The First Thousand Years of British Names

Part 1

by Tangwystyl verch Morgant Glasvryn
(Heather Rose Jones,
copyright 1998, all rights reserved

Some History

We must never forget that names occur in a historical context, and are both shaped and supplied with their raw materials by historical forces. At the time that the Romans first set foot in Britain, in 55 BCE, the inhabitants were largely Brythonic- speaking Celts. (This article won't concern itself with the Picts in the north of Britain.) The Celts had come to the British Isles in several waves. The earliest had come perhaps six centuries before. In a separate wave, Goidelic-speaking Celts had bypassed Britain to settle in Ireland, and we find Goidelic influences on British names primarily along the western seacoast, due to the influence of visitors, peaceful and not. The last wave of Celts came to the southeast of Britain during the century and a half immediately preceding Caesar's little foray and were Belgae from the northern part of Gaul -- presumably Gaulish-speaking, but Gaulish was a close relative of the language spoken in Britain at that time. The same names and name elements can be found in both continental and insular inscriptions and so for our purposes we can consider the resulting mix "British" [1]. One consequence of this is that when searching for appropriate early British names, it is valid to consider Gaulish sources as a "second-best" option. Although I won't be covering its contents here, D. Ellis Evans' Gaulish Personal Names is a title to be aware of.

Caesar's rather abortive "invasion" of 55 BCE likely had little affect on British names. There was no Roman presence established at that time and trade with the continent was already well- established at that point. The Claudian conquest, begun in 43 CE, brought with it a permanent presence of soldiers, beaurocrats, and merchants, and marks the beginning of a serious Roman influence on the names of the island. And for a time, recorded names become more and more Romanized until it is difficult to know from the name alone in what part of the empire a particular inhabitant of Britain might have been born.

In the early 5th century, the last connections between Britain and what was left of Rome dissolved, but the Roman legacy in names continued in the form of a handful of borrowed given names, but mostly in the use of the Latin language as the primary written medium, via the Christian church. The church itself had brought a dowry of names: Biblical names, the names of saints and apostles. The next major influx came with the Germanic invasions in the south and east during the 5th and 6th centuries. Not all the interactions with the British were hostile, and both Anglo- Saxons and their names made their way into Celtic Britain eventually. The next major historical impact on British (and by now we can call them Welsh) names comes after our period of interest, with the Norman invasion at the end of the 11th century.

The Languages

The British language was not recorded until the Romans brought writing to the island and the Romans appear to have had little interest in native languages. To be literate, during this era, was to be literate in Latin, and apart from proper names little, if anything, was recorded of the British language. Fortunately, our subject here is proper names. But we must always remember that the earliest British names were filtered through the Latin language when they were recorded. Thus we find that the British nominative (i.e., the "basic" form of the word) ending -os is most often found as the very similar Latin ending -us, and names appearing in the genitive (i.e., "possesive") are usually given Latin declensions. We might expect that such things as word order have also been Latinized.

The British language of the early first millenium was, like Latin and unlike modern Welsh, a highly inflected language with case suffixes on the nouns. (Verbs are a bit beyond our interest here.) Words (and elements in compounds) were stressed on the next to last syllable, just as they are in modern Welsh. Three major events mark the transition from British to Old Welsh, around the 6th century [2]. One is the loss of those case suffixes. The second is the loss of some unstressed internal syllables (the case suffixes were lost largely because they were unstressed). These changes can be seen in a name like the British Cunobelinos. The name is a compound of two elements, each with its own stress, something like CUNo-beLINos. When the unstressed final syllables of each element were lost, the name ended up something like Cunbelin. At that point, the name would continue to be stressed as if the lost syllables were there: CUN-belIN. Eventually -- by the 11th century, possibly a little earlier [3] -- some natural tendency to accent the penultimate syllable reasserted itself and the accent shifted: con-BELin.

Another change that had been happening prior to the transition from British to Old Welsh -- probably around the mid-5th century [4] -- is the lenition or softening of consonants intervocalicly (i.e., between vowels). Thus our British name Cunobelinos was, by the end of the 5th century, being pronounced something more like Cunovelinos, because the "b" comes between two vowels. This softening did not begin to be reflected in the spelling for the most part until the transition from Old Welsh to Medieval Welsh. This is an important fact to remember for those choosing names during the Old Welsh period who wish to pronounce them "correctly". This softening happened to the same set of sounds as the "soft mutation" in modern Welsh, which makes it a little easier to keep track of. The major sounds to be concerned with are:

Spelled Pronounced
p b
t d
k /c g
b v (approximately)
d soft th (mod. Welsh "dd")
g y (approximately -- in modern Welsh this disappears entirely when lenited, but in Old Welsh there were still traces)
m v

Other changes that occur include shifts in particular vowels and the development of voiceless fricatives from double or geminate consonants, but I'm trying to keep this relatively simple.

The evidence suggests that by the heyday of Roman Britain the dominant language, at least in urban areas, was Latin. So when we find names being recorded with Latin filius and the like, we need not interpret this as their being "translated" for the purpose of writing them down. They could simply be being written in the ordinary language of their bearer. But at the same time, it is likely that outlying areas -- such as the part that later became Wales -- retained the native language as primary. The best evidence for this is that Welsh, unlike Gaulish or Iberian, did not transmogrify into a Romance language. So even throughout the Roman occupation, we could expect to hear British forms being used side by side with Latin ones.

There is some unexpected light thrown on the Latin versus vernacular question by a group of bilingual Latin/Irish inscriptions appearing in Britain in the 5th and 6th centuries [5]. The Latin inscriptions appear in Roman letters, using Latin relational terms and Latin inflections. The Irish inscriptions are in ogham characters and appear not only with Irish relational terms and inflections, but with spellings that reflected the difference between Irish and Latin pronunciations. Since the names in question are clearly Irish in many cases, we have good evidence that names in Latin inscriptions might be altered from their spoken form when recorded and that the use of words such as filius in inscriptions did not necessarily mean that was what was used every day.

After the political importance of the Latin language followed the legions back over the channel, it remained the major linguistic context in which names were written. This was due to the place of Latin in the Christian church, and to the place of the church in the creation of written documents. Here again we can postulate that, in everyday use by secular people, the context language was the vernacular. But due to the Latinity of written records up through this time, those who want to use a vernacular-context name in the SCA must usually reconstruct parts of it on theory alone.

Roman Names

In order to understand the ins and outs of Romano-British names, you first need a grounding in the traditional Roman naming system. At the time of the Empire, the "standard" name of a Roman citizen was the so-called "triple-name". There were elaborations, but we'll stick to the basic form. The three parts were the praenomen, nomen, and cognomen. The praenomen is sometimes thought of as equivalent to our idea of a "given name", but this is somewhat misleading. The total number of different praenomina in use was never large, and many families restricted themselves to particular subsets. The set of praenomina was also quite closed by the time we are considering; no new ones were being invented. This meant that the praenomen was fairly useless for the purpose of distinguising one individual from the next.

The nomen was the name of the gens or tribe to which you belonged. It was inherited, just as a surname is now. Nomina of Latin origin end in -ius, although other endings can be found on nomina "invented" for new families in the provinces.

The third part of the name was the cognomen. These originated as personal nicknames -- adjectives of physical appearance or temperment, nouns in the ablative such as naso "with the nose", or nouns indicating origin or nationality. But as the tribes denoted by the nomen grew to unmanageable size, cognomina came to be adopted as part of the hereditary name to mark particular subunits of the tribe. In parallel with that process, a man might still be given a personal nickname, which would also be called a cognomen. Additionally, a recorded name sometimes included a patronym using the father's praenomen, such as Marci filius "Marcus' son".

Two modifications of this system are worth noting. It was a fairly ordinary event for a man -- even as an adult -- to be adopted into a different family. It was customary upon that event for him to take up his adoptive father's three names but to add afterward an adjectival form of his original nomen, changing the -ius ending for -ianus. Of more particular importance to names in Britain was the naming convention for naturalized citizens. When a non-Roman was granted citizenship, he would normally pick a praenomen at will, use the nomen of the person to whom he owed his citizenship (frequently a local governor or general), and keep his original native name as a cognomen. We see a classic example of this in the British king Cogidubnos who was won over to the side of Rome by a package deal from the Emperor Claudius that included citizenship and took as his Roman name "Titus Claudius Cogidubnus". Simple logistics, then, insure that particular nomina would be common among the Britons while others might never be introduced there at all.

During the early stages of the Roman conquest, the triple name formula would be rather rare among native Britons. Citizenship might be granted to kings such as Cogidubnus for political reasons, but ordinary people would not have access to citizenship unless they earned it in some fashion, for example by serving in the army. The usual term of army service was 25 years, so the earliest this could have an effect would be around the end of the 1st century, and even then it was more usual for retired soldiers to settle where they had been serving -- and where they had likely acquired a wife and children -- than to return home. The policy on the frontiers was not to allow conscripts to serve close to home (it no doubt cut down on desertion and made local rebellions a little less nervous-making) so while there are rare inscriptions mentioning British soldiers in Britain in the mid-2nd century for instance [6], retired soldiers would not have been making a serious impact on Romano-British names until the mid-3rd century when forces were being recruited locally [7]. A more likely way for a Briton to acquire citizenship might have been to be an important man in a Romanized town. When a town was chartered as a municipium it was normal for the local magistrates to be granted citizenship [8] and so during the 2nd century, such places as Londinium, Verulamium, Eboracum, and others would have home-grown citizens.

This is not to say that there was any shortage of Roman citizens in Britain during the early years of the occupation, but these would mostly have been immigrants and this article concentrates on the names of Britons. Nor was citizenship the only way for a Briton to acquire a Roman name. A man who probably lived in the early 2nd century was recorded as Nectovelius f[ilius] Vindicis. He was a Brigantian, bearing a British name, but whose father had been given a Roman name (Vindex) [9].

We have been speaking above of men's names. Women got the short end of the stick in names as well as many other things in Rome. Traditionally, the pattern was for a woman to bear a feminine inflection of her father's nomen, or rarely praenomen. Given several sisters, this could easily lead to confusion and a second name was often used that designated birth order: Maxima and Minor or Secunda and Tertia. But by the time of the Empire, women were bearing more individual names. In at least one British example we see a woman bearing a feminine version of a nomen followed by a native British name Lollia Bodicca. The best interpretation is that here we have a family name (a nomen indicating that her family received their citizenship from Lollius Urbicus in the mid 2nd century) and a personal name [10].

Go to Part 2

Editted and published by Arval Benicoeur