Beowulf 1-227 ("Dawn of Things Ferocious") read in Early Mercian

If you like this video and want to help me make more things like it, consider making a pledge at my patreon / azforeman
Round the holidays I try to do something big and different. So here's the first 227 lines of Beowulf. "Dawn of Things Ferocious" (seemed more tonally appropriate than its competitors "Unheimlich Maneuvers" & "Grendel Stole Christmas")
The internet is already full of Beowulf-recitations. But the extant Beowulf MS is demonstrably not in the dialect of composition. So I produced a reading in something likely closer to what the original might have been like: an archaic form of Mercian, based on the orthography of the Épinal-Erfurt-Corpus glosses. (A Northumbrian origin can't be excluded, but balance of evidence suggests Mercian.)
It's unlikely that the poem was originally performed as in this recording, with a speaking voice. More probably it was chanted/sung. But I have no idea where I'd even begin working out a chanted or sung version. I'll stick with the thing I at least have some knowledge of
What you hear in this reading might be called a "reconstructive imagining" of one form of Early Mercian, from around 700 or so, toward the early end of when Beowulf might have been composed (ranging from ca. 690 to 725). The on-screen text is meant to show what a theoretical early MS might have been like, though it is itself in a subtly different type of archaic OE than my actual reading. Its orthography is based on some of the earliest Old English attested in Latin characters (that's why there's no þorn or eð letters, for example.)
I want to emphasize that this is an act of speculative imagination. This reconstruction is far more speculative than, say, my Shakespeare readings are. It's not just that (unlike Early Modern English) we have zero direct evidence for pronunciation to work with beyond sound changes, loanwords, spellings and other scribal behavior. Nor even that, compared to Middle English, the resolution we are working with in terms of dialectal diversity is severely reduced.
This reading assumes Mercian back-umlaut and only the lower half of "Mercian second fronting", like the dialect of the Omont Leaf. On the other hand, it reading retains a contrast between /ĭ ī ǣ/ in inflectional vowels.
Other retained features of the Épinal glosses include: /β/ from Proto-Germanic *b between vowels, inherited /æ/ before nasals. I decided to lexicalize the Anglian smoothing reflex of "ea" before /rC/, so it is sometimes /e/ and sometimes /æ/. Also final fricative devoicing is not complete.
There are several Mercian features here I could point out. Many are relatively minor (like "for" always being disyllabic "foræ" even when unstressed, or the preterite of weak class II verbs having /a/ instead of /u~o/). A major feature absent from the West Saxon taught to students is a /ø:/ vowel represented by œ as in "cwœn" (queen).
I made several other assumptions/choices: I realize palatalized "cg" as [ɟɟ], "sc" as [sç] and "c" as [c~cɕ]. The "g" is a stop only after /n/, a fricative/approximant elsewhere, including initially. These are archaic realizations compared to how OE is normally pronounced. It is difficult to be precise about when the sound-changes involved took place, though it's unlikely that "c" affricated before ca. 650.
There's a reason why the onscreen text is in a slightly different dialect than the recording. The unknowability is real. Don't take this as reconstructing the original dialect of Beowulf but as a general, approximate idea of the type of English proper to the period and broad dialect region in which Beowulf is thought to originate. There's great variety in the choices one could make in such a reconstruction consistent w/ the evidence for the poem's dialectal/chronological origins. We can be pretty sure that the poem at some point existed in writing in a dialect with back-umlaut, and that the Beowulf poet was sensitive to vowel-length in inflections, but we don't know to what degree second fronting was operative, let alone how smoothing affected "ea" or whether he retained /β/.
For a bit more on that see this page on my patreon:
/ 94137461
This reading relies heavily on an unpublished edition of Beowulf by Nelson Goering. To him I am also grateful for inspiring this video w/ his own Mercian retroversion of the first 11 lines, & for putting up w/ my interminable questions about OE diachrony & the Épinal-Erfurt Glosses.
This one was a lot of work. I tried to be careful but in my attempted retroversion into early Mercian, something probably escaped notice. (Edit: it did. I forgot back umlaut in "-stapa" and "clifu", and also the wrong vowel in "reht" and the reduplicative preterite in "heht" and other stuff. Oh screw it. I went and re-recorded. If somehow you're as anal-retentive as I am about diachrony, you can hear the updated recording on my patreon page.)
The translation in the video is my own

[@kaptnhansenpresidentjamaic9577]

2:15 is my favourite chapter of this reading. Even though I don’t really speak Old English, I understand enough to paint this scene in my mind. The music is an excellent choice too. “Men ne cunnon secgan to soðe, sele-rædende, hæleð under heofenum, hwa þæm hlæste onfeng.”

[@oldenglishandlyre]

I love your OE readings, I'll be listening to this over and over. Thank you for making it !

[@oldenglishandlyre]

So glad you called hiim Beow not Beowulf, he is not the main hero, Beowulf doesn't turn up til later.

[@iberius9937]

6:38 Awesome transition!!

[@kwwiedenfeld]

Anyone who really wants to understand J R R Tolkien need to listen to this

[@a.z.foreman74]

It's actually true that Tolkien had a particular fascination, even an obsession, with Old Mercian and the medieval linguistic history of the Midlands. Mercian is the dialect of Old English he used as a proxy for the language of the Rohirrim in LOTR, for example. (The "Mark" as in "Riders of the Mark" is in fact derived from the same word that yields the word "Mercia".)

[@iberius9937]

I'm noticing some peculiarities here compared to other readings I've heard of Anglo-Saxon, such as the uvular fricativization of g in "gōd", exactly like in Dutch.

[@a.z.foreman74]

I'm using a more archaic pronunciation there. We don't know precisely when the development to the more familiar modern realizations took place. Although the development of palatalized intervocalic to a glide must have been early indeed, the development of initial non-palatalized from [ɣ] to [g] must have been later than this, otherwise modern English "yard" would be pronounced "jard". But we can't say for sure one way or another whether initial [ɣ] had become [g] by the 700s.

[@mongke7858]

Masterful

[@iberius9937]

The Mercian dialect is interesting! Sounds almost like Dutch minus the retracted S.

[@siraco4278]

Hey could you explain me what a retracted s even is I have heard of the term before but I can't distinguish it from the normal s I hear in english. Dutch is my native language aswell so i apparently use it on a daily basis.

[@a.z.foreman74]

It depends on your dialect of Dutch. In dialects spoken in the southern Netherlands, and also generally in Belgium, the /s/ tends to be like the English or French /s/. Though there's a good deal of variation even there. Listen to the Wiktionary recording of the guy pronouncing the word "zo". He has a very audibly retracted /s/. en.wiktionary.org/wiki/zo#Dutch