If you have never read the entrancing historical novels of Rosemary Sutcliff (1920-92), you have a huge treat in store. Ostensibly written for children, her books give an uncannily powerful sense of ordinary life in Celtic Britain from the Iron Age through the Dark Ages. They also offer an inspiring historical mythology. In the 5th century, she suggests, the fading of Roman rule in Britain meant snuffing out the candle of civilization, replacing the rule of light with centuries of darkness, of ignorance and pagan barbarism. In books like The Lantern Bearers and Sword at Sunset, Sutcliff depicted the last warriors of that old civilization, the soldiers of the dying light, whose desperate struggles would eventually be commemorated worldwide in the Arthurian legends.

In less romantic form, that idea of the collapse of a once-glorious civilization long dominated accounts of the end of Roman Britain. The narrative has a potent religious dimension, for Roman Britain was Christian, with a full framework of dioceses and churches. Yet during the 5th century, that older structure was utterly uprooted across most of what would become southern and eastern England, raising the troubling question that accompanies the death of any church: how can God permit his faith to be swept from any land? In this instance, though, the wider church did ultimately triumph. From Western Britain, Celtic Christians like Patrick spread their faith to Ireland, where it flourished mightily. By the 7th century, the heirs of those early Irish converts were carrying that faith—and the light of learning and civilization—across the British Isles and over much of Western Europe. Faintly and tenuously at first, the light returned.

And yet, for all the achievements involved in saving civilization, the older Celtic peoples of Roman Britain survived only in sadly reduced circumstances. By the Middle Ages, the British who had once dominated most of the island clung on to political power in the poorer western regions that we today call Wales, with an outlier in Cornwall.

Assailed from many sides, that story—that mythology—has been looking distinctly battered in recent years, and it is wonderful now to have a definitive scholarly survey of these controversial and baffling centuries. Over a brilliant 40-year career, Thomas Charles-Edwards has shown himself a master of the history of this era. He has an un-paralleled grasp of the sources, both literary and archaeological, and an enviable ability to integrate material from Continental Europe and Ireland as well as Britain itself. Covering an era that tends to attract flaky pseudo-scholarship, Charles-Edwards never fails to present lucid reasoning based on rock-solid source material and eminent good sense. Wales and the Britons is likely to be the standard work on its subject for a generation to come. It is an extraordinary book.

The problem in understanding the end of Roman Britain is that we are gazing into a Dark Age, dark in the sense of lacking virtually any reliable contemporary sources. Apart from the jeremiad of Gildas, around 540, some memorial inscriptions, and two precious documents by Patrick himself, Britain in the 5th and 6th centuries has left almost nothing in the way of written records.

Throughout most of British history, then, scholars have necessarily relied on Gildas' horrific account. Briefly, his narrative can be reconstructed something like this. The Roman Empire officially withdrew its protection from Britain in 410. For several decades, the British maintained an independent regime, but barbarian dangers forced them to import friendly barbarian allies to bolster their defenses against more lethal enemies. In the 440s, the system broke down, as the Germanic "allies" treacherously rebelled. A bloody war destroyed the surviving Roman civilization across what is now southern and eastern England. Many British perished, amidst scenes of ghastly slaughter. The more fortunate survivors fled to the north and west, or overseas to the region that became the new Britain beyond the seas, "Brittany." Most others continued under Anglo-Saxon rule, in conditions of degrading servitude.

Ethnic wars between the British and Anglo-Saxons continued for decades, culminating in a decisive British victory. By 540, Gildas can regard the Germanic invaders as a bygone threat, and concentrates instead on denouncing his sinful British fellow-countrymen. From the late 6th century, though, the Anglo-Saxon advance revived, and by 630 the Germanic interlopers occupied the great majority of what we would now call England.

The Gildas-derived story is profoundly uncongenial for modern scholars, who struggle to reconcile it with archaeological evidence. They see little evidence that the old Roman towns still stood in anything like their ancient glory by 450, and in large areas of the country they search in vain for significant evidence of an Anglo-Saxon presence. In terms of material remains, they see few signs of a sharp break in settlement at the appropriate time.

Politically too, modern critics dislike the Gildas story because of its harsh condemnation of the "primitive" barbarian cultures that overthrew advanced Roman Britain: what kind of arrogant value judgment is this? Modern writers accuse their older counterparts of deeply conservative political agendas, of using their narrative to condemn foreign immigration as such, to condemn "lesser breeds without the law." Over the past 30 years, a new paradigm has proposed an almost comical tilt toward political correctness, in which the Anglo-Saxons are seen as a small influx of benevolent immigrants, who merged peacefully with the large majority population of the older British. In fact, we are told, the British were so taken by these nice newcomers that they decided to adopt their lifestyles and even their language, in a splendid example of early multiculturalism. Old British became new English.

If I sound cynical, it is for one overwhelming reason, namely language. In France, Spain, and Italy, new Germanic immigrants really did merge with older Roman populations, adopting the language of those deeper-rooted majorities. Latin evolved into French, Spanish, and Italian. The screaming exception to that rule is Britain, where English wholly supplanted both Latin and Celtic British. Only about 30 loan-words of Celtic origin have ever been identified in English, including all its varied dialects: "crag" is the least obscure. Has a majority population ever moved from one language to another minority tongue with virtually no survival of the original speech, no linguistic interference? Nor is there a word of British Latin: all the early Latin loan-words come from later church sources. (Intriguingly, Charles-Edwards shows convincingly that without the conquest, British Latin would have evolved into a Romance language parallel to Spanish or Italian).

We find no evidence in Anglo-Saxon England of any minority population continuing to speak British long after the conquest, and certainly not in the south or east. Could such a total and rapid abandonment of the old language have been accomplished voluntarily, as is suggested by the "choosing lifestyles" theory? Do we have to imagine a Romano-British father coming home one day and telling his family, "Well, I've decided. We are going to become Germanic, and as of tomorrow, we must speak nothing but that language. So tell the kids, not a word more of this Celtic! We'll start by working on our strong verbs. Oh, and I'm changing my name from Claudius to Aethelfrith." Is it not vastly more likely that British numbers were reduced massively and violently, and the survivors reduced to such brutal slavery that their language had absolutely no impact on their new Germanic masters? The Anglo-Saxons inflicted a kind of apartheid system on the wealhs, the foreigners—the Welsh. Gildas, in short, was exactly right in substance. If not literal genocide, then it was cultural annihilation.

I venture any criticisms delicately here, because Charles-Edwards is such a master of linguistic evidence, and he says, undoubtedly rightly, that many former speakers of British Celtic or Latin did take up the new English languages. But the conditions of that transition are open to much debate.

Critically too, the church evaporated in England as it did not in emerging France or Spain. The effects of endemic warfare were vastly aggravated by severe and recurrent plague, likely accompanied by famine as economic structures collapsed. These disasters destroyed the old urban framework. Latin-speaking elites were eliminated or fled, including the church leadership. The British church that was well established in 420 had vanished without trace by the time new Roman missionaries appeared in the 590s. That is in stark contrast to Western Europe, where Latin survived as the language of the cities, however reduced those were, and of the bishops who effectively ruled them, alongside the new Germanic warlords.

In southern and eastern Britain, though, cities, bishops and Latin perished together, between (say) 450 and 480. Nor, tragically, had monasticism evolved early enough in Britain to provide alternative centers of spiritual power and prestige, which might have kept the church alive.

Obviously the faith did not vanish overnight, and we see odd traces of Christian continuity under pagan Anglo-Saxon rule. A pilgrim shrine like St. Albans survived, as did half-forgotten cults of early martyrs, and we have some scattered place-names like "Eccles" (ecclesia). But when, about 600, St. Augustine of Canterbury wanted to meet the leaders of the old British church, he had to travel far to the west to do so. In southern and eastern England, organized Christianity was no more.

Matters were far different in the north and west, where Christian British kingdoms lasted for centuries. Celtic statelets continued in Rheged (Cumberland), Elmet (around Leeds), and Strathclyde (Dumbarton), and of course in Wales and Cornwall. Much of Charles-Edwards' book concerns the structure of these societies, drawing heavily on legal material as well as epic literature.

It is particularly from the poetry that the author traces changing concepts of ethnicity and national identity. He offers a fine reading of the Gododdin, a classic work from c. 600 that records a doomed attempt by northern Welsh warriors to storm an Anglo-Saxon fortress in Yorkshire. Among other things, the poem is the first to mention a highly evocative name: of one fallen hero, we are told, he fought magnificently, but he was no Arthur. The 10th century Armes Prydein Vawr ("Prophecy of Great Britain") proclaims the hope that the Welsh will join their Norse and Celtic allies to overthrow the loathed dominion of the evil English. The prophecy is placed in the mouth of that ancient poet Myrddin, whom later Europeans would know as Merlin.

The new Celtic societies offered an odd mixture of the barbarian and the Roman. Yes, the lords lived in a kind of heroic Celtic world, surrounded by their devoted war-bands, and glorified by bards. But they also defined themselves as Roman. In areas of Wales, old Roman estates and land law persisted into the 7th and 8th centuries. British princelings kept alive old Roman titles, and bore such ancient names as Honorius, Constantine, and Candidianus. Evidence for continuing links with the Mediterranean world comes in the form of imported pottery, jewels and coins, but also of personal names. British princes gave their sons such voguish Byzantine names as Theodore (which became Tudor); Eugenius (Owen); and Maurice (Morris or Merrick). They were also thoroughly Christian, operating in a landscape densely littered with monasteries, shrines, and schools, some of which powerfully influenced their daughter churches in Ireland.

Charles-Edwards also offers one heart-breaking irony. However passionately the British defined themselves as Roman, the surviving Romans themselves did not acknowledge that status. Continental Christians despised these outlying British and Celtic peoples as barbarians, and so, horror of horrors, did the newly converted Anglo-Saxons, who themselves demanded their share in the Roman inheritance. Anglo-Saxon clergy had only contempt for the Celtic Christians, who were so obviously heretical in such basic matters as the dating of Easter. (Actually, the only sin of the Celtic churches was in using older calendars.)

As the Anglo-Saxon kingdoms evolved into the mighty English state of the Middle Ages, so Wales was ever more marginalized, until it was eventually conquered and assimilated. Not until the Tudor period would the Welsh reassert themselves culturally, to the extent that the united nation would once more adopt the archaic name of "Britain." It was a British Empire that spanned the globe.

I conclude with an apt quotation from Rosemary Sutcliff. One of her characters declares, as the Roman legions depart: "It may be that the night will close over us in the end, but I believe that morning will come again. Morning always grows out of the darkness, though maybe not for the people who saw the sun go down."

Philip Jenkins is Distinguished Professor of History at Baylor University's Institute for Studies of Religion. He is the author most recently of Laying Down the Sword: Why We Can't Ignore the Bible's Violent Verses (HarperOne).

Copyright © 2013 by the author or Christianity Today/Books & Culture magazine.
Click here for reprint information on Books & Culture.