The Brythonic or Brittonic languages (Welsh: ieithoedd Brythonaidd/Prydeinig, Cornish: yethow brythonek/predennek, Breton: yezhoù predenek) form one of the two branches of the Insular Celtic language family; the other is Goidelic. The name Brythonic was derived by Welsh Celticist John Rhys from the Welsh word Brython, meaning an indigenous Briton as opposed to an Anglo-Saxon or Gael. The name Brittonic derives ultimately from the name Prettanike, recorded by Greek authors for the British Isles. Some authors reserve the term Brittonic for the modified later Brythonic languages after about AD 600.
The Brythonic languages derive from the Common Brittonic language, spoken throughout Britain south of the Firth of Forth during the Iron Age and Roman period. North of the Forth, the Pictish language is considered to be related; it is possible it was a Brythonic language, but it may have been a sister language. In the 4th and 5th centuries emigrating Britons also took Brythonic speech to the continent, most significantly in Brittany. During the next few centuries the language began to split into several dialects, eventually evolving into Welsh, Cornish, Breton, and Cumbric. Welsh and Breton continue to be spoken as native languages, while a revival in Cornish has led to an increase in speakers of that language. Cumbric is extinct, having been replaced by Goidelic and English speech. The Isle of Man may also have had a Brythonic language that was replaced with a Goidelic one. By emigration there are also communities of Brythonic language speakers in England, France, and Y Wladfa, the Welsh settlement in Patagonia.
The names “Brittonic” and “Brythonic” are scholarly conventions referring to the ancient Celtic speech of Britain and its descendents, in contrast to the “Goidelic” Celtic varieties that originated in Ireland. The terms also refer to the ancestral language, Common Brittonic or Brythonic. Both were created in the 19th century to avoid the ambiguity of earlier terms such as “British” and “Cymric”. “Brythonic” was coined in 1879 by the Celticist John Rhys from the Welsh word Brython. “Brittonic”, derived from “Briton” and also earlier spelled “Britonic” and “Britonnic”, emerged later in the 19th century. It became more prominent through the 20th century, and was used in Kenneth H. Jackson’s highly influential 1953 work on the topic, Language and History in Early Britain. Jackson noted that by that time “Brythonic” had become a dated term, and that “of late there has been an increasing tendency to use Brittonic instead.” Today, “Brittonic” often replaces “Brythonic” in the literature. Rudolf Thurneysen used “Britannic” in his influential A Grammar of Old Irish, though this never became popular among subsequent scholars.
Comparable historical terms include the Medieval Latin lingua Britannica and sermo Britannicus and the Welsh Brythoneg. Some writers use “British” for the language and its descendents, though due to the risk of confusion, others avoid it or use it only in a restricted sense. Jackson, and later John T. Koch, use “British” only for the early phase of the Common Brittonic language. Prior to Jackson’s work, “Brittonic” (and “Brythonic”) were often used for all the P-Celtic languages, including not just the varieties in Britain but those Continental Celtic languages that similarly experienced the evolution of the Proto-Celtic element /kw/ to /p/. However, subsequent writers have tended to follow Jackson’s scheme, rendering this use obsolete.
Knowledge of the Brythonic languages comes from a variety of sources. For the early languages information is obtained from coins, inscriptions and comments by classical writers as well as place names and personal names recorded by them. For later languages there is information from medieval writers and modern native speakers, together with place names. The names recorded in the Roman period are given in Rivet and Smith.
The Brythonic branch is also referred to as P-Celtic (like Gaulish) because the Brythonic reflex of the Proto-Indo-European phoneme *kw is p as opposed to the Goidelic c. Such nomenclature usually implies an acceptance of the P-Celtic hypothesis rather than the Insular Celtic hypothesis (for a discussion, see Celtic languages).
Other major characteristics include:
The family tree of the Brythonic languages is as follows:
|Western Brythonic||Southwestern Brythonic|
Brythonic languages in use today are Welsh, Cornish and Breton. Welsh and Breton have been spoken continuously since they formed. Cornish nearly died out during the 19th and 20th centuries, retained only by a few elderly people and some families as a language of the home, but a process of revitalisation since 1904 has seen numbers of natural speakers increase. Also notable are the extinct language Cumbric, and possibly the extinct Pictish although this may be best considered to be a sister of the Brythonic languages. The late Kenneth H. Jackson argued during the 1950s, from some of the few remaining examples of stone inscriptions, that the Picts may have also used a non-Indo-European language, but some modern scholars of Pictish do not agree.
The modern Brythonic languages are generally considered to all derive from a common ancestral language termed Brittonic, British, Common Brythonic, Old Brythonic or Proto-Brythonic, which is thought to have developed from Proto-Celtic or early Insular Celtic by the 6th century BC.
Brythonic languages were probably spoken prior to the Roman invasion at least in the majority of Great Britain south of the rivers Forth and Clyde, though the Isle of Man later had a Goidelic language, Manx. Northern Scotland mainly spoke Pritennic, which became Pictish, that may have been a Brythonic language. The theory has been advanced (notably by T. F. O’Rahilly) that part of Ireland spoke a Brythonic language, usually termed Ivernic, before it was displaced by the Q-Celtic Irish language, although the authors Dillon and Chadwick reject this theory as being implausible.
During the period of the Roman occupation of Southern Britain (AD 43 to c. 410), Common Brythonic borrowed a large stock of Latin words, both for concepts unfamiliar in the pre-urban society of Celtic Britain, such as urbanisation and tactics of warfare, and for rather more mundane words which displaced native terms (most notably, the word for “fish” in all the Brythonic languages derives from the Latin piscis rather than the native *ēskos, which may survive, however, in the Welsh name of the River Usk, Wysg). Approximately 800 of these Latin loan-words have survived in the three modern Brythonic languages. Romano-British is the name for the Latinised form of the language used by Roman authors.
It is probable that at the start of the Post-Roman period Common Brythonic was differentiated into at least two major dialect groups – Southwestern and Western (in addition we may posit additional dialects, such as Eastern Brythonic, spoken in what is now Eastern England, which have left little or no evidence). Between the end of the Roman occupation and the mid 6th century the two dialects began to diverge into recognisably separate languages, the Western into Cumbric and Welsh, and the Southwestern into Cornish and its closely related sister language Breton, which was carried from the south west of Great Britain to continental Armorica. Jackson showed that a few of the dialect distinctions between West and Southwest Brythonic go back a long way. New divergencies began around AD 500 but other changes which were shared occurred in the 6th century. Other common changes occurred in the 7th century onward and are possibly due to inherent tendencies. Thus the concept of a common Brythonic language ends by AD 600. It is thought that substantial numbers of Britons remained in the expanding area controlled by Anglo-Saxons, but the only information on their language may be obtained from place names. Over time it is thought they gradually adopted the English language.
The Brythonic languages spoken in what is now Scotland, the Isle of Man and what is now England began to be displaced in the 5th century through the influence ofIrish (Scots), Norse and Germanic invaders. The displacement of the languages of Brythonic descent was probably complete in all of this territory, (except Cornwalland the English counties bordering Wales), by the 11th century (date of extinction in various parts of the territory is debated).
The regular consonantal sound changes from Proto-Celtic to the Welsh language and Cornish language may be summarised in the following table. Where the Welsh and Cornish graphemes have a different value from the corresponding IPA symbols, the IPA equivalent is indicated between slashes. V represents a vowel; C represents a consonant.
|Proto-Celtic consonant||Late Brythonic consonant||Welsh consonant||Cornish consonant|
|*-VdV-||*-d-? -ð-||dd /ð/||d|
|*-j||*-ð||-dd /ð/||-dh /ð/|
|*-kk-||*-cc-||ch /x/||gh /h/|
|*-m-||*v? m?||f /v/||v|
|*-nd-||*n / nn||n, nn||n, nn|
|*-nt-||*nt / nh||nt, nh||n, nn|
|*s-||*h-, s||h, s||h|
|*-t-||*-d-? -t-?||d||dh /ð/|
|*-tt-, *-ct-||*th? *tt?||th /θ/||th /θ/|
|*sw-||*hw-||chw /xw/||hw /ʍ/|
|*VwV||*w||dd /ð/||dh /ð/|
|final vowel||Vh||Vch V/x/||Vgh V/h/|
Place names and river names
The principal legacy left behind in those territories from which the Brythonic languages were displaced is that of toponyms (place names) and hydronyms (river names). There are many Brythonic place names in lowland Scotland and in the parts of England where it is agreed that substantial Brythonic speakers remained (Brythonic names, apart from those of the former Romano-British towns, are scarce over most of England). Names derived (sometimes indirectly) from Brythonic include London, Penicuik, Perth, Aberdeen, York, Dorchester, Dover and Colchester. Brythonic elements found in England includebre- and bal- for hills, and carr for a high rocky place, while some such as combe or coomb(e) for a small deep valley and tor for a hill are examples of Brythonic words that were borrowed into English. Others reflect the presence of Brythons, such as Dumbarton – from the Scottish Gaelic Dùn Breatainn meaning “Fort of the Britons”, or Walton (several) meaning a ‘tun’ or settlement where ‘walha’ (Welsh/Brythons) still lived.
The number of Celtic river names in England generally increases from east to west, a map showing these being given by Jackson. These names include ones such as Avon, Chew, Frome, Axe, Brue and Exe, but also river names containing the elements “der-/dar-/dur-” and “-went” e.g. “Derwent, Darwen, Deer, Adur, Dour, Darent, Went”. In fact these names exhibit multiple different Celtic roots. One is *dubri- “water” [Bret. “dour”, C. “dowr”, W. “dŵr”], also found in the place-name “Dover” (attested in the Roman period as “Dubrīs”); this is the original source of rivers named “Dour”. Another is *deru̯o- “oak” or “true” [Bret. “derv”, C. “derow”, W. “derw”), coupled with two agent suffixes, *-ent- and *-iū; this is the origin of “Derwent”, ” Darent” and “Darwen” (attested in the Roman period as “Deru̯entiō”). The final root to be examined is “went”. In Roman Britain there were three tribal capitals named “U̯entā” (modern Winchester, Caerwent and Caistor St Edmunds), however the meaning is unknown. It may mean “the favoured/chosen” or “(place) of arrivals/comings”.
Some have argued that Celtic has acted as a substrate to English for both the lexicon and syntax. It is generally accepted that linguistic effects on English were lexically rather poor aside from toponyms, consisting of a few domestic words, which may include hubbub, dad, peat, bucket, crock, noggin, gob (cf. Gaelic gob), nook; and the dialectal term for a badger, i.e. brock (cf. Welsh broch, and Gaelic broc). Another legacy may be the sheep-counting system Yan Tan Tethera in the west, in the traditionally Celtic areas of England such as Cumbria. Several Cornish mining words are still in use in English language mining terminology, such as costean, gunnies, and vug.
Those who argue against the theory of a Brythonic substratum and heavy influence point out that many toponyms have no semantic continuation from the Brythonic language. A notable example is “Avon” which comes from the Celtic term for river abona or the Welsh term for river “afon” but was used by the English as a personal name. Likewise the River Ouse, Yorkshire contains the word usa which merely means water and the name of the river Trent simply comes from the Welsh word for a trepasser (an over-flowing river) It has been argued that the use of periphrastic constructions (using auxiliary verbs like doand be) in the English verb (which is more widespread than in the other Germanic languages) is traceable to Brythonic influence although some find this very unlikely and prefer a hypothesis of North Germanic influence rather than Celtic. For example in literary Welsh we can have Caraf = I love and Yr wyf yn caru = I am loving where the Brythonic syntax is exactly mirrored in English, whereas in the Germanic sister languages of English we only have one form, for example Ich liebe in German. (Note that in colloquial German, a progressive tense has elvolved which is formally very similar to that of English and those found in Celtic languages, e.g. Ich bin am Arbeiten “I am working”, literally: “I am at the working”. A similar structure is also found in modern Dutch. These parallel developments suggest that the English progressive is not necessarily due to Celtic influence.)
Some researchers (Filppula et al., 2001) argue that English syntax reflects more extensive Brythonic influences. For instance, in English tag questions, the form of the tag depends on the verb form in the main statement (aren’t I?, isn’t he?, won’t we? etc.). The German nicht wahr? and the French n’est-ce pas?, by contrast, are fixed forms which can be used with almost any main statement. It has been claimed that the English system has been borrowed from Brythonic, since Welsh tag questions vary in almost exactly the same way.
Brythonic effect on the Goidelic languages
Far more notable, but less well known, are the many Brythonic influences on Scottish Gaelic. Like English, periphrastic constructions have come to the fore, but to a much greater degree. Scottish Gaelic contains a number of apparently P-Celtic loanwords, but as there is a far greater overlap in terms of Celtic vocabulary, than with English, it is not always possible to disentangle P and Q Celtic words. However some common words such as monadh = Welsh mynydd Cumbric *monidh are particularly evident. Often the Brythonic influence on Scots Gaelic is indicated by considering the Irish Gaelic usage which is not likely to have been influenced so much by Brythonic. In particular, the word srath (Anglicised as “Strath”) is a native Goidelic word, but its usage appears to have been modified by the Brythonic cognate ystrad whose meaning is slightly different. The effect on Irish has been the loan from British of many Latin-derived words. This has been associated with the Christianisation of Ireland from Britain.