For Phil. Be well and keep on archiving!

‘Yes,’ said Gandalf, ‘these doors are probably governed by words… These doors have no key. In the days of Durin they were not secret.’

On the Doors of Durin are various signs: the emblems of Durin, the tree of the high elves, and the star of the house of Fëanor. Above the door are inscribed two lines of “interlacing elvish letters,” the smaller bottom line tells that Celebrimbor wrote them (and, I surmise, drew also the signs on the door). From The Silmarillion we learn that Celebrimbor was the grandson of Fëanor. 

The interlacing elvish letters are the signs of Fëanor: he it was who invented them and his name is given to them – as the English words below the drawing say, the elvish words are written in “the Fëanorian characters.” In Appendix E we read the name of this elvish language – Sindarin – but learn, primarily, of what Fëanor named the Tengwar and we variously call  marks, signs, characters, and letters of the alphabet.

So the riddle above the gateway – Speak, friend, and enter – is an elvish riddle, written in Sindarin, the sounds of which are, on the page of The Lord of the Rings above, recorded twice: in the Tengwar in the drawing and (in a calligraphic hand) in the Roman (or English) letters in the box below the drawing.

The Tengwar of Fëanor are thus comparable to our ABC: a system of visual signs that record the sounds of a spoken language.

The far-reaching meaning of an alphabet contrives to shape the riddle inscribed above the door, giving an additional twist to the losing of the password ‘in translation’ into the common tongue (Westron, the modern English of the story).

The elvish letters are translated speak, friend, and enter. Once upon a time, in the Second Age, a literate traveler who reached the cliff face on the border of Hollin spoke the words that revealed the door and the signs above it and then spoken as instructed by those signs and entered.

On the level of story, the historical context of the scene in which the Ring-bearer stands without the Doors of Durin and Gandalf the wizard reads the signs on the door, the riddle of this door turns on the meaning of friend —

— The One Ring is without the Doors of Durin, the dwarf-door with elvish signs without inscribed by the elf who made the Three; hidden signs that ancient words make visible, inscribed on this dwarf door by Celebrimbor, grandson of Fëanor, Ring-maker, friend of Sauron the Fair.

Yet the riddle above the door turns in the first instance on both translation and our unnoticed modern private – silent – reading. Tolkien draws on the gulf between silent reading and spoken language to draw out the chasm that stands between us (and the Company) in history, with Hollin a long-deserted and newly ominous landscape where once in a Second Age of Myth it was happening.

One who would enter (minno) must read mellon (friend) aright (aloud, untranslated), which is just what is instructed in the first word of the formal riddle: pedo (speak!).

The riddle as encountered in the story is only a riddle to silent readers – or overly sophisticated translators. We learn what the inscription says through Gandalf’s translation, i.e. we learn that the inscription includes the (English) words Speak friend and enter before we hear the original sounds that constitute the word ‘friend’ in Sindarin, mellon. But if  Gandalf had simply stood before the Doors of Durin and read the inscription aloud in the original Sindarin then, presumably, the doors would have opened.

The Hobbit was written to be read aloud. The sequel was composed after The Hobbit was published, and was written for (largely) silent readers. The riddle takes on what I can only call a semiotic flavour because the signs on the door tells us (in the cryptic way that a riddle does) to abandon our usual reading practice and speak!

(In Bree, Frodo Baggins seems to read Gandalf’s letter silently.)


Actually, this is not the first encounter with Fëanorian letters. Tolkien seems to have invented the Tengwar when writing The Hobbit and the first known usage of Fëanor’s letters is on the jar in the bottom left in this illustration (note the additional Anglo-Saxon runes below that indicate Thror and Thrain).

Jim Allan (Introduction to Elvish, p. 246) reads these partially obscured letters as English: gold th— Thrain: accursed be the thief. (I guess the obscured word starting with th is Thror).

On the title page of The Lord of the Rings we find, at the top, elvish runes – akin to those we find on Balin’s tomb and in the Book found beside it in the Chamber of Mazarbul – and below are the Fëanorian letters (the whole making a long title comparable to that made by the Anglo-Saxon runes on the cover of the first edition of The Hobbit).

And in Bag-end, when Bilbo’s magic ring is cast into the fire the fiery letters of the Ring-verse are revealed in the same script, of which Gandalf says that while the language is that of Mordor, the letters “are Elvish, of an ancient mode” (FR 66).

In Rivendell Gandalf quotes Isildur, who describes the letters on the Ring as “an elven-script of Eregion” (FR 329) – Eregion, it will be recalled, is the ancient name of Hollin, on the border of which stands the western gate of Moria.

Wait! What are the letters of Fëanor doing on the One Ring? Surely this is a queer mixture of sub-creative art and evil. Well, a first answer is given by what Gandalf says of the Palantíri: “Fëanor himself, maybe, wrought them… But there is nothing that Sauron cannot turn to evil uses” (TT 779-80). This is not quite the whole story, though, and I will return later to the incongruity between the lust for power embodied in the Ring and the beautiful calligraphy of the fiery letters inscribed by Sauron on its surface.

What makes the Fëanorian letters particularly interesting is that they are not a fantasy but the real thing – Tolkien attributes to Fëanor his own invention. So, in contrast to the Silmarils that Fëanor later made, the Rings of Power made by his grandson, and the Palantíri that Gandalf says he “maybe” made, we are not here dealing with a fantasy of art “delivered from many of its human limitations” (as Tolkien put it in his famous letter to Milton Waldman). Here we have an instance of ‘real elvish magic’ – or, more properly speaking, elvish sub-creative art – that is within our mortal reach. And I will argue in the conclusion that by studying these elvish letters we gain insight into Tolkien’s conception of the elvish craft that made the jewels, the rings, and the stones.

Fëanor is said in Tolkien’s later writings to have made the letters when he was young, “before the days of his discontent” (SilmarillionMorgothWar of the Jewels, Part IV; Vinyar Tengwar 39). This biographical detail may simply reflect Tolkien’s thought that the linguistic study that lay behind the invention of the letters was not compatible with all the turbulence that arose after Fëanor turned his mind to the Silmarils. But in the conclusion I will suggest that Tolkien came to regard Fëanor’s study of language and invention of letters as a sort of apprenticeship that provided a step toward the making of the Silmarils (and, maybe, also the Seeing Stones).

But we can only begin by turning to Appendix E, where we are told that the Fëanorian letters were not in origin an ‘alphabet,’ which latter Tolkien describes as “a haphazard series of letters,” each with an independent value of its own, recited in a traditional order that has no reference either to the shapes or the functions of the individual letters. A footnote explains that, of the English ABC, the Eldar would have found intelligible the relation between the signs P and B but not their separation in the traditional sequence, nor that of F, M, V.

What we are reading here is Professor Tolkien the philologist and it is worthwhile (if rarely attempted) situating what he says in the context of his professional studies. So, in order to appreciate the significance that he attached to the Fëanorian letters I will introduce two texts that Tolkien must certainly have known in a professional capacity.

Isaac Taylor was one of the greatest comparative philologists of the generation before Tolkien (he was a friend of the Oxford Assyriologist A.H. Sayce). In his magnum opus on the alphabet he first explains how the materials that allow the history of the alphabet to be established have only recently become available and then traces all the known alphabets of the world back to the ancient Egyptians who, in contrast to the Babylonians, the Assyrians, the Medes and the Japanese, advanced beyond syllabaries (visual symbols of syllabic sounds) by way of the key step of the conception of the consonant:

Easy as it seems to ourselves, who are familiar with it, the notion of a consonant, a sound that cannot be sounded except in conjunction with some other sound, different from itself, is by no means so simple as it may appear. It involves the decomposition of the syllable into its ultimate phonetic elements – the mental isolation, for instance, of the unpronounceable sound t, which is common to the articulations tea, tie, toc, and two, and yet is not identical with any of them. (I, 62)

Fëanor’s invention of letters begins from this understanding. He is reported to have said:

… words may be analyzed into their tengwi, but I would say rather that they have one or more chambers, and the vowel is the room in each, and the consonants are the walls. One may live in a space without walls, but not in walls with no space: kt is only a noise, hardly audible in normal speech, but ket may have significance. Our fathers therefore in building words took the vowels and parted them with the consonants as walls… (Vinyar Tengwar 39: Note 3)

A question arises from this analogy, namely what exactly is it that dwells within the ‘house’ that is a word? Well, obviously, meaning – but what this is or what the relationship between house and dweller might be is mysterious, to say the least. Here Tolkien found the art of Ring-making, or so I’ll suggest in the conclusion.

In Taylor’s history, the ancient Egyptian alphabet is older than the pyramids but the ancient Egyptians themselves never employed it properly. The “first true alphabet,” he says, was the ancient Semitic alphabet that was derived from it. Only now were all the old hieroglyphs and syllabics discarded and a writing system established that relied solely “on one single sign for the notation of each consonantal sound” (156).

Incidentally, Taylor, who was an Anglican clergyman, seems to be suggesting – though he never puts it like this in his book – that it was the ancient Israelites who came out of Egypt under Moses who made the step to a “true alphabet” – thereby providing the technological basis for the word of God to be recorded in the books of the Hebrew Bible. (Taylor’s friend Sayce, another Rev., was meanwhile putting the comparative method to work teasing out the relationship between the books of the Bible of the ancient Israelites and the much older civilizations of Babylon and Egypt on the kingdom’s northern and southern borders).

But Taylor insists that the Semitic alphabet was not a “full alphabet” because it did not deal with the vowel sounds (only some centuries later was a system of diacritical dots devised for this purpose). This final step, he says, was taken by Indo-European language-speakers, like the Greeks, who adapted the Semitic alphabet of the Phoenicians but made visual signs also for the vowels (157). The very name alphabet provides a clue to the origin of our own letters – for it is made of the first two letters of the Hebrew alphabet, aleph and bet.

(The English language has been written in two alphabets both derived from the ancient Egyptian by way of an ancient Semitic alphabet. The Romans adapted the alphabet of the Greeks, which they had derived from the (Semitic) Phoenicians, and the letters you are reading now are Roman. But before the English ever came to Britain the alphabet of some more southern people had been adapted in the North to derive the various ancient runic scripts – which look as they do because they are designed to be scratched on stone and wood. But this leads us to Balin’s tomb and Tolkien’s pivotal decision at the very close of the seminal ‘fourth’ (1939) phase of writing to substitute for the inscription to the Lord of Moria a second invented (runic) alphabet in place of the Anglo-Saxon runes of the The Hobbit).

Fëanor’s original letters are like the early Semitic alphabet in that the letters are only consonants. But Tolkien not only attributes to Fëanor a diacritic method of indicating vowels, as we have already seen, after completing The Lord of the Rings he went out of his way to explain that Fëanor held that vowels were each independent tengwi, or word-building elements, though different in function from the consonants. In other words, Fëanor devised what Taylor calls the “full alphabet” and Tolkien the “full writing,” with vowels that were subsequently used for languages such as Sindarin, in which the diacritic method of indicating vowels was inconvenient (Vinyar Tengwar 39). It is such “full writing” (in Sindarin) that we find above the Doors of Durin.

Placing Fëanor’s Tengwar in the context of our own mortal history illuminates why Tolkien attributed the invention of the elvish letters to the same craftsman who made the Silmarils. Taylor introduces his book with the observation:

… if we set aside the still more wonderful invention of speech, the discovery of the Alphabet may fairly be accounted the most difficult as well as the most fruitful of all the past achievement of the human intellect. It has been at once the triumph, the instrument, and the register of the progress of our race. (I, 2-3)

The enormity of this step is indicated on the elvish side by the fact that Tolkien insists that, while the Tengwar of Fëanor “were largely a new invention,” nevertheless they were preceded by, and “owed something to the letters of Rúmil” (RK 1467, Appendix E). On the other hand, Fëanor’s Tengwar are superior to even the “full alphabets” in use in our own day.

From Taylor we learn that the invention of the alphabet entails the analysis of the sounds of everyday speech (which we routinely hear and understand without attending to the actual sounds themselves). With the rise of comparative philology in the nineteenth century, however, systematic study of the sounds of a language (phonology) saw the classification of the consonants according to the physiological conditions of their production (position of tongue, shape of mouth, breath) and established various laws of the relationship between these sounds and their changes in time.

In his later writings, Tolkien situates Fëanor’s invention of the Tengwar in the context of the study of language by the elvish loremasters and suggests that Fëanor himself learned much of the (very different) language of the Valar. Commenting on the TengwarAllan (Introduction, p. 241) infers that Fëanor:

noticed that the sounds of a language – its phonemes – tend to be different from one another in systematic ways and so can readily be described in terms of their relationship with one another. He wished to indicate this clearly by devising a writing system in which the characters varied in shape on a systematic basis, and so the sound relationships in the spoken tongue could be mirrored by visual relationships in the written language.

In other words, Fëanor’s achievement goes beyond the “full alphabet,” the history of which is told by Taylor, for it rested upon an understanding of the spoken sounds of language in general (rather than of a single language), and as such is on a level with the experimental scripts of the professional linguists of Tolkien’s own day.

This is in part why, in Appendix E, we are told that  the Fëanorian letters were not in origin an ‘alphabet.’ The scare quotes are vital to the meaning – elsewhere Tolkien describes the Tengwar as an “alphabetic system” (WJ, 396). Part of what Tolkien means is indicated by his subsequent disparagement of the English ABC as “haphazard.” This is the perspective of the scientific phonologist who recognizes the traditional alphabets as shot through with the accidents of history, such that the order and shape of the letters are essentially random conventions.

This is what Tolkien has in mind when he says, in Appendix E, that the elves would have found intelligible the relation between the signs P and B, but not their separation in the alphabetical arrangement. What he is pointing to is the similarity in visual form (vertical line and – one or two – bows) and the respective sounds and their physical conditions: both P and B denote sounds made by the blocking of the vocal tract (so that airflow ceases) by an occlusion made by the lips. In our ABC this “intelligible” relation is an accidental product of history.

Like various nineteenth- and twentieth-century linguists, Fëanor invented an alphabetical system in which the shape and arrangement of the letters has some systematic principle behind it.

The Tengwar are not an ‘alphabet’ in part because they are systematic rather than haphazard, but also because they are intended to provide, not a single alphabet but a variety of different alphabets tailored to individual languages. The signs above may be “adapted at choice or convenience to represent the consonants of languages observed (or derived) by the Eldar. None of the letters had in itself a fixed value.” (Appendix E)

You can read about the different applications in Appendix E (and many sites online). Here it is sufficient to illustrate the way the relationship between sound and visual sign work (and here I follow Allan’s Introduction, p. 239):

General principles:

(i) Bow to the right, opening downward: sounds made in the front of the mouth – t, p, d, f.

(ii) Bow to the left, opening upward: sounds made in the back of the mouth – ch/k, j/g, k/w, g/gw.

Particular applications:

In both Sindarin and Westron a single bow: voiceless sounds (t, f, k, kw) and a double bow: voiced sounds (d, v, gw).

But in Quenya a single bow: non-nasal sounds (t, f, r) and a double bow: nasal sounds (nd, mp, n).

So to repeat, Fëanor’s alphabetical system is envisaged as a version, not so much of the historical alphabets the genealogy of which Taylor traces to ancient Egypt, but of the invented alphabets of Victorian and Edwardian students of language. For purposes of illustration we may briefly compare the Tengwar with the system of signs devised by Alexander Melville Bell and revised by the great Oxford philologist Henry Sweet:

Henry Sweet: ‘Sound Notation’ (1880), in Collected Papers of Henry Sweet, arranged by H.C. Wyld, Oxford, Clarendon Press, 1913, pp. 285-343.

Sweet’s essay on ‘Sound Notation,’ published three years before Taylor’s two volumes on the alphabet, begins:

The problem of sound notation is as old as civilization itself, but it is only of late years that that of scientific sound notation has become urgent. There is now a general conviction among philologists of the necessity of a general alphabet, but with utter discord of opinion as to the means of attaining it. (285)

Sweet makes similar complaints against the Roman (i.e. English) alphabet as do the elves – fundamentally, he says, the visual symbols are arbitrary: its “elementary symbols have no definite relations either to one another or the sounds they represent” (289). He argues that the Roman letters should be put aside by professional phonologists and “an entirely new organic alphabet” introduced, in which the letters all “stand in a definite relation to one another and to the sounds they represent” (290).

To answer this need, Sweet proceeds to give a revised version of the phonetic symbols developed by Alexander Melville Bell (whose son invented the telephone), which represents the positions of the speech organs in the articulation of sound (Bell had developed this system in his 1867 Visible Speech to help the deaf learn to speak). Sweet reproduces Bell’s system before proceeding to his revisions.

Being unable to reproduce these signs on my computer I give also a reproduction of the beginning of Sweet’s exposition.

And here are some sentences, written in Sweet’s revised version of Bell’s system, with a phonetic equivalent in Roman letters and the conventional English spelling below.

I’m not suggesting that Bell’s system (or Sweet’s revisions) are somehow a precursor of Fëanor’s Tengwar.

Actually, it would be desirable to compare the two systems, but I am not linguistically capable of doing so. (Comments welcome!) My intuition, for what it is worth, is that the Tengwar work on a deeper level and their superiority of principle is manifested in the much smaller number of signs (but what are we to make of the additional letters – to what extent do they point to a deficiency of Fëanor’s invention?)

All I’m claiming here, however, is the two following points of comparison:

Firstly, the two are horses from out of the same stable, that is, both are products of a sophisticated modern (i.e. Victorian) analysis of the sounds of speech and the idea that an alternative system of signs could ‘better’ represent these sounds than the “haphazard” and “arbitrary” signs of traditional alphabets.

But secondly, Tolkien’s intent is fundamentally different to Sweet’s. I’m not talking about mechanics – the meta-alphabet that can be applied to different languages (Tengwar) versus the single phonetic alphabet that can capture all sign systems promoted by Sweet. I mean that Sweet begins by invoking the name of science and applies principles of economy to his knowledge of phonetics. Tolkien, by contrast, employs his knowledge of phonetics in the name of art:

I devised the tengwar for a special purpose; and they are based as much on knowledge of the art of writing as on my moderate knowledge of phonetics… The tengwar were developed to fit a (fictional) historical situation, and their primary purpose is calligraphic. (Personal communication to Laurence J. Krieg, quoted in Introduction 263).

Fëanor was a craftsman and not a scientist, and for Tolkien ‘craft’ entails an aesthetic dimension absent in modern science and machine production. Tolkien’s development of the Tengwar is in fact analogous to his invention of the elvish languages. Tolkien was by no means the only philologist or linguist of his day who invented languages.  From 1900 onward professional philologists, like Otto Jespersen, also invented new languages with the idea of creating an international language of scientific communication. But where Jespersen was serving the cause of science and employing principles of economy and efficiency, Tolkien declared his invention of language a private hobby that satisfied aesthetic criteria and scorned Jespersen’s efforts as factory-produced language, made by a nutritional expert rather than a cook (for more see Fimi on JTR). .

And here we can return to the question asked above of what the letters of Fëanor are doing on the One Ring. The answer given above – that (as Gandalf says) Sauron can turn anything to evil uses – is correct. But this does not negate the incongruity. If Sauron had developed a ‘black alphabet’ to record the ‘black speech’ it would no doubt have looked rather different – more like Sweet’s sentences above, perhaps, than the beautiful letters actually inscribed on the One Ring. (But is the Black Speech of Mordor grammatically akin to Jespersen’s Novial?) In any case, the hidden inscription in an Elvish script on the gold ring is a symbol of the magical pattern of Tolkien’s fairy-stories: the elves supply the sub-creative art while the Enemy steals (Silmarils), counterfeits (Rings of Power), borrows (Tengwar) or takes control (Seeing Stones) of this ‘elvish magic.’


Fëanor’s Apprenticeship

He stood still enchanted, while the sweet syllables of the elvish song fell like clear jewels of blended word and melody. (FR 310)

Did Tolkien consider Fëanor’s making of the Tengwar a stepping stone to his subsequent making of ‘magical’ jewels? I think the answer is that he did, and that he began to think this in the latter part of 1939, in the wake of his March lecture on fairy stories and during the phase of writing that saw the Company arrive at the Doors of Durin. In other words, I suspect that, like the other signs on the Door of Durin, the Fëanorian letters are an indication of Tolkien’s emerging ideas about the history of the Second Age and the nature of the Rings of Power.

To explain why I believe this is the case it is helpful to step back for a moment from the alphabet and focus on the spoken word itself. As a point of entry I introduce a third work from Tolkien’s professional studies:

H.C. Wyld, The Growth of English (1907).

We have already encountered Wyld as the editor of Henry Sweet’s collected papers. He would later become Tolkien’s predecessor as the Merton Professor of English Language and Literature at Oxford. But all we require from him is the commonplace observation about language set out in his introductory discussion. According to Wyld, words have an inside and an outside:

the inner side, which is the meaning which we wish to express, and the external side, which consists of the speech sounds whereby we express that meaning. (p. 5)

Wyld is simply setting out a then conventional expression of the nature of language. Now, I am not aware of any expressions to this effect in Tolkien’s writings before 1939, but after this date we find a theological version of the same idea. For example, in ‘The Notion Club Papers’ (composed between 1945 and 1946, as Tolkien was still composing The Lord of the Rings) language is at one point defined as the fusion of sensible token and mental significance, with the comment that such a fusion is an action “peculiar to an embodied mind; an essential characteristic, the prime characteristic of the fusion of incarnation” (Sauron 202; see also WJ 397, Peoples 397-9).

What we seem to have here is an application to spoken words of the theological idea of sub-creation expressed in On Fairy-stories (1947) with regard to the making of fantasy: “we make in our measure and in our derivative mode,” and we do so “because we are made: and not only made, but made in the image and likeness of a Maker” (OFS 66). In other words, just as we are minds (or souls) incarnated in bodies, so language is an incarnation of inner meaning in sensible sounds.

At this point it is worth recalling Fëanor’s explanation of the difference between consonants and vowels by way of an analogy of words with a dwelling consisting of walls and space – which invites the question of who or what dwells within this building (and here I must resist the temptation to a digression on the curious ways in which Tolkien likely contemplated the meaning of his spontaneous sentence about a hobbit who lived in a hole in the ground).

So what happens when we introduce alphabetical signs to Tolkien’s incarnational vision of the spoken word?

All the complicated discussion of the analysis of spoken sound according to the position of the tongue in the mouth and so on lends itself to the idea that written letters provide an additional casing – that a written word is a spoken word given an additional ‘body.’ From this understanding the act of reading is a translation of signs back into the spoken sounds – and for long periods of history this does seem to have been how people read (Saint Augustine has a famous account of how peculiar it was to see someone reading silently to himself). But this is not how you or I read. As your eye scans these letters and your mind receives their sense their may be audible echoes somewhere in your head but the meanings are received directly from the written signs.

In other words, the invention of writing provides an alternative ‘housing’ for inner meaning – just as spoken sounds embody meaning so too do written letters. This gives us another perspective on Taylor’s implicit connection between the invention of the alphabet and the writing of the Old Testament books and identifies Fëanor’s invention of letters as a first step in the development of an incarnational art. For an incarnational analogy is just what we find if we turn to the account of Fëanor’s making of the Silmarils in the Silmarillion: these jewels are said to be made of a crystal substance that “was to the Silmarils but as is the body to the Children of Ilúvatar: the house of its inner fire, that is within it and yet in all parts of it, and is its life” (Morgoth 95).

I give the reference above to the HoMe volume as the source of the account in the Silmarillion because the HoMe series reveals that it was composed in the 1950s, after the completion of The Lord of the Rings. In writings from the early 1930s (see Lost Road 125 and 249) Fëanor is simply said to have placed the light of the Two Trees within the Silmarils. In addition, it is only in the later writings that the account of the making of the Silmarils is preceded by an account of Fëanor’s invention of letters.

Now, an essay of the 1950s uses similar incarnational language with regard to Sauron’s forging of the One Ring:

Melkor ‘incarnated’ himself (as Morgoth) permanently. He did this so as to control the hroa, the ‘flesh’ or physical matter of Arda. He attempted to identify himself with it. A vaster, and more perilous, procedure, though of similar sort to the operations of Sauron with the Rings. (Morgoth 399-400)

So the making of the Silmarils and the making of the One Ring both involve an incarnational art (or operation) that is analogous to the fusion of meaning with spoken sounds and written letters.

So far as I can make out, the first hints of this idea appear in the first draft of the Council of Elrond, composed around autumn 1939 in that phase of writing that concluded at Balin’s tomb in Moria. Here we find Gandalf explaining that Sauron “put into that Ring much of his own power” (Shadow 397), while to Boromir’s question why the enemies of the Enemy cannot use the Ruling Ring against him, Elrond replies: “it belongs to Sauron and is filled with his spirit” (Shadow 403).

When Tolkien drew the Doors of Durin these thoughts were perhaps not yet settled in his mind – as noted in the previous discussion, it was only in Lórien that the history of the Rings of Power and the relationship between the One and the Three was finally settled. So, as in the discussion of the Doors of Durin, I suggest that the signs on the door reflect, not Tolkien’s solution to a riddle but rather his posing of a riddle to himself.

A complete discussion of the relationship between letters and the ‘magical’ art of the elves would now turn to the Palantíri. Where the incarnational account of the Silmarils build upon a much earlier story, and the nature of the One Ring was in large degree already determined by the story of The Hobbit, the Seeing Stones were an imagination that arose once Tolkien’s idea of incarnational elvish art was already established. The analogy between the Stones and the written word, both of which allow communication over vast distances and give access to an otherwise lost past, is striking. But to tease out the meanings of this analogy would involve attending to the telepathic abilities of the higher beings of Arda, in which both sound and letters are dispensed with and the meaning in one mind directly apprehended by another. This is obviously a component of the same investigation into Tolkien’s vision of the nature of language (and takes us to the second transcription of Vinyar Tengwar 39), but as this page is already very long I postpone this discussion until we have not only passed through Moria but also the realm of the Lady Galadriel, and, shoulder to shoulder with the Riders of Rohan, disposed of Saruman.