Handle with care

There’s been a lot of publicity this week around the decision by Oxford Reference to make their Dictionary of Family Names in Britain and Ireland freely available online for a limited period.

Now, I’m not going to deny for a minute the magnitude of the work, nor in any way denigrate the efforts of the team behind the production of this vast digital tome, but I do have some problems with this approach to the study of surnames. It all seems like a retrograde step and smacks of the sort of methodology which (I thought) George Redmonds (among others) had kicked into touch many years ago.

The Dictionary is a massive undertaking and gathers together information about more than 45,000 names including ‘every surname that currently has more than 100 bearers, and those that had more than 20 bearers in the 1881 census’. The introduction to the work goes on to say:

Each entry contains lists of variant spellings of the name, an explanation of its origins (including the etymology), lists of early bearers showing evidence for formation and continuity from the date of formation down to the 19th century, geographical distribution, and, where relevant, genealogical and bibliographical notes, making this a fully comprehensive work on family names.

This is a big claim (or, rather, a number of big claims) and there’s one aspect in particular that troubles me – the bit about the Dictionary containing explanations of each surname’s origins.

The idea that you can explain the origin of a surname by accessing ‘lists of early bearers’ of the name is exactly the sort of muddled thinking that Redmonds dealt with so effectively more than 20 years ago:[1]

Each surname is in one sense unique, beginning with one person or family at a particular time and in a particular place. It is there and then that the meaning and origin must be investigated, no matter how common-place the name might seem. This present chapter is restricted, therefore, to questions of origins and meaning, looking in depth at certain surnames which are generally considered to have straightforward etymologies. It will be shown that some of the traditional explanations are wrong and also that etymology and meaning are not always the same thing. The truth is that the etymology of any surname is unsatisfactory unless it takes into account the particular local circumstances in which that name evolved. Those circumstances can be very distinctive, requiring the researcher to combine the skills of the historian, the linguist and the genealogist.

The Dictionary’s claim to be a ‘fully comprehensive work on family names’ is thrown into question when we consider the complete absence in their introduction of the concept of surnames ‘evolving’, but instead a focus on ‘formation and continuity’.

Any experienced family historian will appreciate that surnames are particularly unstable, not just in their spelling but also in their adoption by our ancestors. Aliases (explored in depth by Redmonds) were far more common than we might imagine and it’s not at all unusual for one branch of a family to go on to use one of the ‘available’ surnames, while another adopts the alternative. How then can we equate surnames so solidly with a particular family? And then there’s the whole question of illegitimacy and multiple marriages leading to our ancestors ending up with completely different surnames to the ones they began life with.

If we look at the entry for my own surname we see some good examples of why this approach simply doesn’t work:


Entry for Annal in the Oxford Dictionary of Family Names in Britain and Ireland.
Accessed 16 May 2020.

For a start, each of the references to instances of the name is taken from a transcribed source. It’s entirely understandable why they took this approach; data mining from a freely available source like the IGI (FamilySearch) is, realistically, the only way of creating a vast database such as this but, as any decent researcher will tell you, the IGI is, as its name might suggest, an index, not, in itself, a primary source. And if we take one of the instances referenced above – Thomas Annall , 1616 in IGI (Birmingham, Warwicks) – we can quickly see that the surname in the original is actually ARNALL – which instantly appears more likely to be a variant of the name ARNOLD. Indeed a very quick check reveals that the baptism of this Thomas ‘ARNALL’ the son of Thomas, fits neatly into a run of baptisms of children of Thomas ARNOLD at the same parish church (St Martin, Birmingham). There is no connection whatsoever to the name ANNAL (or variants) here and its inclusion is therefore (unintentionally) confusing and misleading.


Baptism of Thomas Arnall, St Martin, Birmingham. Accessed on Ancestry.co.uk. Original held by Library of Birmingham. Reference Number: DRO 34; Archive Roll: M95

This highlights one of many problems with the methodology at play here. As a member of the Guild of One Name Studies, I’ve been carrying out research into the name ANNAL and its variants for the best part of 40 years and I know that my surname evolved separately in two distinct regions of Scotland; namely, Fife and Orkney. Indeed, I can trace every ANNAL or ANNALL living in the UK today back to one of these two areas. And in both cases, I have been able to prove beyond reasonable doubt that the name evolved from the surname ANNAND or ANNAN – a name not even offered as a speculative variant in the Dictionary.

Yet Orkney doesn’t even get a mention in the Dictionary. Obliquely, it does as I am almost certain (still working on it!) that the John Annall, 1774 in IGI (Deal, Kent) was the son of a man who came from South Ronaldsay in Orkney. And I know that the William Annall, 1763 in IGI (Well, NR Yorks) was the grandson of a man who moved from Easter Anstruther in the East Neuk of Fife to settle in the parish of Well, North Yorkshire in the late 1730s.

The ‘Main GB location 1881’ is given as: WR Yorks; Angus; Caithness

This is both misleading and inaccurate. Of the 200 or so ANNALs listed in the 1881 census, 100 of them were living on the Orkney island of South Ronaldsay! The Yorkshire ones are, as I’ve said, descendants of the Fife ANNALs, as are the Angus (Dundee) ones, while the Caithness ANNALs (all 7 of them) had recently hopped across the Pentland Firth from their native Orkney.

The Hampshire/Sussex families are, I am certain, entirely unconnected and all living descendants of ANNALs from that area today have the surname ANNALS or ANNALLS. Again, this is an area that I need to explore further but I suspect that the second proposed etymology, namely that the name is a ‘variant of the obsolete English surname Ennal’ is quite probably correct.

Another example of the shortcomings of this approach to surname studies comes with a surname that I’m currently researching. The name MEEKS is quite common in the area around Biggleswade in Bedfordshire, extending into the neighbouring counties of Cambridgeshire and Hertfordshire. The name has a local and well-documented ‘variant’, MINKS, yet the Dictionary entries for the names MEEKS and MINKS give no indication that the names (at least in the Biggleswade area) were fully interchangeable.

George Redmonds died soon after The Dictionary of Family Names in Great Britain was published (no, I’m NOT linking the two events!) and I don’t know if he had a chance to read or comment on it. But I do know that he was less-than-impressed with the Dictionary’s most famous forerunner, namely Reaney’s Dictionary of British Surnames. Redmonds had this to say of Reaney’s work:[2]

His attitude to genealogy seems to have been somewhat ambivalent; for example in the Introduction to his dictionary he said specifically that the purpose of such a reference work was “to explain the meaning of the names, not to treat of genealogy and family history”…

…[he] did not stop to consider that each hereditary surname is unique, and that its etymology and meaning should never be taken for granted. The truth is that without some sort of genealogical evidence it can be unwise to link modern surnames with those found in medieval sources.

Reaney’s approach reminds me of the methods used by 18th and 19th century antiquarians in their ‘archaeological’ studies. You dig into the ground, pick out what you like the look of and utterly ignore the context (not to mention, that you’ve just destroyed the site for future archaeologists).

Of course surname research isn’t a destructive process but it can be a misleading one. My concern is that the unwary casual armchair researcher will be sucked in by the ‘quick fixes’ inherent in the Dictionary. It seems to me to have placed itself not a million miles away from those shopping centre stalls where you can buy a cheap, gaudy plaque with the history of your surname (not forgetting your coat of arms).

So, by all means have a look at the Dictionary and see what it says about your surname. But don’t take it as gospel and never treat it as a substitute for thorough, meticulous genealogical research.

The Dictionary of Family Names in Britain and Ireland can be accessed free of charge until 21 May 2020.

© David Annal, Lifelines Research, 16 May 2020

[1] Surnames and Genealogy: A New Approach, George Redmonds (Federation of Family History Societies, 2002) p.31
[2] ibid., Introduction p.3

This entry was posted in Surnames and tagged , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s