21st Century Classical Greek -- authenticity and a head's up

This is a head’s up about more changes in accessing my blog. After April 15 the rules on Twitter change so here’s how you can keep reading my posts or read posts you didn’t know about. Generally:

1.         Twitter: You will have to Follow me to see future notices about posts. I am not now and never have been “verified” (had a blue check) and after the 15^th my tweets will no longer show up on the “For You” tab, only on the “Following” tab.

2.         Web. In a browser you can bookmark the blog. On a smart phone, use your search engine. Find pajheil blogspot and click on the link. Then go to the bottom and look for the “website version” link. Click on that and put it on your home screen.

3.         You can subscribe to the blog from the blog site.

4.         My Friends on Facebook will see the links to the post come up.

There’s one more scheduled post on Greek after this one and then, as with other subjects, I’ll post when I think I’ve found something new.

In Book II 101.5 I get to point out a way to tell if part of a writer’s work has been forged.

ὁ δὲ τήν τε Χαλκιδικὴν καὶ Βοττικὴν καὶ Μακεδονίαν ἅμα ἐπέχων ἔφθειρε, καὶ ἐπειδὴ αὐτῷ οὐδὲν ἐπράσσετο ὧν ἕνεκα ἐσέβαλε …

Thucydides has a habit of using this double sigma. We also find it in Josephus’ Antiquities of the Jews, 18.82:

προσποιησάμενος δὲ τρεῖς ἄνδρας εἰς τὰ πάντα ὁμοιοτρόπους τούτοις ἐπιφοιτήσασαν Φουλβίαν τῶν ἐν ἀξιώματι γυναικῶν καὶ νομίμοις προσεληλυθυῖαν τοῖς Ἰουδαϊκοῖς πείθουσι πορφύραν καὶ χρυσὸν εἰς τὸ ἐν Ἱεροσολύμοις ἱερὸν διαπέμψασθαι, καὶ λαβόντες ἐπὶ χρείας τοῖς ἰδίοις ἀναλώμασιν αὐτὰ ποιοῦνται, ἐφ᾽ ὅπερ καὶ τὸ πρῶτον ἡ αἴτησις ἐπράσσετο.

This form of the verb shows up in books 17 and 18 of Antiquities – but in section 8.94 it is ἐπράττετο and in book 16 we have διεπράττετο. Writers don’t normally change their spelling in the middle of a book, unless they are representing dialect. Normally I would have a character in a book say “can’t”, but if I was representing southeastern US dialect I might put “cain’t”.

Book 18 65-84 are the stories of Paulina and Fulvia. The change in grammar identifies them as insertions. This happens all the time. It’s one of the reasons that you can’t take the text of classical writings for granted. Another is the fact that whole works have been composed and then foisted onto the public with a famous writer’s name. The Embassy to Gaius of Philo is probably authentic; the twelve volumes about Judaism aren’t, and the gross errors in them show that they are not. Anecdota probably was not written by Procopius as claimed.

Now look at 63 and 64.  There is an identical section in Wars of the Jews, in only one copy, where it is known to be a forgery. Let’s look at the context. Josephus died in 100 CE. The doctrine of the “christ” was strictly a Christian issue. The phrase o khristos in would be meaningless to both Josephus and his audience, who were upper class Romans. There’s no sense putting the phrase there without an explanation of what it means. The same issue applies to references to Shabbat and New Moon in Amos and Hoshea; they don’t explain what it means because their audience are practicing Jews. Josephus’ audience are not practicing Christians. It takes a practicing Christian to make this insertion.

In addition, the insertion has pollous, “many”, the classic weasel wording that people mark with {{citation needed}} on Wikipedia. How many is many? Fifty? In an empire of maybe 8 million?

Furthermore, Origen had a copy of Antiquities, being a Greek geek, and his copy did not have this material. You’d think it would, but you’d be wrong.

Finally, pollous is grammatically incorrect. I exchanged emails in 2009 with a gentleman at Cambridge University, where they certainly ought to know their Greek, and he said it’s wrong. Somebody who didn’t know Greek put these two sections in simply because this part of Antiquities is set in the time of Pontius Pilate.

Josephus isn’t the only victim. So is Thucydides. Notes on III 84 say it is an interpolation. Charles Smith points out the moralizing tone. Thucydides is not a crusading reporter, he’s Sergeant Friday: “Just the facts, ma’am.”.

This comment, unknown to Smith, reinforces that Thucydides was working like the recorder of an oral tradition. Olrik’s mentor Grundtvig specifically worked with ballads that originated in the oral environment, made their way into literate culture, and then back. The literate redevelopers inserted moralistic material, typically the refrains, and these were stripped out when their work got back into the oral environment.

And remember the other week when I showed that in medieval times, people were still meddling with written material.

So that’s another reason we’re learning Greek and starting with Thucydides – or Herodotus or Xenophon. Until you know how Greek reads in Classical prose, you are at the mercy of writers whose work has been doctored to say what somebody wanted it to mean.

I'm just saying -- sometimes a great notion

is not a great notion, it's somebody's preferences. 

Somebody's tweet in my timeline asked for items for a powerpoint presentation which seems to be about what people wish they had known before they did their dissertations. I had two suggestions. One is how to avoid fallacies. I've talked about that on my Fact-Checking blog. A lot.

The other was how you tell your advisor to take a flying leap when they want you to use the latest jargon instead of plain English.

I got a like on that one.

So the point being that while STEM dissertations have to rely on physical evidence and mathematical calculations, non-STEM (are you listening, you liberal arts people?) go by fads, fancies, and the favorite whims of the department heads and advisors.

It's like that Elliott Gould movie where he's defending his dissertation, and one of the evaluators jumps down his throat about not addressing that particular person's hobby horse, warped as it might be and not available in the literature Gould had access to. (His dissertation was saved by a campus riot interrupting the defense.)

That's aside from the classics field where authoritarianism reigns. I've pounded on that on my Classical Greek page and it's still true in the 21st century. A couple of guys at Cambridge who were teaching Classical Greek looked at the stack of handouts they used in their classes and decided they had a book in there somewhere. Problem is, their point of departure was Smyth's century-old 800 page work which, as I know from studying it, is incomplete, inaccurate, and outdated. Oh, sure, the 21st century book throws in aspect as a sop, but the authors fail to see that they have failed the test of Occam's Razor. If they know what it is. Of course, Ockham studied at Oxford, not Cambridge, so there's that. 

So people who are about to defend their dissertations know that their advisors are pushing them to create ephemera, works doomed to extinction once the fad passes, not build a body of solid research to influence future studies. And that is a very sad thing to have to say about people who have put in six or ten years on a paper chase.

I'm just saying....

21st Century Classical Greek -- concepts that are dead or ought to be

The change to aspect destroys a number of concepts from the old grammars, and is supported by the evidence. Thucydides never uses “imperfect tense” in a context of an interruption of an action. When an educated man who speaks a language on the streets, uses it in a way that contradicts the authoritarian grammars written by non-native speakers, the latter have to be wrong.

Several things fall out of this.

1.                  We don’t need to explain “present tense” in a past situation. Instead, the nuance is doing something out of habit or as part of an existing situation.

2.                  “Future perfect” doesn’t exist. The grammars admit that it is a passive, which is allowable since it is a flavor of the imperfective. The reduplication that created this term only occurs in Attic Greek, not in other dialects.

3.                  We can explain that passive morphology can’t exist for resultative, perfective aspect verbs. They are a one-and-done, not a continuous situation.

4.                  The scarcity of “future tense” as a conjugated verb shows that it represents a promise of future action. Most situations that have not yet happened use an imperfective conceptual i.g. as a complement to some conjugated verb or personal gerundive. The rest tend to use an imperfective conceptual oblique as something likely to happen.

We get rid of some cognitive dissonance.

1.                  Imperfective aspect being the default verb form, not a past tense, we lose the cognitive dissonance of an “aorist imperative” which commands some future action.

2.                  The cognitive dissonance of the term “aorist infinitive” disappears by eliminating terminology that references timing; the “infinitive” is an impersonal gerundive.

3.                  Reported speech copies aspect (and modality), not tense. Reported speech and questions using a “present tense” for something that clearly is in the past of the person reporting the speech, disappears.

Implementing the mantra “context is king” eliminates some concepts.

1.                  Reflexivity and causality are features of context, not of morphology.

2.                  Likewise the “cognate accusative” is poorly defined and tries to shift nuances from context onto morphology.

The “genitive absolute” and its relatives in other cases need a thorough overhaul. The descriptions are vague and differ from one grammar to the next. The examples are non-existent, mis-quoted, mis-cited, don’t coordinate with the description – or there is no citation at all meaning that we can’t track it to a surviving text. We don’t know why writers would stick a gerundive into these positions and just sticking a label on it (that comes from Latin where there are only two examples according to the Stolz and Schmalz grammar of 1910) is useless.

As a science, linguistics should follow the rule of Occam’s Razor: as long as all the data is covered, and represented accurately, the simplest explanation is preferable. With the demonstration that timing is part of context, not of morphology, a tense description of Greek becomes a useless complication. Therefore tense is also a dead concept. Taught strictly from an aspectual viewpoint, Greek becomes much easier to learn and understand. This is not an endorsement of a Nostratic family of languages encompassing Greek and the Semitic languages. It is simply an observation of function.

The 21^st century shift is difficult because every educational institution in the world is still running on old-think. It comes from the historical authoritarianism in the field of classics, not from the textual evidence. It has to end, if linguistics is going to be a real science.

21st Century Classical Greek -- passive redux

Under what circumstances does Thucydides use passives intead of ergatives?

One of our first passives was I 2.6:

καὶ παράδειγμα τόδε τοῦ λόγου οὐκ ἐλάχιστόν ἐστι διὰ τὰς μετοικίας ἐς τὰ ἄλλα μὴ ὁμοίως αὐξηθῆναι:

When a populaton grows, there is no one agent responsible. Also, no one agent was responsible for each of the multiple possible reasons why the population increased.

I 8.1

καὶ τῶν θηκῶν ἀναιρεθεισῶν

The opening of the tombs for cleaning up Delos was done but who did it is irrelevant.

III.85

ὕστερον δὲ οἱ φεύγοντες τῶν Κερκυραίων （διεσώθησαν γὰρ αὐτῶν ἐς πεντακοσίους）

some general bunch fled for their lives, comprising about 500 people.

But then we have I 54.1

οἱ δὲ Κερκυραῖοι τά τε ναυάγια καὶ νεκροὺς ἀνείλοντο τὰ κατὰ σφᾶς ἐξενεχθέντα ὑπό τε τοῦ ῥοῦ καὶ ἀνέμου,

The agent here is inanimate. We can’t put a name to inanimate actors, even if they were the principle movers in what happened. We can’t use a fully transitive form because that would be executive voice and an inanimate agent can’t form intent for doing something. Ergative is likewise unsuitable because of how often it leans toward a deliberate action.

This is different from Biblical Hebrew. BH uses agentless verb forms in narratives when it’s irrelevant or understood who the agent is and in the latter situation it’s usually Gd. As narratives, which rely for their interest on action, BH doesn’t use agentless verb forms often. Neither does Thucydides. Scholars may use them to imply that anybody knows whatever idea they’re pushing – but then if that were true no scholar would need to write it except as a reference to a supporting source. Go back over some of the scholarly books you own and see how often this happens.

Sooo history -- Gildas and Arthur 3

Sooo Gildas in his Excidio claims the Anglo-Saxons wiped out the Celts, but we have literary evidence from shortly after Gildas’ death that he was wrong. We also have a possible record of an oral tradition, the existence as well as the features of which argue for an assimilated Celtic-Saxon culture. This is supported by modern science, which also identifies a gap in Gildas that we can’t let slide.

From archaeology, we know that a wealthy layer of people survived the Roman period. They didn’t congregate in cities and they didn’t maintain Roman outposts; they lived in villas scattered about the region. They imported luxury African Red Slip Ware through Cornwall. They had peaceful relations with the Angles for the most part. But they didn’t convert the Angles. When Augustine of Canterbury came to convert the Angles, he pushed the Welsh clergy to help and they refused. They refused to knuckle under to Roman Christian customs for a long time; the Christian culture they missionized in Europe was the Welsh rite practiced by St. Columba.

The British protected certain parts of the country, the sacred sites in the Witham Valley, the cathedral of Canterbury built in 507, and other churches and the market towns around them. But this probably only involved a set battle in a couple of cases; there’s always somebody hopped up on testosterone and adrenaline that you have to dump on his ass. When Cynric became “king” of Gwyllt, it’s probably closer to the truth that he became an ealdorman subordinate to the British, than that he conquered the region.

The main piece of modern evidence to go by is something you will never hear of in the web’s fake historical videos about the Arthurian period. DNA evidence argues against “Gildas”. In London, refounded by the Angles after Gildas’ death, nearly 50% of the male population has Celtic Y-chromosome DNA, an inheritance from Celtic forefathers, not Celtic foremothers who bore children to Angles. The percentages are higher in Wales and Ireland of course, but a value this large in this region with this history, disproves the genocide notion. Gildas couldn’t have known about DNA, of course, but he could only support the notion of the Angles and Saxons as outright enemies by living so cloistered that he had no knowledge of the peaceable relations most of them had with Britons.  

Fake historical videos will try to convince you that Gildas was writing in the time of the first and worst Krakatau eruption. One says that the “thick black cloud” in Gildas refers to the same sort of clouding as occurred in the eruption of the 1880s. This claim is a fallacy called quoting out of context. When you return those words to their context, you find a “thick black cloud of vices.” This is part of Gildas’ jeremiad, not a weather report.

What’s more, Krakatau is too far away from Britain to produce a thick black cloud. The video ignores the other catastrophe of the same period, the eruption of Ilopango in Central America. Its plume particles would have been carried directly to Britain on the jet stream as it existed at the time.[1] The same subtropical jet stream that now creates the “English Riviera” where palm trees can survive. What resulted?

Well, it wasn’t the catastrophe the fake video wants you to think. Irish missions to the continent went on for 300 years, even after the Irish adopted the Roman Easter calculation in the 700s. The British exemption from Goth, Vandal, and Hunnish immigration meant they still had copies of documents that had been destroyed on the continent, and the Irish and Scots brought them back. Missions like that require logistics – lodgings and food for the missionaries and their pack animals – and while monks have lower standards of living than most people, starvation is starvation and death.

After the eruption, Pope Gregory sent Augustine to convert the Anglo-Saxons, and his entourage needed sustenance on their 900-mile overland trek. Sure, when they reached Britain they may have found the population reduced by hunger, and by the plague brought from Ethiopia by merchants carrying that luxurious African Red Slip Ware. They probably didn’t read Gildas’ fulminations; they probably didn’t think twice about whether there had once been more people in Britain than they found upon arrival. Once the Britons refused to cooperate, the Augustinian ministry probably ignored them. The mission documented the converts about whom Gregory wanted to hear, not the Christians who already existed.

Gildas didn’t write about Augustine’s mission; he died before Augustine arrived. Augustine never made it to Lindisfarne. Anglo-Saxon Bryneich converted to Christianity in the 600s CE, and the first Christian king there lost a war to a Celtic British Christian allied to an Anglo-Saxon pagan. So much for the Anglo-Saxons wiping out the British Celts, so much for the British Celts hating the pagan Anglo-Saxons and, since the Anglo-Saxon ally had control of the Severn Valley as well as Mercia, so much for anything like a system of apartheid between Christians and pagans in Britain. Bede gives the history of the Bryneich battle; claiming the Celts were wiped out ignores his data, a fallacy called sampling bias and a failure of the Test of Occam’s Razor.

Fake historical videos are a great jumping-off point for research, but obviously you can’t trust them to present all the relevant data, even if they have it. New historical data and new insights from new tools turn up constantly; claims have to be revisited periodically, or they turn into urban legends. It’s a lot of work, but it has to be done to get at the truth of history.

---

[1] Paired oxygen isotope records reveal modern North American atmospheric dynamics during the Holocene. Zhongfang Liu, Kei Yoshimura, Gabriel J. Bowen, Nikolaus H. Buenning, Camille Risi, Jeffrey M. Welker & Fasong Yuan. Nature Communications volume 5, Article number: 3701 (2014) https://www.nature.com/articles/ncomms4701

21st Century Classical Greek -- Thucydides redux

If you are me, and you just found Herodotus using progressive aspect in ergative structures, you would now obsessively go back over Thucydides and Xenophon to see if you can catch them doing it. Why is it “if”? because languages change grammar over time. That’s why you hate reading Shakespeare in high school English literature. The grammar has changed – the vocabulary has changed even more – and you need inches of footnotes at the bottom of every page to tell you what the characters said.

And here it is, Thucydides II.65.8:

καὶ οὐκ ἤγετο μᾶλλον ὑπ᾽ αὐτοῦ ἢ αὐτὸς ἦγε,…

It’s a progressive eventive; Pericles was not led by them (the dimos) but led them himself. The verb is ago which has an imperfective intransitive eventive. It’s a case of identical forms in the progressive for use in different structures – you have to look at the structural context to see what is going on.

And the same in Cyropaedia 1.4.3:

…ἅμα μὲν διὰ τὴν παιδείαν, ὅτι ἠναγκάζετο ὑπὸ  τοῦ διδασκάλου…

It’s a case of identical forms in the progressive, which are used in different structures – you have to look at the structural context to see what is going on. This is the same thing I said very early on in this thread with, I think, afairisetai, if you can remember back that far.

The point being that you cannot determine the grammatical assignment of a word until you  examine the context and identify what role the word plays. What’s more, using “imperfect tense” in such structures has gone unrecognized (Goodwin 1250ff) because a) it does not fit the “ongoing action that is interrupted” definition on which the grammarians were fixated and b) ergativity is a new concept deriving from studies of Sumerian (1850 CEff) and Hurrian (1880CE ff), which were unknown to the old sources. Besides, c) Greek scholars isolated themselves from other linguistic studies and are only now looking outside their pipeline.

Herodotus was 20 years older than Thucydides, and grew up in Halicarnassus which had been under Persian domination for 70 years or so (Thucydides was Athenian). That’s nearly three generations and, if you read my Fact-Checking thread, you would know that the Jews of Babylonia recorded their scripture in Biblical Hebrew even as their next generation was growing up to speak Aramaic. Out of Biblical Hebrew, an aspect language with oblique structures and epistemic morphology and structures, developed Mishnaic Hebrew, a tense language which had neither. But Mishnaic Hebrew was not a hybrid with Aramaic, the way Samaritan Hebrew hybridized with Arabic (as I claim in The Real Difference). There may have been an ergative in Ancient Persian that survives in Kurdish today (Karimi) but the ergative in Classical Greek is not a sign of hybridization with Persian, quite the contrary; its source lies in NE Anatolia in the 4000s BCE.

The point of having an ergative seems to be that you can say somebody specific did something specific on purpose to bring about the ordinary results of that action (using executive voice), where the action had to be done and the person in the hupo phrase just happened to be the person who did it.

This is related to the passive, which can have the connotation that anybody in the same position would have done the same thing, or that the same action is incumbent on everybody in the same position. The latter approaches how Biblical Hebrew uses agentless verbs in both narratives and legal material, and hints at the reason scholastic material uses passives: to imply that every expert in the field agrees on a given definition or evaluation.

There is a purpose to everything in the grammar of a language, which is to communicate with those who use the language and put information across to them. Grammars that don’t teach what the language is trying to communicate, are a waste of time and money. Professors using such grammars tacitly admit that they don’t know what the language is trying to communicate. THAT is why we have to learn languages on our own.

Karimi, Yadgar. May 2012Acta Linguistica Asiatica 2(1):23 https://www.researchgate.net/publication/272651632_The_Evolution_of_Ergativity_in_Iranian_Languages

Sooo history -- Gildas and Arthur 2

Sooo Gildas was a Scots (Irish) monk on Lindisfarne in the 500s CE, writing about events around his monastery that didn’t sit right with him. What does that have to do with Arthurian legend? Here I have to get into the other well-known Arthurian sources from Britain. I can show that the latest of them has implications not just for the evolution of the tradition but also for the history of the region. And I can show that Gildas may have suffered from the same meddling as other ancient writings.

Possibly the second best-known Arthurian source after Gildas, is Bede. Gildas and Bede both lived in Bryneich; Bede was a “Geordie”, living most of his life in Tyne.  Bede refers to the Badon battle (it’s in chapter 16). Bede also refers to Ambrosius Aurelianus, and Gildas does not. Bede repeats the “destruction” claim. Bede does not refer to an ostium or the Sabrina. In many places it looks as if Bede is copying from existing manuscripts, but he does not completely support Gildas. Bede was a Saxon; “Gildas” would have been enraged to have Bede copy from him.

Bede was a doctus, according to a papal declaration, meaning he was formidably educated. Bede went on the type of travels that the Breton biography claims for Gildas, and refers to York and London in part of his history. Bede lived almost a century after Gildas died and, in a life of study, work, and travel, would have access to material the cloistered Gildas would not – but he does not refer to the five kings Gildas discusses. The five kings come from a non-Saxon, Bryneich viewpoint Bede did not have.

Next, we have Nennius of the 800s, a Welshman. It is natural for him to use Welsh place names and his naming of Arthur at Badon would also be natural. But he says Arthur personally killed 940 Saxons. This sort of exaggeration is classical for oral traditions and had as much as 300 years to develop. Nennius does not name the Severn. If, as Henry of Huntingdon supposed, his work was actually written by Gildas, it gives us more reason to suspect that Excidio suffered from an interpolation, which I will get to in a moment. But the resemblance to a record of an oral tradition, which is not present in De Excidio, doesn’t support that “Nennius” was written by Gildas.

Finally, we have the Annales Cambriae, from Norman England prior to the period of the Arthurian Vulgate Cycle (Robert de Boron and his heirs). It says that there were two battles at Badon. Morcant Bulc died at the second one. The first Badon reference names Arthur but not the Severn, any more than Bede does. Given that Gildas is the only one of these ancient sources that supposedly refers to the Severn, we can imagine some later scribe with an Arthurian background (i.e. familiar with the Annales, since Arthur is not referred to in Bede), but knowing nothing about Bryneich, “setting Gildas straight”.  

Having two battles at Badon is an example of how oral traditions grow. It can either represent the importance of the battle, or it can be a derived doublet to give more “stage time” to a character, or it can be a Law of Ascents issue by adding a casualty, and one who was well-known at that, Morcant Bulc. Welding the northeastern Morcant Bulc into a southwestern Arthurian tradition happens when the two cultures have assimilated, which takes centuries. We have centuries, from Gildas’ 570 CE death to the earliest-known manuscript record of the oral tradition in the mid-900s CE, for the Celts of the big island to remain a coherent culture after the reign of King Alfred. That doesn’t happen if the Anglo-Saxons wiped out the Celts of the big island; the Annales wouldn’t exist in their current wording. They also don’t survive a universally destructive Norman Conquest. So we have to moderate our view of the violence of the two inflows of population or we produce an impossible situation.

And if you can’t let me go with that, I have another fact for you. We know that the ancient compositions we have today may differ from the originals. For example, we know that in Josephus’ Antiquities, Book 18:65-80 is an interpolation that supports Christianity. We know it’s an interpolation because it uses grammar that Josephus doesn’t use in any of his other writing – a verb form used by Thucydides. It’s good Greek, it’s just not how Josephus writes Greek. We know that there’s an interpolation in Jewish Wars that supports Christianity; its grammar is so faulty that Josephus would have been ashamed to be responsible for it.

There is suspicion that Gildas I.23 contains an interpolation.[1] The prophecy that the Angles would rule Britain for 300 years is part of material with an Anglo viewpoint. With his hatred of Angles, Gildas would never provide favorable information. What is more, the “300 years” has two problems. If the starting point is Gildas’ period, the 300 years ends in the 900s, after the death of Anglo-Saxon Christian king Alfred.

If the starting point is the Augustinian mission, as Woolf discusses, the prophecy dates after “Gildas” died about 570, so, ending in the 800s CE. This could show when the Breton biography was written, and the Breton monastery might be responsible for such an interpolation to Gildas. But they wouldn’t have known anything about Bryneich or its five lords; the Rhuys biography doesn’t even refer to the Angles. So I doubt they were responsible for whatever interpolations there might be.

By the way, Augustine had to work alone to convert the Anglo-Saxons. The Welsh clergy wouldn’t help him. By itself this shows that Celtic culture still ruled in the big island instead of being destroyed.

Now, if we’re talking about the endpoint of Anglo domination, that came when the Norse (and Danes) took over, with their destruction of Lindisfarne. This happened in the 800s, a chronological viewpoint from which the 300 years works, but it also requires a favorable view of Anglos that the Norse would not have, including the Normans who invaded Britain later.

Woolf shows that his supposed interpolation has unusual vocabulary compared to the rest of “Gildas”. This is the same issue as Josephus. Taking out this interpolation leaves no information gaps in the surrounding text. If there’s one interpolation to Gildas, another much shorter one is possible to bring the “Sabrina” into it.

So the four main sources of Arthurianism differ in age, content, and accuracy. The features of the Annales Cambriae give them the best chance of being a record of a genuine British oral tradition, assimilating the southwestern and northeastern cultures. They support Augustine’s evidence that the Celtic substrate was in charge after the arrival of the Anglo-Saxons, and made itself felt after the reign of King Alfred. Gildas, Bede, and Nennius make a poor showing as Arthurian sources, compared to the Annales, and the latter two undercut Gildas’ reference to Sabrina – which in any case may be an interpolation. So now what about modern evidence?

---

[1] Woolf, Alex. An Interpolation In the Text of Gildas's De Excidio Britanniae 2002, Peritia https://www.academia.edu/313152/An_Interpolation_In_the_Text_of_Gildass_De_Excidio_Britanniae