Perhaps this is too vague for you — you would like a date? As a respectable academic, I ought to say this is as precise as we can be. But if you press me — all right, then: in a symbolic sense, at any rate, perhaps we might say that English began in A.D. 449, at a place called Ebbsfleet on Pegwell Bay, near Ramsgate in Kent. I cannot give you a specific birthday, but it will have been in the summer rather than winter half of the year.
Even after a language has changed virtually out of recognition, people will not usually give different names to the old and new languages unless there is some strong reason to do so. The most obvious reason would be that the language has split into two or more different later languages in different places. If Italian were the only modern language to have evolved out of the Latin spoken two thousand years ago, we very likely would not call it “Italian”. We would call it “Modern Latin” or just “Latin”, and on the less frequent occasions when we needed to specify Julius Caesar’s language we would distinguish that as “Classical Latin” or “Ancient Latin”. In reality, though, by the time of the collapse of the Roman Empire, Latin was spoken across a wide area of Western Europe, and different changes to it accumulated in different territories. Present-day Italians, Frenchmen, Spaniards, Rumanians, and others all speak languages which have evolved gradually out of Latin, but they do not speak the same modern language; so they need distinctive names for their languages: Italian, French, and so forth.
If we went back one thousand years and heard the language spoken in England then, we would not easily recognize it as connected to our own, modern, English. We could not understand it without a course of study, as one studies modern foreign languages. Nevertheless, the standard name for that language is Old English — it is treated as essentially the same language as modern English, because there has been no significant split into separate descendant languages. There are regional and, in recent centuries, national dialects of English, of course. American English is not quite the same as the English of modern England; but the differences are too marginal to tempt people to think of them as “different languages”. (Perhaps the closest there has been to development of a separate “sister” language was Scots, which while Scotland remained politically independent of England was the official language of government and public life in that country, not just pronounced distinctively but spelled differently from English, with many of its own words, etc. But after Scotland was united with England in 1707, Scots faded away except as a local spoken dialect among many others in the English-speaking world.)
However, if we were to go back another thousand years, from A.D. 1000 back to the time of Christ, we would find that Old English had emerged from splits which turned one language at the earlier time into a number of later languages requiring distinctive names. Old English shared a common ancestry with other languages that linguists describe collectively as “West Germanic”, including Dutch and German. A little further back, West Germanic speech had split off from North Germanic, the ancestor of modern Danish, Swedish, Norwegian, and Icelandic, and from a now-extinct language called Gothic which formed a one-member East Germanic subfamily.
Shortly before the time of Christ, the ancestor of all these languages was a single language that modern scholars call “Proto-Germanic”. (We do not know what its speakers themselves called it — they may not necessarily have thought of their language as a namable thing, or each of the tribes that spoke it may have called it after themselves.) Proto-Germanic split into Proto-West-Germanic, Proto-North-Germanic, and Gothic, early in the Christian era, and a little later Proto-West-Germanic split into the West Germanic languages we recognize today, including English. (If you are interested in taking this story back earlier than Proto-Germanic, another of my pages discusses the remote ancestry of our languages. On Germanic language and life about the time of the language splitting, an excellent recent book is D.H. Green’s Language and History in the Early Germanic World, Cambridge, 1998.)
The split which led to English developing separately from Dutch, German, and other West Germanic languages was a very concrete historical event, though one which stretched over decades rather than happening on one day. This was the emigration of numbers of West Germanic speakers from Continental Northern Europe across the North Sea into the island of Great Britain. Once that happened, the Germanic speakers in Great Britain were physically separated from those on the Continent. We know that social links continued, but in the conditions of the time these could hardly have been intensive enough to keep speech habits on either side of the North Sea aligned with one another. So West Germanic in Great Britain inevitably began to diverge from its Continental sister languages.
(Incidentally, here and below I am carefully using the name “Great Britain” for the island on which I live, because this is the geographical name. The shorter name “Britain” is a familiar, brief alternative name for what is officially called the “United Kingdom of Great Britain and Northern Ireland”, which is a political entity created many centuries after the time we are discussing. Even if I slip into mentioning “Britain”, I shall be referring here strictly to Great Britain, not to any part of Ireland.)
What was the language situation in Great Britain when the Germanic immigrants arrived? If we go back to the beginning of the Christian Era (that is, the years around A.D. 1), shortly before the Romans conquered Britain, then the inhabitants of Great Britain spoke Brittonic or Brythonic, a Celtic language, ancestor of modern Welsh. (In northern Scotland a language with another name, “Pictish”, was spoken; little is known about Pictish — it went extinct long ago — but the best guess is that it was simply the northernmost dialect of Brittonic.) The inhabitants of the neighbouring island of Ireland spoke a related but rather different Celtic language, Goidelic or Gaelic.
(A good recent book on the linguistic history of the British Isles is Glanville Price, ed., Languages in Britain and Ireland, Blackwell, Oxford, 2000.)
The Celtic languages, although ultimately they are distantly related to the Germanic languages and hence to modern English — Celtic and Germanic are two branches of the “Indo-European” language family — are very different indeed from English. They are at least as “alien” as Russian, or Greek, say. If you have never seen anything of a Celtic language, here as a brief example are the opening lines of the Welsh national anthem. To get the full flavour, you should probably be standing in a rugby stadium when Wales is playing England at home, but in sober black and white they run:
Mae hen wlad fy nhadau yn annwyl i mi,
Gwlad beirdd a chantorion, enwogion o fri
I do not know whether there is a “standard” translation into English (there surely must be, because many present-day Welsh people speak, and therefore sing, only English); but a literal translation would be:
The old land of my fathers is dear to me,
Land of bards and singers, famous men of renown
Actually, that is not quite a word-by-word translation. Unlike English and the Continental languages which British people commonly study, Welsh puts the verb at the beginning of a sentence. Mae means “is”, so the first line taken word by word would run “is old land my fathers yn dear to me” — that sounds like a question, but it isn’t. “Is X Y” is the Welsh way of asserting “X is Y”. The word yn has no corresponding word in English: it shows the function of the following adjective — if you say “the cheerful boy”, “cheerful” is llawen, but in “the boy is cheerful” the word “cheerful” would be yn llawen. All this is just to show how very “foreign” a Celtic language is. One could hardly guess the meanings of any of the words above, except perhaps mi = “me” – and you might strike lucky with chantorion. (Thank you to Julian Ward-Davies for correcting an error in my Welsh.)
By the way, calling Wales a “land of bards” is more than an empty boast. Poetry remains a very important activity for the Welsh. The cynghanedd system of rhyme-and-alliteration which began to emerge when much of what is now England still spoke Welsh, and was formalized in the 14th century, is described (by the 1993 Princeton Encyclopedia of Poetry and Poetics) as “the most sophisticated system of poetic sound-patterning practised in any poetry in the world”. A bardic chair is awarded annually at the Welsh National Eisteddfod (or poetry competition) to the year’s most successful practitioner of cynghanedd. In 2001 Mererid Hopwood became the first-ever female winner, and she has written an excellent book on the topic, Singing in Chains (Gomer, Llandysul, 2004). In it she mentions that the publication which annually prints the winning Eisteddfod poems and the judges’ assessments (Cyfansoddiadau, “Compositions”) sells about 6000 copies within a fortnight – which, scaled to the relative numbers of speakers of the two languages, would be equivalent to the proceedings of an English poetry competition selling 600,000 copies in Britain (or three million copies in the USA). That would be the day.
It would be tempting to write a good deal more about Welsh. But our subject in this page is English.
In A.D. 43 Romans invaded Great Britain and conquered most of it — they never controlled or, probably, wanted to control the barren Scottish Highlands; the border between the Roman province of Britannia and the independent tribal areas of Scotland shifted at different periods. The Romans never went to Ireland.
Obviously, the Romans brought Latin with them. But the fate of Latin in Britain was quite different from what happened on the Continent. Gaul (the territory now called France), and much of Spain, were also Celtic-speaking when the Romans arrived, but after two or three centuries of Roman rule Latin had completely displaced Celtic in those lands. Everyone from top to bottom of society ended up speaking a form of Latin, which is why French and Spanish are Latin-descended languages today. In Roman Britain, on the other hand, Latin was only ever a minority language. The leaders of native British society used it to interact with their rulers. I suppose simpler men who joined the legions would have learned it as part of military life. But in over three centuries Latin never spread throughout the population as it did on the Continent. The natives did adopt words for novel elements of culture from the Latin-speakers through whom they learned about them (though, after two thousand years of evolution as words of Welsh, the Latin origin of this layer of vocabulary is often heavily disguised in modern Welsh). For instance, the Welsh for “glass” is gwydr, from Latin vitrum (the letter v in Latin represented a w sound); “arrow” is saeth (from Latin sagittam). But the Britons did not give up speaking their Celtic language and switch to Latin.
(I do not know why the fate of Latin was so different in Britain versus the Continent. Some historians think that, for the Romans, Britannia was “a province too far”, a continuing net drain on Roman resources which they would have done better not to take on. Possibly their administrators were spread thinner here than on the Continent and hence had less intimate cultural impact on the locals?)
Centuries later, in A.D. 410, the need to defend Rome and the heartland of the Empire against mounting attacks by Germanic tribes led to a withdrawal of troops from Britain, and they never returned. A relatively wealthy and civilized island, essentially early-Welsh-speaking with a Latin-speaking minority, lay undefended. It was obviously attractive to the tribesmen on the far side of the North Sea, who had never been ruled by Rome and whose agricultural societies were poorer and more primitive. They started moving in.
The classic account of this historical event is a book traditionally ascribed to an 8th-century Welshman, Nennius. Because this account had little competition, elements from it have continued to echo in our “national mythology” to this day. It tells how two Saxon brothers, Hengist and Horsa, landed in Kent in 449. The local British king Vortigern foolishly thought he could use them and their men as allies in his tussles with British rivals, and presented them with the Isle of Thanet to live on. But once the cuckoos were admitted to the nest they grew — more and more Saxons were invited over despite Vortigern’s protests. Horsa died in the fighting, but Hengist’s power spread. Eventually, Nennius describes Hengist and the Saxons using trickery to put Vortigern’s court to death, though this part of the account is usually regarded as a fiction.
The names Hengist and Horsa are both Germanic words for “horse”, which seems to have been a totemic animal for the Saxon tribe. In Hengist’s honour, the badge of the County of Kent even now is a white horse; Lower Saxony in Germany, where he started from, uses the same symbol.
It is this account of Hengist and Horsa’s arrival which I used to offer a specific birth date for our language. They landed at Pegwell Bay. (If you go there nowadays, you can see a replica Viking boat which was sailed from Denmark in 1949 for the 1500th anniversary.) Symbolically, this disembarkation was the point when the speech habits that were destined to turn into English became separated from those that turned into modern Dutch and German.
In reality, one cannot take this very seriously. Nennius’s history may have been reasonably accurate in some respects, but it was a very partial account of the immigration. The Germanic speakers who came to Britain belonged to several distinct tribes. The Saxons, who lived in what is now northern Germany, inland from the North Sea coast, were one of these, but equally significant were the Angles who came from the area where modern Germany and Denmark meet (and there were other, smaller groups).
The Saxons settled in the far south of Great Britain: my county of Sussex is the land of the “South Saxons”, Essex to the north of the Thames Estuary (where Essex Girls come from) is the land of the “East Saxons”, and so on. (The tribal name “Saxon” meant “men of the seax”, or sword. A seax is usually depicted these days as a long, curved, scimitar-like thing, for instance in the badge of the county of Essex, though in fact I think it may have been more of a short hunting knife.) The Angles settled further north, up to and beyond the current Scottish border. (Their name derives from the place “Angeln”, part of the area where they came from.)
It is because these were the two main immigrating tribes that Old English is sometimes called Anglo-Saxon. Nennius wrote about the Saxons and ignored the Angles. Hengist and Horsa’s arrival may have been a significant moment in the process of Saxon immigration in particular, but there is no real reason to see it as inaugurating the “Germanization” of Great Britain in general. In fact it is likely that some of these Germanic tribespeople had begun infiltrating a little before the Romans finally pulled out.
Roughly at the same period that the Anglo-Saxons were moving in to south-east Britain, at the opposite end of the island Irishmen were invading and settling in the Highland areas and Western Isles of Scotland. This was how the Gaelic version of Celtic came to Great Britain. Eventually Scots Gaelic was to develop norms of its own and has come to be regarded since the Middle Ages as a separate language from Irish, but those languages are still very closely related.
I do not know whether it was more than a coincidence that these two waves of migration happened about the same time. The area the Irish moved into had never come under Roman rule, so that wave was not a matter of Imperial defences being dismantled.
Because Gaelic is the only non-English language to survive in Scotland today, some people seem vaguely to associate it with Scotland as a whole. That is a mistake. The more populous parts of Scotland were never Gaelic-speaking (though the whole of Scotland was ruled by Gaelic-speakers fairly briefly, around A.D. 1000). Lowland Scots were Welsh-speakers, until they became English-speakers. Lowland Scotland is much more closely related, linguistically and I would presume also genetically, to England than to the historical population of the Highlands and Islands. It just happened to be politically separate until recently, as Canada is separate from the USA.
As the 5th and 6th centuries wore on, the Anglo-Saxons pushed westwards across Great Britain. About the end of the 5th century, the legendary British King Arthur (he of the Round Table — but in reality he was probably a general rather than a king) won a series of battles against them, culminating in Mount Badon, fought possibly about 490, possibly near Bath. But despite their impact in legend, in reality these were temporary successes which did not stem the tide in the long term. In 577 the Saxons won the battle of Dyrham, just east of Bristol, giving them access to the Severn shore, and hence cutting the Britons in the south-west peninsula off from their fellow Britons north of Severn. After the battle of Heavenfield, 634, Great Britain was entirely controlled by the Anglo-Saxon settlers except for some outlying, hilly areas in the north and west: the Scottish Highlands and Islands, Cumberland, the territory we now call Wales, and Cornwall.
Cumberland was in time absorbed into England. The other areas have continued to retain elements of Celtic culture into modern times, although all of them were eventually to become politically subordinate to the English-speaking centres of London and Edinburgh. The Welsh language still flourishes as the everyday speech of significant parts of Wales, although Cornish, the closely-related language of Cornwall, finally became extinct in the 18th century; Scots Gaelic is clinging to life rather precariously today.
Linguistically, what I find remarkable about this Celtic background to the emergence of the English language was the nature of the influence that Celtic had on English. Namely: it had no influence whatever. If the evidence of the English language were all we had to go by, one would think that the Anglo-Saxons had immigrated into an empty land, or that they had carried out thoroughgoing “ethnic cleansing” of its British inhabitants.
Elements of the latter did occur. At Pevensey, near where I live, in 491 the Saxons besieged the British of the area, who had retreated within the walls of the Roman fort there. (The fort still stands — until I moved house in 1997 I could see it from my window.) The Saxons managed to break in, and massacred every man, woman, and child. But historians and archaeologists have argued in recent years that this was exceptional. Some Britons must have retreated westwards into the areas such as Wales which remained British-controlled (and many crossed the Channel to found a new Celtic-speaking domain of “Brittany” in what is now north-west France), but plenty are said to have remained living within Anglo-Saxon territory, as slaves or perhaps in some cases working pieces of land independently. (The word “Welsh” comes from the Anglo-Saxons’ name for these people: wealhas, “foreigners”. Their own name for themselves was — and is — the Cymri.)
The Anglo-Saxons took over many place-names from the Britons. They founded their own scattered agricultural settlements (for these uncouth pagan farmers, the remains of Romano-British cities were scary places, best avoided), so English villages and towns have Germanic names; but many rivers, for instance, still carry the names which the Celts gave them. Thames is a British name meaning “dark river”. My colleague Richard Coates has shown recently that there is a greater density of Celtic-derived place-names in England than was traditionally recognized, though more towards the west than in the east. (See R. Coates and A. Breeze, Celtic Voices, English Places, Shaun Tyas, Stamford, Lincs., 2000.) You would think that Anglo-Saxons interacting with Britons even in a master/slave relationship, within a land where they were until recently strangers, could not fail to pick up some ordinary non-name words from the Britons’ vocabulary. If the Britons were subordinate, would the Anglo-Saxons not sometimes have taken British wives, whose children would have learned some Welsh at their mother’s knee before growing up as members of Anglo-Saxon society?
In later centuries, English became a language which was unusually hospitable to vocabulary borrowed from languages of other peoples and races that English-speakers lived among. When the British ruled India, they not only used many Indian words while out there but brought plenty of them back to Britain where there were no Indians to reinforce their use — “bungalow”, “curry”, “gymkhana”, “juggernaut”, and many other words are standard parts of our modern vocabulary, used by people who often have no idea that their roots are Indian. I believe that American English similarly has a fair number of words borrowed from Red Indian languages. But from the language of the British there is next to nothing.
In regions near the Celtic fringe there are some British borrowings in local dialects. I spent most of my childhood in Somerset: we had a word “coombe” for a kind of valley, short and steep-sided — it was very frequent as part of the proper names of particular valleys, but you could hear it used as a common noun with a small C, also. This was the Welsh cwm, “valley” (Welsh uses the letter W for the “oo” vowel). But in English that is just a local usage — not many Londoners would know what a coombe is.
One way in which some Celtic words came into Standard English was through Irish monks who came as missionaries to the pagan Anglo-Saxons. The word “cross”, for instance, is from Irish (and was earlier borrowed into Irish from Latin crux): it displaced the native Anglo-Saxon word “rood” — people still talk about the rood screen in a church, but outside that context the word is now virtually extinct. But these of course represent the Gaelic, not the British, branch of Celtic. (As it happens, in the case of “cross” there is not much difference: “cross” is cros in modern Irish, croes in Welsh.)
Of words in standard English which were borrowed from British Celtic, there are so far as is known a grand total of three:
The existence of three words in a way points up how remarkable it is that there are not far more. Evidently Anglo-Saxons were capable of getting their tongue round British words. If they found it useful to take over the local word for a bin (whatever precise sort of receptacle the word referred to at the time), how come they did not find themselves borrowing their slaves’ words for a mass of local tools, foodstuffs, or the like? It is as if the Anglo-Saxons marched through Britain saying at every turn, “Tell me, my good man, what is this river called? Thanks, that’s very useful — thump, you’re dead.” But as far as we know, they didn’t.
[Added in 2010:] In the last few years there has been a sudden flurry of suggestions that although English took over practically no vocabulary from Celtic, some of the ways in which English grammar differs from that of other Germanic languages could stem from Celtic as spoken by British natives who learned English imperfectly from the invaders. Theo Vennemann of Munich, for instance, notes that questions which sister languages would answer with a straight yes or no tend in English to be answered also, or instead, with a short statement – “Have you seen …?” “Yes, I have”, or just “I have”. An English bridegroom responds to the priest’s question with “I do”; in Germany he simply says “Ja”. Vennemann thinks this feature of English might be inherited from Celtic languages, which (like many languages in other parts of the world) had no words for “yes” and “no”. But the structural features discussed in this connexion are few and limited, and the evidence for Celtic origin seems fairly weak. (See Richard Coates’s review of a book edited by Markku Filppula and others in Language, vol. 86, pp. 441–4, 2010.)
Admittedly, since the first version of this page was put on the Web, our understanding of the social consequences of the Anglo-Saxon invasions has been modified somewhat by new research at University College London (see Michael Weale et al., “Y Chromosome Evidence for Anglo-Saxon Mass Migration”, Molecular Biology and Evolution, vol. 19, 2002). Geneticists there, in collaboration with Dutch and American colleagues, have compared the genomes of a sample of modern Englishmen with samples of Welshmen and of men from the Netherlands and Norway. They found that, even today, Englishmen are much closer genetically to the Dutch than to the Welsh. So it seems that far fewer Welshmen than previously supposed can have remained in the territory conquered by the Anglo-Saxons, as slaves or otherwise. The portion of the genome studied is transmitted through the male line only, so this in itself doesn’t tell us whether native womenfolk remained. But when we add the lack of Celtic influence on the language, perhaps the most plausible explanation is an orderly retreat by the ancient Britons, men women and children together, before invaders that they weren’t able to resist. Possibly they hoped to regroup in the West and win back the lands they had left, but it just never happened.
[Added in 2007:] I ought now to mention that a new book, Stephen Oppenheimer’s The Origins of the British, argues at length that the above standard account of how English came to exist as a separate language does not just require modification in detail but is entirely and radically wrong. He is managing to make quite a splash in the media, too! It is clear to me that it is Oppenheimer who is wrong, not the consensus, but explaining why gets slightly more technical than appropriate for this page. I have placed a critique of Oppenheimer on a separate page.
An obvious question one can ask about the origin of a language is what its earliest written text was.
The Germanic tribes already had a form of writing back in the Proto-Germanic period: the runes. Runes were letters which in some cases are clearly related to letters of our alphabet, but the shapes were adapted for carving on wood, with curves changed to straight lines. Cuts along the grain of wood are hard to see and can split the wood, so in the runes all horizontals were eliminated. A letter T has two diagonals sloping down from the top of the upright, rather than a straight crossbar.
Of course, inscriptions on wood will usually not have survived. And other early runic inscriptions on stone or metal are often so fragmentary that they are unreadable, undatable, or both. If a coin is stamped with just a king’s name, even if the name is English we can hardly call this an example of written English — if the moneyer had added more words, who knows whether they would have been in English or in Latin (the obvious alternative)? Gold rings have been found at different places with the same long runic inscription, thought possibly to be a spell to staunch blood:
Ærkriuflt Kriuriþon Glæstæpontol
But although the rings may be old, this is like “abracadabra” — these words are not meaningful Old English. (By the way, the letter æ represents an “a”-like vowel, and þ is a curvy version of the rune for the “th” sound. The Anglo-Saxons went on using this last runic letter even when they shifted to our Roman alphabet, because the Roman alphabet had no convenient single letter for “th”. Quaint old signs beginning “Ye Olde ...” are really “Þe olde” — the “th” rune looked like a Y to people who no longer used it.)
The oldest known inscription that appears to consist of actual Old English words appears on a gold “bracteate” or medallion found in 1981 at Undley Common, Suffolk. It comprises the two words mægæ medu written in the Anglo-Saxon version of the runic alphabet, thought to mean something like “reward for a kinsman” (as a description of the medallion); it has been dated to AD 450–480, so not long after English separated from the Continental Germanic languages. (These two words are preceded by a form gægogæ, which is probably not a word in the ordinary sense. It might be an abracadabra-like magic formula, or some have suggested that it was a warcry – though how awestruck anyone would be by troops yelling “gægogæ” is perhaps open to question.)
Before 1981, the item people commonly cited was a roe-deer ankle-bone found in Norfolk and bearing a single word interpreted as “roe-deer”, also dating to the fifth century. But this was written in runes of the Scandinavian variety, and there is no strong reason to identify the word as English rather than Scandinavian.
Leaving such fragmentary items aside, at one time there did seem to be a fairly clear answer to the quest for the oldest extended example of written English: the Bewcastle Cross.
The Bewcastle Cross stands in the graveyard attached to the parish church of the tiny hamlet of Bewcastle, in the hills in the far north-east of Cumberland — about as deep into the “back of beyond” as it is possible to get in modern England. (It was more important in the past: Bewcastle was the site of a fort on a now-disused Roman road.) I have a photograph of the Bewcastle Cross on my office wall, which I took last time I went there, at the beginning of the 1990s. Although originally it was indeed a cross, the crosspiece was lost in the 17th century: these days it is a pillar, square in cross section, with its four faces having carved panels of Christian figurative art, stylized ornamentation, and a number of passages written in runes.
Some of this writing is certainly Old English rather than Latin. The reason why it has been taken as the oldest monument of English is that the text was read as referring to two known historical individuals, Alhfrith, who became under-king of Deira (modern Yorkshire) in about 655, and his wife Cyniburg. The inscription was read as commemorating Alhfrith’s death; he probably died before 671. That dating would give the Bewcastle Cross priority.
It would be very satisfactory if this were right. Bewcastle is a lonely but beautiful, numinous place. It is the kind of place where one would like our language to have a tap-root.
Alas, it probably isn’t so.
The problem is that the inscriptions are very hard indeed to make out. The Bewcastle Cross has been facing the storms and ice of the North Pennines for well over a thousand years: its runes are now at best faint and hard to distinguish from random weathering, and often entirely illegible. Worse, students or guardians of the monument in past centuries have not always respected the difference between cleaning the moss and dirt from the lines, and using their penknives to “improve” the lettering. Forty years ago the Bewcastle inscriptions, together with earlier accounts of them by antiquarians, were exhaustively studied by the leading expert on English runes, R.I. Page — Page’s 1960 article “The Bewcastle Cross” is reprinted in his book Runes and Runic Inscriptions, Boydell & Brewer, 1995. Page believed that the reading “Alhfrith the king” was an over-imaginative misinterpretation (and Cyniburg’s name alone is no evidence for dating, because it was a common one). Without Alhfrith, dating could be based only on artistic or lettering style, and these do not suggest so early a period.
(The best book on English runes, by the way, is R.I. Page’s Introduction to English Runes, Methuen, 1973; and in 1987 the British Museum Press published a shorter, lavishly illustrated book by the same man called just Runes.)
Of course, a question about “oldest text” can be taken another way — not “What is the oldest physical object we can see today that includes some writing in the language?”, but “What is the oldest passage of wording that has come down to us?”, even if the oldest surviving record of it was written much later. (In some cultures, literature has been transmitted by memory and repetition for long periods before ever being written down.)
So far as I know, the oldest known English text in this second sense — at any rate the earliest piece of literature by an identifiable author — is “Cædmon’s Hymn”, composed not too far in time from the period to which the Bewcastle Cross seemed to belong but probably doesn’t.
Over the century or so after Saint Augustine arrived from Rome in 597, England gradually converted from paganism to Christianity. Missionaries from Ireland were preaching a kind of Christianity different from Rome’s; in 664 King Oswy of Northumbria — father of the Alhfrith who seemed to be named on the Bewcastle Cross — organized a conference at Whitby Abbey, on the Yorkshire coast, to decide which version of the religion England should opt for. (The decision went to Rome.) Cædmon — date of birth unknown, died 680 — was a lay brother at Whitby, who was told in mystic visions to produce religious poetry.
Here are the first few lines of Cædmon’s first composition. (Cædmon naturally spoke the Northumbrian dialect of Old English; we have his Hymn recorded both in his dialect and in the Wessex dialect which was to become the standard national language — I shall quote the latter version.)
Nu we sculan herian | heofonrices Weard,
Metodes mihte | and his modgeþonc,
weorc Wuldorfæder; | swa he wundra gehwæs,
ece Dryhten, | ord onstealde.
In modern English:
Now we must praise the maker of the kingdom of heaven, the power of the Creator and his thought, the work of the glorious Father; how he, the eternal Lord, established the beginning of all wonders.
(I cannot divide the modern translation up into lines matching the Old English, because the word order is different — “all wonders” comes at the end of the third line, before “the eternal Lord” at the beginning of the fourth.)
Old English poetry does not rhyme. It consists of lines in two halves (separated here by the “|” mark), which take their artistic unity from stressed syllables beginning with a repeated consonant. So the first line here is an H line, the second an M line, the third a W line. The fourth is a vowel line (based on syllables starting with no consonant, e.g ece, ord). To anyone who would like to get a feeling for how this poetic technique worked in practice, I would recommend Simon Armitage’s translation into modern English of the long 600-year-old poem Sir Gawain and the Green Knight (Faber, 2007); Armitage renders this extraordinary science-fiction-like piece of literature into idiomatic present-day language, but in the same alliterative metre. It has been hailed widely as an outstanding example of literary translation.
If you go to Whitby today and climb the 199 steps up to the Abbey on top of the cliffs, you will see a monument to Cædmon. And a beautiful piece of ancientry it looks, too. But if you think it is remarkably well preserved, that is because it was put up only recently.
[Added in 2016:] Since drafting the above, I have learned that at least one prose document is a bit older than Cædmon’s Hymn, namely the Law of Æthelberht, king of Kent, probably written not long after AD 600 (though the oldest surviving manuscript dates from the 12th century). The Law of Æthelberht is the earliest law code in any Germanic language – because it dates from soon after Augustine’s arrival in Kent, Patrick Wormald suggests that it may have represented an attempt to establish that kingdom’s claim to be a civilized society comparable to Rome. It begins:
Þis syndon þa domas þe Æðelbirht cyning asette on AGustinus dæge
These are the decrees which King Æthelberht set in Augustine’s time
and continues with a long series of scales of monetary compensation for various injuries or offences against people of different social ranks. You can see a digitized version of the surviving manuscript online (the words quoted above appear at the top of “Leaf 11”, in red).
Christianity is a religion of the book, and the Conversion meant much more writing going on, largely in our own Roman alphabet. For several centuries, Roman letters and runes were used side by side. (People often imagine that the Church was hostile to runes as relics of pagan devilry, but that was by no means so.) With so much being written, it becomes impossible for me — and perhaps for the real experts — to identify one oldest physical monument or manuscript among a mass of possibilities.
The Anglo-Saxons were ignorant barbarians when they arrived in Britain, but they certainly raised their game over the following centuries. In particular, the Venerable Bede turned English into a language suitable not only for religious writing but for intellectual discourse much more widely.
Bede was born in 672 or 673, on the domain of the monastery at Wearmouth in north-east England, and spent his life from age 7 as a monk there and at the nearby monastery of Jarrow. He died in 735. He has been called “undoubtedly the most learned man of his time” anywhere in Europe. Bede knew the Earth was a sphere. He was the first person to establish that a year is not exactly 365¼ days long, so that the calendar needed modifying (nine centuries later, it was modified). Among other things, Bede was the first historian to use the “A.D.” system for dates — the year when I am writing is numbered 2001 because, in effect, the Venerable Bede decided it should be. (It was Bede, incidentally, who took down Cædmon’s poetry from dictation.)
And Bede was also the first to use English as a medium of scholarly writing. Thanks to Bede, and also to promotion of English-medium education by King Alfred two centuries later, English became a language serviceable for serious, official and scientific purposes, at a stage when the spoken languages of other European nations were low-status affairs, and anything that mattered had necessarily to be written in Latin. English scholars could use Latin, of course — so far as I know there was no time, from the Roman occupation down to about thirty years ago, when having an education in this island did not include learning Latin. But, when they were not addressing an international audience, they did not need to.
If you visit Durham Cathedral, you can see Bede’s tomb. (You would be hard put to miss it.)
Since the Bewcastle Cross is apparently not the “first” it once seemed, as already said it is beyond my power to identify the oldest single datable inscription in English. But, if you want to go and see one very early example of our language, I know which way I would point you. The Ruthwell Cross, at the village of Ruthwell in Dumfriesshire, Scotland, is another stone cross with runic inscriptions. Unlike the one at Bewcastle, though, it is inside its church and well preserved (even though Cromwell’s men knocked it about a bit in the 17th century). Also, the inscription is not a brief, dry epitaph: it is a long extract from a poem, The Dream of the Rood — a poem worth reading, not just an exercise in worthy piety. (It is about the agony and humiliation of a god stripped and tortured by mortal men; “rood” means cross, remember.) Not all the poem is included, but we know the whole from a separate source, a manuscript which is now in an Italian museum. Furthermore, the artistic standard ... well, go and judge for yourself. The Ruthwell Cross has everything.
There is no way of dating it exactly, and at one time R.I. Page thought it could possibly be as late as A.D. 850, though in recent writing he treats it as older than that. It probably is not the very oldest English inscription in existence — but on the other hand, I do not know of another that is certainly older.
Pity that such a splendid monument of the English language isn’t in England. But it’s the Scots’ language too.
England is named after its people — England, originally “Englaland”, is the land of the English or land of the Angles. The name England/Englaland was not used until about A.D. 1000, centuries after the people and the language were being called “English”, Anglisc. But there were two main tribes who arrived on this island fifteen centuries ago. Why did the Angles beat the Saxons in the name game?
Oddly enough, for the British people we displaced, it is just the other way round. The Welsh call an Englishman a Saes and the English language Saesneg (names of languages all end in - eg in Welsh). In Gaelic, too, England is Sasunn, an Englishman is a Sasunnach (which English people spell, incorrectly, as sassenach). All the Celts name us after the Saxon half of the Anglo-Saxon combination. Yet, in our own language, the “Anglo-” half has prevailed.
(In passing, let me mention that there is one further, mysterious verbal root. The country, England, is called in Welsh Lloegr. I do not know where this word comes from, and so far as I am aware no-one else knows either. I have seen it suggested that Lloegr is the “lost land” — that is a very romantic idea, but so far as I know it has no basis at all in Welsh etymology. If any Celticist reading this does know what the name Lloegr comes from, I would be very grateful if you would get in touch.)
The fact that Saxons were more visible than Angles to the Celts almost certainly had to do with where they settled. It was Saxons in the south-west who were chasing Ancient Britons into the hills of Wales and Cornwall, and probably Saxons who would have been most likely to sail to Ireland. The Angles’ main coastal holdings faced back towards the Continent. That explains Saes and Sasunnach. But it does not explain why we say “English”.
One explanation I have heard sets this, too, down to the Venerable Bede. Bede, as a Northumbrian, was himself an Angle rather than a Saxon. So he naturally called his main book (which was a history of the English nation up to the time when he finished writing it in 731) Historia Ecclesiastica Gentis Anglorum — Ecclesiastical History of the Nation of the Angles. He surely would have been aware that some of the folk he was writing about, the ones living down around London, did not think of themselves as “Angles”; but, if you live in Jarrow, London feels like a distant irrelevance in the 21st century, never mind the 8th. And so that became the general name for the whole nation and its language.
Whether Bede’s usage was actually the decisive factor I do not know. Other sources say that the people and nation were already regularly being referred to in Latin writings as Angli, Anglia in the 7th century. Possibly Bede was just following a chance trend that had already become established.
If the name is down to Bede, he did us a good turn. In the east, the men of the seax turned through the evolution of sounds into the men of Essex. In the south, they turned into the men of Sussex. So, if the Saxon side of the Anglo-Saxon combination had happened to become the dominant name, I would probably be living in Sexland, and we would be speaking Sexish.
We English are a highly respectable nation; those names would not have suited us. “English” will do very nicely. Here’s venerating you, Bede!
This is not the place to attempt a full account of the history of English. But two further early factors that should not go unmentioned are the Danish invasions, and the Norman Conquest.
After the Anglo-Saxons had got control of England, they themselves had to contend with hostile invaders: “Vikings”, that is Danes, speakers of a North Germanic language, who won many victories against the Anglo-Saxon kingdoms beginning about 800, until the situation was taken in hand and brought to a compromise resolution by Alfred the Great (reigned 871-99) at the Treaty of Wedmore, 878. (Wedmore is in the area where, the story goes, in escaping from the Danes Alfred took refuge in a peasant woman’s hut. She left him in charge while she went out, thinking she could probably trust the King; but, preoccupied with the problem of defending England against the Danish onslaught, Alfred let her cakes burn — and got angrily scolded when his hostess got back.) Under this Treaty, Danes took control of the eastern sector of England, but there were no expulsions of population; for English inhabitants of the “Danelaw”, I think it was more like a change in who they paid taxes to, though it meant that Scandinavians came in to settle alongside them.
The linguistic significance was that it brought a large proportion of the English population into regular contact with Danish speakers, and many Danish words came into English — not just ordinary words, for instance loan, flat, take, but even “grammar words” such as they, are, both. Because North and West Germanic were not so very different, these borrowings fitted the pattern of English words fairly closely; once accepted into English, they looked as though they had always been part of the language. (That is not like the words that educated people coined on the basis of Latin roots, which often look quite distinct from the native vocabulary, with more syllables and alien vowel and consonant combinations — for instance that last pair of words, consonant and combination, could neither of them possibly be Germanic.)
Some linguists have claimed that the contacts with Danish had larger consequences for English. Modern English is unusual among European languages for the fewness of its inflexions. We distinguish singular from plural nouns, and we conjugate verbs as eat, eats, eating, ate, eaten — but that is about as far as it goes. Every verb in French, for instance, has a panoply of separate 1st, 2nd, and 3rd person singular v. plural forms in each of numerous tenses; there is nothing approaching that in modern English. Old English did have more inflexion. One suggestion is that it was the contact with Danish that wore it down. English and Danish were close enough that the roots of words tended to be understood across the language barrier, but the inflexion systems were different and hence not intelligible, so English speakers abandoned their inflexions. On this account, Old English turned into something like the “creole” languages which have developed more recently in distant parts of the world where speakers of very different languages have come into contact — for instance in New Guinea, where “me no like loseim fellas belong me” means “I don’t like losing my friends”.
To me this sounds implausible. I cannot see why such a radical hypothesis is needed to explain a case of a language becoming simpler. And, if it is needed, then I wonder how the alleged creole managed to spread among the many English people there must surely have been who rarely or never had occasion to speak to Danes — the ones living outside the Danelaw, for instance. In any case, Old English inflexions were already quite limited relative to its Indo-European forerunners, before the Danes invaded. But I have not studied these theories, so I am not really entitled to voice an opinion.
Lastly, the most famous single date in English history is 1066, when the Normans under their leader Duke William sailed to England and conquered us at Hastings. (More exactly, the battle occurred, on 14th October, at what is now the village called — naturally — Battle, five miles inland from Hastings, and again not far from my home.) On Christmas Day that year, Duke William was crowned King William I of England, and he proceeded to divide up the territory of England among his Norman followers.
(I realize that in writing the above paragraph I betrayed my own allegiance. I have Norman as well as Anglo-Saxon blood — I have much more Norman blood than the average Englishman, because my mother came from a Channel Island family. Nevertheless, I cannot contemplate Hastings without feeling that the wrong side won. Most of what I know suggests that this was a case of a higher civilization subjugated by a lower, cruder one. So I automatically side with the losers. But it is a fait accompli by now. The battle today is against a newer brand of Continental domination.)
The Normans spoke a dialect of French — though, ironically, for them this was a new language. Racially, they were Scandinavians (“Norman” means “north-man”), descendants of Vikings who had been raiding France as other Scandinavians had raided England. In 911 the French king Charles the Simple had granted them the nucleus of what became the Duchy of Normandy, apparently thinking (rather as Vortigern had thought with Hengist) that he could use them as allies against other raiders, and that once things settled down they would go back home. This, I believe, is why the French nicknamed Charles “Simple”. Of course the Normans stayed, and they expanded Normandy beyond its original limits. But, unlike the Anglo-Saxons in Britain, they very quickly abandoned their Germanic language in favour of the local French dialect of their new subjects. By the time of Hastings, 150 years later, although the Normans still gave their children Germanic names, they spoke only French. Hence French became the language of English government.
England being a larger and wealthier territory, the Normans reoriented themselves to see this as their central domain and Normandy as more peripheral; shortly after 1200 the French reconquered Normandy. French, which was only ever a language of the élite in England, gradually lost ground to English. French ceased to be the official legal language of England in 1362; a few decades later, King Henry IV (reigned 1399 to 1413) spoke no French at all. But in the intervening centuries, large numbers of words had been borrowed from French into English. To quote the hackneyed example, beasts were pigs and cows and sheep while a Saxon peasant was herding them, but they became pork and beef and mutton (porc, boeuf, mouton) when served at the lord’s table.
This is perhaps the single most distinctive aspect of the complexion of modern English when contrasted with the languages which are its nearest relatives. German has a large native vocabulary, and it has some international learned words coined from Latin or Greek roots — Natur, Psychologie. But it has only a sprinkling of words borrowed from French or other modern languages. English has many more Latin-derived words than German, and it also has a vast stock of French vocabulary alongside the native wordstock.
Old English was like modern German, only more so. Rather than borrowing alien words when it encountered them (mainly in connexion with Christianity), Old English normally invented equivalents using its own roots. The word discipulus in the Latin Bible might occasionally be borrowed into Old English, as discipul, but usually it was translated as leorning-cniht, “learning-youth”. (I take this example from Leslie Blakeley.) In modern English we say “disciple” and think nothing of it. After the Norman Conquest accustomed English to borrowing words wholesale from French, in due course English became a language strikingly receptive to word-borrowings from anywhere and everywhere.
And that is how English came to be. After the Middle Ages, the language naturally continued to evolve, as languages always do. But there were no further sudden, radical changes to its internal structure or contents. The drama in its subsequent history was its spread across the world.
For instance, the English language first arrived in North America on the morning of Saturday 24th June, 1497, on board the ship Mathew out of Bristol. The Mathew’s captain, John Cabot, the discoverer of North America, was an Venetian (his original name was Giovanni, or in Venetian dialect Zuan, Caboto), but his crew will have been mainly or entirely West Country men. Cabot’s original voyage cannot be the reason why the American accent of English sounds more like West Country English (to my ears, at least) than like any other English regional dialect — the brief 1497 landfall on Cape Breton Island in what is now Nova Scotia did not lead to colonization; but Bristol continued to be our main Atlantic-facing port for many centuries, so plenty of English-speakers who did end up in America permanently will have come from that area.
(Many people think of Christopher Columbus as having discovered America in 1492, but in that year he only reached the West Indies. Columbus and Cabot both first set foot on the American mainland — South America in Columbus’s case, North America in Cabot’s — about the same time in 1498. I went to school in Bristol, and what we were taught about the name “America” was that on returning from his voyage of discovery Cabot named the place after one of his sponsors, Richard Ameryk, who was the King’s customs officer at Bristol. There is an alternative story according to which the continent is named after an Italian mapmaker. You pays your money and you takes your choice.)
The English which arrived in America half a millennium ago, and the Englishes which immigrated into other distant parts of the world in subsequent centuries, were not precisely identical to English as it is spoken in any part of England at the beginning of the 21st century. That is one reason why separate national versions of English are recognized today. But, relative to the diversity of the world’s languages, the differences between “British English”, “American English”, and so forth are tiny — minor dialect differences within a single language. By the end of the Middle Ages, to all intents and purposes the English language was a finished structure.
Hasn’t it done well since!
last changed 14 Nov 2017