Tolkien's Old English

[Lord of the Rings]

Have there been any studies that anyone knows about looking into Tolkien's use of Old English? I mean in general, and not just with the Rohirrim - he wrote quite a lot of prose and verse that's been reproduced in The History of Middle-earthand elsewhere (e.g. S&G), and I'm looking for anything that's analysed it, especially in terms of his use of dialect and archaism.

[halfir]

<DIV marginheight="1" marginwidth="1" topmargin="1" leftmargin="1" ="WebWizRTE">Other than the work done by Tom Shippey in The Road to Middle -earth and his Beorhnoth essay in Roots and Branches- both of which I am sure you know I can only think of three essays - Drout, Holmes, and Bolintineanu (who also has written the OE entry in the JRR Tolkien Encyclopedia) in Jane Chance's edt. Tolkien and the Invention of Myth in the section on Old English - but I don't think they address in any great detail your points about dialect and archaism.However, they may be of some use and interest to you.Edited by: halfir

[Olme]

There needs to be an edition of all of Tolkien's Old English.



I think there's an article in _Tolkien 2005_ that talks about his poetry in Old Germanic languages... and obviously the largest portion of that poetry is in OE.

[Dorwiniondil]

There needs to be an edition of all of Tolkien's Old English.

Indeed there does. Do I hear anyone offering to do it?

[Olme]

By the way, I heartily approve of the new sig, LotR!

[Olme]

The writing I was thinking of was this:



"Tolkien's writings in Old Germanic languages"<br style="color: rgb0, 0, 102; font-family: Georgia, 'Times New Roman', Times, serif; font-size: 14px; line-height: normal; text-align: left; : rgb252, 252, 232; ">Maria Artamanova

[Lord of the Rings]

Thanks for the reference, Olme! I'll have to take a look.



An edition of his OE - that would be great, though I wonder just how big a project that would be. We have a fair amount that's been reproduced already in one place or another, and I wonder how much more there is floating about unpublished.

[Olme]

Yeah... kind of like saying it's a comprehensive edition just to have something new come out (or finding out just how much there is). I have a feeling there could be quite a bit more out there.

[Troelsfo]

Not Old English, of course, but despite my utter ignorance of Gothic, I actually enjoyed Lucas Annear's article in Tolkien Studies VIIItitled‘Language in Tolkien's “Bagme Bloma”’ (Bagme Blomabeing an alliterative poem in Gothic that Tolkien published in the collection Songs for the Philologistswhich was privately published in 1936)— both the linguistic part (where of course I have to take Annear's word at face value) and the analysis.

[Olme]

Glad to year you enjoyed the essay! I hope it made the poem a bit more accessible. (and LotR played no small role in boosting the quality of that article).



-Olme (Lucas Annear)

[Troelsfo]

I guess there's no point in my trying to summarize it here, then



As I said, I know no Gothic at all (for some reason it's been completely overlooked in the physics curriculum at the University of Copenhagen— I can't figure why … it's a grave oversight ), but it's precisely for this reason that I was so impressed with the article:I remember that it was with some self-conscious hesitation that I began reading, but that was soon overcome, andI was amazed how much I could get out of it, and in particular how well I could follow even the specific linguistic discussion. So, full credits for making the poem accessible to the linguistic layman!

(EDIT:) Realizing that the needs of the linguist or philologists are very different from those of the layman, I think that something along the same lines about Tolkien's compositions in Old English would be very useful. Discussions about how he used the language as well as what he used it for might lead to an improved understanding of why he used it in these situations. What did Tolkien feel or believe that he achieved better by writing in Old English than he could by writing in a more modern language?

Edited by: Troelsfo

[Olme]

(EDIT:) Realizing that the needs of the linguist or philologists are very different from those of the layman, I think that something along the same lines about Tolkien's compositions in Old English would be very useful. Discussions about howhe used the language as well as whathe used it for might lead to an improved understanding of whyhe used it in these situations. What did Tolkien feel or believe that he achieved better by writing in Old English than he could by writing in a more modern language?

Well, I don't see why we couldn't go ahead and do exactly that here! Granted, I'm sure there are a lot of the people outside of the Plaza who would like to read something like this but won't since they dont' post/read here (hence the usefulness of an edition). The nice thing about forums is that it takes so little time to get feedback and work through ideas.
To take a look at OE specifically in context of the Legendarium, we could start sort of from the beginning in HoMe. Would that be appealing to folks here? I can't remember the first instance of OE in HoMe though, and I'm away from my books at the moment.
Also, both me and LotR I'm sure would be happy to discuss Gothic whenever you'd like, as we're both pretty big fans

[Olme]

Also, Troelsfo, at the University of Wisconsin Tolkien Society meeting a couple of nights ago I saw your name in an article in Mythlore as we passed it around. I knew I recognized the name but couldn't remember where from. Well, the Plaza of course!

[Lord of the Rings]

I'm away from a fair portion of my books now, including the BoLT (though I do have copies of HoME 5-9 here), but if memory serves we do have at least some Old English there. If nothing else, there's the interspersing of words like Wealas, 'Britons'in English passages - this might 'just' be a means to evoke the world of the Anglo-Saxons by using older forms of ethnic terms rather than a more modern word like 'Welsh'. A detailed look at that could be interesting.
What I was particularly interested in looking at when starting this thread was the variety of OE Tolkien used. OE has come down to us in several dialects - the most common is West Saxon (which we can further differentiate into Early and Late varieties). This is common not just because people wrote original compositions in the dialect, but a lot of works were copied under the West Saxon kings and were often 'translated' into that dialect. A very large number of works show a mixture of dialect features because of things like this,Béowulfincluded.
But there are other dialects preserved in a more ore less distinct form, namely Kentish and the Anglian dialects Mercian and Northumbrian. Of these the most relevant for Tolkien is surely Mercian, most classically preserved in a text known as the Vespasian Psalter, which he used a fair amount (for the Rohirrim, for example: inFerthu Théoden hál, feris a specifically Mercian form (cf Campbell s. 744)). The use of Mercian forms has obvious connections with Tolkien's sense of home, since Mercian was the form of OE spoken in the regions of England Tolkien most closely identified with.
Still, Tolkien doesn't justwrite in Mercian, andhe also uses 'normal' West Saxon forms a fair amount, especially in chronicles. And sometimes he seems to be attempting to give specifically older forms of Old English, using variation of age rather than dialect. Sometimes a given text, such as these linesfrom Dorwiniondil's All in the Booksriddle that prompted this thread, appears in different forms. A good hard look at what he uses when and why might be illuminating.
&lt;Edit: the reference to 'Campbell' is to Old English Grammarby A. Campbell, revised 1968, the standard Anglophone reference grammar.&gt;



Edited by: Lord of the Rings

[Troelsfo]

Well, if you will undertake such a study here, I'll be reading along all the way, cheering you on— hopefully asking some clever questions at times


I do think there's some bits of Old English in The Book of Lost Tales, but I don't recall any substantial compositions: some names and titles, perhaps a few lines of poem. The first substantial composition in Old English is, as far as I remember, the Old English versions of the Annals of Valinor and the Annals of Beleriand in The Shaping of Middle-earth.

[Olme]

I think you're right. Once I'm done with these two exams today, I'll pull it off the shelf!

[Troelsfo]

Now, I've wanted a good excuse to go back to the ideas of this thread again, and with the Old English in the Notion Club Papers in fresh memory, I also came across this delightful little thing and thought that a bit of humour would be just as good an excuse as any other (and yes, despite being a man of the natural sciences, I didn't need to check a translation, but of course the OE cynerīce is not so far from the corresponding Danish word).

[Dorwiniondil]

[Lord of the Rings]

(A few grammatical mistakes, but hilarious nonetheless )

[Almarëa Mordollwen]

*is late to the party, but still finds this highly amusing*That's probably a pretty accurate picture of what I look like at the moment ...

[Olme]

Love it!

[Troelsfo]

And as a service from Lingwë a list of “reconstructed lexis in Tolkien's Middle English vocabulary”:

http://lingwe.blogspot.dk/2012/11/re...ns-middle.html

For those who are not on top of their linguistic lingo (as I am not), this means all the words that Tolkien had reconstructed for the Middle English vocabulary that he wrote for Kenneth Sisam’s collection of Fourteenth Century Verse and Prose;the asterisk-forms as many Tolkien fans will be familiar with from, among other places, Tom Shippey's books.

[Tyrhael]

I am not sure this is relevant, but LotR mentioned "a lot of prose and verse that's been reproduced in The History of Middle-earth and elsewhere" ... I can provide a bit more of "elsewhere," though as there's no analysis I don't know if it's what he's looking for.

I've been checking out PE20, and it has a few bits of Tolkien's Old English written in the Qenya Alphabet (one an original line, another an excerpt from Beowulf with "a few minor deviations from the received text"). I don't know if these are of any interest, but I shall provide them just for the purpose of data.

On p. 78, Arden Smith transliterates JRRT's swē bogas & spraҟan elebēama (normalized as Swē bogas & sprancan elebēama), which they translate as "Like (boughs and branches) of olive-trees," believing it to be a reference to Psalm 128:3.

On p. 70, Smith transcribes JRRT's Beowulf excerpt (rendered phonetically) as:
ƕæt wē gārdena on ġeārdaȝum
þeōdkyniḡa þrym ġefruȝnon
hū ða æðəliḡas ellən fremədon.
Oft scyld scēfiḡ sceaðena þrēatum
manəgum [?mǣġðum >] mǣȝ̇ðum medosetla' oftāχ
eȝzode eorlas, oþþæt him ǣȝƕylc

Smith notes that "After rendering Tolkien's version in normal Old English spelling, a few minor deviations from the received text may be noted: on for in (l. 1), gefrugnon for gefrūnon (l. 2), manegum for monegum (l. 5), medosetla' for meodosetla (l. 5), and oftāh for oftēah (l. 5). More importantly, however, the final line of Tolkien's text jumps from line 6a to line 9a." I do not know whether this indicates that Tolkien changed the rendition towards "his" Old English, if it is supposed to represent a dialectal variant, or if there is another explanation. Regardless, here's the data.

[Lord of the Rings]

I haven't gotten around to ordering PE20 yet, but it looks like it would definitely be worthwhile - that Béowulf excerpt is really fascinating!

I'm not sure entirely what Tolkien was doing with all the changes. Substituting on for in seems a bit peculiar, since in was the normal Mercian form of that preposition. Using on makes the text more West Saxon: that is, more like 'classical' Old English, but less like Tolkien's favoured dialect, and less like the dialects Béowulf was likely composed in (it was almost certainly either Mercian or Northumbrian, despite some modern doubts on this matter). Though once I'm back home I'll double check a few things, since it's possible that temporal in/on was a bit different from the spatial adposition. Other changes, like ġefruᵹnon for gefrúnon are just writing the older form of a word (or what's ambiguously an older form or a dialectal variant, like medo- for meodo-).

Tolkien's oftāχ is interesting, and seems to be a textual emendation using an entirely different verb from what the manuscript has. The MS oftéah is the past tense of a verb oftéon, a contracted form of older *oftéohan, meaning something like 'took X away from Y'. As far as I can tell, Tolkien's oftáh can't be a form of this word in any dialect, and seems instead to be the past tense of a quite different verb, also appearing as oftéon in the present, but contracted from *oftíhan. This oftéon² has a slightly more legalistic meaning 'proceed successfully against'. This would mean something more like 'Scyld Scéfing . . . successfully went against bands of foes, against many kindreds of mead-halls'. I'm not entirely sure why Tolkien would make this change though - the MS reading works fine, and Tolkien's verb creates some syntactic problems with the cases of the nouns in that phrase. Still, very interesting, and probably worth looking into more. I wonder if anyone else has ever proposed a similar emendation.

The skipping around in the last line is really strange. On the one hand, it might reflect Tolkien's ideas about the history of the poem, cutting out (or moving to later?) the rather sudden reference to Scyld Scéfing's fosterage, first finishing up the account of his exploits as king. But that breaks the metre - oðþæt him ǽghwylc is of a metrical type that properly speaking should only occur in the first half a line, never the second half (type A3, in the usual classification). Anyone with a fairly reasonable grasp of the metre should be able to see it doesn't scan (it's pretty jarring, actually), so I'm a bit confused why he wrote it this way.

[Tyrhael]

Well, that's all the Old English that's in PE20 (most of them are either English or Latin), though as far as Germanic goes there is one excerpt of Old High German in the Qenya Alphabet; "the text is an excerpt from the Liber euangeliorum (Evangelienbuch), a ninth-century Gospel harmony in verse, written in the South Rhemish Franconian dialect of Old High German by Otfrid von Weissenburg. The passage in question is located at 7. 1, 5, lines 3-12" (108-9). I could type up Smith's transliteration along with the "Braune/Ebbinghaus edition" version if you're interested; other than that, however, I can't say there's any more Germanic content; PE20 is more about Tolkien's use of a tengwar-precursor, with transliterations / analyses of the texts he wrote in it, much like how the Alphabet of Rúmil texts are presented in PE13 (which has some Middle English sarati in it, i.e. lines from Hali Meidenhad and Pearl). There are 40 documents analyzed in PE20; Smith summarizes the types as "Tables and descriptions of the alphabet [...] Samples of verse are very common: excerpts from "The Adventures of Tom Bombadil" [...] "Errantry" [...] Lewis Carroll's "The Walrus and the Carpenter" [...] Beowulf [...] Otfrid's Evangelienbuch [...] and various nursery rhymes and folk songs [...] Prayers, hymns, and other elements of the Roman Catholic Mass also appear frequently [...] Transcriptions of letters to friends and associates appear [...] Other text types include a draft of a university statute [...], a fictional narrative [...], a diary entry [...], philosophical musings [...], and a political observation. Most of the texts in the corpus are written in English. Latin is the next most common language, [...]" with mention of the two Old English and one Old High German.

I'm not sure if it's relevant, but the swē bogas one is from July 9, 1931, and the Beowulf excerpt is likely from 1931 as well.

I have some thoughts/hunches about Germanic content in other PEs, but those'll have to wait until another day (it's past 2 a.m.).

[Lord of the Rings]

I've finally decided to have a proper think about the dialects of Old English Tolkien uses, and especially about Shippey's claim that the Rohirrim speak a specifically Mercian dialect of Old English. I think in the end that he's right, but for the wrong reasons, and the Old English Tolkien puts in the mouths of the Eorlingas is generally archaic rather than specifically dialectal, with only one obviously Mercian form.

Shippey makes his observation in a footnote:

Not many have noted that [the names and tongue of the Rohirrim] are not in the 'standard' or 'classical' West Saxon dialect of Old English but in what is thought to have been its Mercian parallel: so Saruman, Hasufel, Herugrim for 'standard' Searuman, Heasufel, Heorugrim, and cp. Mearc and *Marc. In Letters, p. 65, Tolkien threatens to speak nothing but 'Old Mercian'.
-The Road to Middle-earth, p. 123, note

'Old Mercian' was one of the four main dialects of Old English that we have preserved in our records, the others being Northumbrian, Kentish, and the 'classical' West Saxon. Shippey's Mercian proposal would make a lot of sense on a number of levels. The Anglo-Saxon kingdom of Mercia covered most of the areas of England that Tolkien felt a particularly strong attachment to (including both his childhood home and his university), and the Middle English dialects he was most concerned with were outgrowths of Mercian Old English. Béowulf seems to have a few Mercian elements, and many have suggested that it was composed somewhere in Mercia. So if Tolkien were to choose any one dialect of Old English for use in his fiction, Mercian would be an unsurprising choice.

Jason Fisher accepted Shippey's comments in his essay 'Horns of the Dawn: The Tradition of Alliterative Verse in Rohan', and added a further suggestion that Tolkien drew on an Old English text known as the Vespasian Psalter as the best example of Old Mercian that has survived (p. 15). Both Shippey's and Fisher's suggestions are interesting, but neither really hold up under scrutiny. (Btw, I haven't read all of Fisher's essay - I don't have his book, and neither does my library. So I've only read as far as the 'preview' on Google Books allows, through page 16. This seems to cover the main discussion of dialect, but it's possible he expands on relevant issues in the endnotes.)

Shippey's first three examples of 'Mercian' words used by the Rohirrim - Saruman, Hasufel, and Herugrim - differ from their 'classical' counterparts by a single sound change known as 'back mutation'. This term encompasses both the change of *a > ea and *e > eo in these words. The problem is that back mutation actually operated more​ prevalently in Mercian than in any other dialect! In fact, and contrary what Shippey wrote, Hasufel would be the 'classical' West Saxon form, while Heasufel would belong to the Mercian dialect; much the same goes for Saruman. In all dialects, Heorugrim would have been standard, not Tolkien's Herugrim.

So these forms don't demonstrate that Tolkien was writing in Old Mercian; actually Herugrim seems odd for any dialect, and so offer us a bit of a puzzle. They certainly don't have anything to do with the Vespasian Psalter, where back mutation is rampant.

The answer to the puzzle is that back mutation isn't a marker of dialect, but of age. It's a relatively late sound change, probably occurring only after the Anglo-Saxons had started writing down some texts (it's hard to give a specific date, unfortunately). So the very oldest Old English didn't have back mutation, while the later dialects did. Forms like Saruman don't tell us that 'Rohirric' was any particular dialect of OE, but they do tell us that it was old. 'Old' meaning maybe the Old English of 650-750 AD, basically the Age of Bede or a bit earlier, though it's a bit hard to date these things precisely.

What about Shippey's *Marc for West Saxon Mearc? I'm actually not sure that we should take Tolkien's use of 'Mark' as genuinely representing a 'Rohirric' word *Marc - but if we do, it would have to belong to the Northumbrian dialect. Since other 'Rohirric' words are demonstrably non-Northumbrian, this would be a bit odd. I think they would have called their country the *Merc, in all likelihood, with Mark simply being the Modern English form of this word. There's an early modern English sound change of *er > ar, seen (for example) in Modern fart from Middle English fert, Old Mercian *fertan - so Old Mercian Merc would correspond precisely to Modern Mark. Anyway, if 'Mark' even gets us anywhere at all, it's not to Mercian Rohirrim. (Actually Shippey's probably right about this after all - see next post.)

Looking at all the 'Rohirric' words and names in The Lord of the Rings that I could find (85 words or phrases - I wrote them in a spreadsheet, which means I can sound fancy by giving precise numbers ), there are a few more helpful clues to identifying the dialect. Words like Erkenbrand, Elfwine, and Dernhelm show various linguistic features that rule out Kentish and West Saxon - but they aren't decisive between the two remaining dialects, Mercian and Northumbrian. (The features in question are a bit complicated to go through, particularly in Elf- and Dern- where multiple changes have happened in succession - but if anyone wants details I can talk about them.).

The question of how to decide between Mercian and Northumbrian is a bit more difficult. These two dialects are fairly similar, and are sometimes referred to together as 'Anglian' Old English. Incidentally, Modern English is a descendant of various Anglian sub-dialects; the 'classical' West Saxon fell out of use after the Norman Conquest, and represents something of a side-branch in the history of English.

In deciding between the two, I can only find one really reliable word, mearas, which seems to point to Mercian. There's not really anything distinctively Mercian about it, except that it is not Northumbrian; in that dialect, the word would have been *maras. (Specifically, the Primitive Old English word *mærhas evolved into mearas in most dialects, but in Northumbrian it underwent a special sound change, called 'retraction', to become first *marhas and later, with loss of *h, maras.). It would be nice to have more than this one word to go off of, but it seems conclusive as far as it goes. This would mean that Shippey was entirely right to identify the dialect as Mercian, but not on the basis of the words he gave as evidence for that.

There is one other word that could also be distinctively not just Mercian, but a specific sub-dialect of Mercian: fer, in the phrase Ferthu Théoden hál 'fare thou, Théoden, well'. The imperative fer is actually attested in the Vespasian Psalter, where it is derived from an older form *fær by a sound changed known as 'second fronting'. As I said above, Tolkien did not simply take the dialect of the Vespasian Psalter wholesale, but we might (if all we had was this one word) think that he was writing in a hypothetical precursor to Vespasian Mercian: one in which second fronting had already occurred, but before the date of back mutation.

But this doesn't seem to be the case, since there are a few 'Rohirric' words that ought to show second fronting if it had applied, but don't. The names Haleth and Fastred are is particularly telling. Tolkien doesn't use the symbol æ in The Lord of the Rings, but these clearly represent Old English hæleþ and fæst-red. In the Vespasian dialect, they it would be heleth and festred. (Tolkien's avoidance of the letter æ is unhelpful here - it makes it hard to tell whether things like Hasufel are meant to be general Mercian Hasufel, or pre-Vespasian-Mercian Hæsufel, with second fronting.) So based on haleth fastred being used instead of heleth festred, it seems clear that Tolkien wasn't using a (pre-)Vespasian dialect, or at least not consistently. And since there are other possible explanations* for ferthu as well, it's probably best to conclude that this isn't what Tolkien was going for at all. (Edit: Again, a rather silly error on my part: second fronting did not apply before *l, so haleth is not useful evidence. Fastred still seems diagnostic though. I can only plead that I made the mistake of staying up much too late writing this post, and didn't do the sensible thing of reading it over in the morning before postin.)

*(Specifically, in the collocation *færthu 'fare thou', the two words were joined very closely in Old English, a bit like the blurring of words in Modern English phrases like I'd've. In Old English, the sound sequence *-ær- develops differently depending on whether a consonant immediately follows in the word or not. If *færthu was mushed together enough to count as 'one word' for the sound changes in question, then ferthu would be the normal development in all Mercian dialects, not just the Vespasian dialect with second fronting. This is nonsense. I clearly should've gone to bed by this point: the form ferthu can only be a pre-Vespasian form, not a general Mercian one. I'll make a fuller post at some point to try to look into this; for now, I'll just mention that a mixture of forms with and without second fronting among the Rohirrim mirrors the forms with and without second fronting in the Épinal and Efurt Glossaries. Note that the forms like haleth show unambiguously non-second fronted vowels, so there doesn't seem to be a 'pure' dialect at work here. As I mentioned, a lot of forms are ambiguous, since Tolkien's 'hasufel' could equally be hæsufel or hasufel.)

But even if Tolkien wasn't following the Vespasian Psalter, as Jason Fisher suggested, it doesn't mean he wasn't getting his cues from particular Old English texts. Something very much like the archaic Mercian he puts in the mouths of the Rohirrim is in fact attested in a couple of very old texts: the Erfurt and Épinal Glossaries. These texts are rather boring to anyone but a philologist, being basically Latin-Old English dictionaries, but the dialect the Old English words are in seems to be a very archaic form of Mercian, displaying pretty much all the early Mercian sound changes that 'Rohirric' does, but (like 'Rohirric') pre-dating the change of 'back mutation'. Any of the 'Rohirric' words I've taken a look at could appear in these glossaries and not be out of place.

So I think Fisher was right to look at specific texts, but it's probably the archaic Mercian glossaries that Tolkien turned to, rather than the more classically Mercian Vespasian Psalter​.

(I hope this isn't overly obscure! I tried to avoid too many technical terms, but it seemed like things got even more confusing if I didn't at least refer to sound changes like 'back mutation' and 'second fronting' by their names. Unfortunately dialectology is inevitably a little bit technical, and there were a lot of possibilities to consider, and I'm not always great presenting these sorts of things intelligably.)

[Lord of the Rings]

I did a little more digging into the draft material for Rohirric names, and I found out a little more about 'Mark' - it looks like Shippey was probably right that Tolkien intended the Rohirrim to call their own country 'Marc', not 'Merc' as I suggested. In the earlier drafts of 'The Battle of the Pelennor Fields' Tolkien went through a couple of versions of the alliterative poem that closes the chapter, both of which include the 'Marculf' as one of the fallen Eorlingas. This is clearly 'Mark-wolf, Wolf of the Mark', and shows clearly that Tolkien envisioned the endonym of Rohan to be either *Marc or *Mærc.

If the language of the Rohirrim is basically the language of the early glossaries, then it would actually make sense for their country to be the *Mærc in their own tongue. The older common Anglian form would have been *Mearc, which was first 'smoothed' to give *Mærc, and then 'raised' to the typical Mercian form *Merc. This intermediate stage, between smoothing and raising, is what the early glossaries generally represent, though their language is perfectly consistent. So if Tolkien was following the early glossaries and extrapolating, *Mærc would indeed be the right form to use (even though only the earlier or non-Mercian mearc and the later Mercian merc are directly attested in Old English). In Tolkien's spelling conventions in The Lord of the Rings, early Old Mercian Mærc would be written Mark or optionally Marc.

[Lord of the Rings]

Looking at the 'point' of dialect investigation, I think there's actually a fair bit to say. Shippey's conclusion that Tolkien was using Mercian instead of the 'classical' dialect (the one you'll learn in any teaching grammar of Old English) is interesting, since it shows Tolkien's concern with the language of his local region, and ties in nicely with Middle-earth's intimate connections to the Marches of England.

But this regional connection could have been emphasized more by using the dialect of the Vespasian Psalter, as Jason Fisher suggested. This sub-dialect of Mercian was very quintessentially representative of that region, and was basically the precursor to the 'AB' dialect of Middle English that Tolkien felt a very strong connection to. His choice to use a more archaic version of Mercian in a sense detracts from the regional connection, at least in a direct linguistic sense.

But it helps reinforce other connections, which do ultimately lead back to Mercia. The most prominent one is to Béowulf. Like with 'Rohirric', the text sample quoted by Tyrhael from PE20 uses a very archaic form of Old English to render the opening lines of the poem. So a form like medosetla in line 5 shows a form of Old English predating back mutation - the manuscript has a later, mutated meodosetla. This clearly reflects Tolkien's supposition that Béowulf was composed in the 'Age of Bede', the first flourishing of Christianized Anglo-Saxon culture from (very roughly) 700-750 AD. Much of the PE20 sample shows Tolkien making the language more archaic to fit this time period.

I don't think it's coincidence that the 'Rohirric' speak a version of Mercian also basically appropriate to the first half of the eighth century. As Shippey has pointed out many times, there are innumerable points of detail in the Rohan chapters that tie in to Béowulf, from the niceties of gaining an audience with a king, to the paved road and golden roof, to the very phrases used. It's been recognized that the Rohirrim are perhaps not literal Anglo-Saxons in historical detail, but that they are very much Anglo-Saxons as they imagined themselves in their own poetic traditions. One might say that the Rohirrim are not literal Old Mercians in historical detail, but that they are indeed Old Mercians as they imagined themselves in their own poetry, in their own speech.

(This is made somewhat more complicated by the presence of the form sceaþena in the Béowulf sample in PE20, which seems to be a Northumbrian form. But I'm not sure how significant this is, and I don't think it should be taken as evidence that Tolkien favoured a Northumbrian origin for the poem. After all, sceaþena is just a retention of the manuscript spelling that Tolkien failed to 'Mercianize', so it might just be a slip.)

The archaic-Mercian/alliterative connection is something that Tolkien strongly emphasized. His student S.R.T.O. d'Ardenne, certainly reflecting views Tolkien would have sympathized with, commented that 'the alliterative tradition, and much of the vocabulary that goes with it, is in general associated with the West', and draws a connection between Mercian OE, 'AB' Middle English, and the alliterative tradition (Þe Liflade ant te Passiun of Seinte Iuliene, p. 179). And Tolkien draws a very strong connection between Mercian, alliterative verse, and Béowulf specfically in The Notion Club Papers. There is one bit in that story where Ælfwine alludes to Mercian poetry: 'There are others in the hall. Men of the Marches I hear by their speech; and they were used to boast of their songcraft, before the Danes came' (HoME IX, p. 273). Tréowine, one of these 'Marchers' (i.e. Mercians), goes on to recite a bit of verse that is evidently supposed to be in an archaic form of his dialect, since it was hard for the people of Wessex listening in to understand him. Tréowine's verses are very clearly modelled on Béowulf:

Hwæt! wé on geárdagum of Gársecge
fyrn gefrugnon of feorwegum
to Longbeardna londgemǽrum
tha hí ǽr héoldon, íglond micel
on North-théodum, nacan bundenne
scírtimbredne scríthan gangan . . .

It's interesting to compare this with the PE20 version of the opening of Béowulf - Tolkien makes both of these versions archaic relative to the manuscript of Béowulf in the same way, for example PE20 and Tréowine both using gefrugnon instead of the later form gefrúnon found in the manuscript.

I think all these things together point to a persistent interest on Tolkien's part in the ancient dialect of Mercia, which he regarded as the golden age of English alliterative verse and the original home of Béowulf. The Rohirrim are drawn into that, not simply as speakers of Mercian OE, but as speakers of archaic Mercian, and with their customs, architecture, and values as well as their language being drawn primarily from Béowulf, the crown jewel of old Mercian poetry.

[Dorwiniondil]

This is very interesting. It has led me to look again at variants of Cædmon's Hymn:
http://www.heorot.dk/bede-caedmon.html#appendix-1

A couple of Northumbrian ones, plus a variety of West Saxon, from different MSS ....

[Lord of the Rings]

Cædmon's Hymn is a great text for comparing dialect variants (the Lord's Prayer is also useful). It's too bad we don't have a proper Mercian version, though if we did I think it would probably look something like this (in the language of the Vespasian Psalter):

Nú sculun hergan heofenríces weard
meotudes mæhte and his módgeþonc
werc wuldurfeadur, swé he wundra gihwes,
éce dryhten, ór astalde
He ǽrest scóp ælda bearnum
heofen tó hrófe, hálig sceppend.
Þá Middangeard moncynnes weard
éce dryhten efter téode
fírum foldu fréa ælmihtig.

I think it's especially telling that Tolkien didn't use this Vespasian-dialect in The Lord of the Rings, since he did use it for one of the longest bits of Old English in The Notion Club Papers. The Tengwar text that Ramer got 'old Rashbold' to translate is explicitly said to be 'in Old English of a strongly Mercian (West-Midland) colour, ninth century' (p. 257), and in fact the text itself does follow the language of the Vespasian Psalter very closely (though Tolkien actually possibly makes his text just a tad later, or else more West Saxon, in one or two details).

[Olme]

This thread has really had some great developments. Very insightful discussion, LotR. Though I've reread a lot of Shippey over the last few years, you've shown that I really ought to go all the way through works like Road again with a much more learned and critical eye.

I happen to be piecing my way through Beowulf and the Critics right now where Drout talks about Tolkien's almost casual dating of Beowulf. In Tolkienien fashion, and as you've made pretty clear with dating the language of the Rohirrim, it's not too surprising that as we did into things (in this case Rohirric) we find that Tolkien is making statements (however direct, indirect, or serious) about real-world issues with his fiction. In this case, having the language of the Rohirrim date to the first half of the 8th century, as you suggested.

[Lord of the Rings]

it's not too surprising that as we did into things (in this case Rohirric) we find that Tolkien is making statements (however direct, indirect, or serious) about real-world issues with his fiction.

I think this is especially true when his fiction touches on things like Béowulf and the 'mythology for Anglo-Saxon England' that Drout has talked about. Tolkien thought about those issues very deeply, and touched on them in all sorts of ways.

Incidentally, I hope I haven't come across as too hard on Shippey - he's a really great scholar, and he knows Old English and Old Norse literature inside and out. But I do think some of claims about language should be poked a bit more. I just realized I haven't really said a good word about him in this thread, even though he's usually very insightful about these things.


I've corrected a comment in one of my earlier posts, where I rather absurdly claimed that Éowyn's phrase Ferthu Théoden hál might not reflect the sound change of 'second fronting' in the vowel of fer. It does - there is, I believe, simply no other way to get that vowel except through second fronting. This is an interesting contrast with Fastred (not ˣFestred), which definitely doesn't show second fronting. (Haleth actually isn't a good example, since second fronting doesn't take place before l.)

This kind of mixture of forms, without any obvious phonological conditioning, is actually fairly characteristic of the two glossaries I mentioned before as possible models, the Épinal and Erfurt. So 'hawthorn' in Épinal is hæguthorn, but haguðorn in Erfurt (19); Erfurt also shows second fronting elsewhere, such as hæƀuc (497) for non-fronted hafoc (I've normalized slightly, though not in any consequential way; the parenthetical numbers are to the relevant lines of the glossaries). Alistair Campbell characterizes the change of second fronting as 'fairly common' in the early glossaries (Old English Grammar, s. 168), which might apply well enough to 'Rohirric' as well (in our tiny sample of two diagnostic forms, one has second fronting, and the other doesn't).

[Olme]

Incidentally, I hope I haven't come across as too hard on Shippey - he's a really great scholar, and he knows Old English and Old Norse literature inside and out. But I do think some of claims about language should be poked a bit more. I just realized I haven't really said a good word about him in this thread, even though he's usually very insightful about these things.

Well, I think it goes without saying that any criticism of Shippey is is preceded by, said or unsaid, a profound respect for the guy.

That being said, I think any scholar should be happy to see people "poking at their claims". So long as the goal is furthering our knowledge of a given field, I think they'll only be happy to see advances being made, as you've done here.

[Lord of the Rings]

Here's a fairly 'rough and ready' annotated list of the Old English that I could find in The Shaping of Middle-earth, The Lost Road, and The Notion Club Papers. Corrections and additions welcome. Some of the comments reflect things already brought up in this thread, but I wanted this list stand on its own. I've only got the repaginated Del Rey edition of HoMe IV, so if anyone felt like supplying standard page numbers for that section, that might be helpful for reference.


The Shaping of Middle-earth

The Quenta, Appendix 1

Pennas – Part of a translation of the Quenta into Old English. 64 lines, basically West Saxon, but note especially elfe for ielfe.

'Lists' – Collections of names from the Legendarium given with Old English equivalents. Shows a mixture of dialect forms along the lines often seen in poetic texts. At least one word, eldre for yldre, seems Mercian (or at least Anglian) beyond the norm for poetry.

'Geneological Tables' – More names, which appear to mostly be West Saxon.


The Earliest Annals of Valinor, Appendix

Version I – Appears to be straightforward West Saxon (note especially ielfe).

Version II – Generally West Saxon, but with several more Anglian characteristics than Version I.

Version III – As Christopher Tolkien notes, this version is (with a few exceptional forms) in the 9^th century Mercian dialect represented in the Vespasian Psalter.

Version IV – A very short text in West Saxon.


The Earliest Annals of Beleriand, Appendix

Beleriandes Géargesægen – Seems to be of a piece with OE Version I from AV, nearly exclusively West Saxon in form.


The Lost Road and Other Writings

The Lost Road

Chapter II, p. 43: Westra lage wegas rehtas, nu isti sa wraithas – Not Old English, but a reconstructed form of early Germanic.

Chapter II, p. 44: Thus cwæth Ælfwine Wídlást . . . – Alliterative verse, fairly typical West Sexon.

Unwritten chapters, p. 84: Monað modes lust mid mereflode . . . – Basically West Saxon, like the language of the original poem The Seafarer that this verse is modelled on (that is West Saxon, with a few Anglian features as usual in poetry).

Note on 'The Song of Ælfwine', p. 103: Fela bið on Westwegum uncúðra . . . – Same as p. 44, but in a generally later form linguistically.


Sauron Defeated

The Notion Club Papers, Part II

Night 66, p. 242, miscellanious names – Thes first several names don't show features distinctive to one dialect or another. The 'archaic' forms hebaensuil and frumaeldi show spellings and sound changes appropriate to the archaic Mercian of the Épinal and Erfurt Glossaries. The 'very antique indeed' Wihawinia would appear to be reconstructed early Germanic.

Night 66, p. 243: westra lage wegas rehtas, wraikwas nu isti – Reconstructed early Germanic.

– westweg wæs rihtweg, wóh is núþa – Old English of indeterminate dialect (not Vespasian).

– Monath módaes lust mith meriflóda . . . – Archaic Mercian, in dialect and spelling similar to the language of the Épinal and Erfurt Glossaries. Much earlier than 'Lichfield Mercian'.

Night 66, p. 244: Þus cwæð Ælfwine Wídlást Éadwines sunu . . . – Apparently generally West Saxon, like the version in The Lost Road, p. 44, but with the distinctly Mercian form ælfa (contrasting with the phonologically similar eldo, which along with elfa is a southern form, though not the 'classical' West Saxon ieldo, ielfa).

Night 68, p. 257f. – As Rashbold writes, 'To sum up: it is in Old English of a strongly Mercian (West-Midland) colour, ninth century I should say.' The language is very similar to that of the Vespasian Psalter, which is what one might call 'classical Mercian', or 'Lichfield Mercian' (after the important centre of power and learning in the region of Birmingham). The few differences, such as gewarhton, are mainly explicable as either being slightly later (-on for older -un), or else as showing dialectal changes later reflected in the 'AB' dialect of Middle English (-arh- for -orh-). See Þe Liflade ant te Passiun of Seinte Iuliene, ed. S.R.T.O. D'Ardenne, p. 142 (wurchen) and 169f. (warpen).

Night 69, p. 268: Uton efstan nú, Tréowine! . . . – In good early West Saxon; see comments for the conversation on p. 271.

Night 69, p. 271: Westu hál, Ælfwine! . . . – The conversation between king Éadweard and Ælfwine is in the early 10^th century 'Alfredian' Wessex dialect appropriate to the time and place being depicted.

– Hwæt! Éadweard cyning Ælfredes sunu . . . – Seems to be straightforward West Saxon poetry.

Night 69, p. 272: Monath módes lust mid mereflóde . . . – Although the poem has been changed slightly, linguistically the text remains the same as the version from The Lost Road, p. 84.

Night 69, p. 273: Hwæt! Wé on geárdagum of Gársecge . . . – Tréowine, who recites these lines, is said to be marked as a 'Marcher' through his speech, and this poem was 'dark to some of our younger men of Wessex'. The few lines quoted have perhaps a couple of relatively archaic features, but there is nothing that would point particularly either to or away from Mercian – the sorts of words that would be diagnostic just don't appear.

Night 69, p. 277 – Béo thu blíthe æt thisse béorthege! – Not obviously dialect; presumably West Saxon.

Fragments, p. 277: Swéte is blóstma brǽþ begeondan sǽ – West Saxon (see notes to p. 271).

Earlier versions of Edwin Lowdham's Old English text, p. 313ff. – The earlier versions are all in a relatively late West Saxon dialect.

[Dorwiniondil]

Ah yes. Memories of a hot, stuffy lecture theatre in Loughborough ...

[toms]

Olme, this is just to say that you are absolutely right: no scholar SHOULD object to having his claims poked a bit, especially if they're wrong. I've sent a considerably longer reply to Mandos on the Marhwini thread, with an apology for making a hash of the comment on Libram-Moreno in the Fisher Introduction. Was it age, weariness, or drink? All three, quite likely. But I am not familiar with the Plaze system yet, so refrain fom more extended comment on the 'Mercian' topic till I am sure these posts are getting through. (Tom Shippey)

[Tyrhael]

I quite agree, Olme - this thread has had some great developments indeed. :) LotR, you asked for "additions," and I happen to recall some more Old English in The Treason of Isengard, ch. XXVI., i.e. Aragorn and Legolas' "strongly 'Beowulfian' reception" starting Abidath cuman uncuthe! Hwaet sindon ge [and so on]; see notes 5 and 10 as well. I imagine there's more OE content in TI in the form of earlier names of places and people (like Eodoras and Winseld) as well, but I confess I'd prefer to re-read LotR for the first time in several years (after I finish my similar reread of The Hobbit first) before delving too deeply into Tolkien's earlier drafts of that tale, so I don't plan to go through making a list just yet.

Other OE I can think of at the moment is Ælfwine's usage (via Tolkien) of ælflæden for Quenya and ælfgeþéode for Beleriandic Sindarin in PE19, in addition to "(iv) In ky the k was fronted and dentalized (as in ñy > ny). The resultant sound is here transcribed ty. It was probably closely similar to t in English tune. Ælfwine said that the sound as heard in Tol Eressea was like that of the c in English cild and ceaf" (73), regarding which C. Gilson notes "On the pronunciation of OE palatal c: 'All that we know for certain is that OE. had a guttural and a palatal k, that the former was sometimes written k and the latter always c, and that the two k-sounds had separate characters in the OE. runic alphabet.' Joseph Wright, Old English Grammar, 1925, p. 162" (ibid).

I seem also to remember an OE equivalent of the Noldorin Lheben teil brann i annon ar neledh neledhi gar godrebh, but I can't seem to find where I put my copy of A&I, so I'll have to edit this in a few hours. I also seem to recall David Salo 'correcting' Tolkien's usage of OE for that passage, but not where - probably one of the mailing lists. I'll look after supper and edit this post. I'm curious whether you'd think his criticism justified or not.

Edit: The OE for Thrór's map is 'Fif fóta heah is se duru ond þrie mæg samod [?] þurhgangend' (A&I:150), and David Salo's comments are Elfling 00221. Carl Hostetter also commented on it in TolkLang 7.60, seven years prior.

[Lord of the Rings]

On the OE for Thrór's map, I wonder if a re-examination of the original would be worthwhile. Even in A&I you can see that there's something small written above/after samod, which would be interesting if it could be read more clearly. And the map is reproduced quite a lot more clearly in The Art of The Hobbit, p. 51, where there seems to be a distinct graph of some sort at the end of mæg. It looks like just one letter, but I wonder (given how much Tolkien sometimes trailed off at the ends of words) if it was just a squiggle meant to indicate en - that would make the verb mægen, a correct subjunctive plural. Even if it just reads mæg or mæge (both singular forms), this wouldn't be completely awful, since OE sometimes has a bit of fluctuation in agreement with entities viewed as collectives, as I suppose a bare numeral might be (though a plural verb would be much better).

The lack of an infinitive is not really an issue as such - auxiliary verbs can often omit the dependent infinitive of verbs of motion. It's maybe a little strange to add in a present participle (it's not a past participle like Hostetter said), but it could be read basically as 'three should be able to [go] together walking through'. But the form þurhgangend is still not entirely correct, since the inflection (whether singular or plural) should be þurhgangende. Or perhaps what he wrote (since the last vowel seems rather a-like to me, at least) was þurhgangand, that is, a perfectly acceptable infinitive þurhgangan with an extra d on the end (perhaps because of a momentary intrusion of the participial nd in his mind).

So the thing does have the appearance of being written hastily, and with some mistakes. Whether these mistakes are purely graphical (truncation of -en and -e, or accidental extra -d), and what Tolkien intended linguistically was correct, or whether he wrote exactly what he intended but made morphological/syntactic mistakes, I'm not sure.

To sum up, it seems to me to literally read:

fíf fóta heah is se duru ond
þrie mægẹ¹ samod ^[xxxxx] þurhgangạnd<sup>2

1</sup>Or possibly mægṇ
²Or possibly þurhgangend

'Five of-feet high is the door and
three should-be-able together through-go[ing?]'

This can't be entirely correct. It could be made grammatical by adjusting the last word to either þurhgangende or þurhgangan (the second one being a more normal OE construction, but requiring a more unusual error in Tolkien's writing). It could be improved stylistically by reading mægṇ̣/mægẹ̣ (the dots underneath represent letters that, at least looking at the reproductions, could be reasonably read more than one way) as mægen​, which seems perfectly reasonable to me given how reduced the final squiggle is. I have no idea what the small word after samod says.

Edit: Tolkien also either got the gender of duru wrong or again truncated the end of a word (extremely so - it looks pretty clearly like se to me), since the article before it should be feminine seo, not masculine se.

[Findegil]

And the map is reproduced quite a lot more clearly in The Art of The Hobbit, p. 51, where there seems to be a distinct graph of some sort at the end of mæg. It looks like just one letter, but I wonder (given how much Tolkien sometimes trailed off at the ends of words) if it was just a squiggle meant to indicate en - that would make the verb mægen, a correct subjunctive plural. Even if it just reads mæg or mæge (both singular forms), this wouldn't be completely awful, since OE sometimes has a bit of fluctuation in agreement with entities viewed as collectives, as I suppose a bare numeral might be (though a plural verb would be much better).

We have the image file from which the reproduction on p. 51 of The Art of The Hobbit was made, so are able to enlarge and attempt to enhance in Photoshop the Old English inscription. The mark following mæg appears to be no more than a flick of Tolkien's pencil as he moved his hand to the right, i.e. not even a partially formed letter or combination of letters.

It's conceivable that the word (and it seems to be a word) written very small above samod could be mægen, but that would be no more than a guess, and maybe also a result of the power of suggestion. Unfortunately, the image can be enlarged only so much before pixelating.

Or perhaps what he wrote (since the last vowel seems rather a-like to me, at least) was þurhgangand

This is entirely possible, and we would now (seeing the inscription at enlargement) tend to agree that the final vowel is most likely a rather than e.

So the thing does have the appearance of being written hastily, and with some mistakes.

The inscription was indeed written in haste, probably with little thought, certainly not for publication. And one can make only so much of a scribble.

It's worth adding that although we did our independent transcription of the Old English from the original map when writing Artist and Illustrator, before including a note about it we consulted Rhona Beare's analysis in Parma Eldalamberon 6, and were in touch with Carl Hostetter, who with Pat Wynne corrected some of Dr. Beare's readings. Dr. Beare read þurhgangend thus, as did Carl, Pat, and Arden Smith (with whom we were in contact as well), though they too felt it was dubious. We also wrote at that time to Christopher Tolkien, who likewise questioned various forms, and who himself couldn't make out the stray words of the inscription, in the reproduction of the map he had at hand.

Wayne & Christina

[Lord of the Rings]

The mark following mæg appears to be no more than a flick of Tolkien's pencil as he moved his hand to the right, i.e. not even a partially formed letter or combination of letters.

Ah, there goes one idea then. Thanks for taking a closer look!

And one can make only so much of a scribble.

Very true. Though given that Tolkien's Old English was often very good, even when writing quickly (as in some of the HoMe IV Annal writings), it seemed worth it to not dismiss this as ungrammatical too readily. But there's no point in making too much of a hasty jotting either.

[Troelsfo]

I have only now found the time to catch up with the developments in this thread since New Year — and how exiting it has been!

One of the things that always strike me when I read about Old English is how my knowledge of modern Germanic languages (my native Danish and some Swedish, German and Norwegian in addition to English) helps me understand things better, but also how this can confuse an attempt to understand things better. Well, if I ever get rich enough to stop working for a few years, I'll be back in university, but this time to study Germanic philology

All this mainly just to say that though I really don't have anything to add, I do read along and learn a lot, so keep up the good work! (Oh, and LotR, don't worry about speaking of back-mutation and fronting: you make the meaning quite easily understandable in context, and of we are to learn about this it is, in my opinion, far better to have the correct concepts introduced from the start).