by 55.9 million people, but names shared by fewer that five people have been excluded from this list. The database was established in 1998, and births are continually added, but the 1.5 million deaths between 1998 and 2002 have not been weeded from the system. The database is also said to include a level of duplicate entries and mis-spelt surnames, as well as people currently living abroad and temporary visitors to England and Wales. However, experience suggests that multiplying the result for your surname by 0.93 will give a good idea of the living population for your surname, and multiplying by 3.5 will give the population since the start of parish registers in the 16th century.
Data source: Surnames, England & Wales, National Statistics 2002. Crown copyright material is reproduced with the permission of the Controller of HMSO under Licence C02W0001623
CREDITS: The original database was developed as a spin-off from the Llangynfelyn Digital History Project. The database and web enquiry page were built by Nigel Callaghan of Technoleg Taliesin Cyf.
Approaching my research using all these tools I have been able to gather evidence which suggests that:
1. The Westmorland lines (no matter how the name is spelt) have been positively linked throughout the north including Yorkshire, Lancashire, Durham, Shropshire, The Midlands and Lincolnshire. I now hope to collect DNA evidence to support this.
2. The CUMSON in Worcester gives me a headache as I can’t link him to anyone. Is he yours?
3. As part of my One Name Study I have an office filled with the names of people still to be added – when I started I thought I was doing a small study of 5 names. Now I know that they all had catarrh on census night and none of them could spell their names correctly! Do help me try and sort them out.