Sauron Defeated (53 page)

'So I think Arry must be right. It is a case of linguistic coincidence or congruence, and the key is not to be found in Anglo-Saxon. We need not bother with the connexions of English Earendel in the other related languages, like the proper names Orendel, and Aurvendill, or Saxo's Horwendillus.'(5)

'But is not Auriwandalo actually recorded as a name in Langobardic?' said Markison, who has a finger in most pies of learning. 'Odd how the Langobards crop up.'

'It is,' said Lowdham.

'Hm, yes, and there is a connexion between these names and the stars,' said Jeremy. 'Didn't Thor throw Aurvendil's toe up into the sky, Arry?(6) And Earendil certainly had a connexion with a star in the strange tongue. Somehow I feel sure of that.'(7)

'Yes, that's so,' said Lowdham; 'but in the unknown language it was only a legendary connexion, not a linguistic one, I think.

Earendil meant Sea-friend.(8) I am quite sure of that, because -

well, perhaps I'd better go on where I left off.

'From the time of my father's departure...

The following passage in E / F 1 was retained in the revised typescript F 2 (p. 237 - 8) as far as 'some great tale of Numenor' almost without change, and there is no need to repeat it. The only difference between the texts is in the name of the 'cone-shaped mountain', and this is a difference very important in determining the relation of the texts of The Drowning of Anadune to those of The Notion Club Papers. Where F 2 has 'Desolate is Minul-Tarik, the Pillar of Heaven is forsaken! ' the name in E is Menelminda, changed in pencil to Meneltyula, while in F 1 it is Menel-tubel, changed to Menel-tubil.

From 'some great tale of Numenor', however, all three texts diverge among themselves, and the major divergence is between the manuscript E and the first typescript F 1. I continue now therefore with the text of E (cf. pp. 238 ff.).

'But most of the word-recollections are, as it were, casual; as casual as the words caught by the eye from a lexicon when one is looking for something else. It was a long time before I began to note them down, and use them for the language I was amusing myself by "making up". They did not fit, or rather they took control and bent that language to their own style. In fact it became difficult to tell which were my invented words and which the ghost-words; indeed I've a notion that "invention"

gradually played a smaller and smaller part. But there was always a large residue that would not work in.

'I soon found, as I got to know more, that some of the ingredients were Anglo-Saxon, and other things: which I'll mention in a minute. But when I weeded them out there was still a large amount of words left over, and in worrying over these I made a discovery. They belonged to another ghost-language, and to one that was related to the other. I could perceive a good many of the laws or rules of change: for the Numenorean style was in most points the older, more archaic, while the other had been altered (as if by contact with our western shores) to a style much more like that of the older north-western tongues.'

'I don't follow all this,' said Stainer. 'Nor do I,' said both Markison and Guildford. 'Give them some of the examples you gave me, Arry,' said Ramer.

'Well,' said Lowdham hesitating, 'if I can remember any of the examples where the relationship is clear to lay folk (it is often rather complex). Yes, lome is 'night' (but not 'darkness') and lomelinde is 'a nightingale': I feel sure of that. In the second language it is dumh, later du; and duilin. I refer them to a Primitive Western domi, domilinde. Alda means a 'tree' - it was one of the earliest certain words I got - and orne when smaller and more slender like a birch or rowan; in the second language I find galad, and orn (plural yrn): I refer them to galada, and orne (plural ornei). Sometimes the forms are more similar: the Sun and Moon, for instance, appear as Anar, Isil beside Anaur (later Anor) and Ithil. I liked first the one language and then the other in different linguistic moods,(9) but the older seemed always the more august, somehow, the more, I don't know ... liturgical, monumental: I used to call it the Elven-latin; and the other seemed more resonant with the loss and regret of these shores of exile' - he paused - 'but I don't know why I say that.'

'But why Elven-latin?' asked Markison.

'I don't quite know,' said Lowdham. 'I certainly don't mean Elves in any of the more debased post-Shakespearean sort of ways. Actually the language is associated in my mind with the name Eresse': an island, I think. I often call it Eressean.(10) But it is also associated with names like Eldar, Eldalie which seem to refer to, well, something like Ramer's Enkeladim.'(11)

'That breaks my dream!' cried Jeremy.(12) 'Of course! Now I know. It wasn't a library. It was a folder containing a manuscript, on a high shelf in Whitburn's second-hand room,(13) that funny dark place where all sorts of unsaleable things drift. No wonder my dreams were full of dust and anxiety! It must have been fifteen years ago since I found the thing there: Quenta Eldalien, being the History of the Elves, by John Arthurson (14)-

in a manuscript much as I've described it. I took an eager but hasty glance. But I had no time to spare that day, and I could find no one in the shop to answer any enquiries, so I hurried off.

I meant to come back, but I didn't, not for almost a fortnight.

And - then the manuscript had vanished! They had no record of it, and neither old Whitburn nor anyone else there remembered ever seeing any such thing. I recall now what a catastrophe it seemed to me at the time; but I was very busy with other work, and soon forgot all about it.'

'It certainly looks as if more than one mind had been working back along similar lines,' said Ramer. 'Several minds indeed; for our expert is at fault for once. Lewis also mentions the name somewhere.'

'So he does!' cried Jeremy. 'In a preface, was it? But he was quoting from someone, I think, from a source that hasn't been traced. And he used the form numinor. All the other sources have numenor, or numenore' - that's so, isn't it, Arry?'(15)

'Yes,' said Lowdham. 'nume is West, and nore is kindred, or land. The ancient English was Westfolde, Hesperia.(16) But you wanted to know why Elven. Well, I got that from another line, too. You remember I mentioned that Anglo-Saxon used to come through mixed up with this other queer stuff? Well, I got hold of Anglo-Saxon through the ordinary books, of course, fairly early, and that confused the issue; though some words and names came through to me that are not in the dictionaries...'

From here to the end of Night 66 the version in the original manuscript E is very close to the final form (pp. 242 - 5), though some elements are lacking, notably Lowdham's description of the ancient slowness and sonorousness of diction (p. 242): following Frankley's

'Unless, of course, you back up their theories' Lowdham goes on: 'As a matter of fact, I think they do. At least, here is a bit that came through very early, long before I could interpret it; and it has been repeated over and over again in various forms:

Westra lage wegas rehtas wraithas nu isti...'(17)

The Old English lines beginning Monad modes lust are in later spelling, but have the same form as that in F 2 (see pp. 243-4 and note 50, and p. 272). There is no reference in E to the date of the

'coming through' of these lines, nor to its being an evening of high wind.

The remarkable feature of this original version is of course that Lowdham's two 'ghost-languages' were Quenya and Sindarin (or rather, the language that would come to be called Sindarin). Lowdham's account in this version thus maintains the linguistic experience of Alboin Errol in The Lost Road (cf. note 9): 'Eressean as he called it as a boy ... was getting pretty complete. He had a lot of Beleriandic, too, and was beginning to understand it, and its relation to Eressean'

(V.45).

The first typescript version F 1 follows the manuscript E at the beginning of the section just given ('But most of these word-recollections ...', p. 302), in Lowdham's description of how the

'ghost-words' 'soon took control and bent my [invented] language to their own style'; but when he comes to tell that as he sifted the 'large residue [of words] that would not work in' he made a discovery, his discovery is totally different from that in the original text. This is where Adunaic first appeared. It may be that my father had been long cogitating this new language; but even if this is so, it would seem that it had not reached a form sufficiently developed to enter as Lowdham's 'second language' in manuscript E. In fact, I doubt that it is so.

It seems to me to be overwhelmingly probable that Adunaic actually arose at this time (see further p. 147).

I give here the text of F 1 from this point (corresponding to the E

text on p. 302 and the final text F 2 on p. 238).

'I found, when I got to know more, that some of the

ingredients were Anglo-Saxon and other related things: I'll deal with that in a minute; it was not a large part. Working over the rest, collecting and sifting it, I made a discovery. I had got two ghost-languages: Numenorean A and B. Most of what I had got at an early period was B; later A became more frequent, but B

remained the most common language, especially in anything like connected passages; A was chiefly limited to single words and names, though I think that a lot of it is incorporated in my invented language.

'As far as I could or can see, these languages are unrelated, though they have some words in common. But in addition to these tongues there remains a residue, and I now see that it consists of some echoes of other later tongues that are later than Numenorean A and B, but are derived from them or from their blending. I can discern some of the laws or lines of change that they show. For the Numenorean tongues, I feel, are archaic and of an elder world, but the others are altered and belong to Middle-earth.'

'I don't follow all this,' said Stainer. Most of us felt the same, and said so.

'Couldn't you give them some of the examples that you gave to me, Arry?' said Ramer. 'Some of the important names, and a word or two; it would be clearer with something definite to go on.'

Lowdham hesitated. 'I'll try,' he said. 'But I shan't be able to give many examples of the later changed forms; the relations would seldom be clear, even to philologists, without many instances side by side in writing.

'Well, take the name Numenor or Numenore. That belongs to language A. It means Westernesse, and is composed of nume

"west" and nore "folk" or "country"; but the B name is Anadun, and the people are called Adunai. And the land had another name: in A Andore, and in B Athanati; and both mean land of gift>. There seems no connexion between the two languages here; but in both menel means "the heavens". It occurs in the B name Menel-tubil that I mentioned just now.

And there seems to be some connexion between the A word Valar, which appears to mean something like "gods", and the B

plural Avaloi and the place-name Avalloni.

'The name Earendil, by the way, belongs to language A, and contains eare "the open sea" and the stem ndil "love, devotion".

The corresponding B name is Pharazir, made of pharaz and the stem iri- [changed in ink on the typescript to: Azrubel, made of azar "sea" and the stem bel-]. A large number of the names seem to have double forms like this, almost as if one people spoke two languages. If that is so, I suppose the situation could be paralleled by the use of, say, Chinese in Japan, or indeed of Latin in Europe. As if a man could be called Godwin and also Theophilus or Amadeus. But even so, two different peoples must come into the story somewhere.

'I don't know if you want any more examples; but the words for the Sun and Moon in A are Anar and Isil (or in their oldest form Anar and Ithil); and in B they are Uri and Nilu. These words survive in not much changed shapes in the later languages that I spoke of: Anor (Anaur) and Ithil, beside Uir, Yr and Nil, Njul. Again the A and B forms seem unconnected; but there is a word that often occurs and is nearly the same in both: lome in A, and lomi in B. That means "night", but as it comes through to me I feel that it has no evil connotations; it is a word of peace and beauty and has none of the associations of fear or groping that, say, "dark" has for us. For the evil sense I do not know the A word. In B and its derivatives there are many words or stems, such as dolgu, ugru, nulu.

'Well, there you are. I hope you are not all bored. I love these languages. I call them Avallonian and Adunaic.(18) I find first the one and then the other more attractive, in different linguistic moods; but A, the Avallonian, is the more beautiful, with the simpler and more euphonious phonetic style. And it seems to me the more august, somehow, the more ancient, and, well, sacred and liturgical. I used to call it the Elven-latin. But the Adunaic is more resonant with the loss and regret of Middle-earth, these shores of exile.' He paused, as if he heard echoes from a great distance. 'But I do not know why I say that,' he ended.