Professional Documents
Culture Documents
Quenya Course
Quenya Course
Helge K. Fauskanger
Copyright policy: This course can be downloaded from Ardalambion (http://www.uib.no/people/hnohf/), but not re-posted elsewhere on the net without permission from the author, H. K. Fauskanger. Print-outs may be made for personal use (private groups included).
A Typeset by Johan Winge using L TEX.
Contents
0 Introduction 0.1 The question of copyright . . . . . . 0.2 What is Quenya like? . . . . . . . . 0.3 The sources . . . . . . . . . . . . . . 0.4 A word of warning regarding parts of 0.5 Spelling conventions . . . . . . . . . 1 Lesson One 1.1 General remarks . . . 1.2 Basic terms . . . . . . 1.3 The sounds of Quenya 1.4 Pronunciation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 12 18 19 26 31 39 39 40 42 44
. . . . . . . . . . . . . . . . . . . . . the corpus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
2 Lesson Two 65 2.1 The Noun . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 2.2 The article . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 3 Lesson Three 74 3.1 Dual number . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 3.2 Stem variation . . . . . . . . . . . . . . . . . . . . . . . . . . 76 4 Lesson Four 81 4.1 The Adjective . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 5 Lesson Five 92 5.1 The Verb . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 5.2 More about adjectives . . . . . . . . . . . . . . . . . . . . . . 100 6 Lesson Six 105 6.1 The past tense . . . . . . . . . . . . . . . . . . . . . . . . . . 105 7 Lesson Seven 116 7.1 The future tense . . . . . . . . . . . . . . . . . . . . . . . . . 116 7.2 The Aorist . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
8 Lesson Eight 127 8.1 The perfect tense . . . . . . . . . . . . . . . . . . . . . . . . . 127 8.2 Pronouns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 9 Lesson Nine 142 9.1 The innitive . . . . . . . . . . . . . . . . . . . . . . . . . . . 142 9.2 The negative verb . . . . . . . . . . . . . . . . . . . . . . . . 145 9.3 Active participles . . . . . . . . . . . . . . . . . . . . . . . . . 147 10 Lesson Ten 10.1 Adverbs . . . . . . . . . . . . . . . . . . . 10.2 The pronominal endings -nt and -t . . . e 10.3 Innitives with object pronouns . . . . . . 10.4 The past tense of intransitive verbs in -ya 10.5 Passive participles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 . 152 . 155 . 157 . 159 . 162
11 Lesson Eleven 170 11.1 Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170 11.2 The Genitive . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 12 Lesson Twelve 183 12.1 The Possessive . . . . . . . . . . . . . . . . . . . . . . . . . . 183 12.2 Various notes . . . . . . . . . . . . . . . . . . . . . . . . . . . 188 12.3 Verbal or Abstract nouns . . . . . . . . . . . . . . . . . . . . 191 13 Lesson Thirteen 13.1 The Dative case . . . . 13.2 The Gerund . . . . . . 13.3 The pronoun we . . 13.4 An indenite pronoun . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 . 197 . 202 . 206 . 207
14 Lesson Fourteen 212 14.1 The Allative and Ablative cases . . . . . . . . . . . . . . . . . 212 14.2 Equ and auta: two peculiar verbs . . . . . . . . . . . . . . . 217 e 14.3 Possessive pronominal endings . . . . . . . . . . . . . . . . . . 220 15 Lesson Fifteen 15.1 More about possessive pronominal endings 15.2 The Locative case . . . . . . . . . . . . . 15.3 Relative sentences . . . . . . . . . . . . . 15.4 Third Person obscurities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232 . 232 . 236 . 238 . 242
16 Lesson Sixteen 16.1 The Instrumental case . . . . . 16.2 Verbs with an unaccented vowel 16.3 The imperative . . . . . . . . . 16.4 The nai formula . . . . . . . .
. . . . + -ta . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . .
249 249 253 257 258 263 263 266 268 272
17 Lesson Seventeen 17.1 Quenya demonstratives . . . . . . . . 17.2 Inecting the Last Declinable Word 17.3 U-stem nouns . . . . . . . . . . . . . . 17.4 Ordinals . . . . . . . . . . . . . . . . . 18 Lesson Eighteen 18.1 Independent pronouns 18.2 Impersonal verbs . . . 18.3 U-stem verbs . . . . . 18.4 The various uses of l a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
276 . 276 . 282 . 284 . 287 . . . . 293 293 295 299 300
19 Lesson Nineteen 19.1 Pronouns in imperative phrases 19.2 Emphatic pronouns . . . . . . . 19.3 Question-words . . . . . . . . . 19.4 Postpositions . . . . . . . . . .
20 Lesson Twenty 305 20.1 The verb to be . . . . . . . . . . . . . . . . . . . . . . . . . 305 20.2 Ma: An interrogative particle? . . . . . . . . . . . . . . . . . 311 20.3 Sa introducing nominal clauses . . . . . . . . . . . . . . . . . 313 A Keys 317
Lesson 0
Introduction
Of all the languages invented by British author and philologist J.R.R. Tolkien (18921973), the most popular has always been Quenya. It also seems to be the most highly developed of all the languages Tolkien devised. Indeed only two of them Quenya and Sindarin are so complete that one can with some ease write substantial texts in them without resorting to massive invention of ones own. Until recently, Sindarin was poorly understood, and its complex phonology may daunt fresh students (especially if they have no linguistic training). My advice to people who want to study Tolkiens linguistic creations would denitely be that they start with Quenya. Knowing this tongue will facilitate later studies of the other languages, including Sindarin, since Quenya represents just one branch of the Elvish language family: The Elvish languages are not independent entities, but all evolved from a common ancestral tongue, and in many respects, Quenya stands closer to this primitive original than the other languages. In reality as opposed to this ctional context, Tolkien knew well what kind of style he was aiming for, and having sketched a primitive Elvish language, he cleverly devised sound-shifts that would produce a tongue with the desired avour: Quenya resulted from his youthful romance with Finnish; he was, in his own words, quite intoxicated by the sound and style of this language when he discovered it (The Letters of J. R. R. Tolkien, p. 214). However, it should be emphasized that Finnish was an inspiration only; Quenya is in no way a garbled version of Finnish, and only a few words of its vocabulary display any semblance to the corresponding Finnish words. (See Harri Perlas discussion at http://www.sci./alboin/nn que.htm; a the writer is a Finn himself.) Tolkien also mentioned Greek and Latin as inspirations; we can evidently add Spanish to the list as well. The ctional or internal history of Quenya is synopsized in my regular Ardalambion Quenya article (see http://www.uib.no/People/hnohf/quenya. htm) and does not have to be repeated in any detail here. Very briey, within Tolkiens mythos Quenya was the language of the Elves that dwelt in Valinor in the Uttermost West; being spoken in the Blessed Realm, it
was the noblest tongue in the world. Later one of the clans of the Elves, the Noldor, went into exile in Middle-earth, bringing the Quenya tongue with them. In Middle-earth it soon fell out of use as a daily speech, but among the Noldor it was ever preserved as a ceremonial language, and as such it was also known to Mortal Men in later ages. Hence in The Lord of the Rings we have Frodo delivering the famous Quenya greeting elen s l menn la u omentielvo, a star shines on the hour of our meeting, when he and his friends run into some Elves (and the Elves are delighted to meet a scholar in the Ancient Tongue). If one studies Quenya as a way to immerse oneself in Tolkiens ction, it may indeed be best to picture oneself as a mortal student in Middle-earth in the Third Age, about the period covered in The Lord of the Rings. (Picturing oneself as an Elvish native speaker in Valinor back in the First Age may be overly ambitious.) The particular form of Quenya taught in this course is by intention precisely the late Exilic or Third Age variant. This is the kind of Quenya exemplied in The Lord of the Rings, with Galadriels Lament (Namri ) as the most substantial a e example. Numerous enthusiasts have brought forth a limited, but steadily growing body of Quenya literature, especially since a substantial amount of vocabulary nally became available with the publication of The Lost Road in 1987, fteen years after Tolkiens death. Thanks to this and the fteen other books of Middle-earth material that Christopher Tolkien in the period 197796 edited from the manuscripts his father had left behind, we now know very much more about Tolkiens languages than we ever did during the lifetime of their inventor. We certainly cant sit down and readily translate the works of Shakespeare into Quenya, but we do know a few thousand words and can infer the general outlines of the grammar Tolkien envisioned. Still, you cannot really become uent in Quenya, not matter how hard you study what is presently available. But it is eminently possible to write quite long Quenya texts if one deliberately eschews the unfortunate gaps in our knowledge, and we can at least hope that some of these gaps (especially regarding grammatical features) will be lled in by future publications. In the future, we may be able to develop Quenya into a more fully useable language. But we must obviously start by carefully internalizing the information provided by Tolkiens own material, as far as it is available to us. Many have wanted a regular course or tutorial, with exercises and all, that would allow them to study Quenya on their own with some ease. One such eort has been made before: Nancy Martsch Basic Quenya. All in all, this was certainly a good work; the fact that material that has been published after it was written now reveals certain shortcomings, cannot be held against the author. However, many would like to have a more updated course, and I have repeatedly been approached by people suggesting that I would be the right person to write it. It is of course nice when others call me an expert on Tolkienian linguistics; actually I would say that it is dicult 7
to be an expert in these matters, due to the scarcity of source material. Nonetheless, I have been so privileged that I have been able to spend much time studying these matters (starting more than ten years ago), and I see it as my duty to record and pass on whatever insights I may have gained. Hence in the end I sat down and started writing this course, intended for beginners. (This conveniently allows me to ll the uncritical, vulnerable minds of fresh students with my interpretation of Quenya grammar, which interpretation I inevitably hold to be the best and most accurate. Ha ha ha.) However, this course does not seek to imitate a Linguaphone-like format with long dialogues etc. to help the student to acquire basic uency in various situations relating to everyday life. This would be quite pointless in the case of an art-language like Quenya, which is to be used for carefully prepared prose and poetry rather than casual chatting. Rather these lessons take the form of a series of essays on various parts of Quenya grammar, reviewing and analyzing available evidence in an attempt to reconstruct Tolkiens intentions, with some exercises appended. Why study Quenya? Obviously not because you are going to Valinor on holiday and need to be able to communicate with the natives. Some may want to study this language to somehow get in better accord with the spirit of Tolkiens authorship. He referred to . . . what I think is a primary fact about my work, that it is all of a piece, and fundamentally linguistic in inspiration. [. . .] It is not a hobby, in the sense of something quite dierent from ones work, taken up as a relief-outlet. The invention of languages is the foundation. The stories were made rather to provide a world for the languages than the reverse. To me a name comes rst and the story follows. I should have preferred to write in Elvish. But, of course, such a work as The Lord of the Rings has been edited and only as much language has been left in as I thought would be stomached by readers. (I now nd that many would have liked more.) [. . .] It is to me, anyway, largely an essay in linguistic aesthetic, as I sometimes say to people who ask me what is it all about. (The Letters of J. R. R. Tolkien, pp. 219220) In light of such strong statements made by the author, studying his invented languages cannot be dismissed as some kind of silly escapism for romantic teenagers. It must be considered a crucial part of scholarship relating to Tolkiens authorship, or indeed his work in general: The languages constructed by Tolkien are part of his output as a philologist, not necessarily less serious than his writings on pre-existing languages like Anglo-Saxon; notice that he refused to call his fundamentally linguistic work a mere hobby. One may call Quenya and the other languages works of art, but no matter what word we use to describe them, in the end it all boils down 8
to this: Tolkien was not just a descriptive linguist, passively exploring and contemplating pre-existing tongues he was a creative linguist as well. Obviously uency in Quenya or Sindarin is not a prerequisite before you can say anything intelligent about Tolkiens narratives; yet it is clear that some critics and scholars have woefully underestimated the crucial role of the invented languages, nding themselves unable to take even very clear statements like the one quoted above wholly seriously. To fully appreciate the scope and intricacy of Tolkiens linguistic sub-creation one has to actively study it for its own sake. It should certainly be able to command interest for its own sake. Some years ago, recognized Tolkien scholar Tom Shippey observed that . . . its clear that the languages Tolkien created are created by, you know, one of the most accomplished philologists of all time, so there is therefore something interesting in them, and I think also in them there is poured much of his professional knowledge and thought. [. . .] Ive often noticed that there are really very valuable observations about what Tolkien thought about real philology buried in the ction. And I would not be at all surprised if, you know, there were valuable observations buried in the invented languages. So there may be, in fact, something which emerges from it. [From an interview conducted during the Arda
symposium in Oslo, April 35 1987, published in the journal Angerthas, issue 31.]
Even if one does not believe that there are great new philological insights waiting to be unearthed from the structure of Tolkiens languages, I cannot see why conducting detailed studies of these languages should necessarily be seen as escapism, or at best a somewhat silly pastime for people who are too lazy to nd something better to do. The languages constructed by Tolkien have been likened to music; his biographer Humphrey Carpenter observes that if he had been interested in music he would very likely have wanted to compose melodies; so why should he not make up a personal system of words that would be as it were a private symphony? One may study one of the languages Tolkien painstakingly developed as one may study a musical symphony: a complex work of many interrelated parts woven into intricate beauty. Yet the symphony is xed in its form, while a language can be innitely recombined into ever new texts of prose and poetry, and yet retain its nature and avour undiminished. One of the attractions of Quenya is that we can compose linguistic music ourselves just by applying Tolkiens rules, so Carpenters comparison is too limited: Tolkien did not just make a symphony, he invented an entire form of music, and it would be a pity if it were to die with him. Of course, others may want to study Quenya to immerse themselves in Tolkiens ction, with no pretensions of scholarship of any kind: Tolkiens 9
vision of the Elves (Quendi, Eldar) is no doubt the main achievement of his authorship, and Quenya was at least in the somewhat biased opinion of the Noldor the chief Elvish tongue, the noblest, and the one most nearly preserving the ancient character of Elvish speech (The War of the Jewels p. 374). But one may grope towards Elvishness in a deeper sense than just trying to immerse oneself in ction. Happily abandoning the all too classical idea of Elves as tiny, overly pretty fairies, Tolkien instead achieved the vision of Elves as something more: I suppose that the Quendi are in fact in these histories very little akin to the Elves and Fairies of Europe; and if I were pressed to rationalize, I should say that they represent greater beauty and longer life, and nobility the Elder Children (The Letters of J. R. R. Tolkien, p. 176). The quintessence of Tolkiens vision of Elvishness is contained primarily in the languages, for to the Eldar the making of speech is the oldest of the arts and the most beloved (The Peoples of Middle-earth p. 398). In a way, the study of Quenya can be a quest for this vision of something beautiful and noble beyond the normal capability of our mortal and nite selves: The Elves represent, as it were, the artistic, aesthetic, and purely scientic aspects of the Humane raised to a higher level than is actually seen in Men (Letters, p. 176). The seeking for such a higher level transcends all ction. Tolkiens inner vision of this level he translated partly into pictures, much more prominently into narratives, but (to him) more importantly still, into the words and sounds of language. In Quenya his vision of Beauty lives on, awaiting those capable of comprehending and appreciating it. On their web-pages, the Swedish Tolkien-linguists of the Mellonath Daeron group try to justify their study of Tolkiens languages: Our activity has been described as the ultimate luxury. We study something that does not exist, just for fun. This is something you can aord when you have everything else; food, shelter, clothes, friends, and so on. The Tolkien languages are well worth studying for their high aesthetic values alone. And knowledge of the languages is a key to a fuller appreciation of the beauty of Tolkiens sub-creation, his world, Arda. I heartily agree with the last two sentences, but I cannot agree that Quenya or Sindarin does not exist. Obviously we are not talking about physical, tangible objects, but that goes for any language. These are not ctional languages, but languages as real as Esperanto or any other constructed language. Tolkien himself noted about his languages that they have some existence, since I have composed them in some completeness (The Letters of J. R. R. Tolkien, p. 175). Unlike Esperanto, Quenya is however strongly associated with a ctional internal history. (Tolkien once stated that Esperanto had been more successful if there had been an Esperanto mythos to go with it!) The associated 10
mythos certainly enriches Quenya and helps us to understand what kind of linguistic avour Tolkien was aiming for, and the fact that this language has a role to play in the most famous fantasy novels ever written obviously provides it with much free publicity that Esperanto can only dream of. Yet it must be emphasized that Quenya does exist as an actual entity in our own world, and as mentioned above, it indeed has a steadily growing literature, mostly in verse: The texts presently in existence must already be hundreds of times more comprehensive than all the Quenya texts Tolkien ever wrote himself. He endlessly rened the structure and imaginary evolution of his invented languages, but he wrote remarkably few substantial texts in them. Though he stated that he should have preferred to write in Elvish (see quote above), he actually wrote about the Elvish tongues rather than in them. Delight lay in the creation itself, Christopher Tolkien observes (Sauron Defeated, p. 440). His father made the languages just because he loved making them, not because he needed to use them for any specic purpose. To be sure, Tolkien wrote a number of poems in Elvish, but they amount to very little compared to the thousands of pages he wrote about the structure of his languages. Tolkien had his fun in sheer invention; that was his privilege as the original creator. However, I daresay quite few people are capable of deriving much pleasure from mere passive contemplation of the structure of a language, or from reading the grammar of an invented language as if it were some kind of novel. I imagine that most people who want to study Quenya have some intention, however vague, of putting this knowledge to use by writing Quenya texts themselves, or at least by reading other peoples texts (at the very least Tolkiens own). Really learning any language in any case requires active participation: Even if you wouldnt dream of ever publishing anything in Quenya but rather want to assess Tolkiens Elvish for purely academic purposes, you will still have to work yourselves through some exercises to internalize grammar and vocabulary. Such exercises are provided in this course. My favorite angle on the study of Tolkiens languages is probably this (building on the musical analogy suggested by Carpenter): Id say we are in somewhat the same situation as if a genius composer were to invent a new form of music, writing a great deal about its structure, but making relatively few actual compositions some of them not even published during the lifetime of the composer himself. Yet these few compositions gain a steadily growing international audience, an audience that would very much like to hear more much more music of this kind. The original composer being dead, what are we to do? There is only one way to go: We must carry out a thorough study of both the published compositions and the more theoretical writings, to make out and internalize the rules and principles for this kind of music. Then we can start to compose ourselves, making entirely new melodies that yet comply with the general structure devised by the 11
original inventor. This, of course, has a rough analogy when it comes to Tolkiens narratives as well. Tolkiens themes and principles of story-telling have been taken over by generations of new authors, resulting in the modern fantasy genre though it would not be very controversial to say that far from all authors have been able to live up to the high standards set by the master. In somewhat the same manner, the quality of the numerous post-Tolkien Quenya texts varies greatly. In the case of some early attempts, written when very little source material was available, it is now easy to spot various shortcomings and misinterpretations of what Tolkien really intended. Today, with much more material available, I would say it is possible to write texts that Tolkien probably would have recognized as at least roughly correct Quenya (though I think reading Quenya texts not originating with himself would have been a strange experience for him; his invented languages were originally something very private). This course should in any case be useful no matter what your angle on this study may be whether you want to learn Quenya to immerse yourself in Tolkiens ction, to better appreciate a crucial side of his authorship, to learn about the intricate creations of a talented linguist, to accept the intellectual challenge of trying to master a sophisticated system, to go on a meditative quest for Elvishness, or simply to enjoy Quenya aesthetically. None of these are mutually exclusive, of course. Whatever your angle is, I hope you would like to have a part in making Quenya literature grow and ourish. Another Tolkien quote may be in place here: No language is justly studied merely as an aid to other purposes. It will in fact better serve other purposes, philological or historical, when it is studied for love, for itself (MC:189).
0.1
This is an issue I shall have to devote quite a few paragraphs to, though this will probably surprise any fresh, innocent student who has never given much thought to this at all. However, debates revolving around copyright issues have sadly caused a great deal of bitterness among students working in the eld of Tolkien-linguistics; such debates essentially blew apart the TolkLang mailing list, leading to the establishing of Eling instead. If Tolkiens heirs or their lawyers ever read what follows, I hope they are not oended. This really is not about stealing anything from them, but about directing attention to one highly important part of Tolkiens work and help people learning about it, so that it can live and grow and stand as a lasting testimony to his eorts, and as a dynamic memorial to himself. Talking about his father, Christopher Tolkien in a TV interview described Quenya as language as he wanted it, the language of his heart. Students of Quenya merely want this special
12
part of Tolkiens heart to live on. Nobody is trying to make any money or otherwise prot from this. (If the Tolkien Estate, or rather HarperCollins, might ever want to publish this course in book form, I would be happy to let them do so, and I would not expect to receive any royalties.) In 1998 and early 1999, on the TolkLang list, lawyer W. C. Hicklin vociferously argued that publishing unauthorized grammatical descriptions of a Tolkien-language would be a blatant violation of the copyright of the Tolkien Estate, asserting that any such publication would undoubtedly make the Estate react with money, guns and lawyers. (One hopes the part about rearms was a gure of speech.) I cannot agree with such an interpretation of copyright law, especially considering that what we know about Quenya we have for the most part learnt by studying the examples we have not by reading Tolkiens explicit grammars, that still have not been published. I cannot imagine that when studying available Quenya texts, is it illegal for us to put our conclusions into words and tell others about them. If this is what copyright means, then all sorts of scholarly commentary and literary criticism immediately go down the drain. While Hicklin said he reported the position of Christopher Tolkien (whom he claimed to know on a rst-name basis), the Tolkien Estate itself has so far declined to present its opinion on these issues, even when asked to do so by TolkLang moderator Julian Bradeld. It may be noted that copyright law is not the area Mr. Hicklin specializes in, and I think he pressed the concept of character rather far by asserting that every individual word in the invented languages must be considered a literary character of Tolkiens, apparently on par with such characters as Aragorn or Galadriel. Mysteriously, Hicklin still agreed that it is OK to write fresh texts in Tolkiens languages, though in Hicklins world this would seem to be the analogy of writing new stories involving Tolkiens characters (which everybody agrees would be a copyright violation). Hicklins obvious problems in putting together a consistent argument, as well as subsequent legal inquiries conducted by myself and others, have led me to the conclusion that copyrighting a language as such would be quite impossible. The language itself is not to be likened to a xed text in or about it; it is an entirely abstract system, and for anything to enjoy copyright protection it must rst of all have a xed form to be protected. Arguing that the very grammatical structure and vocabulary of the language is its xed form is no use, for this is an abstract system, not a form. Any actual text about (or in) a language is indeed protected, but not the language itself. To return to the analogy of our genius composer who invents a new form of music: His copyright to his own compositions, and to his writings on this form of music as xed texts, cannot and should not be disputed by anyone. But he or his heirs cannot well assert that publishing entirely new compositions, or wholly original descriptions of the principles of this kind of music, would somehow violate their copyright. This course is written and published (for free on the Internet) by me as 13
a private person. The Tolkien Estate has not been asked to endorse it or even comment on it, it is in no way ocial, and I must take full responsibility for the quality of the contents. No disrespect is intended when I point out that any endorsement by the Estate would not have meant much in the way of a quality guarantee, since certain earlier works on Quenya that were published with explicit permission from the Estate can now be seen to contain certain obvious shortcomings and misinterpretations. There is little reason to believe that Estate lawyers or Christopher Tolkien himself are capable of judging the quality of a Quenya grammar (and likewise no reason to hold this against them; learning Quenya from the primary sources is a long and challenging study reserved for the especially interested). In such a situation I hope and believe that the Tolkien Estate respects the right of scholars to carry on their studies undisturbed, and to present the results of such research especially when the relevant publications are entirely non-commercial. Despite the strong claims made by Hicklin and a very few others, there is presently no concrete evidence that the Estate or Christopher Tolkien see such studies as a violation of their copyright. If they do, let them contact me and we will talk. The interpretation of Quenya grammar that it here set out is based on a study of the available sources, mostly analysis of actual Quenya text, and on exegesis of the relatively few explicit notes on grammar that are presently available. I hold it to be obvious that this is primarily a work of analysis and commentary (presented in a didactic fashion), and in terms of copyright, discussing the structure of Quenya cannot be much dierent from discussing (say) the plot structure of The Lord of the Rings: In either case it is clear that anything I can say must ultimately be based on Tolkiens writings, but the resulting study still is not a derived work in terms of copyright law. What we are doing here is not retelling Tolkiens ction (though I will certainly refer to it but then from the perspective of a critic, or better commentator, to demonstrate how Tolkiens ction and language-construction interlock). Primarily we will be studying one of Tolkiens languages as an actual rather than a ctitious entity. The fact that this language was rst presented to the world in a context of ction does not make it a ctional language, and use or discussion of it is not necessarily derivative ction. As already mentioned, Tolkien himself observed that his languages as such have some existence simply because he had actually devised them they do not exclusively reside within a ctitious context (The Letters of J. R. R. Tolkien, p. 175). Much of Quenya vocabulary is not wholly original; Tolkien readily admitted that the vocabularies of his Elvish languages were inevitably full of . . . reminiscences of pre-existing tongues (The Peoples of Middle-earth p. 368). Though usually not so obvious that it is disturbing to those who want to study Quenya as a highly exotic language, the fact remains that the knowledgeable easily discern Indo-European (and sometimes even Semitic) 14
words and stems underlying many of Tolkiens invented words. This is not to be seen as some sort of failure of imagination on Tolkiens part; he noted that it is impossible in constructing imaginary languages from a limited number of component sounds to avoid such resemblances adding that he did not even try to avoid them (Letters, pp. 384385). Even where no plausible real-world inspiration for a Quenya word can be cited, the fact still remains that there is no legal tradition whatsoever to allow a person coining new words to somehow claim them as his personal property. Tolkien himself was aware that names cannot be copyrighted (Letters, p. 349), and then one cannot well copyright common nouns, verbs, adjectives or prepositions either, precluding unauthorized use of them. Some words in common use today, such as robot, rst occurred in a context of ction. One cannot therefore claim that they are ctional words, protected on par with ctional characters, and not to be used, listed or explained without explicit permission from the one who rst coined them (or his heirs). Legal inquiries conducted after Hicklin made his amboyant claims have conrmed that words as such automatically enter public domain the second they are coined, and nobody can monopolize them or claim exclusive ownership to them. You can register a word as a trademark, of course, but that is something entirely dierent: Apple Computers cant stop anyone from using apple as an everyday word. It is also irrelevant that the manufacturer of some kind of fantasy game had to remove all references to balrogs, for here it is not the Sindarin word balrog, but balrogs as characters that lie in Tolkiens copyright. The fact that Tolkien coined the word alda for tree hardly implies that trees are his literary characters. It is not just a tree growing in Middle-earth that can be termed an alda; the word works just as well if I write a Quenya poem about a tree growing outside my house. I agree, though, that Quenya and the other languages enjoy some protection in their capacity as parts of the Middle-earth setting. If anyone were to write new fantasy stories involving Elves speaking a language called Quenya, and there were samples demonstrating that this is indeed Tolkiens Quenya, this would obviously be the same kind of plagiarism as if any fantasy writer were to borrow a city called Minas Tirith, and the description in the book made it clear that this city happened to be built on several levels and was overlooked by a white tower. But again: this course is most certainly not intended as derivative ction. This is about studying and using one of Tolkiens languages largely irrespective of the ctional context as such though since I also aspire to present Quenya as a part of Tolkiens authorship, I shall of course have to mention, refer to and sometimes even quote the narratives as well as presenting mere technicalities. Nonetheless: It is obviously untrue that Tolkiens languages cannot in any way be separated from his ctional world (as Hicklin seemed to claim). Vicente Velasco was for instance able to write a Quenya poem (R anna) commemorating Princess Diana after her tragic death, but this does not imply that the accident where she was killed 15
must actually be a plot point in a Tolkien novel. Indeed Tolkien himself made a Quenya translation of the Lords Prayer, a text that obviously belongs to our own reality and could not occur within the Middle-earth setting. When discussing copyright issues, we must distinguish very clearly between the ctional context and actual use of systems or ideas described within this ction; the latter is quite irrelevant for a discussion of copyright. By way of comparison: I fully agree that if anyone were to write new fantasy stories involving a race of small people with hairy feet living in underground structures called smials, then this writer would clearly plagiarize Tolkien and possibly even violate his copyright. But I cannot imagine that I violate anyones copyright if I dig out a smial in my garden or for that matter, if I have a head-to-feet hair transplant. Similarly, one shouldnt feel free to write fantasy stories about Elves speaking Quenya, but actually using the linguistic structures dreamed up by Tolkien to write new texts that by their contents have nothing to do with his ction cannot be a violation of copyright. The new Quenya texts are copyright to no one but their writers. Happily, Tolkiens heirs seem to agree to this; at least they have never tried to stop anyone from publishing their Quenya poems. If the Estate has no problems with this, I can only assume that their lawyers also agree that it is perfectly legal for anyone to write Quenya grammars or compile Quenya wordlists. Otherwise we should be left with the rather absurd notion of a language that can be used, but not taught or in a scholarly way described. I cannot imagine that the Estate would assert that the by now quite large number of Quenya texts that are not written by Tolkien and have nothing to do with his ction cannot be subjected to grammatical or lexicographical studies simply because they happen to be written in Quenya. This would be an attempt to block and veto certain kinds of scholarship relating to an entire body of literature, and I dont think this could be sustained, legally or even morally. I dont know that Tolkiens heirs disagree. I have no intention, however, of disputing the Estates copyright to Tolkiens actual writings (on the languages or otherwise), and though it is an interesting exercise to reconstruct the Elvish original supposedly underlying some of Tolkiens poems or stories, one should not publish Elvish translations of a great amount of continuous Tolkien text. All of his texts lie in the copyright of the Tolkien Estate until it expires in 2023 (or was it 2048?), and publishing substantial translations or close retellings thereof would require the permission of the Estate: No matter how exceedingly esoteric the target-language is, any translation is still directly derived from Tolkiens own, copyrighted text. Neither should one write long stories set in Tolkiens world; that would be a violation of copyright no matter what language you use. However, making translations of a limited amount of Tolkien text can probably pass as fair use (but please dont publish your own Quenya rendering of the Ring Poem; there are all too many competing versions already. . . ) Neither is there much reason to believe that the Estate would take 16
any action against short Quenya-language novelettes even if they do seem to be set in Middle-earth, since it should be obvious that the real purpose is to demonstrate the use of Quenya, not to write new stories to compete with Tolkiens own (I wouldnt publish even such novelettes in any way that could conceivably be seen as commercial publishing, though). Poems about persons or events in Tolkiens world (like Ales Bicans Roccalassen or Song to Eowyn) I think can pass as a branch of commentary or synopsis, as long as you dont include any fresh ction of your own. But please dont push even that too far; Tolkiens heirs are in their good right when they assert their copyright to his stories. In exercises made for this course, I have in any case deliberately avoided any direct references to persons, places or events in Tolkiens ctional world (except for one reference to the Two Trees because the Quenya word provides such a good example of dual number). Instead of referring to Tolkiens ction I have in most cases resorted to a wholly generic fantasy world or medieval world; there is nothing to preclude the possibility that this is Tolkiens world, but nothing concrete to conrm it, either. There are lots of Elves and Dwarves in these exercises, but though we inevitably use words like Eldar and Naucor for these peoples when talking about them in Quenya, they are really just generic Elves and Dwarves. Feel free to imagine that these Elves are Tolkiens Eldar if you like, but there is nothing that denitely ties them to any specic mythos. Despite the fact that I dont think the Tolkien Estate could legally stop people from doing pretty much what they want with Quenya as an actual language (separated from Tolkiens ction), I urge students to use whatever knowledge they may obtain in a respectful way. We should feel some kind of moral obligation, or even gratitude, towards Tolkien as the creator of this language. Quenya as we know it is the result of decades of painstaking work and endless renement; its creator intended it to have an august or even sacred avour, and it is not to be used for unworthy or downright silly purposes. (Please dont publish your Quenya compositions on toilet walls, for instance.) There is an old TV interview where Tolkien says he would not necessarily mind others knowing and enjoying his invented languages, but he would not like to see any of them turned into some sort of secret lingo used to exclude others. This is a wish I urge any and all students to respect. As a student and user of Quenya one should also be committed to preserving the integrity of Tolkiens system, taking great care not to distort it or needlessly dilute it. Occasionally we have to coin new words, but in such cases one should eschew arbitrary invention and instead work from Tolkiens own stems, using his methods of derivation. Wrote Tolkien, Of course the L[ord of the] R[ings] does not belong to me. It has been brought forth and must now go its appointed way in the world, though naturally I take a deep interest in its fortunes, as a parent would of a child. I am comforted to know that it has good friends to defend 17
it (The Letters of J. R. R. Tolkien, p. 41314). Perhaps he would have felt the same way about the invented languages exemplied in the book he is talking about: They have been brought forth and do already go their way in the world, studied and even used by many but now Quenya and the other languages must live their lives independently of their parent, for he is no longer among us. So let students and users be their good friends and defend their systems, true to the vision of the man who spent a lifetime developing them. And this brings us back to the structure of Quenya itself.
0.2
What kind of language is this, structurally speaking? It seems that Finnish provided considerable inspiration not only for the sound-patterns, but for the basic structure as well. Tolkien described Quenya as a highly inected language (The Road Goes Ever On p. 69). That is, words appear in many dierent forms depending on their precise function in any given grammatical context. The diering forms are for the most part constructed by employing a plethora of endings, endings with meanings that in English would often be expressed as separate words instead. Hence an English translation of a Quenya text will normally consist of more words than the Quenya original: In Unnished Tales p. 22, 51, we learn that three words of Quenya may well require a seven-word English translation: Anar caluva tielyanna = the sun shall shine upon your path. Some may see this as evidence that Quenya is a more ecient language than English, but whether one uses one long word or several shorter words to express a given meaning is not very crucial. (It may be noted that if one counts syllables instead of words, it is suddenly English that is the most ecient language in the example above: The English text has one syllable less than the Quenya version!) Quenya should be enjoyed for its own qualities, not by comparing it to other languages. But the word tielyanna upon your path illustrates the main dierence between English and Quenya: small independent words like your or upon frequently become endings instead in this example -lya and -nna, respectively. Is Quenya a dicult language? Speaking of Quenya and Sindarin, the two main languages of his mythos, Tolkien wrote that both languages are, of course, extremely dicult (Letters:403). Undoubtedly there are many presently unsuspected complexities waiting for us in the vast amount of unpublished material. But as far or short as our knowledge goes today, I certainly wouldnt call Quenya extremely dicult. It may be an involved and intricate construct, but certainly less complicated than Sindarin, and the acquisition of Quenya as we know it is in no way a superhuman feat. Any devoted student should be able to achieve basic mastery of the grammatical system in relatively short time, weeks or even days rather than months.
18
General knowledge about linguistics would certainly be helpful in such a study, but hardly a prerequisite; in this course I have tried to make the explanations so simple that any reasonably bright teenager should be able to understand what is going on. (Bearing in mind that some people who want to study Quenya are quite young, I have tried to pre-suppose virtually no knowledge about linguistics, and I will explain even elementary linguistic terms more knowledgeable students may feel that I sometimes go into boring baby-talk.) It must still be understood that it is not a streamlined Esperanto we are dealing with here. Tolkien deliberately tried to make his languages naturalistic; hence there are some irregular verbs and the like, though I would say their number is quite manageable. Quenya probably stands about midway between an absolutely regular Esperanto and a typical real language with its spate of complexities and irregularities, yet perhaps closer to the former. Indeed Quenya is probably too simple to be entirely credible as a supposedly non-constructed language, at least if we compare it to the messy languages of Mortal Men in our own age. But then Quenya wasnt really non-constructed within the scope of the ctional history either; it was constructed and rened by the Elves, and the Eldar know their tongue, not word by word only, but as a whole (The Peoples of Middle-earth p. 398). So perhaps the Eldar, being very much conscious of the structure of their speech, would tend to make languages with a relatively tidy grammar. Anyway, from the viewpoint of students it is dicult to regret the absence of more irregular forms to be memorized, so if this simplicity does indeed make Quenya less credible as a natural language, Tolkien is easily forgiven!
0.3
The sources
We know that Tolkien wrote literally thousands of pages about his languages. Unfortunately and here I must ask fresh students to brazen themselves for their rst big shock, though the shocking fact has already been alluded to very little of this material is available to us. However, Christopher Tolkien has apparently tried to make arrangements for its publication. Throughout most of the nineties, he was sending photo-copies of his fathers linguistic manuscripts to a group of Americans often (but unocially) referred to as the Elfconners, apparently because of their prominence on the cons or conventions of ELF, the Elvish Linguistic Fellowship. However, the most outspoken member of the group seems to have convinced himself beyond refutation that the term Elfconners was always meant to be derogatory, associating it with conning or deception. As pointed out by TolkLang moderator Julian Bradeld, it may be that this member of the group is inventing insults against himself, but currently it is politically correct to refer to this group simply as the Editorial Team,
19
abbreviated ET. Whatever we call them, the group consists of Christopher Gilson, Carl F. Hostetter, Patrick Wynne and Arden R. Smith (in recent years, Bill Welden has also joined in). Before they started to receive Tolkien manuscripts, these people quite regularly published the Tolkien-linguistic journals Vinyar Tengwar (edited by Hostetter) and Parma Eldalamberon (edited by Gilson), generally maintaining a high standard. This, we must assume, was the reason why Christopher Tolkien wanted them to publish his fathers linguistic manuscripts in the rst place. The very strange and most unfortunate fact is that after they started receiving Tolkien manuscripts for publication, the groups rate of publication dropped disastrously. They began to receive manuscript copies in 1991; a decade later they have managed to get a few hundred pages of new material into print (most of it wordlist material pertaining to the very earliest stages of Tolkiens work, far removed from the LotR scenario). Some of us are not impressed. What little material has appeared has been nicely presented, but with the present rate of publication, the completion of the project must be very far o indeed. In 1996, Christopher Gilson stated that next year, his group planned to publish fairly comprehensive grammars for the two main languages of Tolkiens mythos. In early 2002 they nally published some material relating to certain stages of Gnomish, an early conceptual ancestor of the language Tolkien later called Sindarin; we are still waiting for a really substantial amount of material about Quenya. Most other deadlines that the members of Gilsons group have set for themselves has proved equally worthless, and since about 1998 they have largely refrained from stating any deadlines at all. Still, we must hope that in ten (or twenty, or thirty. . . ) years, we will know more but if the Editorial Team is able to ever commence ecient, regular publication of Tolkiens material, they have yet to demonstrate this ability. We must work, then, from sources already available sources that often touch on the languages more or less incidentally. The linguistic aspect of Tolkiens authorship luckily permeates his works to such an extent that if you bring together all the scattered pieces of information and analyze them thoroughly, you will be able to gure out much about his languages even without access to his explicit grammars. Unfortunately this method of study will leave many gaps in our knowledge, gaps most irritating to people who try to actually use these languages. In other cases, the material is so scarce that we can formulate not just one but all too many theories about what the underlying grammatical rules look like, and we dont have any further examples that would allow us to identify the correct theory. Nonetheless, we do know a great deal about Quenya, though some of our knowledge is more tentative than we would like. A survey of the sources is in place here; at least I should explain the abbreviations used in this work. The primary narrative works, The Lord of the Rings (LotR, 195455) and The Silmarillion (Silm, 1977) need no further introduction. (Of course, 20
there is also The Hobbit, but this book contains little linguistic information, and hardly anything at all about Quenya.) Most of the Elvish names of people and places found in LotR (such as Aragorn, Glorndel, Galadriel, Minas Tirith) are Sindarin, but there are substantial samples of Quenya as well. In LotR, we nd one of the longest Quenya texts known, the poem Namri near the end of chapter VIII (Farewell to Lrien) in Book Two a e o in the rst volume, The Fellowship of the Ring. Also known as Galadriels Lament, this is the poem commencing with the words Ai! lauri lantar e lassi s rinen. . . u Various shorter samples of Quenya are also sprinkled throughout LotR, such as Frodo speaking in tongues in Shelobs lair (Aiya Erendil Elenion a Ancalima! he cried, and he knew not what he had spoken), the praise that the Ringbearers received on the Field of Cormallen (part Sindarin, part Quenya), Elendils Declaration as repeated by Aragorn at his coronation, and Treebeards greeting to Celeborn and Galadriel. The Quenya parts of the Cormallen Praise (as I shall refer to it), as found in volume 3, Book Six, chapter IV (The Field of Cormallen), go like this: A laita te, laita te! Andave laituvalmet! . . . Cormacolindor, a laita trienna! (Cf. a Sauron Defeated p. 47.) This is translated in The Letters of J. R. R. Tolkien, p. 308: Bless them, bless them, long we will praise them. The Ring bearers, bless (or praise) them to the height. In the next chapter (V) we have Elendils Declaration, repeated by Aragorn at his coronation: Et Erello Endorenna ut lien. Sinome marua u van ar hildinyar tenn Ambar-metta. This is translated in the text as out of the Great Sea to Middle-earth I am come. In this place will I abide, and my heirs, unto the ending of the world. Treebeards Greeting in the chapter after that (VI) goes a vanimar, vanimlion nostari, translated a both in Letters p. 308 (o beautiful ones, parents of beautiful children) and Sauron Defeated p. 73 (fair ones begetters of fair ones; this rendering is the more literal). Quenya material (though mostly isolated words only) also occur in the Appendices to LotR, in particular Appendix E. In the Silmarillion, we also have a few short Quenya sentences. In chapter u 20 there are some battle-cries: Ut lien aur! Aiya Eldali ar Atanatri, e e a ut lien aur! The day has come! Behold, people of the Eldar and Fathers u e of Men, the day has come! Auta i lm! The night is passing! Aur o e e entuluva! Day shall come again! Near the end of chapter 21 there is the cry a T rin Turambar turun ambartanen, o Trin master of doom by u u doom mastered but Unnished Tales p. 138 indicates that turun should rather read tur n (evidently shortened from a longer form tur na, the u u nal -a dropping out because the next word also begins in a-). The Silmarillion Appendix, Elements in Quenya and Sindarin names, also mentions many words belonging to these two languages. In the case of other sources, a more summary survey will suce, since 21
these books and journals (unlike LotR and Silm!) have not appeared in too many editions and translations. Hence I can simply refer to the relevant book and page when quoting from them, and that reference will hopefully be precise enough. We will list them by the abbreviations used hereinafter: RGEO: The Road Goes Ever On (our page references are to the Second Edition of 1978, ISBN 0-04-784011-0). The rst edition was published in 1968; this is thus one of our very few sources outside LotR that were published in Tolkiens own lifetime, which lends it extra authority (for when something had been published, he would normally consider it a xed and unchangeable part of the mythos). While RGEO is basically a song cycle (Tolkiens poems with music by Donald Swann), Tolkien also included quite extensive notes on two Elvish poems occurring in LotR, Namri and the Sindarin hymn A Elbereth Gilthoniel a e (RGEO:6676). Besides writing them out in Fanorian script, he also e provided an interlinear translation of both; this allows us to know with certainty which word means what. He also rearranged Namri a e into a clearer prose version, as an alternative to the poetic version in LotR providing us with a unique opportunity to compare poetic style and prose style in Quenya. Hence I will sometimes refer to the prose Namri . a e UT: Unnished Tales (1980, ISBN 0-04-823208-4). A posthumously published collection of material supplementing and sometimes eshing out the stories of LotR and Silm, though as the title implies, not all of it was ever nished by the author. Of particular interest to students of Elvish is Cirions Oath found in UT:305: Vanda sina termaruva Elennanro alcar enyalien ar Elendil vorondo voronw. Nai o e e tiruvantes i hrar mahalmassen mi N men ar i Eru i or ily a u e mahalmar e tennoio. The (not entirely literal) translation given a in the text goes: This oath shall stand in memory of the glory of the Land of the Star, and of the faith of Elendil the Faithful, in the keeping of those who sit upon the thrones of the West and of the One who is above all thrones for ever. Tolkien added some interesting notes about the Quenya words (UT:317), allowing us to analyze the Oath itself. Letters: The Letters of J. R. R. Tolkien (1981, ISBN 0-04-440664-9). Edited by Humphrey Carpenter, Tolkiens biographer, this collection of letters also contains some linguistic information. Readers of LotR occasionally wrote to Tolkien asking questions touching on the samples of Quenya and Sindarin found in that work, and this being Tolkiens favourite subject, he often wrote fairly detailed answers. Among other things, Letters provides translations of some Elvish samples that are not translated in the LotR itself, e.g. Aiya Erendil Elenion Ancala ima = hail Erendil brightest of stars (Letters:385; we have already a quoted the translation of the Cormallen Praise in Letters:308). 22
MC: The Monsters and the Critics and Other Essays (1983, ISBN 0-04-809019-0). This book contains Tolkiens essay A Secret Vice (MC: 198223), in which he sets out his thoughts and theories about language-construction in general. He also included some Elvish poems, most notably Oilima Markirya or The Last Ark, which is listed in several versions. The version of Markirya that is most interesting to people studying the kind of Quenya exemplied in LotR, is found in MC:221223 (including some valuable annotation). Having edited and published Silm, UT and MC from the papers his father had left behind, Christopher Tolkien commenced what would become a highly ambitious project. In the period 19831996, he published a series of no less than twelve volumes, demonstrating how his father had developed his world-famous narratives over many years. The History of Middle-earth series (HoME) presents the many layers of manuscripts, chronicling how the Silmarillion and LotR as we know them gradually came into being, and also presenting other materials relating to Tolkiens vast mythology. For convenience I will list all the volumes of HoME by their standard abbreviations, though I will not actually quote from each and every one of them: LT1: The Book of Lost Tales 1 (1983, ISBN 0-04-823231-5) LT2: The Book of Lost Tales 2 (1984, ISBN 0-04-823338-2) LB: The Lays of Beleriand (1985, ISBN 0-04-440018-7) SM: The Shaping of Middle-earth (1986, ISBN 0-04-440150-7) LR: The Lost Road (1987, ISBN 0-04-440398-4) RS: The Return of the Shadow (1988, ISBN 0-04-440669-X) TI: The Treason of Isengard (1989, ISBN 0-261-10220-6) WR: The War of the Ring (1990, ISBN 0-261-10223-0) SD: Sauron Defeated (1992, ISBN 0-261-10305-9) MR: Morgoths Ring (1993, ISBN 0-261-10300-8) WJ: The War of the Jewels (1994, ISBN 0-395-71041-3) PM: The Peoples of Middle-earth (1996, ISBN 0-216-10337-7) Each of these books provide clues to the structure of Tolkiens languages, though often in a somewhat incidental fashion (Christopher Tolkien included relatively little of his fathers narrowly linguistic writings, which being highly technical would be of limited interest to the general readership). For people interested in Tolkiens languages as they appear in LotR, the most important volumes of HoME are LR, WJ and SD, which any serious student of these languages should have in his or her private library. The only long Quenya text occurring in HoME, F riels Song, is found in LR:72 but more importantly, these books reproduce three important source documents that I 23
will often refer to simply by name (as do most students of Tolkiens linguistic creation). Therefore, they will be briey described here. We are talking about the Etymologies and the essays Quendi and Eldar and Lowdhams Report. 1. The Etymologies (called Etym for short) is found in LR:347400. (I should mention that there are dierent editions of LR around, so there is unfortunately more than one pagination; my page references are to the edition normally used by Tolkien-linguists.) To casual readers undoubtedly the most baing document in the entire HoME series, this is our most important single source of Elvish vocabulary. However, it is not a regular dictionary. It is an alphabetical list of about six hundred primitive bases or roots, the various entries listing actual words derived from these roots as they appeared in later Elvish languages (sometimes the actual ulterior primitive Elvish form is also mentioned, closely reecting the base itself). For instance, under the entry alak (LR:348), itself dened rushing, we nd this series: *alk-w swan: Q alqa; T alpa; ON alpha; N alf. Tolkiens idea is a that the Primitive Elvish word alk-w developed into Q[uenya] alqa, a T[elerin] alpa, O[ld] N[oldorin] alpha and N[oldorin] alf. The Etymologies was written in the second half of the thirties, and the spelling and general concepts dier somewhat from the scenario of the LotR as published in the early fties. (If we were to update the sample just quoted, we must read Sindarin for Noldorin, and Quenya alqa and Noldorin/Sindarin alf should rather be spelt alqua and alph, respectively both words, so spelt, are actually attested in later writings.) Despite the fact that the Etymologies in some respects reects a somewhat outdated linguistic scenario, Tolkien undertaking important revisions after he wrote Etym, it is still a gold-mine of information (and as we have just demonstrated, it can to some extent readily be updated in accordance with Tolkiens later ideas). Of all the languages Tolkien mentioned in Etym, Quenya is in any case among the tongues that were not very signicantly aected by his subsequent revisions. (In the case of Noldorin, on the other hand, he would tinker with its phonology and imaginary evolution, and drastically alter its internal history, to produce Sindarin as we know it from LotR.) 2. Quendi and Eldar (sometimes Q&E for short) is found in WJ:360 417. This is ostensibly an essay on the Origin and Meanings of the Elvish words referring to Elves and their varieties. With Appendices on their names for other Incarnates. This ground is certainly covered, but luckily (from our point of view!) there are many digressions, appendices and notes that provide much extra information about the Elvish languages as Tolkien had come to see them in the post-LotR period: This essay dates from ca. 195960. Christopher Tolkien felt that 24
one substantial section departed too radically from the stated subject of the essay, and edited it out (see WJ:359, 396). Luckily, the omitted section was later published in the journal Vinyar Tengwar, issue #39. When I quote from Quendi and Eldar, I will therefore sometimes refer to WJ and sometimes to Vinyar Tengwar (VT). Digressive though the section that appeared in VT may be, it is of course of immense interest to people studying Tolkiens languages. 3. Lowdhams Report, or in full Lowdhams Report on the Adunaic Language, can be found in SD:413440. As the title implies, this report is mainly concerned with another language than Quenya: Adunaic (in the LotR appendices spelt Adnaic), the vernacular of Nmenor. However, u u a little information about Quenya, which in this report is referred to as Avallonian, can also be gleaned the two languages sometimes being compared or contrasted. (Lowdham is just a ctional character of Tolkiens. Tolkien sometimes presented even quite technical information about his languages as if he were merely quoting or referring the observations and viewpoints of various people residing within his mythos. Among his favourite ctional sources we nd Fanor, e the greatest but also the proudest of the Noldor, Rmil the sage of u Tirion, and Pengolodh the loremaster of Gondolin: Many of Tolkiens characters seem to share their authors interest in mysterious scripts and strange languages.) The sources so far mentioned are the ones published or edited by Tolkien himself or by his son except for Letters, that was edited by Humphrey Carpenter. In addition there are a few works edited and published by others. Some very brief scraps of information can be extracted from J. R. R. Tolkien Artist and Illustrator, edited by Wayne Hammond and Christina Scull. The results of the Editorial Team, scarce but not unimportant, must also be mentioned. The journal Vinyar Tengwar (VT), edited by Carl F. Hostetter, had its golden age in the period 198893, when the editor managed to uphold continuous bimonthly publication. When Hostetter and the other ET members in the early nineties started to receive original Tolkien material of the utmost interest to be edited and published, the rate of publication mysteriously dropped to about one issue a year, and this situation has continued throughout the latter half of the nineties and into the new decade. Not all of the few issues that have been published include any new Tolkien material, and those that do are usually devoted to very short scraps (that are moreover samples of very early material that is often clearly not LotR-compatible). A few issues do stand out, however, and one of them has already been mentioned: In issue #39, July 1998, Hostetter published the part of Quendi and Eldar that Christopher Tolkien left out of WJ, as well as the companion essay Osanwe-kenta (the latter is not strictly linguistic by its subject, but Tolkien nonetheless mentioned quite a few Quenya words). Some useful ma25
terial also appeared in issue #41, July 2000, lling certain annoying gaps in our vocabulary (in particular regarding the verb can) and providing interesting new information about the formation of the present tense. In January 2002, various Quenya translations of the Pater Noster and the Ave Maria were published in issue #43; Tolkien the Catholic produced more than one Elvish version of these texts. The other main results of the ETs editing eorts for the most part consist of wordlist material: the Gnomish Lexicon (GL) and the Qenya Lexicon (QL, a.k.a. the Qenyaqetsa, abbreviated QQ). The GL also mentions a few Qenya words (just like the QL mentions some Gnomish words; the languages are not infrequently compared or contrasted). Regarding Qenya as opposed to (LotR-style) Quenya, see below. These Lexicons were published in issues #11 and #12 of the journal Parma Eldalamberon, 1995 and 1998, respectively. They were originally written during World War I, when the earliest form of Tolkiens mythos began to take shape: The QL manuscript is usually dated to 1915, the GL to 1917. Substantial excerpts were published already in 198384, Christopher Tolkien drawing heavily on the Lexicons in the appendices to LT1 and LT2. Prexed to each Lexicon, Parma also published some related material: a never nished Gnomish Grammar in #11, and some phonological descriptions for Qenya in #12. Of the actual Quenya samples so far mentioned, I shall often refer to Namri, Treebeards Greeting, Elendils Declaration, Cirions Oath, F a e riels Song and Markirya simply by title or name, not always providing a reference to book and page. From the discussion above the student will know where they are found (if you feel the urge to check the accuracy of my quotes!) In most other cases I will provide a reference when I quote something, since it will usually be found in one of the sources that do allow a precise reference to book and page (since there is not a spate of dierent editions with diering pagination around). When I refer to entries in the Etymologies (in LR), I simply quote the entry-head, which can easily be located in all the various editions (irrespective of diering pagination).
0.4
Scattered in the sources listed above we have a total Quenya corpus that would amount to very roughly 150 pages if it were all brought together (though most of this would unfortunately just be unconnected wordlist material; the samples of actual text are much rarer and could probably be tted into no more than two or three pages). But here a word of warning is in place: If you want to learn the kind of Quenya that you have encountered in LotR, not all of the samples found in this corpus are entirely reliable even though they are certainly genuine Tolkien. To avoid what is poten26
tially a quagmire of confusion, the student should immediately internalize one fact: The kind of Quenya exemplied in LotR is not the only kind of Quenya there is. If you start analyzing all the samples of Quenya that we now have, you will soon realize that they do not form a homogenous mass. Most samples certainly look much the same, never straying too far from the Finnish-inspired word-shapes, but much of the early material (never published during Tolkiens lifetime) can be shown to employ or presuppose words, inectional endings and grammatical rules that dier from the system of LotR-style Quenya. By all means, no sample is entirely dierent from LotR-style Quenya but in material predating the mid-thirties, neither is there any sample that is entirely identical. Tolkien was, so to speak, all too good when it came to devising languages. Fixing them in one clear-cut and unchangeable form was an almost impossible task form him. There were ever new ideas that he wanted to work into their structure, even if these ideas contradicted and obsoleted things that he had written earlier. We can be sure that his ctional character Lowdham speaks for Tolkien himself (SD:240): In making up a language you are free: too free [. . .] When youre just inventing, the pleasure or fun is in the moment of invention; but as you are the master your whim is law, and you may want to have the fun all over again, fresh. Youre liable to be for ever niggling, altering, rening, wavering, according to your linguistic mood and to your changes of taste. With the posthumous publication of many of Tolkiens writings, we got evidence for plenty of niggling, altering, rening, and wavering on his part. It is now evident that Quenya appeared in many incarnations, and while they all share the same general style and would probably look the same to a fresh student, they actually dier in many details of grammar, vocabulary and even phonology. A powerful demonstration of the extent of Tolkiens revisions is provided by the Markirya poem, which exists in one version dating from the early thirties (MC:213) and another that is about forty years younger, dating from the last decade of Tolkiens life (MC:221223). Both versions have (almost) the same meaning, but the late version is in the full sense of the word a translation of the former, not a mere rewriting: Only a few words and inectional endings are common to both texts. Since Tolkien in pre-LotR sources typically used the spelling Qenya instead of Quenya (though the intended pronunciation is the same), I and others often use Qenya (preferably in quotes) as a name of early variants of Quenya that are more or less dierent from the form that appears in LotR and later sources. The rst version of Markirya I would thus call a Qenya poem; only the later version is Quenya as we know it from LotR. Some other poems reproduced in MC (Nieninqe and Earendel, pp. 215216), as 27
well as an alternative Last Ark poem of another meaning than the classical Markirya (MC:221), are also decidedly Qenya rather than Quenya. These texts may certainly be enjoyed for their own qualities, but as sourcematerial for students trying to gure out the structure of LotR-style Quenya they exclude themselves. As we would expect, the language generally becomes more and more similar to its nal form the closer we get to Tolkiens writing LotR. For instance, the relatively late text F riels Song is almost, but not quite LotRstyle Quenya. However, one should not have a simplistic view of these things, thinking that Tolkien started out in 1915 with a language that was wildly dierent from the Quenya of LotR and that it gradually evolved into LotRstyle Quenya in a nice and tidy evolutionary line. The scarcity of published material does not allow us to follow the process in any detail, but it is already evident that Tolkien kept changing his mind back and forth, not only doing revisions but frequently also undoing them later: Indeed some of the very earliest material, written during WWI, gives an overall impression of being more similar to LotR-style Quenya than certain Qenya poems of the early thirties. It may seem that Tolkien, rather than boldly progressing toward LotR-style Quenya, made a series of detours on the way, sometimes venturing o into radical revisions that eventually proved dissatisfying and were rejected. Yet in other cases certain revisions proved durable, Tolkien evidently perceiving them as genuine improvements but the whole process was wholly unpredictable, for in a game like this there could be no imaginable objective criteria for what constitutes an improvement: As Tolkien had Lowdham saying, Your whim is law. Something really close to LotR-style Quenya seems to have made its rst appearance in the latter half of the thirties, with the writing of the Etymologies. But it is not to be thought that everything was completely settled even after LotR had been written and published in the rst half of the fties; Tolkien indeed used the opportunity to tinker just a little with even the published samples of Quenya in this work when a revised edition appeared in 1966 (and even more niggling was certainly going on behind the scenes). Seven years later he died, and there is little reason to believe that he ever managed (or even seriously tried) to x Quenya and his other languages in one denite cut-in-stone form sorting out every detail. Members of the Editorial Team have indicated that Tolkiens later manuscripts bear witness to one last phase of intense experimentation, but apparently, no nal or denite version of the language ever emerged. This was not necessarily a failure, like a composer never managing to nish his great opera: Unceasing change, often frustrating to those who study these languages, was inherent in this art, Christopher Tolkien observes (SD:440). In another place, he remarks regarding his fathers work on the languages that it seems indeed that they very attempt to write a denitive account produced immediate dissatisfaction and the desire for new constructions: so the most beautiful manuscripts 28
were soon treated with disdain (LR:342). Insofar as delight lay in the creation itself, Tolkien could not write a denitive account, or his fun would be past and over. Nonetheless, if compared to Tolkiens intense experimentation in the twenty years from 1915, Quenya does seem to have entered a somewhat stable phase in the second half of the thirties. Over the next decade Tolkien wrote LotR, which included some samples of Quenya as it now appeared (most notably Namri ). With the eventual publication of LotR in 195455, these a e forms became a xed part of the mythos (despite Tolkiens slight tinkering in the 1966 revision). Having published LotR, Tolkien obviously could not revise his languages anything as freely as he could before. Reportedly, there are hints in his post-LotR manuscripts that he indeed felt somewhat constrained. But this relative stability would later be good news for people wanting to learn or study the Quenya, Tolkiens more-or-less nal decision on how this language had really been back in the remote ages chronicled by his narratives. Some (including me) have referred to this as mature Quenya. Others feel this term to be unduly disparaging to the earlier forms of Quenya or Qenya, since the inevitable implication is that they were somehow immature and inferior. Artistically, subjectively speaking I do think the nal form of Quenya is more attractive than Tolkiens earlier experiments, and there can be no doubt that this is the kind of Quenya that most students will primarily want to learn not the earlier variants that Tolkien himself rejected. For that matter, this is certainly the version of Quenya that Tolkien himself would have wanted us to study; if it had been up to him, we would never have seen any other versions! He took the utmost care to ensure that his mythos would remain free from internal contradictions, and he would never have recognized contradictory variants of Quenya as being somehow equally valid. Indeed it should be noted that elderly Tolkien referred to his earliest form of Qenya as very primitive (PM:379). Hence we have no choice but to treat the early material with considerable caution, and there is little reason to believe that Tolkien would have been greatly insulted if others were to say (or indeed to agree!) that his early Qenya variants are not quite as attractive as his later, carefully rened version(s) of the language. Even so, in this course I have opted to speak, not of mature Quenya, but rather of LotR-style Quenya. The latter term must be wholly uncontroversial. The language that this course teaches is of course LotR-style Quenya, as well as it can be approximated at the present stage but there is no point in pretending that the various earlier Qenya variants never existed. I will indeed refer to some of their features, to give the student some idea of what kind of variations occur in the material. Apart from such academic considerations, the early material is something we may fall back on where material closer to (and ideally postdating) the writing of LotR is insucient for our needs. In particular, we may cannibalize the Qenya material for 29
useful vocabulary items, in each case making sure that the words we carry over into LotR-style Quenya t this tongue (i.e., they must not clash with later words of dierent meaning, and the shape of the words themselves must t the phonology and derivational system of the language as Tolkien eventually came to envision it). After all, all the incarnations of Q(u)enya in the entire period from the language was invented in 1915 and until Tolkiens death in 1973 may well be seen as endless variations on somewhat the same themes. Therefore it is in a way only tting that in our attempts to develop a useable form of Quenya, we take everything into consideration. But as for the overall grammatical and phonological structure, we must give priority to Tolkiens vision as it manifested in LotR and writings postdating this work: If we have any respect whatsoever for Tolkiens intentions, the form of Quenya that we attempt to crystallize must be LotR-compatible. Little can be easy or clear-cut in this strange corner of Language. Reconstructing the structure of Quenya is like trying to piece together a huge puzzle of far-sundered pieces. Many pieces are simply missing, vast amounts of material being inaccessible to scholars (and to make matters worse, the ones who are supposed to be publishing it often seem far more concerned about concealing it). Moreover, because of Tolkiens frequent revisions you cant even be sure that all the pieces you do nd belong to the same puzzle at all. Some clearly do not t and can be ignored; many other fall in the category of doubtful, and you dont really know what to do with them. In this course I will mention some of the variations and present my hopefully qualied guesses as to what we should accept as authoritative and what is probably best ignored. Indeed, due to the general lack of explicit grammatical information from Tolkien, I will not always present Quenya grammar with condence and authority; rather you will often see me review whatever evidence is available and try to make out some rules that we can follow when putting together our own Quenya compositions. But in a way this is precisely what I should like to do anyway, so as to acquaint students with the kind of deductions that the eld of Tolkienian linguistics is all about at this stage. Concerning some material I published earlier, Ive had (gentle) complaints to the eect that I merely listed my conclusions without showing what they were based on, somewhat dogmatically asserting that this is how it is, take my word for it. I think this style was unavoidable in a brief presentation, but here I will in many cases avail myself of the opportunity to go back to the primary sources and really demonstrate what kind of deductions underlie everything. Precisely because Tolkiens Quenya is a somewhat uid entity, xed in general outline but with endless contradictory variations when it comes to the details, we can to some extent feel free to crystallize our own standard (not making it more dicult than we have to). As long as we piece together a usable system from elements Tolkien provided, even though there is no way we can accept all the known variations within a single, unied system, 30
the resulting language will be real Quenya to the extent such a thing can exist.
0.5
Spelling conventions
Over the decades, Tolkiens spelling of Quenya varied in certain details. As discussed above, just about every aspect of Quenya was somewhat variable, but unlike the unstable grammar, the spelling variations are not very consequential: In theory our alphabet is not the writing native to Quenya anyway. Tolkien was merely hesitating on how to best render into our own letters the supposed original Elvish script (the Tengwar, also called Fanorian writing a singularly beautiful script that Tolkien dee vised with the same loving care as the languages themselves). In this course, a consistent spelling has been imposed on the material, mostly based on the spelling used in LotR (I say mostly based because the spelling used in LotR is not entirely consistent either, but it is close!) Concerning the spelling used in LotR, Tolkien wrote: The archaic language of lore [namely Quenya] is meant to be a kind of Elven-latin, and by transcribing it into a spelling closely resembling that of Latin (except that y is only used as a consonant, as y in E[nglish] Yes) the similarity to Latin has been increased ocularly (Letters:176). I will outline the spelling conventions used in this work. What follows is not something a fresh student needs to carefully internalize. People who want to study Quenya should nonetheless be aware of the major spelling inconsistencies in the primary sources. Guided primarily by the spelling Tolkien used in LotR, I have regularized the following features:
is to be distributed over the Internet; vowels with macrons may be replaced by various weird symbols if the software of the recipients is not overly fond of linguistics.
C rather than K
If you bothered looking up the reference I gave for the sentence Anar caluva tielyanna above (Unnished Tales p. 22), you may have noticed that in the source, the middle word is actually spelt kaluva instead. In Quenya, k and c represent the same sound (pronounced K ); Tolkien just couldnt make up his mind which letter to use. In pre-LotR sources, such as the Etymologies and the early Qenya Lexicon, he mostly used k (though in a few cases, c pops up in these sources as well). Since the original inspiration for Quenya was Finnish, and Finnish orthography employs the letter k, it is not surprising that Tolkien originally preferred that grapheme. But as is evident from Letters:176 quoted above, he later decided that in LotR, he would spell Quenya as similar to Latin as possible. Guided by Latin orthography, he started to use the letter c instead of k: I decided to be consistent and spell Elvish names and words throughout without k (Letters:247). For instance, the word for metal had been spelt tinko in the Etymologies (entry tinko), but in LotR Appendix E, the same word with the same meaning appears as tinco instead. Hence we regularize k to c throughout. It is a curious fact that Tolkien, even in sources that postdate LotR, in many (indeed most) cases reverted to the use of k. His writings are quite inconsistent on this point. A word for Dwarf is given as Kasar with a k in WJ:388; yet on the next page Tolkien switches to c when quoting the Quenya name of Moria: Casarrondo (Dwarf-cave or Dwarf-hall). A word for house appears as ka in WJ:369 (karya his house), but in MR:250 o o the same word is spelt with a c in the compound cacalina light of the o house (an Elvish expression for the soul inside the body). In some late notes published in VT41:10, Tolkien mentioned a word ruskuite foxy using the letter k rather than c, but immediately afterwards he wrote down a word calarus polished copper using c rather than k. From the posthumously published Silmarillion we remember names like Melkor and Tulkas, but in MR:362, 382 the spellings used are Melcor and Tulcas. The Quenya word for horse is spelt rocco in Letters:282, but in Letters:382 we have rokko instead. Imitating Tolkiens persistent indecision in this matter would be quite pointless or even confusing. For instance, the Quenya word for bed is given in LR:363 as kaima, but in Namri in LotR, the obviously related a e word lies is spelt caita. Maintaining the inconsistent spelling out of some kind of misunderstood reverence would obscure the relationship between the words; to go with caita, the word for bed clearly ought to be spelt caima. I should mention that there are those who would regularize the material to k instead, discarding the spellings used in LotR in favour of the orthography 32
Tolkien uses in many other sources. This is only a matter of taste, and in the C or K question all writers can essentially make their own choice, but I will normally adhere to the LotR spelling. After all, the LotR is a rather central work regarding the setting Tolkien placed his languages in.
NOTE: But in the case of the title of the Markirya poem, I tend to retain k simply because the word markirya or ark only occurs in the early, Qenya version of the poem. It is not found in the later Quenya version, though I dont know what we should otherwise call it. So in this case I will leave the k in to mark this as an early Qenya word, though a form marcirya would surely work in LotR-style Quenya as well and this is the spelling I would use if I ever needed the word ark in an actual Quenya text. I guess I would normally also retain k in some names that we are very familiar with from the Silmarillion: Melkor, Tulkas, Kementri and a few others. But the Silmarillion also a employs forms like Calaquendi (rather than Kalaquendi), so there is little consistency in this work.
33
or axo bone in MC:223 so x must be seen as his nal decision in this matter. In LotR Appendix E, Tolkien refers to the combinations ts, ps, ks (x ), that were favoured in Quenya; this also seems to suggest that ks is to be represented by x in normal spelling. (No actual example of a Quenya word containing x/ks seems to occur in LotR, but as mentioned above, we have Helcarax in the Silmarillion.) e
N rather than N
In many sources, Tolkien uses the symbol n, which should not be pronounced as in Spanish orthography (e.g., as in seor ). In the transcription n [is n used for] the Fanorian letter for the back nasal, the ng of king (MR:350). e Unlike English, Quenya could originally have this ng at the beginning of words (as well as in other positions where it may also occur in English). A prominent example is the word Noldo, plural Noldor, which is so spelt in many sources. But in LotR Appendix E, Tolkien wrote that this ng or n has been transcribed n (as in Noldo) according to the pronunciation in the Third Age. The list of Tengwar names in the same Appendix conrms the development Tolkien hinted at here: the pronunciation of certain symbols of Tengwar writing was slightly changed as the long Ages of Middle-earth went by. The letters that were originally called ngoldo and ngwalme (= noldo, nwalme) were later called noldo and nwalme instead; since the letters were named after actual Quenya words containing the sound denoted by the letter, this reects a development whereby initial n- becomes normal n-. Already in the Etymologies of the mid-thirties, Tolkien hinted at a similar development: In the entry ngar(a)m, the word for wolf was listed as armo, narmo, n which is evidently to be understood as an older and a later form. MR:350 mentions a word nol lore, knowledge that is spelt with initial n- in the e gol, where it is glossed wisdom), but in the Etymologies as well (entry n Silmarillion Appendix (entry gl ) it is spelt nl. This would be the later, u oe Third Age form. We go for the Third Age form everywhere, regularizing n to n throughout. (Notice, though, that in Tengwar writing the distinction between the symbols transcribed n and n was upheld even after they had both come to be pronounced n. But this is not a problem as long as we write Quenya in our normal alphabet.) Undoubtedly the combinations ng and nc in the middle of words are also technically ng and nc, as in anga iron or anca jaw, but this pronuncation comes naturally to speakers of English and does not have to be expressly represented in writing. As far as is known, Tolkien never used the letter n before g or c in Quenya words, but only n.
34
S rather than
This is a case somewhat similar to n vs. n: Tolkien imagined that Quenya as spoken in Valinor possessed , more or less like the sound spelt th in English think. (In Valinorean Quenya it was strictly a little more s-like than the English sound, pronounced with the tip of the tongue against the upper teeth only, not between the upper and lower teeth as in English.) However, in the dialect of the Noldor, this s-like eventually turned into normal s, merging with preexisting ses (a change Fanor vehemently but vainly opposed: see e PM:331339 for an eminent example of how intertwined Tolkiens languages and narratives can be). Quenya as a ceremonial language in Middle-earth always had s, since only the Noldorin dialect was known there. In WJ:484, Tolkien mentions inde as the Quenya word for grey, pale or silvery grey, but adds that in the Noldorin (N) dialect, this became sinde. In WJ:319, we nd elma as a word for xed idea, will; in this case the later Noldorin form selma is not mentioned there or elsewhere, but we would still use the latter form here, since we are aiming for the kind of Quenya that was used in Middle-earth in the Third Age.
The diaeresis
In many cases, Tolkien adds a diaeresis, two dots, above a vowel, for instance , , in the names Erendil, Enw. This is only to clarify the a o e a o e pronunciation, primarily for readers used to English orthography. It should be emphasized that the diaeresis is not in any way necessary to write correct Quenya. Tolkien wrote about the spelling that it is only a device of e transcription, not needed in the original that is, in the supposed original Tengwar writing (PM:343). It is not really needed in the transcription either Tolkien never used it in the Etymologies and it can safely be left out in e-mail. Indeed some scholars advocate leaving it out altogether in all media, perceiving it as a superuous graphic encumbrance useful only to people who dont know the rst thing about Quenya (and to people used to the orthographies of such languages as German, Swedish or Finnish, it can be downright misleading). But I dont know; I guess I like to see the diaeresis in carefully presented texts, even if it doesnt tell me anything I dont know beforehand. It adds an exotic tint to the texts, and also represents a nod in the direction of the visual impression made by written Finnish, since Finnish orthography employs letters like a and o that however denote sounds distinct from normal a, o, which is not the case in Quenya spelling. If we are to use the diaeresis, it should however be used in a consistent way. In WJ:425, Christopher Tolkien comments on his fathers very variable use of it, so some regularization is required. (Christopher Tolkien himself has been regularizing his fathers spelling in some quotations; for instance, in PM:371 he cites the Quenya word ross ne rain, dew from the entry e
35
ros1 in Etym., but there the word is actually spelt rosse with no diaeresis.) The nal - in (say) Enw is meant to remind the reader that nal - is e o e e not silent, as it usually is in English orthography. Final e is never mute or a mere sign of length as in English, Tolkien noted in LotR Appendix E (this very sentence providing two examples of this feature of English spelling, namely mute and mere). He added that to mark this nal e is often (but not consistently) written . As he says, this spelling is not used consistently, e whether in LotR or in other sources cf. some of the words already quoted: quesse, sinde, nixe. Hereinafter, we will however be consistent about this: quess, sind, nix. (Notice, however, that the diaeresis is not used in e e e words where the nal e is also the only vowel, as in short words like te them or ve as, like both of which occur in LotR. From time to time I see some overeager dot-fan produce spellings like t and v, but while this e e is not in any way harmful, it is quite superuous: Tolkien never uses such spellings.) Since only a nal -e receives the diaeresis, the dots normally go if you add an ending to the word (or use it as the rst element in a compound), since the -e is then no longer nal. An attested example of this is provided by the word lmatyv sound-taste (individual pleasure in wordforms), the plural of a a e which is spelt lmatyver (MR:215216). We do not see **lmatyvr, a a a a e for because of the plural ending -r, the vowel e before it is not nal anymore. (Throughout this course, a double asterisk ** is used to mark a wrong form.) Appendix D in LotR likewise indicates that the plural form of enqui (the e Eldarin six-day week) is to be spelt enquier rather than **enquir. e Besides nal , we shall use the diaeresis to clarify the pronunciation of e the combinations ea, eo and oe (sc. to indicate that both vowels are to be pronounced clearly separate: e-a, e-o, o-e; hence for instance a is not to e be drawn together like ea in English heart). In the case of e + a and e + o, the diaeresis is placed above the e as long as it appears as a lower-case letter: a, o. If, however, it is to be capitalized, the dots move to the next e e letter instead: E, E (as in Erendil, Enw). Tolkiens own writings are a o a o e not consistent in this matter; we adopt the spelling used in LotR and the Silmarillion. Sometimes he places the diaeresis above a capital letter as well; for instance, the Quenya name of the universe in some texts appears as Ea (e.g. MR:7), though according to the system we just sketched it should be E a as in the published Silmarillion. (Gross inconsistency is seen in Letters:386, where Tolkien refers to the attempt of Erendil to cross Ear [the ocean] a it must be either Earendil, Ear OR Erendil, Er!) Conversely, Tolkien a a sometimes places the diaeresis over the second vowel in the group even when the rst vowel is not capitalized, resulting in spellings like e (UT:305, 317); a we would rather spell it a (as Tolkien himself did elsewhere; see VT39:6). In e a footnote in MR:206, Christopher Tolkien observes that his father wavered between Fanro and Fenro (the Quenya form of the name Fanor ); e a a a e according to the system here outlined, it should be Fanro. e a 36
In the case of oe (a very rare combination), we place the diaeresis over the , as in the example lond in LotR Appendix D (this is the name of the e e e middle day of the year in the calendar of the Elves). In Appendix E, Tolkien explicitly stated that the fact that oe is disyllabic is often indicated by writing . . . o . e In some sources, the combination ie is also broken up with a diaeresis, resulting in spellings like Ninna (name of a Vali or goddess), for ine e stance in MR:49. Yet this spelling is not used in the published Silmarillion, that simply has Nienna. The LotR itself is somewhat ambiguous on this point. In Appendix A we have the names Telperin and Silmarin so spelt e e (though Unnished Tales p. 173 has Silmarien). However, the most substantial Quenya text in LotR, Namri, does not use the diaeresis in this a e combination this text has tier, not tir, for paths (though the latter e spelling occurs in RGEO:67). In accordance with this example, as well as Nienna in the Silmarillion, we will not use the diaeresis in the combination ie. If, however, the group -ie occurs at the end of a word, the e receives the diaeresis because it is nal (wholly irrespective of the fact that it is also part of the combination ie), in accordance with the rule established above. Hence Namri, Vali rather than Namrie, Valie, and if the rst elea e e a ment of Nienna occurs by itself, we will spell it ni this is the word for e tear. Removing the plural ending -r from tier paths likewise produces ti path, since - becomes nal. e e In many post-LotR sources, Tolkien also started to break up the combination oa by means of a diaeresis (apparently to warn the reader that oa is not drawn together as in English load ). Hence we have spellings like hra o body (MR:350 and passim). Cf. also some of the words quoted above: karya, cacalina. However, in LotR Tolkien simply wrote oa. Contrast o o the spelling loa used in LotR (Appendix D: The Eldar also observed a short period or solar year . . . usually called loa) with the spelling la in o MR:426 (where the word occurs in the plural: lar upon lar = years o o upon years). Regularizing in accordance with the system used in LotR, we will not use the diaeresis in the combination oa. Hence we will here use spellings like hroa body, coa house etc. Hroa without a diaeresis is actually found in MR:399400 (and VT41:13), so we are not tampering with Tolkiens spelling, just crystallizing a standard by choosing one of the options his writings provide and carrying it through consistently. This, as I have tried to demonstrate, is true of all the regularization I impose on the material.
[As for the English orthography employed in this work, it follows the authors homemade Mid-Atlantic convention. Therefore you will nd British spellings like avour, colour (not American avor, color ) next to American spellings like analyze (not British analyse). The orthography of my native Norwegian does not in all cases match the pronunciation, but it comes rather closer to the spoken language than any variant of English spelling
37
does. Therefore, please allow me to gently mock the pointless variations introduced by wellmeaning reformers of English orthography. It is actually beyond help, so why bother?]
38
Lesson 1
The sounds of Quenya. Pronunciation and accentuation.
1.1
General remarks
Quenya as an actual entity in our own world exists primarily as a written language: Quenya enthusiasts tend to be widely scattered and must generally share their compositions via some written medium only (indeed I shall normally refer to users of Quenya as writers rather than speakers). Nonetheless, any student should obviously know what pronunciation Tolkien imagined, as well as his intentions can be approximated now. There exist a very few recordings of Tolkien himself reading Quenya texts. In a late TV interview, Tolkien writes out and pronounces the greeting elen s l menn omentielvo. More notably, he made two dierent recordings la u of Namri (sung and spoken). The spoken version is also available on the a e net: http://www.salon.com/audio/2000/10/05/tolkien elvish/index.html (under Poem in Elvish). A few lines of this version of Namri dier a e from their LotR counterparts: The recorded version has inyar unti nar o ve rmar aldaron / inyar ve lint yulmar vnier instead of yni a e a e untim ve rmar aldaron! / yni ve lint yuldar (a)vnier as in o e a e e a LotR. The recording was made before the book was published (and hence before the nal revisions). A much later recording, with the same text as in the book, also exists. I have not heard it, so I cannot comment further. The very few extant recordings are interesting, but they are not our chief source of information. Most of what we know about Quenya pronunciation is based on Tolkiens written notes about how his languages should be pronounced, predominantly the information provided in LotR Appendix E. (Indeed Tolkiens actual pronunciation in the recordings is not always quite awless according to his own technical descriptions, but then he was not a native speaker of Quenya.) Any natural language has a phonology, a set of rules dening what sounds are used, how they vary and behave, and how they can be combined. This goes for any well-made invented language as well. Quenya is most denitely 39
not a haphazard jumble of sounds; Tolkien carefully constructed its phonology both as an evolving entity (classical Quenya gradually developing from Primitive Elvish) and as a xed form (dening the kind of Quenya that was used as a language of lore and ceremony in Middle-earth). Tolkien had Pengolodh, the sage of Gondolin, observe that Elvish tongues tended to use relatively few sounds for the Eldar being skilled in craft are not wasteful nor prodigal to small purpose, admiring in a tongue rather the skilled and harmonious use of a few well-balanced sounds than profusion ill-ordered (PM:398). None of the sounds used in Quenya are particularly exotic from a European viewpoint, but they are combined in an exquisitely tidy manner. Compared to Tolkiens Elvish, many real languages indeed appear rather messy.
1.2
Basic terms
Let us get some basic terms into place (people with linguistic training need not spend much time on this section). The sounds of any language can be divided into two broad categories, vowels and consonants. The vowels are sounds made by letting the air stream freely through the mouth: Dierent vowels are produced by modifying the position of the tongue and the lips, but the stream of air is not directly obstructed. If one draws out various vowels, pronouncing aaaaa. . . or eeeee. . . or ooooo. . . , it is easy to feel how the air streams quite unhindered though the mouth: One merely congures the tongue and lips to shape the desired sound. Vowels can be more or less open or closed: You only have to notice the position of the tongue and lower jaw when pronouncing aaah. . . as contrasted with their position when you pronounce ooooh. . . to understand what is meant by this. The vowel a (as in English part) is the most open, while the vowel u (as in English rude) is the most closed. Other vowels fall between. Vowels can also be more or less rounded, mainly depending on the position of the lips; the vowel u (as just described) is said to be rounded because it is pronounced with the lips pouted. A vowel like o (as in English sore) is actually pronounced much like the a of part, but o is rounded and a is not making the vowels audibly distinct. When pronouncing vowels, the stream of air is only modied (by means of devices like the ones just described). It is never actually hindered. In the case of the consonants, the air is however more actively obstructed. Thus, Tolkien can inform us that one early Elvish term for consonant was tapta tengw or just tapta, meaning impeded element or impeded e one (VT39:7). In the most extreme cases the stream of air may even be completely halted for a moment: This is easily perceived in the case of a consonant like p, which is pronounced by bringing the lips into contact, momentarily cutting o the stream of air from the lungs and allowing a
40
pressure to build up inside the mouth. Then the lips are suddenly parted again, releasing the air in a small explosion and this explosion constitutes a p. Such plosive consonants include t, p, k and their counterparts d, b, g (sc. hard g as in gold, not as in gin). They are all formed by halting and then suddenly releasing the air various places in the mouth. Instead of halting the air completely one may also let it zzle through a small opening, as when f is pronounced by forcing the air out between the lower lip and the upper teeth; such friction sounds are called fricatives (or spirants) and include consonants like f, th, v. And there are yet other options on how to manipulate the stream of air, such as rerouting it through the nose to produce nasal consonants like n or m. The concept of voicing should also be understood. Humans (and, it would seem, Elves) come with a kind of buzzing device installed in their throats, namely the vocal chords. By making the vocal chords vibrate, one may add voice to the stream of air before it enters the speech organs proper. The presence or lack of such voicing is what distinguishes sounds like v vs. f. If one draws out a sound like ffff . . . and suddenly turns it into vvvv. . . instead, one will feel the buzzer in the throat kicking in (put a nger on your glottis what in men is called the Adams apple, less protuberant in women and you will actually feel the vibration of the vocal chords). In principle, the device of voicing could be used to double the number of sounds we are able to produce, since they could all be pronounced either with vibration in the vocal chords (as voiced sounds) or without such vibration (as unvoiced sounds). In practice, most of the sounds of speech do not appear in unvoiced versions. Many sounds would barely be perceptible without the voicing (n, for instance, would be reduced to little more than a weak snort). Normally all vowels are voiced as well, certainly so in Quenya (though in Japanese, vowels may lose their voicing in certain environments). But I have already referred to d, b, g as the counterparts of t, p, k; they are counterparts in the sense that the former are voiced and the latter are not. One characteristic feature of Quenya (at least the Noldorin dialect) is the very limited distribution of the voiced plosives d, b, g; they occur solely in the middle of words, and then only as part of the consonant clusters nd/ld/rd, mb, and ng. Some speakers also pronounced lb instead of lv. (Possibly Tolkien imagined dierent rules for the poorly attested Vanyarin dialect of Quenya: The Silmarillion refers to a lament called Aldudni e e made by a Vanyarin Elf; this word has puzzled researchers since the middle d occurs in a position that would be quite impossible in Noldorin Quenya.) Syllables: Made up of vowels and consonants, speech is not an undierentiated outburst of sound. Rather it is perceived to be organized into rhythmic units called syllables. The shortest possible words are necessarily monosyllabic, having only one syllable like English from or its Quenya equivalent ho. Words of more than one syllable, polysyllabic ones, form longer strings of rhythmic beats. A word like faster has two syllables (fas-ter ), a word like 41
wonderful has three (won-der-ful ), a word like geography has four (ge-ogra-phy), and so on though obviously we cant go much further before the words would be felt to be impractically long and dicult to pronounce. Some oriental languages, like Vietnamese, show a great preference for monosyllabic words. But as is evident from the English examples just quoted, European languages often employ longer words, and Tolkiens Quenya makes extensive use of big mouthfuls (as does Finnish). Consider words like Ainulindal or e Silmarillion (ve syllables: ai-nu-lin-da-l, sil-ma-ril-li-on). An unine ected Quenya word typically has two or three syllables, and this number is often increased by adding inectional endings, or by compounding.
1.3
In Quenya, the basic vowels are a, e, i, o, u (short and long). They may also be combined into diphthongs, groups of two basic vowels pronounced together as one syllable: There are three diphthongs in -i (ai, oi, ui) and three in -u (au, eu, iu, though the diphthongs eu and iu are quite rare). The consonants of Third Age Quenya may be listed as c (= k), d, f, g, gw, h, hy, hw, l, ly, m, n, nw, ny, p, qu, r, ry, s, t, ty, v, y and w (this listing is not wholly uncontroversial; the consonant system of Quenya can be plausibly analyzed in more than one way). In Elvish writing, the Tengwar orthography also upholds the distinction between some consonants that by the Third Age had come to be pronounced alike and thus merged altogether ( merging with s, while initial n fell together with n see the discussion of spelling conventions). In the transcription and spelling employed in this course, the former presence of lost distinct consonants is reected in two cases only: hl and hr, that were originally unvoiced l and r, but later they merged with normal l, r (and are therefore not included on the list of Third Age Quenya consonants above). Thus we will spell, say, hr e (winter) v in this way despite the fact that Tolkien imagined the typical Third Age pronunciation to be simply r e (with a normal r). v Though the consonants hy, gw, hw, ly, nw, ny, ry, ty, and qu (and hr, hl) must here be written as two letters (as digraphs), they should evidently be taken as unitary sounds: Their pronunciation will be discussed in greater detail below. The digraphs in -w represent labialized consonants, while the digraphs in -y stand for palatalized consonants; again, see below for further discussion of these terms. It should be understood that qu is simply an aesthetic way of spelling what would otherwise be represented as cw (most people will agree that Quenya looks better than Cwenya), so qu, like nw, is a labialized consonant. When counting syllables one must remember that there is no actual vowel u in qu; u here stands for w. A word like alqua (swan) thus has only two syllables: al-qua (= al-cwa). One must not think al-qu-a and conclude that there are actually three syllables. In
42
Tengwar writing, qu is denoted by a single letter, and in most early sources, Tolkien also used the single letter q to represent it. Double consonants: Some consonants also occur in long or double versions; double vs. single consonants may be compared to long vs. short vowels. The obvious cases, sc. the double consonants directly represented in orthography, are cc, ll, mm, nn, pp, rr, ss and tt (e.g. ecco spear, colla cloak, lamma sound, anna gift, lappa hem of robe, yarra- to growl, ess name, atta two). The group pp is very rare, only attested in mae terial far predating the LotR. In the Note on Pronunciation appended to the Silmarillion, Christopher Tolkien noted: Consonants written twice are pronounced long, thus Yavanna has the long n heard in English unnamed, penknife, not the short n in unaimed, penny. Words like tana that vs. tanna sign, tyel ceases vs. tyell grade, ata again vs. atta two e e should be audibly distinct. It is possible that some of the consonants written as digraphs must also be counted as double consonants when they occur between vowels; e.g. ny = long or double palatalized n (more on this below). Consonant clusters (vs. single consonants): It is dicult to pronounce many sequential consonants, so the languages of the word generally conne themselves to relatively small groups (or clusters) of consonants. The most typical word, from just about any language, is a series of vowels and consonants (single ones or relatively short consonant clusters) alternating the core of each syllable usually being a vowel. Tolkiens Quenya is no exception; this language actually has quite restrictive rules for how consonants and vowels can be combined into syllables and longer words. Even so, consonant clusters are quite common, but they are not distributed as freely as in English. While English and for that matter Sindarin allow consonant clusters at the beginning of words, Quenya does not (SD:417418). A word like scream, commencing with a cluster of no less than three consonants, would be quite impossible in Quenya. Tolkien noted that the name that the Woses or Wild Men had for themselves, Drughu, was adapted to Quenya as R (UT:385). Quenya could not preserve the initial cluster dr - of the origu inal form of this loan-word (even apart from the fact that Quenya could not have d in this position). Quenya does allow a limited number of consonant clusters medially, between vowels in the middle of words; among frequent of favoured clusters Tolkien cited ld, mb, mp, nc, nd, ng, ngw, nqu, nt, ps, ts and x (for cs). Hence we have such typical Quenya-style words as Elda Elf, lamb tongue, tumpo hump, ranco arm etc. Finally, e at the end of words, only ve single consonants may occur: only -l, -n, -r, -s, or -t is permitted in this position (Letters:425; however, most Quenya words end in a vowel). Consonant clusters or double consonants are not normally found at the end of words, though they may occur if a nal vowel drops out (is elided ) because the next word begins in the same or a similar vowel. Hence in LotR we have a nal nn in the phrase l menn omenu tielvo (on the hour of our meeting), but only because this is reduced from 43
l menna omentielvo (this full form occurring in WJ:367 and Letters:424). u The only genuine consonant cluster occurring at the end of a word seems to be nt used a specic grammatical ending (dual dative, to be discussed in later lessons) e.g. ciryant for a couple of ships, formed from cirya ship. Tolkiens earliest Qenya experiments, as recorded in the Qenya Lexicon of 1915, were more liberal in this respect. Qenya allowed more nal consonants and even nal consonant clusters, but as LotR-style Quenya evolved in Tolkiens notes, he tightened up the phonology. Thus he gave the language a more clearly dened avour.
1.4
Pronunciation
Vowels: Quenya vowels are pure. For people who want to pronounce Elvish vowels with some degree of accuracy, Tolkien recommended Italian vowels as a model (as did Zamenhof for Esperanto, by the way). Speakers of English have an ingrained habit of blurring many vowels, especially when they are not fully stressed; hence in a word like banana it is typically only the middle A that comes out as a proper A-sound. The two other As, that are not stressed, are typically made to sound like a blurred, obscure reduction vowel that linguists call a schwa (from a Hebrew word for nothingness; English textbooks sometimes prefer the spelling shwa). But in Quenya all vowels, in all positions, must be clearly and distinctly pronounced; any tendencies to blur them must be strongly resisted. As we remember, Quenya has both long and short vowels, the long ones being marked with an accent: , , , u, vs. short a, e, o, u, i. Long and a e o short vowels must be kept apart and pronounced clearly distinct. Sometimes vowel length is the only thing that makes otherwise similar words distinct: for instance, cu with a short u means dove, whereas c with a long u u means crescent. Long can be sounded as in English father : m hand, nr ame, a a a e quco crow. However, English does not have anything corresponding to a Quenya short a. It is absolutely necessary to master it, for short a is by far the commonest of Quenya vowels. Tolkien noted that it should be more open than the long . What we want is a vowel that by its sound (or a quality) is about midway between the as of English father and English cat but as for its length (or quantity), it should by all means be short as in the latter word. The vowel heard in Spanish padre will do. Speakers of English may pin down a short a by isolating the rst part of the diphthong ai as in aisle.
NOTE: If you have the original Star Wars movie available, listen carefully when Harrison Ford rst appears about 45 minutes in and introduces himself as Han Solo: Ford actually produces a nice Quenya-style short a in Han, making this syllable sound as it would in Quenya words (e. g. hanu a male or handa intelligent; apparently there is
44
even a Quenya word han beyond). But later in the SW movies, the vowel of Han is inconsistently pronounced either with a long a as in English father or with the vowel heard in English cat, which is precisely the vowel to be avoided in Quenya. Linguistic consistency was never the, ahem, force of Star Wars. By the way, do you remember Endor, the green moon where George Lucas placed his reinvented teddy bears in the third movie? Guess what the Quenya word for Middle-earth is! Lucas would surely say that his intention was to pay tribute to Tolkien. . . UPDATED NOTE: Now that Peter Jacksons The Fellowship of the Ring has appeared, I can quote examples from the soundtrack of this movie as well; most people interested in Tolkiens work will surely have seen it, and many are also going to buy it on video or DVD. Good examples of short Elvish a occur in the Sindarin name Caradhras Redhorn as pronounced by Christopher Lee (Saruman) in the scene where his spying crows return to Isengard: So, Gandalf, you try to lead them over Caradhras. . . Lee also gets the short as more or less right in a scene following shortly afterwards, when he stands on the top of Isengard reading a Quenya invocation: Nai yarvaxa rasselya taltuva nottoe carinnar. . . (but the last word sounds almost like crinnar, the rst vowel being long a after all, Chris Lee is not a native speaker of Quenya!)
An extra challenge for speakers of English is to pronounce -a as a full vowel at the end of words. Where English orthography has a nal -a, it is normally pronounced like a schwa. Contrast the English and the Spanish pronunciation of the nal vowel in a name like Sara; in Spanish, the Englishstyle reduction or blurring of the -a does not take place. In one very early source, Tolkien actually stated that Qenya, like English, turned nal, unaccented -a into a schwa (as in English drama, QL:9), but there is nothing to suggest that this idea was still valid decades later when he wrote the LotR. Indeed even the early source just referred to has it that there was one important dialect of Qenya where the weakening of nal -a did not take place. So speakers should try to pronounce a full a in all positions: neither of the as in a word like anna gift should be pronounced as in the English name Anna. Long is another Quenya sound that does not occur in contemporary e English. The long e of English became long i (like Quenya centuries ago ) though because of this descent it is still often spelt ee, as in see. Quenya e has the value of German eh as in Mehr. The pronunciation of ai in English air at least approaches , but this is really a short e followed by a schwa. e Tolkien notes that long should be closer than short e (see LotR Appendix e E), so just lengthening the vowel heard in English end will not be quite sucient. The quality of the vowel should be about midway between the vowels heard in English end and English see, but it should be long like the latter: nn water, r day, mna region. e e e Short e may be pronounced as in English end. In Quenya this sound also occurs in nal position. Since word-nal e is usually silent in English orthography, Tolkien often used the spelling in this position and throughout e this course, this spelling is employed consistently. This is only to remind 45
English readers that in Quenya, this letter is to be distinctly pronounced. But since word-nal e never occurs in spoken English, some speakers tend to substitute i or ey (following English practice in the rare cases of a nal orthographic e being sounded, as when Jesse is pronounced jessi, or karate is pronounced karatey). Quenya e should have the value described above in all positions. It must NOT be pronounced i, nor must there be a y-like sound creeping after it: lm night, mor black, tinw sparkle. o e e e Long is pronounced as in English machine, that same as ee in English see: the Quenya word s (now) is similar in sound. Other examples include n woman and r s ma edge. This long must be noticeably longer than short i, which may be pronounced like in English pit: Titta tiny, imb e between, vinya new. In one early source, Tolkien himself quoted the word pit as an example of short Qenya i (QL:8). Later writings suggest that the quality of the vowel-sound should be like the i of machine, in English often spelt ee start with this sound and shorten it. (Before unvoiced stops, as in feet, ee may be quite short also in English just make sure there is a distinction of length between i and Notice that i is never .) pronounced ai as in English ne = fain. (Quenya n larch has two e syllables, the vowels being those heard in pit [ideally a little closer] and pet, respectively.) Of course, this also goes for nal -i (usually a plural ending). If the student will forgive another Star Wars reference, George Lucas Jedi may be jedai = jed-eye, but Tolkiens Quendi are most denitely not quendai. In Quenya, nal -i should rather be pronounced as in Iraqi, Mississippi. Long may be pronounced more or less as in English sore, but preferably a o little tenser and closer (midway between the vowel-sounds of English sore and English oo as in soon): ml slave, t wool, ma voice. Short o o o o may be pronounced as in English for (when accented), or as in box. The quality of the latter vowel may be just a little too open and A-like according to Tolkiens descriptions. Yet this is the pronunciation he himself used in most cases in the recording of him reading Namri ; it should perhaps be a e attributed to his English accent. Some words with o: rondo cave, olos dream, tolto eight. Of course, Quenya o is never pronounced ow as in English so, also; a word like tolto must NOT come out as tol-tow. Neither must o ever be reduced to a schwa or dropped altogether; be especially mindful of the ending -on, often found in masculine names (and also in plural genitives like Silmarillion; see later lessons). English-style pronunciation of a name like Sauron would result in what a baed Elf might try to represent in writing as Srn (or at best Sren). The nal -on should o o sound rather like the rst syllable of English online, with the vowel fully intact even though it is unaccented in Sauron. In the Jackson movie, the actors usually deliver a good pronunciation of this name; especially listen to how Gandalf and Saruman pronounce it. Good examples of short Elvish o also occur in the name Mordor as pronounced by the same two 46
actors. Long u is the vowel of English brute, in English often spelt oo as in fool : N men west, c crescent, y yo both. It must be distinctly longer u u u than short u, which is pronounced somewhat like the vowel of English put (NOT like in English cut). Ideally, Quenya short u should be a little more rounded than the vowel of put; it should be simply a shorter version of the long u or oo described above: Cundu prince, nuru death, ulundo monster. Notice that Quenya u is never pronounced yu as in English union; ulundo should not become yulundo. Speakers of English must be especially mindful of their vowels when a combination vowel + r occurs. In the combinations ar, or, many speakers of English have a tendency to lengthen the vowel even where it should be short (and many would also let the r drop out, especially when it is followed by another consonant). But in Quenya words like narda (knot) or lorna (asleep), the vowel before the r must be short, as indicated by the absence of the accent mark. It is not permissible to let the pronunciation drift towards n(r)da, l(r)na, no matter how tempting this is to people used a o to English speech-habits. Where the groups er, ir, ur occur (e.g. in words like serc blood, e tirno watcher, turma shield), speakers of English must take care NOT to pronounce the vowels after the fashion of English serve, girl, turn. (I once had an English teacher who described the vowel of girl as one of the ugliest sounds of the English language. She taught English at university level, so she should know though perhaps she wasnt wholly serious. . . ) Short e, i, u should sound just as described above, wholly irrespective of the following r. In LotR Appendix E, Tolkien noted that er, ir, ur should sound, not as in English fern, r, fur, but rather like air, eer, oor (that is, like it would be natural for a speaker of English to pronounce orthographic air, eer, oor however, it should be understood that this would only be an approximation of the ideal pronunciation). In the Peter Jackson movie, the actors struggle to pronounce the nal syllable of the Quenya name Isildur correctly, with variable results. In the ash-back scene where Elrond (played by Hugo Weaving) leads Isildur into Mount Doom and urges him to destroy the Ring, Weavings pronunciation of the name Isildur is very good following Tolkiens guidelines to the letter. Diphthongs: In addition to the basic, unitary vowel-sounds discussed above (what linguists would call the monophthongs), we have the diphthongs combinations of two basic vowels that are run together into one syllable, in many ways behaving like a unitary vowel for the purpose of word-building: The Quenya diphthongs are ai, au, eu, iu, oi, and ui. The diphthong ai is the same that is heard in English aisle. It is NOT like the one in English mail, though English orthographic ai usually represents the latter sound (can anyone think of other exceptions than 47
aisle?) The rst syllable of faila just, generous must not pronounced like the English word fail, since Quenya ai always has the sound of English I, eye: Aica fell, terrible, caima bed, aira holy. Of course, the rst syllable of the latter word sounds nothing like English air ! The diphthong au is pronounced as in German Haus, or more or less as the ow of English cow : aul invention, laura golden, e e taur forest. It is never sounded as in English caught, aura (in e which words au is pronounced rather like Quenya ). In his Note o on Pronunciation appended to the Silmarillion, Christopher Tolkien notes that the rst syllable of Sauron should be like English sour, not English sore. (However, the diphthong in sour is in British English followed by a schwa a faint reminiscence of the otherwise silent nal r. This schwa should not be pronounced in Sauron.) The diphthong eu does not occur in English, but it is not dissimilar to the o of English so. The only dierence is that while the rst part of the diphthong is a schwa in English, it should be a normal e (as in end ) in Quenya. In particular, some British upper-class pronunciations of English o as in so come close to Quenya eu (but the American pronunciation does not). Quenya examples: leuca snake, neuma snare, peu pair of lips. This diphthong is not very common. The diphthong iu may be sounded like yu in English yule, according to the usual Third Age Pronunciation. Tolkien imagined that originally, it had rather been a falling diphthong like the other Quenya diphthongs, stressed on the rst rather than the last element (LotR Appendix E). However, the Third Age pronunciation would be equally valid also within the mythos, and for speakers of English it is easier to achieve. This diphthong is in any case very rare; in the Etymologies it is only attested in a handful of words (miul whining, mewing, e piuta spit, siul incitement and the group tiuca thick, fat, e tiuco thigh and tiuya- swell, grow fat a few more examples of iu could be quoted from Tolkiens early Qenya material). The diphthong oi is easy, corresponding to English oi or oy as in oi l, toy : coira living, soica thirsty, oira eternal. e The diphthong ui Tolkien sometimes compared to the sound occurring in English ruin. This is a rather surprising example, for surely the word ruin is not normally pronounced as containing a diphthong, but as two distinct syllables: ru-in. Rather think ooy as in the English phrase too young: huin shadow, cuil life, uil (long, trailing) e e e plant. Notice that the combination qui does not contain this diphthong; this is just a more visually pleasing way of spelling cwi (e.g. orqui Orcs = orcwi).
48
All other groups of vowels are not diphthongs, but simply vowels belonging to separate syllables, to be pronounced distinctly. In linguistic terms, vowels that are in direct contact without forming diphthongs are said to be in hiatus. Primitive Elvish apparently did not have such combinations, at least not in the middle of words: Tolkien had Fanor concluding that our fathers e . . . in building words took the vowels and parted them with the consonants as walls (VT39:10). But some consonants had been lost in Quenya, so that vowels that were originally so parted had come into direct contact (VT39:6). In Quenya we even have polysyllabic all-vowel words like E (a a name of the universe) or oa (away). The most frequent combinations of vowels in hiatus are ea, eo, ie, io, oa; each vowel should be sounded by itself. Tolkien often emphasizes this fact by adding diaereses or dots to one of the vowels, and in the consistent spelling here imposed on the material, we regularly write a (E), o (E), o. Thus there is no excuse e a e o e for such mistakes as pronouncing a as in English heart or please, or o e e as in canoe or foetus. (Other distortions are apparently also possible: Cate Blanchett simply reduced Erendil to Erendil the one time her version of a Galadriel pronounces this name in the Jackson movie: I give you the light of E[]rendil, our most beloved star. . . Can we have an extra vowel for the a Directors Cut, please?) In this course we do not use the diaeresis in the combinations ie (except when nal) and oa, but as indicated by the spelling i and a in e o certain Tolkien manuscripts, the vowels must be pronounced distinctly and not drawn together as in English piece (or tie), or English load. In accordance with this, Christopher Tolkien in the Note on Pronunciation that he appended to the Silmarillion indicates that the name Nienna is to be pronounced Ni-enna, not Neena as if ie were sounded as in English piece. (Immediately after the line in which she mangles the name Erendil, Cate a Blanchett pronounces the Quenya word namri, farewell. Im glad to a e say that she did a better job with this word, getting the -i more or less e right!) Some words with vowels in hiatus: fa soul, lo shade, lond e e e e year-middle (the middle day of the year according to the Elvish calendar), coa house, ti path. e Consonants: Most Quenya consonants are easy to pronounce for people used to speaking a Western language. These points may be observed: C is always pronounced k, never s; indeed Tolkien does use the letter k rather than c in many sources. Celma channel or cirya ship must not come out as selma, sirya. (This goes for Sindarin spelling as well: When Celeborn is pronounced Seleborn in the Rankin/Bass animated version of LotR, it clearly shows that the moviemakers never made it to Appendix E.) In the groups hw, hy, hl, hr, the letter h is not to be pronounced separately. These are just digraphs denoting unitary consonants: 49
What is spelt hl, hr was originally unvoiced l, r. That is, these sounds were pronounced without vibration in the vocal chords, resulting in what may be described as whispered versions of normal l, r. (If you can isolate the l of English please, you will have an unvoiced l though in this case, it is just incidentally unvoiced because of the inuence from the unvoiced plosive p immediately preceding it. English never has unvoiced l as an independent sound of speech, as Quenya originally did.) In Quenya, these sounds are quite rare; examples include hr e v winter and hlc serpent, dragon. However, Tolkien stated that o e by the Third Age, hr and hl had come to be pronounced as normal voiced r, l, though the spelling hl, hr apparently persisted in writing. What is spelt hw corresponds to English wh in dialects where this is still distinct from normal w (e.g., witch and which are audibly distinct words American English, as well as northern British English, normally uphold this distinction, though it has been abandoned in the British Received Pronunciation). Put simply, hw is a (weak) version of the sound you make when you blow out a candle. Hw is not a very frequent sound in Quenya; this seems to be a quite complete list of the known words where it occurs: hwan sponge, fungus, hwarin crooked, hwarma crossbar, hwerm gesturee code, hwesta breeze, breath, pu of air (also as verb: hwesta- to pu), hwind eddy, whirlpool. e What is spelt hy represents a sound that may occur in English, but that is not normally recognized as a distinct consonant in this language. Hy denotes what by a German term is often referred to as ich-Laut or ich-sound, since it is exemplied by ch in the German word ich (I). To speakers of English it may sound much like sh (one imagines Kennedy training long and hard to avoid Ish bin ein Berliner). Still, as I said, a (weak) version of the sound in question may often be heard in English as well: In words like hew, huge, human, the h may be pronounced like an (obscure) hy. Cf. SD:418419, where Tolkien states that in Quenya or Avallonian, the sound hy is approximately equivalent to . . . h in huge. In LotR Appendix E, Tolkien also pointed out that hy has the same relationship to y as hw (discussed above) has to normal w: one is unvoiced, the other voiced. So another way of arriving at hy is to start with the sound of y (as in you) and produce a voiceless, whispered variant of it. Once you have the sound pinned down, you only have to strengthen it; it should be pronounced with the same force as English sh: Hyarmen south, hyalma shell, conch, hyell glass. It seems that hy mostly occurs e at the beginning of words; ahya- change is presently the sole known example of hy occurring between vowels in the middle of a word. However, h in the combination ht following certain vowels should also be 50
pronounced like hy; see below. In LotR Appendix E, Tolkien noted that speakers of Westron (the supposed original language of the Red Book, that Tolkien translated into English) often substituted the sound of sh for Quenya hy. Speakers of English who dont care about subtle phonological details may of course do the same, turning a word like hyalma into shalma. This would be a pronunciation that existed also within the Middle-earth setting, though it was not quite like the proper Elvish pronunciation (and it does seem best to aim for the latter!) I guess many speakers of English would hardly be able to tell the dierence, though. Incidentally, one can achieve a pretty good hy by starting from sh; just make sure that your tongue is not raised (you may press its tip against the lower teeth to be certain of that). If you try to pronounce sh with the tongue in this position, what comes out ought to sound like hy. Outside the groups hw, hy, hl, hr, the letter h does represent an independent sound, but it is pronounced somewhat dierently in different positions. It seems that originally, Quenya h (at least where it comes from Primitive Elvish kh) was typically stronger than English h that is, a breath-h as in high. In Fanors day it was apparently e pronounced like ch in German ach or Scottish loch, or like Cyrillic X. In phonetic writing, this sound is represented as [x]. But later, at the beginning of words, this [x] was weakened and became a sound like English h. In LotR Appendix E, Tolkien informs us that the Tengwa letter for [x] was originally called harma; naturally this Tengwa was so called because the initial h of this word was an example of the sound the letter denoted, [x]. But when [x] in this position eventually turned into an English-style h, the Tengwa was renamed aha, for in the middle of words, [x] was not weakened. So we can extract these rules: at the beginning of words (before a vowel), the letter h is to be pronounced like English h. But in the middle of words, h is to be pronounced [x]: as between vowels in aha wrath, and likewise before t in words like pahta closed, ohta war, nuhta- to stunt. In one late source, Tolkien noted that in Quenya and Telerin medial [x] eventually became h also in most cases (VT41:9). It may therefore be permissible to pronounce even words like aha with an English-style breath-h. But the group ht must probably always be pronounced [xt]; the weaker breath-h would be barely audible in this position. This rule needs one modication. Likely, h before t was originally pronounced [x] in all cases. Following any of the vowels a, o, and u, this pronunciation persisted, as in the examples pahta, ohta, nuhtaabove. But following the vowels i and e, the original [x] turned into a sound similar to German ich-Laut (German may indeed be Tolkiens inspiration for this particular development in Quenya phonology). 51
Thus in words like eht spear or rihta- to jerk, h should be e pronounced just like the hy described above. Again, Tolkien imagined that human (mortal) speakers of Westron had a tendency to substitute a sound like English sh and say esht, rishta instead. e Quenya l represents more or less the sound of English initial l, as in let (LotR Appendix E). Now why did Tolkien specify that Quenya l is to sound like an initial English l (regardless of its position in a Quenya word)? As Tolkien was well aware, British English l is pronounced somewhat dierently in dierent positions. An initial l, as in let, is pronounced as a so-called clear l and this is the kind of l that should be used in all positions in Quenya (as is also the case in other languages, like German). But when l is not initial, English in most cases employs a so-called dark l, which diers from the the clear l in that the dark variant is pronounced by arching the back of the tongue upwards: Contrast the pronunciation of l in two words like let (clear l) and ll (dark l). Compared to the clear l, the dark l sounds lower pitched, but this sound is to be avoided in Quenya. This may be something of a problem to Americans, since their Ls tend to be rather dark in all positions, even initially (at least as perceived by European ears). Perfectionists should also observe another detail: In Letters:425, Tolkien mentioned l among the Quenya dentals, sc. sounds that are pronounced with the tip of the tongue touching the (upper) teeth. English normally uses an alveolar l instead, that is, a sound pronounced with the tip of the tongue further back, above the teeth rather than touching them. This again makes for a somewhat darker sound. When pronouncing a Quenya l, one should make sure that that the tip of the tongue touches the teeth. Quenya n is like English n. Usually this sound had been n all along, but in some cases it represents older ng as in English king, ding (notice that there is no distinct g to be heard, despite the spelling). Unlike English, Quenya could also have this sound at the beginning of words. As mentioned in the discussion of spelling conventions, Tolkien sometimes used the letter n to represent this older ng, e.g. Noldor. In his letters, Tolkien in one case added a footnote to the word Noldor (so spelt), informing the recipient that the initial N was to be pronounced ng as in ding (Letters:176). This would however be the archaic pronunciation; people speaking Quenya in Frodos day would simply say Noldor: LotR Appendix E clearly indicates that by the Third Age, initial n had come to be pronounced like a normal n, and therefore the Elvish letter for n has been transcribed n. We have adopted the same system here, so the letter n in nearly all cases represents normal English n, regardless of its phonological history in Quenya. I say in nearly all cases because n is still pronounced n before c (= k), g and 52
qu. This is not much of a problem, for it is natural for speakers of English and many other languages to use this pronunciation anyway. In a word like anca jaw the cluster nc is therefore pronounced like nk in English tank, and in a word like anga iron the ng should be sounded like ng in English nger. Notice that Quenya ng occurring in the middle of words should always be pronounced with an audible g (this also goes for the group ngw, as in tengwa letter). It is NOT just the simple n described above, the ng of English king, with no distinct g. (We are of course talking about a hard g here; Quenya ng must never be pronounced nj as in English angel, but always as in nger. The sound of soft g as in English g in does not occur in Quenya.) Quenya r represents a trilled r in all positions; the sound was not lost before consonants (as in [British] English part) (LotR Appendix E). English r is generally much too weak for Quenya. Its weakness is precisely the reason why it tends to drop out before consonants and at the end of words (except where the next word happens to begin in a vowel and by analogy, some speakers of English even introduce an R-sound where a word that properly should end in a vowel comes before a word beginning in a vowel. That is when vanilla ice starts coming out as vanillar ice or, if you like, vanilla rice! Of course, this must be avoided in Quenya.) Quenya r should be trilled, as in Spanish, Italian, Russian etc., or for that matter as in Scottish English. Certain subtleties of Tengwar spelling suggests that in Quenya, r was somewhat weaker immediately in front of consonants (as opposed to vowels) and at the end of words. Nonetheless, it should be a properly trilled, wholly distinct sound even in these positions: Parma book, erd seed, tasar willow, Eldar Elves. The vowel in front of r e should not be lengthened or otherwise aected. In the Jackson movie, the actors portraying Gandalf and Saruman normally pronounce the name Mordor correctly, with trilled rs and short vowels (whereas Elijah Woods Frodo invariably says Md with no trace of any rs!) o o In the movie, Mordor is Sindarin for Black Land, but by its form and pronunciation, the word could just as well be Quenya mordor = shadows or stains (the plural form of mordo). The uvular r that is common in languages like French and German should be avoided in Quenya, for LotR Appendix E states that this was a sound which the Eldar found distasteful (it is even suggested that this was how the Orcs pronounced R!) The consonant s should always be unvoiced, as in English so, geese (LotR Appendix E). In English, s is often voiced to z, even though orthography may still show s. For instance, though the s of English house is unvoiced, it becomes voiced in the plural form houses (for 53
this reason, Tolkien noted that he would have liked the spelling houzes better see PM:24). When pronouncing Quenya, one should be careful not to add voice to s, turning it into z: Asar festival, olos dream, naus imagination. Third Age Exilic Quenya did not possess the e sound z at all. (Tolkien did imagine that z had occurred at an earlier stage, but it had later turned into r, merging with original r. For instance, UT:396 indicates that the plural of olos dream was at one stage olozi, but later it became olori.) Where it occurs between vowels, s often represents earlier (more or less = th as in thin); the words asar and naus mentioned above represent older aar and e nau and were so spelt in Tengwar orthography. e On v and w: We must assume that v and w are properly pronounced as in English v ine and w ine, respectively (but initial nw is strictly not n + w but simply a so-called labialized n; see below). There are some unclear points here, though. LotR Appendix E seems to indicate that in Third Age Quenya, initial w had come to be pronounced v: it is said that the name of the Tengwa letter vilya had earlier been wilya. Likewise, Tolkien indicated that the word vra (personal, private, e own) had been wra in what he called Old Quenya (PM:340). e In the Etymologies, the evidence is somewhat divergent. Sometimes Tolkien has primitive stems in w- yield Quenya words in v-, as when the stem wan yields Quenya vanya- go, depart, disappear. Some times he lists double forms, as when the stem wa (or wawa, waiwa) yields Quenya vaiwa and waiwa, both meaning wind. Under the stem way Tolkien listed a word for envelope as w - vaia, evidently indicating a double form waia and vaia (all of these examples are found in LR:397). In LR:398, there are further double forms, but in the case of the verb vilin (I y) from the stem wil, Tolkien curiously changed it to wilin. Perhaps he suddenly decided to go for the Old Quenya spelling rather than actually rejecting one in favour of the other? The weight of the evidence seems to be that at the beginning of words, w- had come to be pronounced as normal v- by the Third Age; where Tolkien listed double forms in w- and v-, the former is apparently to be taken as the more archaic form. However, I have not regularized the spelling on this point, though where Tolkien himself used or listed a form in v- rather than w- (either alone or as an alternative to w-), I will use the form in v- in this course. (This also goes for vilin!) It is possible, though, that according to the Third Age pronunciation all initial ws should be sounded as v, the original distinction between initial v and w having been lost in the spoken language. It is unclear whether or not Tolkien meant that this distinction was consistently upheld in Tengwar orthography (as when this writing upheld the dis54
tinction between and s even after both had come to be pronounced s). If so, the letter called (wilya >) vilya was still used for v representing older w, while another letter (vala) was used for v that had been v all along. Other than at the beginning of words, the distinction between v and w was upheld even in the Third Age. In the case of the groups lw and lv the distinction could even be emphasized by altering the pronunciation of the latter: For lv, not for lw, many speakers, especially Elves, used lb (LotR Appendix E). Hence a word like elva starlike would often be pronounced elba, and e e it might also be so written in Tengwar orthography. Though frequent, this would seem to be a non-standard pronunciation, and the spellings employed by Tolkien usually indicates the pronunciation lv. Cf. for instance Celvar (or Kelvar, meaning animals) rather than Celbar in the speeches of Yavanna and Manw in the Silmarillion, chapter 2. e In PM:340 Tolkien quotes a Quenya word for branch as olba rather than olva, though. The letter y is only used as a consonant, as y in E[nglish] Yes: Tolkien singled this out as one of the few major departures from Latin spelling in the spelling conventions he used for Quenya (Letters:176). The vowel y, like German u or French u as in lune, does not occur in Quenya (though it is found in Sindarin).
should actually be pronounced French-style, with no aspiration.) So were Quenya t, p, c still unaspirated, since they had been so in the primitive language? Since the primitive aspirated sounds had been changed, adding aspiration to t, p, c would cause no confusion. It should be noted, though, that in the writing system devised by Fanor, there were originally distinct letters e for aspirated sounds: The original Fanorian system also possessed a grade e with extended stems, both above and below the line [of writing]. These usually represented aspirated consonants (e.g. t + h, p + h, k + h) (LotR Appendix E). However, these were not the letters used to spell Quenya t, p, c. So all things considered, I think Quenya t, p, c should ideally be pronounced without aspiration. For people who are used to automatically slip in a h-like pu of breath after these consonants it may be dicult to get rid of it, since they are not really conscious of its presence at all. A phonology teacher once advised me that one way of getting rid of the aspiration is to practice pronouncing t, p, c/k with a burning candle in front of your mouth; the trick is to pronounce these consonants without the ame of the candle ickering (because of the pu of breath that constitutes the aspiration). The voiced counterparts of t, p and c/k, namely d, b and (hard) g respectively, are not aspirated in English. For this reason, people who are used to hearing the unvoiced sounds pronounced as aspirated variants may (wrongly) perceive unaspirated unvoiced plosives as their voiced counterparts. Pronounced without aspiration, Quenya words like tarya (sti), parma (book) or calma (lamp) may sound a little like darya, barma, galma to speakers of English (speakers of French, Russian or Finnish would not be confused). When pronouncing such words, one must not introduce vibration in the vocal chords to produce actual voiced sounds d, b, g. But I should add that the whole aspiration issue is not something a student needs to spend much time on; as I said, the exact pronunciation of Quenya t, p, c is nowhere addressed in published writings. If it is indeed wrong to add aspiration to these consonants, at least one will err little more than Tolkien did himself when reading Namri. a e
phy as in seor. Of course, this sounds very much like senyor, but is n n really a single consonant. This is a palatalized version of n, an n that has n been tinted in the direction of y. English employs one distinctly palatalized consonant, usually represented by the digraph sh (which, of course, is not a cluster s + h); this can be described as a palatalized s. By carefully comparing the pronunciation of s and sh you can perceive the palatalization mechanism operating in your own mouth: A consonant is palatalized by arching the back of the tongue up towards the roof of the mouth (the palate, hence the term palatalized consonant). The relationship between s and sh corresponds to the relationship between n and Quenya ny (or Spanish ). n Besides ny, Quenya also has the palatalized consonants ty, ly, ry (e.g. in tyali play, alya rich, verya bold); these are palatalized counterparts e of normal t, l, r. Regarding ty, Tolkien wrote that it may be pronounced as the t of English tune (see for instance SD:418419 it should be noted that he is thinking of dialects where this comes out as tyoon; this is not the case in all forms of American English). In Gondor, some mortal speakers of Quenya supposedly pronounced ty like ch as in English church, but that was not quite the proper Elvish pronunciation. As for the consonant ly, it would be similar to the lh of Portuguese olho (eye). In LotR Appendix E, Tolkien noted that l (so spelt) should also to some degree [be] palatalized between e, i and a consonant, or nally after e, i . The wording to some degree seems to suggest that we would not have a regular, full-blown palatalized l in these positions (like the sound spelt ly), but in words like Eldar Elves or amil mother, the l should ideally have just a little tint of palatalizing to it. Besides the palatalized consonants, we have the labialized consonants: nw, gw and qu (= cw). These are not really clusters n + w, g + w, c + w. Rather they represent n, g, c (k) pronounced with pouted lips, as when pronouncing w: By the pouting of the lips, the consonant is labialized (this word comes from the Latin term for lip). Quenya qu may certainly be pronounced as in English queen, but ideally it should be pronounced as k and w merged together in a single, unitary sound. (True, there does exist one early source where Tolkien states that qu, though originally being simply k accomp[anied] by lip-rounding, is now sounded practically exactly as English qu a liprounded k foll[owed] by a distinct w sound: See Parma Eldalamberon #13, page 63. However, I think this idea may be superseded by information from a much later source, indicating that Quenya had no initial consonant clusters: SD:417418.) Nw and gw similarly represent merged versions of n/w, g/w. It should be noted that nw is a single, labialized consonant only at the beginning of words, where it represents earlier ngw (sc. what Tolkien might also spell w, using for ng as in king). In the n n middle of words, e.g. in vanwa gone, lost, nw really is a cluster n + w and is so spelt also in Tengwar orthography. However, the labialized consonants 57
qu and gw also occur in the middle of words. In fact, gw occurs only in that position, and always in the combination ngw (not w but gw, n n still using as Tolkien did): Lingw sh, nangwa jaw, sungwa n e drinking-vessel. The question of length: It may seem that when they occur medially between vowels, the palatalized and labialized consonants count as long or double consonants (as if the digraphs represented actual consonant clusters after all). Again using the letter with its Spanish value of a palatalized n n (and not, as Tolkien often did, for ng as in king), one may ask whether a word like atarinya (my father, LR:61) actually represents atarina. If n so, the group ny in the middle of words denotes a long palatalized N. Then the very word Quenya would be pronounced Quena rather than Quenn ya. Another possibility is Queya, the n being palatalized all right, but n there is still a relatively distinct y-sound following it (which there would not be when ny occurs at the beginning of a word). Tolkien reading a version of Namri at least once pronounced the word inyar as inar (but a e n the second time it occurred he simply said inyar with n + y). In any case, the groups ny, ly, ry, ty and qu (for cw) must be counted as either long consonants or consonant clusters for the purpose of stress (see below) though it is also clear that sometimes they must be analyzed as single, unitary consonants.
Stress
Whenever a language has polysyllabic words, speakers of this language may enunciate some syllables more forcefully than others. We say that these syllables are stressed or accented. In some languages speakers dont normally emphasize certain syllables more than others. For instance, the Japanese put about the same amount of stress on every syllable, resulting in what unloving foreigners have referred to as machine gun articulation. But in Western languages, a varying amount of stress is common: Some syllables are stressed, others unstressed. The rules for which syllables are stressed vary wildly, though. Some languages have a very simple system; in French, words that are to receive any stress are always accented on the nal syllable. To the natives, Paris is not PARis as in English, but rather parIS (actually the French dont pronounce the s, but that has nothing to do with the accent). The Finns also have a very simple system, stressing all words on the rst syllable: While some speakers of English may think that Helsinki is most naturally pronounced HelSINK i, the residents of the city will insist on HELsinki instead. Since the Finnish language was evidently Tolkiens foremost inspiration, one might think that he would have copied its simple system of accenting all words on the rst syllable over into Quenya. In the internal or ctional 58
history of the language, he did indeed envision an early period during which Quenya words were so accented (the so-called retraction period, WJ:366). However, this was replaced by a new system already before the Noldor went into exile, so Quenya as a language of lore in Middle-earth employed dierent accentuation patterns, carefully described in LotR Appendix E. This is the system we must use. (It seems that Tolkien actually copied it from Latin!) Words of one syllable, like nat thing, obviously pose no problem; this one syllable is the sole candidate for receiving the stress. The simplest polysyllabic words, those of two syllables, are no problem either: In LotR Appendix E, Tolkien noted that in words of two syllables [the accent] falls in practically all cases on the rst syllable. As this wording implies, there may be a very few exceptions; the only exception known seems to be the word av dont!, that is accented on the nal syllable: aVA. (Even a this one word also appears in the alternative form va, stressed on the rst a syllable according to the normal rule: AV a.) The name of the Blessed Realm, Aman, I sometimes hear people pronounce with the stress on the second rather than the rst syllable but the correct pronunciation must be AM an, if we can trust the rules set out by Tolkien. (AmAN would be Amman, capital of Jordan!) Longer words, with three or more syllables, are slightly more complex when it comes to stress. Many of them are accented on the second to last syllable. However, in some cases the second-to-last syllable isnt qualied to receive the accent: This syllable cannot be accented if it is short. So how do we recognize a short syllable? If it contains no long vowel (no vowel marked with an accent), this is obviously one omen. Then the vowel itself is necessarily short. If this short vowel is followed by only one consonant, or even no consonant at all, this syllable has little chance of receiving the accent. Its one remaining chance of redeeming itself as a long syllable is that instead of a simple short vowel it actually contains one of the Quenya diphthongs: ai, au, eu, oi, ui or iu. Two vowels combined into a diphthong count as having the same length as a normal, unitary long vowel (marked by an accent). But if there is no diphthong, no long vowel, and not even a short vowel followed by more than one consonant, the syllable in question is irredeemably short. If this is the second-to-last syllable in a word of three or more syllables, this penultimate syllable has forfeited all its chances to receive the stress. In such a case the stress moves one step ahead, to fall on the third syllable from the end (no matter what this syllable looks like). Tolkien noted that words of such a shape are favoured in the Eldarin languages, especially Quenya. Examples: A word like vestal wedding is accented VEST al. The seconde e to-last syllable cannot receive the stress because its vowel (the a) is short and followed by only a single consonant (the l); hence the accent moves one step ahead, to the third syllable from the end. Plural forms 59
like Teleri (the Sea-Elves) and Istari (the Wizards) I sometimes hear people mispronounce as TeLERi, IsTARi; applying Tolkiens rules we have to conclude that he actually intended TELeri, IST ari. The short penultimate syllables in these words cannot be accented. A word like Eressa (the name of an isle near the Blessed Realm) some e speakers of English are tempted to accent on the second-to-last syllable (following the stress-pattern of such a place-name as Eritrea!) But since in Er-ess--a the second-to-last syllable is just a short e e not followed by a group of consonants (actually not even one consonant), this syllable cannot be accented and the stress moves to the syllable before it: ErESSa. Other words of the same pattern (with e no consonant following a short vowel in the second-to-last syllable): Eldali the people of the Elves (ElDAli though the word Elda e e Elf by itself is of course accented ELda), Tilion The Horned, name of a Maia (TILion), laura golden (LAURa), Yavanni e e e September (YaVANN i), Silmarillion [The Story] of the Sile marils (SilmaRILLion). But though such words were favoured, there is certainly no lack of words where the second-to-last syllable does qualify for receiving the accent. Examples: Vardas title Elentri Star-Queen is pronounced ElenTARi, since a the vowel in the second-to-last syllable is long. (If this had been a a short a, it couldnt have been stressed since it is not followed by more than one consonant, and the third syllable from the end would have been accented instead: ELEN tari but no such word exists.) The names N menr, Valinr are likewise accented on the long in u o e o e o the second-to-last syllable (whereas in the shortened forms N menor, u Valinor the accent must fall on the third syllable from the end: NUMenor, VALinor ). Words like hastaina marred or Valarauco Power-demon (Sindarin Balrog) are accented hasTAIN a, ValaRAUC o since diphthongs like ai, au can be counted as long vowels for the purpose of stress. The names Elendil and Isildur are accented ElENDil and IsILDur, since the vowel in the second-to-last-syllable, though short, is followed by more than one consonant (the groups nd, ld, respectively). A double consonant would have the same eect as a cluster of dierent consonants; for instance, Elenna (Starwards, a name of Nmenor) is u pronounced ElENN a. (Contrast the adjective elena stellar, of the stars: this must be accented ELena since the second-to-last syllable en is short and therefore unable to receive the accent unlike the long syllable enn in Elenna.)
60
Notice that the one letter x represents two consonants, ks. Therefore, a word like Helcarax (a place-name) is accented HelcarAX (not Hele e CARax as if there were only one consonant following the a in the seconde to-last syllable). Cf. the alternative spelling Helkarakse in the Etymologies, entry karak. As noted above, some combinations should apparently be thought of as single consonants: qu (for cw/kw) represents labialized k, not k + w. Similarly, ny, ty, ly, ry would be palatalized n, t, l, r (the rst = Spanish n ). But in the middle of words, for the purpose of stress, it seems that qu, ly, ny, ty etc. do count as groups of consonants (double consonants or clusters we cannot be certain precisely what Tolkien intended). In WJ:407, Tolkien indicates that the compound word ciryaquen shipman, sailor (made from cirya ship + -quen person) is to be accented cirYAquen. If qu (= cw/kw) were here thought of as a single consonant, labialized k, there would not be a group of consonants following the a and it could not receive the accent: the word would then have been pronounced CIRyaquen instead. So either qu here does count as a cluster k + w, or it represents a long or double labialized k (or even labialized kw followed by w). Bottom line is: pronounce cirYAquen and be relieved that the rest is mainly academic meandering. A few other words including the combinations in question: Elenya (rst day of the Eldarin six-day week, accented ElEN ya), Calacirya or Calacilya (a place in the Blessed Realm, accented CalaCI rya, CalaCI lya). A word of warning regarding the accent mark: Notice that the accent mark that may appear above vowels (, , , u) only denotes that the vowel is a e , o long. While this symbol is frequently used to indicate the stressed syllable, this is not the case in Tolkiens normal spelling of Quenya. (Some may have noted that Pokmon isnt accented on the either, so Tolkien isnt wildly e e idiosyncratic in this department!) A long vowel will often receive the stress, as in the example Elentri above, but not necessarily so: If the long vowel a does not appear in the second-to-last syllable, its length (and the accent mark denoting it!) is quite irrelevant for the purpose of stress. In a word like Ulairi, the Quenya name for the Ringwraiths or Nazgl, the stress u falls on the diphthong ai, not on the u. The spelling palant has mislead r many, making them think that this word is to be accented on t Here is r. something Ian McKellen, playing Gandalf in the Peter Jackson LotR movie trilogy, wrote as the lm was being shot: . . . I have to learn a new pronunciation. All this time we have being saying palanT IR instead of the Old English stress on the rst syllable. Just as the word was about to be committed to the soundtrack, a correction came from Andrew Jack, the Dialect Coach; he taught me a Norfolk accent for Restoration, and for LOTR he supervises accents, languages and all things vocal. Palant being strictly of elvish origin should folr, 61
low Tolkiens rule that the syllable before a double consonant should be stressed paLANT making a sound which is close r to lantern. . . Andrew Jack was right. Palant cannot be stressed on the nal syllable; r virtually no polysyllabic Quenya words are accented in such a way (as I said above, av dont! is the sole known exception). Instead the a in the seconda to-last syllable receives the accent because it is followed by the consonant cluster nt (I should not call this a double consonant like McKellen does, since I want to reserve that term for a group of two identical consonants, like tt or nn but for the purpose of stress, double consonants and clusters of dierent consonants have the same eect). So it is indeed palANT r. (But in the plural form palant where the long suddenly appears in the ri, second-to-last syllable, it does receive the accent: palanTIRi.) In the case of long words ending in two short syllables, the last of these syllables may receive a weaker secondary stress. In a word like h sim mist, e the main stress falls on h but the nal syllable -m is not wholly uns, e stressed. This secondary stress is much weaker than the main accent, though. (Nonetheless, Tolkien did note that for the purpose of poetry, the secondary stress can be used metrically: RGEO:69.)
Speed
Finally a brief note on something we know little about: How fast should one talk when speaking Quenya? The few recordings of Tolkien speaking Quenya are not reliable in this matter; he inevitably enunciates quite carefully. But regarding Fanors mother M e riel he noted that she spoke swiftly and took pride in this skill (PM:333). So fast Quenya is evidently good Quenya. When Tolkien also wrote that the Elves made considerable use of . . . concomitant gestures (WJ:416), one remembers that he had a great love for Italian see Letters:223.
consonants are normally written as digraphs in -w (e.g. nw, but what would be cw is spelt qu instead). H is pronounced [x] (German ach-Laut) before t, unless this combination ht is preceded by one of the vowels e or i, in which case h is sounded like German ich-Laut. Otherwise, h may be pronounced like English h; the digraphs hy and hw however represent ich-Laut and unvoiced w (like American English wh), respectively. The combinations hl and hr originally represented unvoiced l, r, but by the Third Age, these sounds had come to be pronounced like normal l and r. In polysyllabic words, the stress falls on the second-to-last syllable when that is long (containing either a long vowel, a diphthong, or a vowel followed by a consonant cluster or a double consonant). If the second-to-last syllable is short, the stress falls on the third syllable from the end (unless the word has only two syllables, in which case the rst syllable receives the stress whether it is short or long).
Exercises
As far as the most critical subtleties of pronunciation are concerned, I unfortunately cannot make any exercises; we are not in a classroom so that I can comment on your pronunciation. But regarding stress (accent) and the pronunciation of h, it is possible to make exercises. 1. Determine which vowel (single vowel or diphthong) receives the accent in the words below. (It is not necessary to indicate where the entire syllable it belongs to begins and ends.) A. Alcar (glory) B. Alcar (longer variant of the above) e C. Alcarinqua (glorious) D. Calima (bright) E. Oronti (mountains) o F. Untim (uncountable, numberless) e G. Envinyatar (renewer) H. Ulund (ood) e I. Eruil (seaweed) a e J. Ercass (holly) e Extra exercise on stress: While we hear many Sindarin lines in the movie, one of the few really prominent samples of Quenya in Peter Jacksons The Fellowship of the Ring is the scene where Saruman (Christopher Lee) standing on the top of Isengard reads an invocation to bring down an avalanche in order to stop the Fellowship. He says to the mountain they are attempting to cross: Nai yarvaxa rasselya e 63
taltuva notto-carinnar! = may your bloodstained horn collapse upon enemy heads! (not translated in the movie). The actor accents the words like this: nai yarVAXa RASSelya TALTuva nottoe CARinnar. Are all the words accented as they should be, according to Tolkiens guidelines? If not, what is right and what is wrong? 2. Where the letter h appears in Quenya words as they are spelt in our letters, it may be pronounced in various ways. Ignoring the digraphs hw and hy, the letter h may be pronounced A) a breath-h like English h as in high, B) more or less as in English huge, human or ideally like ch in German ich, C) like ch in German ach or Scottish loch (in phonetic writing [x]). In addition we have alternative D): the letter h is not really pronounced at all, but merely indicates that the following consonant was unvoiced in archaic Quenya. Sort the words below into these four categories (A, B, C, D): K. Ohtar (warrior) L. Hrv (esh) a e M. Nahta (a bite) N. Heru (lord) O. Neht (spearhead) e P. Mahalma (throne) Q. Hell (sky) e R. Tihtala (blinking) S. Hlc (snake, serpent) o e T. H e (mist) si
64
Lesson 2
Nouns. Plural form. The article.
2.1
The Noun
Words that denote things, as opposed to for instance actions, are called nouns. The things in question may be inanimate (like stone), animate (like person, woman, boy), natural (like tree), articial (like bridge, house), concrete (like stone again) or wholly abstract (like hatred). Names of persons, like Peter or Mary, are also considered nouns. Sometimes a noun may denote, not one clearly distinct object or person, but an entire substance (like gold or water). So there is much to be included. In most languages, a noun can be inected, that is, it appears in various forms to modify its meaning, or to make it t into a specic grammatical context. For instance, if you want to connect two English nouns like Mary and house in such a way as to make it clear that Mary owns the house, you modify the form of the noun Mary by adding the ending -s, producing Marys, which readily connects with house to make the phrase Marys house. Or starting with a noun like tree, you may want to make it clear that you are talking about more than one singular tree, and so you modify the word to its plural form by adding the plural ending -s to get trees. In English, a noun doesnt have very many forms at all; there is the singular (e.g. girl ), its plural (girls), the form you use when the one denoted by the noun owns something (girls) and the combination of the plural and this ownershipform (written girls and unfortunately not really distinct from girls or girls in sound, but speakers of English somehow get along without too many misunderstandings rest assured that the Quenya equivalents are clearly distinct in form!) So an English noun comes in no more than four dierent forms. A Quenya noun, on the other hand, comes in hundreds of dierent forms. It can receive endings not only for two dierent kinds of plural, plus endings denoting a pair of things, but also endings expressing meanings that in English would be denoted by placing small words like for, in/on, from, to, of, with etc. in front of the noun instead. Finally a Quenya noun can 65
also receive endings denoting who owns it, e.g. -rya- her in mryat her a hands in Namri (the nal -t, by the way, is one of the endings denoting a e a pair of something in this case a natural pair of hands). Having read the above, the student should not succumb to the idea that Quenya is a horribly dicult language (imagine, hundreds of dierent forms to learn where English has only four!), or for that matter start thinking that Quenya must be some kind of super-language (wow, hundreds of different forms to play with while the poor English-speaking sods have to get along with a pitiful four!) English and Quenya organize the information dierently, that is all the former often preferring a string of short words, the latter rather jumbling the ideas to be expressed into one big mouthful. The hundreds of dierent forms arise because a much lower number of endings can be combined, so there is no reason to despair. It is a little like counting; you neednt learn two hundred and fty dierent numeric symbols to be able to count to 250, but only the ten from 0 to 9. Most of the endings a noun can take we wont discuss before in (much) later lessons. We will start with something that should be familiar enough, found even on the puny list of English noun-forms: Making a noun plural going from one to several. In Quenya, there are two dierent plurals. One is formed by adding the ending -li to the noun. Tolkien called this the partitive plural (WJ:388) or a general pl[ural] (see the Etymologies, entry teles). Unfortunately, the function of this plural sc. how it diers in meaning from the more normal plural discussed below is not fully understood. We have a few examples of this plural in our scarce source material, but they are not very helpful. For a long time it was assumed that this plural implied that there were many of the things in question; hence Eldali (formed from Elda Elf) would mean something like many Elves. There may be something to this, but in several of the examples we have, there seems to be no implication of many. It has been suggested that Eldali may rather mean something like several Elves or some Elves, sc. some out of a larger group, some considered as part of this group: The term partitive plural may point in the same direction. However, I will for the most part leave the partitive plural alone throughout this course. Its function just isnt well enough understood for me to construct exercises that would only mean feeding some highly tentative interpretation to unsuspecting students. (I present some thoughts about the -li plural in the appendices to this course.) For now we will deal with the normal plural form instead. Any reader of Tolkiens narratives will have encountered plenty of examples of this form; they are especially common in the Silmarillion. Nouns ending in any of the four vowels -a, -o, -i or -u , plus nouns ending in the group -i, form e their plural with the ending -r. Cf. the names of various groups of people mentioned in the Silmarillion:
66
Elda Elf, plural Eldar Vala god (or technically angel), pl. Valar Ainu spirit of Gods rst creation, pl. Ainur Noldo Noldo, member of the Second Clan of the Eldar, pl. Noldor Vali female Vala, pl. Valier e For another example of -i, cf. tier for paths in Namri ; compare e a e singular ti path. (According to the spelling conventions here employed, e the diaeresis in ti is dropped in the plural form tier because the dots e are there merely to mark that nal - is not silent, but in tier, e is not e nal anymore because an ending has been added and hence the dots go.) Examples of the plurals of nouns in -i are rare, since nouns with this ending are rare themselves, but in MR:229 we have quendir as the pl. of quendi Elf-woman (and also quendur as the pl. of quendu Elf-man; nouns in -u are not very numerous either). This singular word quendi Elf-woman must not be confused with the plural word Quendi that many readers of Tolkiens ction will remember from the Silmarillion, for instance in the description of the awakening of the Elves in chapter 3: Themselves they named the Quendi, signifying those that speak with voices; for as yet they had met no other living things that spoke or sang. Quendi is the plural form of Quend Elf; nouns ending e in - typically form their plurals in -i, and as we see, this -i replaces the e nal - instead of being added to it. In WJ:361, Tolkien explicitly refers to e nouns in -e, the majority of which formed their plurals in -i . As this wording implies, there are exceptions; a few nouns in - are seen e to use the other plural ending, -r, instead. One exception we have already touched on: where the - is part of -i, we have plurals in -ier, as in tier e e paths. Hence we avoid the awkward plural form **tii. Other exceptions cannot be explained as easily. In LotR Appendix E, we have tyeller for grades, evidently the plural of tyell. Why tyeller instead of **tyelli? e LR:47 likewise indicates that the plural of mall road is maller; why not e **malli? It may be that nouns in -l have plurals in -ler because regular e **-li might cause confusion with the partitive plural ending -li mentioned above. Unfortunately, we lack more examples that could conrm or disprove this theory (and so I dont dare to construct any exercises based on this assumption, though I would follow this rule in my own Quenya compositions). The form tyeller confused early researchers; with extremely few examples to go on, some wrongly concluded that nouns in - regularly have plurals in -er. e The name of the early journal Parma Eldalamberon or Book of Elventongues (sporadically published still) reects this mistake; the title incorporates **lamber as the presumed plural of lamb tongue, language, while e we now know that the correct plural must be lambi. Though the error was early suspected and is now recognized by everyone, the publisher never bothered to change the name of the journal to the correct form Parma El67
dalambion (and so, ever and anon, I get an e-mail from some fresh student wondering why my site is called Ardalambion and not Ardalamberon. . . ) In some cases, Tolkien himself seems uncertain which plural ending should be used. In PM:332, the plural form of Ingw Elf of the First Clan [also e name of the king of that clan] is given as Ingwi, just as we would expect; yet a few pages later, in PM:340, we nd Ingwer instead (it is there said that the First Clan, the Vanyar, called themselves Ingwer, so perhaps this reects a special Vanyarin usage?) It may be noted that in Tolkiens earliest Qenya, more nouns in - apparently had plural forms in -er. For instance, e the early poem Narqelion has lasser as the plural of lass leaf, but in e Namri in LotR Tolkien used the plural form lassi. a e As far as I know, the words in the exercises below all follow the normal rule: Nouns ending in -, except as a part of -i, have plurals in -i. e e This leaves only one group of nouns to be considered, namely those that end in a consonant. These nouns, just like those that end in -, are seen to e have plurals in -i. A few examples: Eleni stars, the plural form of elen star, occurs in Namri (and also in WJ:362, where both the singular and a e the plural form are quoted). The Silmarillion has Atani for Men (not males, but humans as opposed to Elves); this is formed from the singular word Atan. According to WJ:388, the word Casar Dwarf has the plural Casari Dwarves. Of these two plural endings r as in Eldar Elves, but i as in Atani (Mortal) Men Tolkien imagined the latter to be the most ancient. The plural ending -i comes directly from Primitive Elvish - a word like Quendi , representing primitive Kwend The plural ending -r arose later: For the . showing of many the new device of r was brought in and used in all words of a certain shape and this, it is said, was begun among the Noldor (PM:402). In primary-world terms, both plural endings were however present in Tolkiens conception from the beginning; already in his earliest work on Qenya, written during World War I, we nd forms like Qendi (as it was then spelt) and Eldar coexisting. The twin plural endings are a feature that evidently survived throughout all the stages of Tolkiens development of Quenya, from 1915 to 1973.
NOTE ON THE DIFFERENT WORDS FOR ELF: As the attentive reader will have inferred from the above, there is more than one Quenya word for Elf. The word with the widest application, within the scope of Tolkiens ction, was Quend pl. Quendi. This e form is at least associated with the word to speak (quet-), and Tolkien speculated that ultimately these words were indeed related via a very primitive base kwe- having to do with vocal speech (see WJ:391392). When the Elves awoke by the mere of Cuivinen, they e called themselves Quendi (or in primitive Elvish actually Kwend since for a long time ) they knew of no other speaking creatures. Eventually the Vala Orom found them under a e starlit sky, and he gave them a new name in the language they themselves had developed: Eldi, often translated Starfolk. In Quenya, this primitive word later appeared as Eldar a (singular Elda). While the term Eldar (Eldi) was originally meant to apply to the entire a
68
Elvish race, it was later only used of the Elves that accepted the invitation of the Valar to come and dwell in the Blessed Realm of Aman and embarked on the Great March to get there (the term Eldar is also applicable to those who never actually made it all the way to Aman, such as the Sindar or Grey-elves who stayed in Beleriand). Those who refused the invitation were called Avari, Refusers, and hence all Elves (Quendi) can be subdivided into Eldar and Avari. Only the former play any important part in Tolkiens narratives. So in later Quenya the situation was this: Quend pl. Quendi remained as the only truly e universal term for all Elves of any kind, but this was a technical word primarily used by the Loremasters, not a word that would be used in daily speech. The gender-specic variants of Quend Elf, namely masculine quendu and feminine quendi, would presumably e be used only if you wanted to speak of a specically Elvish (wo)man as opposed to a (wo)man of any other sentient race: These are not the normal Quenya words for man and woman (the normal words are nr and n presumably applicable to a man or e s, woman of any sentient race, not just Elves). The normal, everyday Quenya term for Elf was Elda, and the fact that this word technically didnt apply to Elves of the obscure Avarin tribes living somewhere far east in Middle-earth was no big problem since none of them was ever seen anyway. Regarding the compound Eldali (which combines Elda e with li people, folk) Tolkien wrote that when one of the Elves of Aman used this e word, he meant vaguely all the race of Elves, though he was probably not thinking of the Avari (WJ:374). Throughout the exercises found in this course, I have used Elda (rather than Quend) as the standard translation of English Elf, regardless of any e specialized meaning it may have within Tolkiens mythos. As I said in the Introduction, in these exercises I largely eschew specic references to Tolkiens mythos and narratives.
2.2
The article
We have time for one more thing in this lesson: the article. An article, linguistically speaking, is such a word as English the or a, an. These little words are used in conjunction with nouns to express such dierent shades of meaning as a horse vs. the horse. Anyone capable of reading this text in the rst place will know what the dierence is, so no lengthy explanation is necessary. In short, a horse refers to a horse that hasnt been mentioned before, so you slip in the article a as a kind of introduction: Look, theres a horse over there! You may also use the phrase a horse if you want to say something that is true of any horse, as in a horse is an animal. If, on the other hand, you say the horse, it usually refers to one denite horse. Hence the is termed the denite article, while a, an, lacking this denite aspect, is conversely called the indenite article. In this respect at least, Quenya is somewhat simpler than English. Quenya has only one article, corresponding to the English denite article the (and since there is no indenite article it must be distinguished from, we may simply speak of the article when discussing Quenya). The Quenya word corresponding to English the is i. For instance, Namri has i eleni for a e
69
the stars. As can be inferred from the above, Quenya has no word corresponding to English a, an. When translating Quenya into English, one simply has to slip in a wherever English grammar demands an indenite article, as in the famous greeting Elen s l menn omentielvo, a la u star shines upon the hour of our meeting. As we see, the rst word of the Quenya sentence is simply elen star, with nothing corresponding to the English indenite article a before it (or anywhere else in the sentence, for that matter). In Quenya, there is no way you can maintain the distinction between a star and just star; both are simply elen. Luckily there isnt much of a distinction to be maintained anyway. Languages like Arabic, Hebrew and classical Greek employ a similar system: there is a denite article corresponding to English the, but nothing corresponding to the English indenite article a, an (and this is the system used in Esperanto as well). After all, the absence of the denite article is itself enough to signal that a (common) noun is indenite, so the indenite article is in a way superuous. Tolkien decided to do without it in Quenya, so students only have to worry about i = the. Sometimes, Tolkien connects the article to the next word by means of a hyphen or a dot: i-mar the earth (F riels Song), icoimas the lifebread (PM:396). However, he did not do so in LotR (we have already quoted the example i eleni the stars in Namri ), and neither will we here. a e The Quenya article is generally used as in English. However, some nouns that would require the article in English are apparently counted as proper names in Quenya, and so take no article. For instance, the sentence Anar caluva tielyanna is translated the Sun shall shine upon your path (UT: 22, 51); yet there is no article in the Quenya sentence. The Sun is not **i Anar, but simply Anar. Clearly Anar is perceived as a proper name, designating one celestial body only, and you dont have to say the Anar any more than an English-speaking person would say the Mars. The name of the Moon, Isil, undoubtedly behaves like Anar in this respect. It may be noted that both words are treated as proper names in the Silmarillion, chapter 11: Isil was rst wrought and made ready, and rst rose into the realm of the stars. . . Anar arose in glory, and the rst dawn of the Sun was like a great re. . . Also notice that before a plural denoting an entire people (or even race), the article is not normally used. WJ:404 mentions a saying Valar valuvar, the will of the Valar will be done (or more literally *the Valar will rule). Notice that the Valar is simply Valar in Quenya, not i Valar. Similarly, PM:395 has lamb Quendion for language of the Elves and e coimas Eldaron for coimas [lembas] of the Eldar not **lamb i Quene dion, **coimas i Eldaron. (The ending -on here appended to the plurals Quendi, Eldar signies of; this ending should not aect whether or not the article has to be present before the word.) With this usage compare Tolkiens use of Men in his narratives to refer 70
to the human race as such: Men awoke in Hildrien at the rising of the o Sun. . . A darkness lay upon the hearts of Men. . . Men (it is said) were at rst very few in number. . . (Silmarillion, chapter 17.) By contrast, the Men would refer, not to the entire race, but only to a casual group of Men or humans. Quenya plurals denoting entire peoples or races seem to behave in the same way. In a Quenya text there would probably be no article before plurals like Valar, Eldar, Vanyar, Noldor, Lindar, Teleri, Atani etc. as long as the entire race or people is considered, even though Tolkiens English narratives speak of the Valar, the Eldar etc. However, if we replace Eldar with its equivalent Elves, we see that the article often would often not be required in English, either (e.g. Elves are beautiful = Eldar nar vany; if you say the Elves are beautiful = i Eldar nar e vany, you are probably describing once particular group of Elves, not the e entire race). Occasionally, especially in poetry, the article seemingly drops out for no special reason. Perhaps it is simply omitted because of metric considerations. The rst line of Namri, ai! lauri lantar lassi s rinen, Tolkien a e e u translated ah! like gold fall the leaves. . . though there is no i before lassi leaves in the Quenya text. The Markirya poem also leaves out the article in a number of places, if we are to judge by Tolkiens English translation of it.
Vocabulary
Regarding Frodo hearing Galadriel singing Namri, the LotR states that a e as is the way of Elvish words, they remained graven in his memory. This may be a comforting thought to students attempting to memorize Quenya vocabulary. In the lessons proper, while I discuss various aspects of Quenya, I will normally mention quite a few words but in the exercises, I will only use words from the vocabulary list that is hereafter presented at the end of each lesson. Thus, this is all the student is excepted to carefully memorize (doing the exercises for the next lessons, you will also need vocabulary intro71
duced earlier). We will introduce twelve new words in each lesson: a tting number, since Tolkiens Elves preferred counting in twelves rather than tens as we do. A unied list of all the vocabulary henceforth employed in the exercises of this course can be found in Appendix A. min one (from now on, we will introduce one new number in each lesson) e Anar (the) Sun Isil (the) Moon ar and (a most useful word that will allow us to have two exercises in one. . . translate
the Sun and the Moon, for instance. . . )
Elda Elf li people (sc. an entire ethnic group or even race, as in Eldali = the People of e e
the Elves).
vend maiden(in archaic Quenya wend) e e rocco horse (specically swift horse for riding, according to Letters:382 aran king tri queen a tasar willow (by its form this could be the plural of **tasa, but no such word
exists, and -r is here part of the basic word and not an ending. This word occurs, compounded, in LotR Treebeard chanting In the willow-meads of Tasarinan [Willow-vale] I walked in the spring. . . )
nu under
Exercises
1. Translate into English (or whatever language you prefer): A. Roccor B. Aran (two possible English translations!) C. I rocco. D. I roccor. E. Arani. F. Min li nu min aran. e e e G. I aran ar i tri. a H. Vendi. 2. Translate into Quenya: I. Willows. J. Elves. 72
K. The kings. L. Peoples. M. The horse under (or, below) the willow. N. A maiden and a queen. O. The queen and the maidens. P. The Sun and the Moon (I promised you that one. . . )
73
Lesson 3
Dual number. Stem variation.
3.1
Dual number
The previous lesson covered two Quenya plural forms: the somewhat mysterious partitive plural in -li, and the normal plural in either -r or -i (mostly depending on the shape of the word). Like quite a few real languages, Quenya also possesses a dual form, that has no direct counterpart in English. Dual number refers to two things, a couple of things. The dual is formed with one out of two endings: -u or -t. Within the ctional timeline imagined by Tolkien, these two endings originally had somewhat dierent meanings, and so were not completely interchangeable. A footnote in Letters:427 provides some information on this. The ending -u (from Primitive Elvish - ) was originally used in the case u of natural pairs, of two things or persons somehow belonging together as a logical couple. For instance, according to VT39:9, 11, the word p lip has e the dual form peu lips, referring to one persons pair of lips (and not, for instance, to the upper lip of one person and the lower lip of another, which would be just two lips and not a natural pair). The noun veru, meaning married pair or husband and wife, has dual form; in this case there does not seem to be a corresponding singular spouse (but we have verno husband and vess wife from the same root; see LR:352). The e noun alda tree occurs in dual form with reference, not to any casual pair of trees, but the Two Trees of Valinor: Aldu. Notice that if the ending -u is added to a noun ending in a vowel, this vowel is displaced: hence the dual of alda is aldu rather than **aldau though a word quoted in PM:138, reproducing a draft for the LotR Appendices, seems to suggest that Tolkien for a moment considered precisely the latter form. There is also an old source that has Aldaru, apparently formed by adding the dual ending -u to the normal plural aldar trees, but this seems to be an early experiment of Tolkiens that was probably long obsolete by the time he wrote LotR. In the dual form peu, the nal vowel of p lip e is apparently not displaced by the dual ending -u. However, Quenya p is e 74
meant to descend from primitive Elvish pee, whereas the dual form peu n is meant to come from peu (VT39:9) so the e of peu was not originally n nal. As for the other dual ending, -t, it according to Letters:427 represents an old element ata. This, Tolkien noted, was originally purely numerative; it is indeed related to the Quenya word for the numeral two, atta. By purely numerative, Tolkien evidently meant that the dual in -t could denote two things only casually related. For instance, ciryat as the dual form of cirya ship could refer to any two ships; ciryat would only be a kind of spoken shorthand for the full phrase atta ciryar, two ships. However, Tolkien further noted that in later Q[uenya], the dual forms were only usual with reference to natural pairs. Precisely what he means by later Quenya cannot be determined; it could refer to Quenya as a ritual language in Middle-earth rather than the vernacular of the Eldar in Valinor. In any case, the Third Age Quenya we aim for in this course must certainly be included when Tolkien speaks of later Quenya, so here we will follow the rule that any dual form must refer to some kind of natural or logical pair, not to two things only casually related. In other words, the dual in -t came to have just the same meaning as a dual in -u. A dual like ciryat 2 ships (curiously spelt ciriat in Letters:427, perhaps a typo) would not in later Quenya be used with reference to any two ships, but only of two ships that somehow form a pair like two sister ships. If you just want to refer to two ships that do not in any way form a natural or logical pair, like any two ships that happen to be seen together, you would not use the dual form but simply the numeral atta two hence atta ciryar. Since the two endings -t and -u had come to carry the same meaning, some rule is needed to determine when to use which. Which ending should be used can apparently be inferred from the shape of the word itself (just like the shape of the word normally determines whether the plural ending should be -i or -r). In Letters:427, Tolkien noted that the choice of t or u [was] decided by euphony, sc. by what sounded well adding as an example that -u was preferred to -t if the word that is to receive a dual ending already contains a t or the similar sound d. Hence the dual of alda is aldu rather than **aldat. It seems that as far as later Quenya is concerned, -t would be your rst option as the dual ending, but if the noun it is to be added to already contains t or d, you opt for -u instead (remembering that this ending displaces any nal vowels). The duals Tolkien listed in the Plotz Letter, ciryat a couple of ships and lasset a couple of leaves (formed from cirya ship and lass leaf) conrm that a words with no t or e d in them take the dual ending -t. Perhaps the ending -u would also be preferred in the case of nouns ending in a consonant, since -t could not be added directly to such a word without producing a nal consonant cluster that Quenya phonology wouldnt allow; unfortunately we have no examples. (If the ending -t is to be used anyhow, a vowel would probably have to be 75
inserted before it, producing a longer ending likely -et. We will eschew this little problem in the exercises below, since nobody really knows the answer.) It is clear, however, that Quenya has a number of old duals that do not follow the rule that the ending is normally -t, replaced by -u only if there is a d or t in the word it is to be added to. The examples veru married pair and peu lips, pair of lips are proof of that; here there is no t or d present, but the ending is still -u rather than -t. Presumably these are fossilized dual forms reecting the older system in which only -u denoted a natural or logical pair. The example peu (pair of) lips suggests that the ending -u is used in the case of body-parts occurring in pairs, such as eyes, arms, legs. (The other ending -t may however be used if certain other endings intrude before the dual ending itself; we will return to this in a later lesson.) The word for arm is ranco; the dual form denoting one persons pair of arms is not attested, but my best guess is that it would be rancu. The compound hendumaica sharp-eye[d] mentioned in WJ:337 may incorporate a dual hendu (pair of) eyes. The Quenya word for eye is known to be hen, or hend- before an ending (the Etymologies only mentions the normal plural hendi eyes, LR:364). In the case of this word the dual ending would be -u rather than -t anyway, since there is a d in hend-. The word for foot, tl, probably has the dual talu (for the shortening of the vowel, see below). a
3.2
Stem variation
This is a subject we shall have to spend some paragraphs on, since even on this early stage of the course we havent been able to wholly eschew it. I will go into some detail here, but students can rest assured that they are not expected to remember all the words and examples below; just try to get a feel for what stem variation is all about. Sometimes the form of a Quenya word subtly changes when you add endings to it. Two such words were mentioned above. If you add an ending to tl foot, for instance -i for plural or -u for dual, the long vowel is a a shortened to a. So the plural feet is tali rather than **tli, the dual a a couple of feet is talu rather than **tlu. In such a case, tl foot may be a a said to have the stem tal-. Likewise, the word hen eye has the stem hend, since its plural is hendi and not just **heni. The stem form does not occur by itself, but is the form you add endings to. When presenting a gloss, I will represent such stem variation by listing the independent form rst, followed by a parenthetical stem form with a hyphen where the ending goes, e.g.: tl (tal-) foot, hen (hend-) eye. a In the case of tl vs. tal-, the variation is apparently due to the fact a that vowels were often lengthened in words of only one syllable, but when the word had endings the word obviously got more than one syllable and so the lengthening did not occur (another example of the same seems to be
76
nr man vs. plural neri men, MR:213/LR:354). Originally, the vowel e was short in all forms. It is usually true that the stem form gives away how the word looked at an earlier stage in the long linguistic evolution Tolkien dreamed up in great detail. Hen eye in its stem hend- reects the primitive base khen-d-e from which it is ultimately derived (LR:364). Quenya could not have -nd at the end of a word and simplied it to -n when the word stands alone (thus, hen in a way represents the impossible full form hend), but before an ending the group -nd- was not nal and could therefore actually appear. Very often stem variation has to do with clusters or sounds that are not allowed at the end of words, but that may appear elsewhere. Cf. a word like talan oor. The plural oors is not **talani as we might expect, but talami. The stem is talam- because this is the form of the Primitive Elvish root-word: talam (LR:390). As Quenya evolved from Primitive Elvish, a rule came into place that only a few consonants were allowed at the end of words, and m was not one of them. The closest permissible consonant was n, and so the old word talam was altered to talan but in the plural form talami (and other forms that added an ending to the word), the m was not nal and therefore persisted unchanged. Another, similar case is lit small bird, that has the stem lic- (e.g. plural lici small birds): The primitive root-word was philik (LR:381), but Quenya did not permit -k at the end of a word, so in that position it became -t. When not nal it remained k (here spelt c). In some cases, the independent form is a simplied or shortened form of a word, while the stem form reects the fuller form. For instance, Tolkien apparently imagined that the word merend feast, festival was often e shortened to meren, but the stem is still merend- (LR:372). Hence the plural of meren is merendi, not **mereni. When it stands alone, the word niss woman is normally reduced to nis (or n with a lengthened e s vowel), but the double S persists before endings: thus the plural women is nissi (LR:377, MR:213). A similar case is Silmarill, the name of one e of the legendary jewels created by Fanor; this is normally shortened to e Silmaril, but before endings the double L of the full form is preserved (Silmarill-); hence the plural is always Silmarilli. In the case of compound words, sc. words made up from several other words, the second element in the compound is often reduced, but a fuller form may turn up before an ending. For instance, the noun Sindel Grey-elf (WJ:384) incorporates -el as a reduced form of Elda Elf. The plural of Sindel is not **Sindeli, but Sindeldi preserving the cluster -ld- seen in Elda. (Since the nal -a is lost in the compound, we cannot have the plural **Sindeldar.) In some cases a word may be contracted when you add endings to it. In such cases the stem-form does not reect the older, more complete form of the word. Such contraction often occurs in two-syllable words containing two identical vowels. For instance, feren beech-tree is reduced to fernbefore an ending, e.g. plural ferni instead of **fereni. WJ:416 likewise 77
indicates that laman animal may be reduced to lamn- before an ending, hence for instance lamni animals, though the unreduced form lamani was also in use. Occasionally, the contracted forms suer further change when compared to the unreduced form; as the plural of seler sister we might expect **selri, but since lr is not a permissible consonant cluster in Quenya, it is changed to ll the actual plural sisters being selli (LR:392). Another form of stem-variation is very poorly attested as far as nouns are concerned, but there are hints to the eect that the nal vowel of some words would change when an ending is added. In Quenya, the nal vowels -o and - sometimes come from -u and -i in Primitive Elvish. At one stage e of the linguistic evolution, original short -i became -e when the vowel was nal; in the same environment original short -u became -o. For instance, the primitive word tundu hill, mound came out as tundo in Quenya (LR:395). But since this change only occurred when the vowel was nal, it is possible that its original quality would be preserved before an ending. The plural hills may well be tundur rather than tundor, though neither form is attested. According to SD:415, the Quenya noun lm night has the stem o e lmi-, evidently meaning that the nal vowel - changes to -i- if you add an o e ending after it. For instance, adding the dual ending -t to lm (to express o e a couple of nights) would presumably produce lmit rather than lmet. o o This would be because lm comes from Primitive Elvish dmi (LR:354), o e o and -i never turned into -e except when nal. Some think certain words in Namri, l a e rinen and s rinen, are attested examples of this phenomenon: u These are forms of l e song and s r wind (the latter is attested by itr u e self in MC:222; the meaning of the ending -nen seen in l rinen and s rinen u will be discussed in a later lesson). If this word originally ended in an -i that became - only later (and only when nal), it may explain why in this word e - seemingly turns into -i- before an ending. We would then say that s r e u e has the stem s ri-. u There seems to be a similar variation between involving the nal vowel -o, that in some cases descend from nal -u in Primitive Elvish; again the primitive quality of the vowel may be resurrected if an ending is added to it. For instance, rusco fox is said to have the stem ruscu-, so if we add the dual ending to speak of a a couple of foxes, the resulting form should presumably be ruscut rather than ruscot. However, there is no extensive treatment of this phenomenon in Tolkiens published writings; indeed the statements made in SD:415 and VT41:10 that lm and rusco have stems o e lmi-, ruscu- are as close as we get to explicit references to it. o The student should not despair, thinking that all sorts of strange things typically happen whenever you add an ending to a Quenya word, so that there is a great potential for making embarrassing mistakes (or at least very much extra stu to memorize). Most Quenya words seem to be quite wellbehaved, with no distinct stem form to remember; you just add the ending and thats it. Where a distinct stem-form is known to exist (or where we 78
have good reason to suspect one), this will of course be indicated when I rst present the word, if it is relevant for the exercises.
Vocabulary
atta two hen (hend-) eye ranco arm ando gate cirya ship aiw bird e talan (talam-) oor nr (ner-) man (adult male of any sentient race Elvish, mortal or otherwise) e n (niss-) woman (similarly: adult female of any sentient race) s sar (sard-) stone (a small stone not stone as a substance or material) alda tree oron (oront-) mountain
Exercises
1. Translate into English: A. Hendu B. Atta hendi (and answer: what is the dierence between this and hendu
above?)
C. Aldu D. Atta aldar (and answer again: what is the dierence between this and
Aldu above?)
E. F. G. H.
2. Translate into Quenya: I. J. K. L. M. N. O. P. Two ships (just any two ships that happen to be seen together) Two ships (that happen to be sister ships) Arms (the two arms of one person) Two mountains (within the same range; Twin Peaks, if you like use a
dual form)
Double gate (use a dual form) Two birds (that have formed a pair) Two birds (just any two birds) Men and women. 80
Lesson 4
The Adjective. The Copula. Adjectival agreement in number.
4.1
The Adjective
The vocabulary of any language can be separated into various classes of words various parts of speech. Tolkiens languages were designed to be denitely of a European kind in style and structure (Letters:175), so the parts of speech they contain are not very exotic, but ought to be quite familiar to any schoolchild in Europe or America. We have already mentioned the nouns, which by a somewhat simplied denition are words denoting things. Now we will move on to the adjectives. Adjectives are words that have assumed the special function of description. If you want to say that someone or something possesses a certain quality, you can often nd an adjective that will do the job. In a sentence like the house is red, the word red is an adjective. It describes the house. There are adjectives for all sorts of qualities, quite useful if you want to say that someone or something is big, small, holy, blue, silly, rotten, beautiful, thin, nauseous, tall, wonderful, obnoxious or whatever the occasion demands. One often distinguishes two dierent ways of using an adjective: 1. You can team it up with a noun which it then describes, resulting in phrases like tall men or (a/the) red book. Such phrases can then be inserted into a full sentence, like tall men scare me or the red book is mine, where the words tall, red simply provide extra information about their companion nouns. This is called using the adjective attributively. The quality in question is presented as an attribute of the noun, or is attributed to it (tall men OK, then we know precisely what kind of men we are talking about here, the tall ones, their tallness being their attribute). 2. But you may also construct sentences where the whole point is that someone or something possesses a specic quality. You dont just presuppose the tallness as when you speak of tall men you want to say that the men are tall, thats the very piece of information you want to convey. This is called using an adjective predicatively: You choose a party you want to say something about, like the men in this case, and then add an adjective to tell 81
what quality this party possesses. The adjective is then called the predicate of this sentence. As the attentive reader already suspects from the example above, there is one more complication: You dont just say the men tall, but the men are tall. Actually sentences like the men tall would be quite OK in a great number of languages (and Quenya may even be one of them), but in English you have to slip in a word like are or is before the adjective when you use it as a predicate: The book is red. The men are tall. This is/are doesnt really add a whole lot of meaning here (there is a reason why so many languages manage without any corresponding word!), but it is used to couple the adjective with the words that tell us what we are really talking about here like the book and the men in our example. Hence is/are is called a copula. In sentences like gold is beautiful, I am smart or stones are hard, it can be perceived the prime function of the copula (here variously manifesting as is, am and are) is simply to connect the following adjectives beautiful, smart, hard with the thing(s) or person we are talking about: gold, I, stones. The copula is an integral part of the predicate of the sentence. This is one of the most important constructions speakers or English have at their disposal when they want to say that X possesses the quality Y. Well, lets get down to Quenya here. When compared to the plethora of shapes that a noun can have, Quenya adjectives are quite restricted in form. The vast majority of Quenya adjectives end in one of two vowels -a or -. The latter ending is the less common and typically occurs in coloure adjectives: Ninqu white, mor black, carn red, varn brown e e e e etc. When an adjective does not end in -a or -, it virtually always ends e in -in, e.g. rin dead, hwarin crooked, melin dear or latin open, free, cleared (of land). The latter adjective is actually listed as latin(a) in Tolkiens writings (LR:368), evidently suggesting that latin is shortened from a longer form latina, both variants occurring in the language. (Perhaps all the adjectives in -in are to be considered shortened forms or full forms in -ina.) Adjectives that do not end in either -a, - or -in are extremely rare; e there is at least teren slender but even this adjective also has a longer form in - (teren). e e Adjectives in -a are by far the most common type. The nal vowel -a may appear by itself, as in lra at, but it is often part of a longer ada jectival ending like -wa, -na (variant -da), -ima or -ya. Examples: helwa (pale) blue, harna wounded, melda beloved, dear, melima loveable, vanya beautiful. The word Quenya itself is in its origin a yaadjective meaning Elvish, Quendian, though Tolkien decided that it came to be used only as a name of the High-elven language (Letters:176, WJ:360 361, 374). In Quenya as in English, an adjective can be directly combined with a noun, describing it. We have many attested examples of adjectives being used attributively like this; they include the phrases lint yuldar swift e 82
draughts (Namri ), luini tellumar blue vaults (prose-style Namri ), a e a e fna cirya a white ship (Markirya), quant tengwi full signs (a term a e used by early Elvish linguists; we neednt discuss its precise meaning here; see VT39:5). In these examples, the word order is the same as in English: adjective + noun. This is apparently the normal, preferred order. In Quenya, it is however also permissible to let the adjective follow the noun. For instance, Markirya has anar p ra for a bleared sun, literally (a) sun u e bleared, and in LR:47 we have mall tra, literally road straight, for e e a straight road (cf. LR:43). Perhaps this word order is used if you want to emphasize the adjective: the context in LR:47 indicates that this is a straight road as opposed to a bent one. However, letting the adjective follow the noun may be the normal word order in the case of an adjectival title that is used in conjunction with a proper name: In UT:305 cf. 317 we have Elendil Voronda for Elendil the Faithful (well, the form found in UT:305 is actually Elendil Vorondo, because the phrase is inected; we will return to the ending -o here seen in a later lesson). Presumably you could also use the more normal word-order and speak of voronda Elendil, but that I guess would simply be a more casual reference to faithful Elendil, not meaning Elendil the Faithful with the adjective used as a regular title. It may be noted that Quenya, unlike English, does not insert the article before an adjective used as a title (not **Elendil i Voronda, at least not necessarily). What, then, about using adjectives as predicates, like red is the predicate of the sentence the book is red? (Contrast the attributive use of the adjective in a phrase like the red book.) The adjective vanwa lost is used predicatively in Namri : Vanwa n . . . Valimar lost is . . . Valia e a mar (a place in the Blessed Realm that Galadriel thought she would never see again). This sentence tells us that the Quenya copula is has the form n. Plural are seems to be nar, attested in an early version of Namri a a e recorded by Tolkien on tape (see Jim Allans An Introduction to Elvish, p. 5). It is generally assumed that these copulas would be used as in English, for instance like this: I parma n carn. The book is red. a e Ulundo n uma. A monster is evil. a e I neri nar hall. The men are tall. e In this lesson as originally published in December 2000, I slipped in a warning at this point: I should add, though, that due to the extreme scarcity of examples we cant be certain what the preferred word order really is. From the example vanwa n . . . Valimar lost is . . . Valimar a in Namri one could argue that n should follow the adjective, a e a so that the book is red should rather be i parma carn n, e a 83
the book red is. It would be interesting to know if n is a would still follow vanwa lost if we relocated Valimar to the beginning of the sentence; should Valimar is lost be Valimar n vanwa, English-style, or perhaps Valimar vanwa n? In a a the examples above and the exercises below I have organized the sentences using the English word order, but Tolkien may have had something more exotic up his sleeve. There is no way of telling before more material is published. I revise this lesson in November 2001, and this summer a few more examples involving the word n is nally became available. There does seem a to be a tendency to place n at the end of the sentence, as in the example a l caritas . . . alasaila n (literally, not to do it unwise is VT42:34). a a Yet the same article that provides this example also cites the formula A n a calima l B (literally, A is bright beyond B) as the Quenya way of exa pressing A is brighter than B (VT42:32). Notice that this formula employs an English-style word-order, with n is preceding rather than following a calima bright. So it seems that sentences like i parma n carn, word a e by word corresponding to English the book is red, may be possible after all. Therefore I have not revised any of the examples or exercises of this course, all of which employ the English word order as far as the copula n is concerned. It seems, however, that the order i parma carn n the a e a book red is must be considered a perfectly valid alternative, and Tolkien may even have intended this to be the more common word order. We must await still more examples. [New update, January 2002: This month some new examples were indeed published. It seems that the exact word order is simply a matter of taste. The example ely na manna thou art blessed from VT43:26 has an e English-style word order, and here the copula is/art appears in the short form na rather than n. I have however maintained n in the exercises of a a this course; the word na has several other, quite distinct meanings. But perhaps the short form na- is consistently preferred when some ending is to be added; cf. the plural form nar are. Of course, the unattested form nr a could be equally valid for all I know.] In F riels Song (a pre-LotR text), the word for is appears as ye rather than n or na, as in a rima ye N menor lovely is Nmenor (LR:72). u u However, both the Qenya Lexicon (QL:64) and the Etymologies (LR:374) point to n instead, and in Namri we have this word attested in an actual a a e text. Etym and the QL are earlier than F riels Song, but Namri is later, a e so would seem that ye was just a passing experiment in Tolkiens evolution of Quenya. In F riels Song we also see an ending for is, -i, appended to e adjectives and displacing their nal vowel: hence in this song we have mri a e for (it) is good, derived from the adjective mra good. This ending a i is transparently related to the independent word ye. I dont think the e 84
system of using the ending -i for is was still valid in LotR-style Quenya, e and I wouldnt recommend it to writers. The ending -i has other meanings e in later Quenya. Another system may well be valid, though: using no copula at all. You simply juxtapose the noun and the adjective, the word is/are being understood: Ilu vanya the World [is] fair (F riels Song), maller raicar roads [are] bent (LR:47). The formula A is bright beyond B = A is brighter than B referred to above is actually cited as A (n) calima l a a B in VT42:32. As suggested by the parentheses, n could be omitted. The a example malle tra a straight road mentioned above could also be intere preted a road [is] straight, if the context allowed it. The nal version of Tolkiens Quenya translation of the Hail Mary, published in January 2002, leaves out several copulas: Aistana ely, ar aistana i yv mnalyo = e a e o blessed [art] thou, and blessed [is] the fruit of thy womb. We must assume that the copula n, nar is not limited to combining nouns a and adjectives, but can also be used to equate nouns: Parmar nar engwi books are things, Fanro n Noldo Fanor is a Noldo. (Notice, by the e a a e way, that the proper Quenya form of Fanors name is Fanro; Fanor is e e a e a Quenya-Sindarin hybrid form used in Middle-earth after his death.) Again it may be permissible to leave out the copula and retain the same meaning: Parmar nati, Fanro Noldo. e a Adjectival agreement in number: Quenya adjectives must agree in number with the noun they describe. That is, if the noun is plural, the adjective must be, too; if the adjective describes several nouns it must also be plural, even if each of the nouns is singular. English makes no such distinction its adjectives do not change but it is not surprising that Tolkien built adjectival agreement in number into Quenya, since this was to be a highly inected language. We have no examples of what happens if an adjective is to agree with a noun in the dual form (or, for that matter, with a partitive plural noun in -li). It is generally assumed, though, that there are no special dual or partitive plural forms of adjectives, but only one plural (or should we say non-singular) form. How, then, is the plural form of adjectives constructed? From the examples now available, it can be seen that Tolkien experimented with various systems over the years. In early sources, adjectives in -a form their plural form by adding the ending -r, just like nouns in -a do. For instance, one very early map of Tolkiens imaginary world (actually depicted as a symbolic ship) includes a reference to I Nori Landar. This evidently means The Wide Lands (LT1:8485; the adjective landa wide occurs in the Etymologies, entry lad. Christopher Tolkien in LT1:85 suggests the translation The Great Lands.) Here the plural noun nori lands is described by the adjective landa wide another example of an attributive adjective following the noun, by the way and since the noun is plural, the adjective takes 85
the plural ending -r to agree with it. This way of forming plural adjectives was still valid as late as 1937 or slightly earlier; we have already quoted the example maller raicar roads [are] bent from LR:47, where the adjective raica crooked, bent, wrong (listed by itself in LR:383) is plural to agree with maller. However, this system cannot be recommended to writers; the evidence is that in LotR-style Quenya, it had been abandoned. Tolkien in a way reached back into the past and revived a system he had used in what may be the very rst Qenya poem he ever wrote, Narqelion of 191516. In this poem, adjectives in -a form their plurals by means of the ending -i. For instance, the phrase sangar umai occurring in this poem apparently means e throngs large = large throngs; the adjective uma large is listed in the e early Qenya Lexicon (QL:97 but in later Quenya, the word uma means e evil instead). Later, Tolkien however introduced one more complication: Adjectives in -a had plurals in -ai in archaic Quenya only. In Exilic Quenya, Quenya as spoken by the Noldor after they had returned to Middle-earth, -ai at the end of words of more than one syllable had been reduced to -. (Cf. e WJ:407 regarding the ending -v representing archaic Q -vai .) So while e the plural form of, say, quanta full was apparently quantai at the older stages of the language, it later became quant. This form we have already e met in one of the examples quoted above: quant tengwi, full signs, e where quanta appears in the plural form to agree with tengwi signs (VT39:5). There is one special case to be considered: adjectives in -a, such as laura e e golden. In archaic Quenya, we must assume that the plural form was simply laurai. But when -ai later became -e, what would be ?laure did e e not prove to be a durable form. To avoid the cumbersome combination of two concomitant es, the rst of them was changed to i. Hence the plural form of laura in Exilic Quenya appears as lauri, as in the rst line of e e Namri : Ai! lauri lantar lassi s rinen. . . Ah! golden fall the leaves in a e e u the wind. . . the adjective being plural to agree with the noun it describes, lassi leaves. As for adjectives in -, they seem to behave like most nouns of the same e shape: - becomes -i in the plural. We dont have very many examples, but e the phrase luini tellumar blue vaults in the prose version of Namri a e seems to incorporate the plural form of an adjective luin blue (actually e not attested in this form, but as observed above, there are many colouradjectives in -). Moreover, in the Etymologies Tolkien noted that the an e adjective mait handy, skilled has the plural form maisi (LR:371). Evie dently the plural form was especially mentioned primarily to illustrate another point: that adjectives in -it have plural forms in -isi, the consonant e t turning into s before i. This particular idea seems to have been dropped later, though: In a much later, post-LotR source, Tolkien wrote hlon ti tengwi, not ?hlon tengwi, for phonetic signs (WJ:395). So perhaps si 86
the plural form of mait could simply be ?maiti as well. e As for the plural form of adjectives ending in a consonant, such as rin dead, we dont seem to have any examples to guide us. It has traditionally been assumed that they form their plurals in -i, just like nouns of this shape do, and this still seems reasonably plausible. So, say, dead men could be rini neri. If any argument can be raised against this assumption, it is that adjectives in -in actually seem to be shortened forms of longer adjectives in -ina. As pointed out above, Tolkien quoted the adjective meaning open, free, cleared (of land) as latin(a), indicating double forms latin and latina. The plural form of latina should obviously be latin, for older latinai. But e what about latin? If this is merely a shortened form of latina, perhaps the plural form would still be latin rather than latini? We cannot know for e certain; in the exercises below I have followed the traditional assumption, using plurals in -i. Adjectives ending in a consonant are quite rare anyway, so this uncertainty does not greatly jeopardize the quality of our own Quenya texts. In what positions do adjectives agree in number? Attested examples like those already quoted, like luini tellumar blue domes, would seem to indicate that an attributive adjective in front of the noun does show agreement. So does an attributive adjective following the noun; the Markirya poem has i fairi nc for the pale phantoms, or literally the phantoms pale e e (nca pl. nc vague, faint, dim to see, MC:223). An adjective separated e e e from the noun it describes also agrees in number, hence laura golden ape pears in plural form lauri in the rst line of Namri, lauri lantar lassi e a e e golden fall leaves (the prose Namri has lassi lantar lauri leaves fall a e e golden). As for predicative adjectives, we lack late examples. In German, adjectives do agree in number when they are used attributively, but adjectives used predicatively do not. Yet the old example maller raicar roads [are] bent in LR:47 would seem to indicate that in Quenya, adjectives agree in number also when they are used predicatively. In later Quenya we should presumably read maller (nar) raic, since Tolkien changed the rules for e how the plural form of adjectives is constructed. So in short, we can conclude that adjectives agree in number with the nouns they describe everywhere whether they appear before, after or separated from the noun, whether they are used attributively or predicatively. There are a few examples that dont quite t in, though. Appendix E of the essay Quendi and Eldar of ca. 1960 contains several well-behaved examples of plural adjectives that are used attributively with the plural noun tengwi signs, making up various phrases used by early Elvish linguists when they tried to analyze the structure of their tongue (as I said above, we neednt concern ourselves with the precise meaning of these terms here). Besides hlon tengwi phonetic signs and quant tengwi full ti e signs already quoted (WJ:395, VT39:5), we have racin tengwi stripped e signs and peny tengwi lacking signs (VT39:6; the singular of the latter, e 87
penya tengw a lacking sign, is attested: VT39:19). In these phrases the e adjectives hlon e phonetic, quanta full, racina stripped, deprived t and penya lacking, inadequate all assume their plural forms, beautifully agreeing with tengwi signs, elements, sounds. So far, so good. But then we turn to the draft material for Appendix E of Quendi and Eldar. Here Tolkien did not let the adjectives agree in number, and we have phrases like lehta tengwi free/released elements, sarda tengwi hard sounds and tapta tengwi impeded elements (VT39:17). We would of course expect leht tengwi, sard tengwi, tapt tengwi, but these are not found. e e e Unless we are to assume that there are several classes of adjectives, some that agree in number and others that dont and I think this is rather farfetched it seems that Tolkien in the draft material used a system whereby an attributive adjective immediately in front of its noun does not agree in number. But when he actually wrote the Appendix, he would seem to have introduced agreement in this position as well, and so we have for instance quant tengwi rather than ?quanta tengwi for full signs. Elvish grame mar could change at lightening speed whenever Tolkien was in his revision mood, so this would not be surprising. The last version of the Markirya poem, which Christopher Tolkien thinks was written at some point in the last decade of his fathers life (196373), is also relevant here. In the phrase fallen towers, Tolkien rst wrote the adjective atalanta ruinous, downfallen in its plural form atalanti, just like e e we would expect. Then, according to Christopher Tolkien, he mysteriously changed atalanti to the singular (or rather uninected) form atalanta, e e though the adjacent noun towers was left in the plural (MC:222). Again Tolkien seems to be experimenting with a system whereby attributive adjectives immediately in front of the noun they describe do not agree in number, but appear in their uninected form. A similar system appears in Tolkiens writings on Westron, the Common Speech of Middle-earth (a language he only sketched). Perhaps he considered introducing such a system in Quenya as well, and we see this idea ickering on and o, so to speak, in his writings? However, the system I would recommend to writers is to let adjectives agree in number also in this position. In Namri in LotR we have the a e phrase lint yuldar swift draughts, and in the interlinear translation e in RGEO:66 Tolkien explicitly noted that lint is a pl. adjective. We e must assume, then, that lint represents older lintai, the plural form of an e adjective linta. If an attributive adjective immediately in front of the noun it describes did not agree in number, swift draughts should have been ?linta yuldar instead. The source where Tolkien explicitly identied lint e as a plural form was published during his own lifetime, and moreover as late as in 1968, possibly postdating even the last version of Markirya. So his nal decision seems to have been that adjectives do agree in number with their nouns also when the adjective appears immediately in front of the noun. One suspects that he spent many sleepless nights carefully considering the 88
89
Vocabulary
Except for the two rst items, all of these are adjectives. Dont worry about the other words occurring in the exercises below; those you have already memorized carefully, following my instructions in Lesson Two. Right? neld three e n is (nar are) a vanya beautiful, fair alta great (= big; the word is used of physical size only) calima bright taura mighty saila wise (we will use this form found in late material; a pre-LotR source has saira
instead)
uma evil e carn red (we suspect that Tolkien the Devout Catholic was thinking about cardinals e
with their red attires; the Italian word carne = [red] meat may also be relevant here. . . )
ninqu white e mor black (cf. the rst element of Sindarin Mordor = Black Land) e rin dead
Exercises
1. Translate into English: A. Mor rocco. e B. Calim hendu. e C. Neld rini neri. e D. Vany aiwi. e E. Tri n taura n a a s. F. I oronti nar alt. e G. Aran taura (two possible translations!) H. I nr ar i n nar sail. e s e 2. Translate into Quenya: I. The white gate. J. A great ship. K. The oor is red. 90
L. One black stone and three white stones. M. Wise kings are mighty men. N. The mighty man and the beautiful woman are evil. O. Elves are beautiful. P. The Elves are a beautiful people.
91
Lesson 5
The Verb: Present tense and agreement in number. Subject/object. The superlative form of adjectives.
5.1
The Verb
As I mentioned at the beginning of the previous lesson, the vocabulary of any language can be separated into various classes of words, or parts of speech. So far we have explicitly discussed the nouns, which denote things, and adjectives, which are words used to describe nouns (linguists would nd these denitions rather simplistic, but they will do for our purpose). Actually we have already touched on a three other parts of speech as well, without discussing them in depth. As part of Lesson Two you hopefully memorized the word nu under, which is a preposition; prepositions are small words or particles like under, on, of, to, in, about etc., often used to provide information about spatial relationships (e.g. under the tree = nu i alda), though frequently they are used in more abstract contexts. With the word ar and we have also included the most typical representative of the conjunctions, words used to connect (or indeed conjoin) other words, phrases or sentences, e.g. Anar ar Isil = the sun and the moon. Still, no thorough discussion of prepositions or conjunctions as such seems necessary: in Quenya they seem to behave pretty much like their English equivalents, so for the most part you simply have to learn the corresponding Quenya words. They are not normally inected in any way. Another part of speech that we have already touched on is far more sophisticated and intriguing: the verb. We encountered one verb in the previous lesson: n is, with its plural form nar are. As verbs go, this one is not a very exciting; it is simply used to coordinate a noun with some sort of predicate that tells us what the noun is: Aran n taura, a king is mighty, a tasar n alda a willow is a tree. As I said in the previous lesson, the copa ula n doesnt really provide much extra information here, except clarifying a the relationship between the various elements of the sentence. Most other verbs (very nearly all other verbs, actually) are however full of meaning. They dont just tell us what someone or something is, but what someone 92
or something does. The Verb brings action into the language. In a sentence like the Elf dances it is easy to identify dances as the action-word, telling us what is going on here. And sure enough, dances is a form of the English verb to dance. This verb may appear in other forms as well; instead of dances we might say danced, which moves the action into the past: The Elf danced. This illustrates an important feature of verbs in European languages: the form of the verb gives information about when the action denoted takes place, in the present or in the past. Some languages also have special future forms. Tolkien built all of these features into Quenya. The dierent time-forms of the verb are called various tenses; we speak of present tense, past tense and future tense. We will only deal with the present tense in this lesson, and return to the others later. (The trinity of present, past and future does not represent a full list of all the tenses there are. We will discuss a total of ve dierent tenses in this course, and I would be very surprised if unpublished material does not describe even more tenses than the ones we know at present.) Here I should slip in a warning: We dont have much explicit information about the Quenya verb. In the so-called Plotz Letter, that Tolkien wrote to Dick Plotz at some point in the mid-sixties, he set out the declension of the noun. Apparently similar information about the verb was to follow; it never did. This is of course most unfortunate. Not that Tolkien took this information to his grave; we know that he did write about these matters, but the relevant writings have not been published. For the time being, we must for the most part try to gure out the grammatical rules by ourselves if we would like our Quenya poems to include verbs. Regarding the present tense, some scraps of information luckily appeared in Vinyar Tengwar #41, July 2000. Combining this info with some linguistic deduction, we can probably make out the main features of the system Tolkien had in mind. As they appear in various sources, Quenya verbs seem to fall into two main categories (though there are some verbs in our corpus that dont readily t into either, even if we exclude the early Qenya material where some really weird things are going on in the verbal system). The rst and largest category is what can be termed A-stems, for they all end in -a. Another term for the same is derived verbs, for these verbs never represent a naked primitive root-word, but are derived by adding endings to this root. The most frequent of these endings are -ya and -ta; much less frequently we see -na or just -a. Examples: calya- to illuminate (root kal) tulta- to send for, to fetch, to summon (root tul) harna- to wound (root skar; primitive initial sk - became h- in Quenya) mapa- to grasp, to seize (root map) (Convention has it that when you list verbal stems as such, you add a 93
hyphen at the end; Tolkien usually does so in his writings. The stem of a verb is a basic form that we start from when deriving other forms, such as dierent tenses.) If these A-stems can be termed derived verbs, the other category consists of the non-derived or primary verbs. These are verbs that display no such ending as -ya, -ta, -na or -a. The verbal stems in question can be termed primary or basic since they essentially represent a primitive root with no additions. For instance, the verb mat- eat comes directly from the root mat- of similar meaning. Tac- fasten represents the root takx, make fast. Tul- come can be identied with the root tul- come, approach, move towards (contrast the derived verb tulta- send for, summon, fetch from the same root, derived by means of the ending -ta). In the case of the roots mel- love and sir- ow, Tolkien didnt even bother to repeat the glosses for the Quenya verbs mel- and sir- (see LR:372, 385). When discussing Quenya verbs, we sometimes need to refer to the stemvowel. This is the vowel of the root-word underlying the verb as it appears in Quenya. In the case of primary verbs like mel- love, it is of course easy to identify the stem-vowel, since e is the only vowel there is (and sure enough, this is also the vowel of the underlying root mel-). In the case of derived verbs like pusta- stop or ora- impel, the vowels of the added ending (here -ta and -a) do not count as stem-vowels. Pusta-, for instance, is derived from a root pus, and its stem-vowel is therefore u, not a. In the vast majority of cases, the stem-vowel is simply the rst vowel of the verb (but not necessarily so, there may be some prexed element). With this we have the necessary terms in place and can nally start discussing the formation of the present tense. To start with the primary verbs, what seems to be the present tense of the verb mel- love is attested in LR:61, Elendil telling his son Herendil: Yonya iny tye-mla, I too, my e e son, I love thee. Here we have the verb describing a present or on-going (in this case quite permanent) action. Another example of a present-tense primary verb can apparently be found in the LotR itself, in the famous greeting elen s l menn omentielvo, a star shines [or, is shining] upon the la u hour of our meeting. S would seem to be the present tense of a verb la sil- shine (with white or silver light), listed in the Silmarillion Appendix. Mla and s show the same relationship to the simple verbal stems mele la and sil-: the present-tense forms are derived by lengthening the stem-vowel (this is denoted by supplying an accent, of course) and adding the ending -a. This conclusion is supported by an example from VT41:13: The verb quetspeak, say there appears in the present tense quta is saying. e Though forms like mla and s may occasionally be translated using e la the simple present tense in English, hence love(s) and shines, is seems that the Quenya present tense properly denotes a continuous or ongoing action that is best translated using the English is . . .-ing construction, as in the example quta just quoted: this is is saying rather than just e 94
says. The conclusion that the Quenya present tense properly denotes continuous actions is also supported by other evidence: The Quenya present tense of the primary verb mat- eat is nowhere attested in published material. However, Tolkien stated that mt was the stem of the continuous aa form, which could be translated is eating (VT39:9; a here denotes long a, in Quenya spelt ). Tolkien actually put an asterisk in front of mt to a aa mark it as an unattested form, so this should evidently be taken as primitive Elvish rather than Quenya. How Quenya evolved from the primitive language can be inferred from many other examples, so we know that mt aa would come out as mta. This form would seem to fall into the same pata tern as mla, s and quta: lengthened stem-vowel and ending -a (and e la e working backwards, we can deduce that Tolkien meant mla, s quta e la, e to be descended from primitive Elvish ml, s a, kwt ). Presumably these e a l ea are all continuous forms; just like primitive mt is eating they apparaa ently emphasize the ongoing nature of the action: S may literally be is la shining rather than just shines. Perhaps the lengthening of the stemvowel somehow symbolizes this ongoing or drawn-out action. In the case of mla in the sentence iny tye-mla, it is more natural to translate I e e e love you rather than I am loving you, but the latter would seem to be the most literal meaning. Then we must consider the second and larger category of verbs, the Astems. In their case, the information from VT41 is of particular value. It seems that the A-stems form their present tense by somewhat the same rule as the primary verbs, but the rule needs a little adaptation to t the shape of an A-stem verb. Our sole attested example is the verb ora- urge or impel. VT41:13, 18 indicates that its present tense is ra (is urging). o e As in the case of primary verbs, the stem-vowel has been lengthened and the ending -a has been added. There is one complication, though: since the verbal stem ora- already ended in -a, this vowel is changed to e so as to avoid to as in sequence: What would be ra-a manifests as ra. Hence o o e we must conclude that verbs like mapa- grasp, seize and lala- laugh appear as mpa, lla in the present tense. a e ae Short A-stems like ora- or mapa- are however of a rather unusual shape, since they add only the simple vowel -a to the original root. As discussed above, A-stems where the nal -a is only part of a longer derivational ending (most often -ya or -ta) are much more common. We have already quoted examples like calya- to illuminate and tulta- to summon (roots kal, tul). Such complex A-stems have a consonant cluster following the vowel of the original root, like ly and lt in these examples. We have no actual example of the present tense of such a verb. If we were to apply the pattern we deduce to exist from the example ra is urging, it would land us on o e forms like ?clya is illuminating and ?t lta is summoning. However, a e u e there seems to be a phonological rule in Quenya prohibiting a long vowel immediately in front of a consonant cluster. It would seem that a word like 95
?t lta cannot exist (but frankly Im not quite sure about ?clya, since u e a e ly/ny/ry sometimes seem to count as unitary palatalized consonants rather than consonant clusters). Lacking actual examples, we can only assume that in such a case the lengthening of the vowel would simply be dropped, so that the present tense of verbs like calya- and tulta- would be calya, tulta e e (though as I just indicated, ?clya may be possible for all I know). This a e would apply wherever there is a consonant cluster following the vowel of the verbal stem. Further examples are lanta- fall, harna- wound and pusta- stop, that would all presumably form their present-tense forms in -a: Lanta is falling, harna is wounding, pusta is stopping. e e e e We must assume that this system also applies where there is a diphthong in the verbal stem, since like a vowel in front of a consonant cluster, a diphthong cannot be lengthened in any way. The present-tense forms of verbs like faina- emit light or auta- pass would presumably be faina, e auta. e We now know enough to start building simple sentences: Isil s the Moon is shining (present tense s formed from the la la primary verb sil- shine) I Elda lla the Elf is laughing (present tense formed from the ae short A-stem lala- laugh) Lass lanta a leaf is falling (present tense formed from the come e plex A-stem lanta- fall; we cannot have *lnta paralleling lla a e ae because a long vowel cannot occur in front of a consonant cluster) Some useful terms can be included here. Once you include a verb in the sentence, denoting some kind of action, you must normally devote another part of the sentence to telling who is doing this action. The party that does whatever the verb tells us is being done, constitutes the subject of the sentence. In a sentence like Isil s the Moon is shining, it is thus Isil la the Moon that is the subject, since it is the Moon that does the shining the verb s tells us about. In a sentence like i Elda mta the Elf is la a eating, i Elda the Elf is the subject, since the Elf does the eating. This very sentence, i Elda mta, has possibilities. We can add one more a element, like the noun massa bread, and get i Elda mta massa the a Elf is eating bread. Now what is the function of this added word? It is the target of the verbal action, in this case what is eaten. The target of the verbal action is called the object, the passive counterpart of the active subject: The subject does something, but the object is what the subject does something to. The subject subjects the object to some kind of action. This action may of course be much less dramatic than subject eats object as in the example above. For instance, it can be as subtle as in the sentence the subject sees the object (ll in with other sense-verbs if you like), where the action of the subject does not physically aect the object in 96
any way. That is not the point here. The basic idea of the subject-object dichotomy is simply that the subject does something to the object, though does something to must sometimes be understood in a wider sense.
NOTE: Notice, though, that in sentences with the copula n/nar is/are, for instance a i alda n tasar the tree is a willow, tasar a willow does not count as the object a of i alda the tree. I alda is the subject all right, since this is the element that does what little action there is in this sentence: the tree is. . . But tasar a willow is not the object, for in this sentence the tree does not do anything to a willow and the hallmark of the object is that something is done to it. Rather than doing anything do a willow, the tree is a willow, and that is another thing altogether: Tasar is here the predicate of i alda, as we discussed in the previous lesson. But if we substitute mta a is eating for n is, we are right back to a subject-verb-object construction: I alda a mta tasar, the tree is eating a willow. If you are unduly troubled by the fact that a this sounds somewhat nonsensical, rest assured that the grammar is ne.
In the case of some verbs, there can be no object. In the case of (say) lanta- to fall, you can have a subject and say i Elda lanta the Elf e is falling. Here the subject doesnt do anything to an object; it is just the subject itself that is doing something. With a verb like mat- eat, it is kind of optional if you want to ll out the sentence with an object or not: I Elda mta (massa), the Elf is eating (bread); this works as a complete a sentence even without the object. But some verbs by their meaning demand an object, and the sentence would be felt to be incomplete without it. If we say i Elda mpa the Elf is seizing, this only raises the question the Elf a e is seizing what? and we must come up with an object to make the sentence complete. In the Plotz letter, Tolkien indicated that in one variant of Quenya, socalled Book Quenya, nouns would have a special form if they function as objects. Singular nouns ending in a vowel would have this vowel lengthened (for instance, cirya ship would become ciry if it appears as the object a of a sentence), and nouns that normally employ the plural ending -r would switch to -i (so ships, as object, would be ciryai instead of ciryar). This special object form, in linguistic terms the accusative case, was supposedly used in (archaic?) written Quenya. However, this accusative does not appear in any actual texts, such as Namri or even the last version of the a e Markirya poem, which must be almost contemporaneous with the Plotz letter. Namri, sung by Galadriel, is perhaps supposed to reect the usage of a e spoken Third Age Quenya. Whatever the case, I do not use the distinct accusative in the exercises I have made for this course (or in my own Quenya compositions). It seems clear that the use of the accusative was far from universal, within or without the ctional context. So I would say cirya(r) for ship(s) even if the word appears as the object of a sentence. With the terms subject and object in place, we can discuss another feature of the Quenya verb. Just like adjectives agree in number with the nouns they describe, verbs agree in number with their subjects. Let us have a closer look 97
at the rst line of Namri, lauri lantar lassi like gold fall the leaves, a e e or literally golden fall [the] leaves. Here the adjective laura golden e appears in plural form lauri to agree in number with the plural noun lassi e leaves, as we discussed in the previous lesson. But the verb lanta- to fall must also agree with its plural subject lassi. The verb lanta therefore takes the ending -r. (The verb itself appears in the so-called aorist tense, to be discussed later; you can think of aorist lantar vs. present tense lantar e as corresponding to English fall vs. are falling, respectively.) The plural ending -r we have already met in the case of nouns, as in Eldar Elves, but nouns may also have plurals in -i, depending on their shape. In the case of verbs, the plural ending -r seems to be universal, no matter what the verb looks like. The ending -r is not restricted to the present tense of verbs, but is seemingly used in all tenses, wherever a plural subject turns up. Essentially we have already met the verbal plural ending in the verb nar are, the plural of n is. (One may ask why n does not turn into ?nr a a a with the long vowel intact. The latter form may very well turn out to be valid, but nar are with a short a is at least less prone to confusion with the noun nr ame.) a More than one subject has the same eect on the verb as a (single) plural subject, the verb taking the ending -r in both instances: I arani mtar the kings are eating (sg. i aran mta the king is a a eating) I aran ar i tri mtar the king and the queen are eating (if you want a a the verb mat- eat to appear in singular present-tense form mta here, a you must get rid of either the king or the queen so that there is just a single subject) On the other hand, it has no eect on the verb if we have a plural object or multiple objects, e.g. i aran mta massa ar apsa the king is eating a bread and meat (apsa cooked food, meat). The verb agrees in number with the subject only. It has generally been assumed that the verb has only one plural form, the ending -r being universal. In other words, the verb would take the ending -r not only where the subject noun appears in the normal plural (ending -r or -i), but also where it is dual (ending -u or -t) or appears in the partitive plural form (ending -li). However, we have no actual examples from LotRstyle Quenya, and in particular I will not rule out the possibility that there may be a special dual form of the verb to go with dual subjects (ending -t as for most nouns, like Aldu s lat rather than Aldu s lar for the Two Trees are shining???) The published material allows no certain conclusion in this question, so I will simply avoid dual subjects in the exercises I make for this course. The last thing we must consider when discussing the verb is the question of word order. Where in the sentence does the verb t in, really? English 98
sentences generally list the subject, the verb and the object (if there is any object) in that order. The attentive reader will have noticed that most of the Quenya sentences above are organized in the same manner. This seems to be the most typical word order in Quenya prose. Examples of the subject and the verb in that order include lassi lantar leaves fall and morni caita e darkness lies [upon the foaming waves] both from the prose version of Namri. But there are also examples of the verb being placed rst, a e e.g. Fingons cry before the Nirnaeth Arnoediad: Auta i lm!, literally o e Passes the night, but translated the night is passing! in the Silmarillion ch. 20. Indeed both of the above-quoted examples of the order subject-verb from the prose Namri instead show the order verb-subject in the poetic a e version in LotR: lantar lassi, caita morni. In English, fronting the verb e is one way of turning a declarative statement into a question, e.g. Elves are beautiful vs. are Elves beautiful?, but this way of forming questions evidently doesnt work in Quenya. (Auta i lm! passes the night! for o e the night is passing! is perhaps an example of dramatic style or aectionate speech; the verbal action is evidently considered more important than the subject that performs it. I suspect that in a less dramatic context, one would rather say i lm auta.) o e Namri also provides an example of a sentence with both subject, verb a e and object: h e unt pa Calaciryo m mist [subject] covers [verb] the si u ri, jewels of Calacirya [this whole phrase being the object] . Yet word order is again quite exible, especially in poetry, as further examples from Namri a e shows. We have object-subject-verb in the sentence mryat Elentri ora a tan, literally her hands (the) Starqueen raised (in LotR translated the e Queen of the stars . . . has uplifted her hands). The sentence ily tier e undulv lumbul, literally all paths downlicked (i. e. covered) shadow, a e e has the order object-verb-subject (in LotR, Tolkien used the translation all paths are drowned deep in shadow). In the prose version of Namri, a e Tolkien interestingly reorganized both of these to subject-verb-object constructions: Elentri ortan mryat, lumbul undulv ily tier. This a e a e a e e is our main basis for assuming that this is the normal order, preferred where there are no poetic or dramatic considerations to be made. In general, one must be careful about putting the object before the subject, for this could in some cases cause confusion as to which word is the object and which is the subject (since the commonest form of Quenya does not maintain a distinct accusative case to mark the object). Such inversions are however quite permissible when the subject is singular and the object is plural or vice versa. Then the verb, agreeing in number with the subject only, will indirectly identify it. In the sentence ily tier undulv lume a e bul we can readily tell that it must be lumbul shadow and not ily e e e tier all paths that is the subject, because the verb undulv does not a e receive the ending -r to agree with the plural word tier. Hence this cant be the subject but the singular noun lumbul shadow can. e 99
5.2
In English and other European languages, adjectives have special forms that are used in comparison. In English, adjectives have a comparative form that is constructed by adding the ending -er, and a superlative form that is formed with the ending -est. For instance, the adjective tall has the comparative form taller and the superlative form tallest. (In the case of some adjectives, English however resorts to the independent words more and most instead of using the endings, e.g. more intelligent and most intelligent instead of intelligenter and intelligentest, which forms are perceived as cumbersome.) The function of these forms is to facilitate comparison between various parties. If we want to say that one party possesses the quality described by the adjective to a greater extent than some other party, we may use the comparative form: Peter is taller than Paul. The superlative form is used if we want to say that one party possesses the quality in question more than all others that are considered: Peter is the tallest boy in the class. In the rst version of this Quenya lesson, as published in December 2000, I wrote: But when it comes to Quenya, there is not much we can say. The published material includes absolutely no information about comparative forms; we dont even have an independent word for more. . . Since then, the situation has happily changed; during 2001 a little more information appeared in the journals Tyali Tyelelliva and Vinyar Tengwar. Now we e e do have a word for more (malda), and we also know a special formula that is used in comparison: A is brighter than B may be expressed as A n calima l B, literally A is bright beyond B (VT42:32). However, the a a word l has other meanings beside beyond, and it will be more practical a to discuss and practice its use in comparison in a later lesson (The various uses of l , Lesson Eighteen). a We will here focus on the superlative form of adjectives instead. One tiny scrap of evidence has long been available: In Letters:278279, Tolkien explained the adjectival form ancalima occurring in LotR. Translating it as exceedingly bright, he stated that this is calima shining brilliant with the element an- added, the latter being a superlative or intensive prex. For this reason, many writers have used the prex an- as the equivalent of the English ending -est, to construct the superlative form of adjectives e.g. anvanya fairest from vanya fair, beautiful (but is should be understood that ancalima remains our sole attested example of an- used in this sense). One may wonder whether the form that is made by prexing an- really is the equivalent of an English superlative, sc. a form of the adjective that implies having the most of the property involved in comparison with certain others. It may be noted that Tolkien translated ancalima, not as brightest, but as exceedingly bright. When he describes an- as a superlative or intensive prex, he may almost seem to mean superlative or 100
rather intensive prex. So perhaps an- implies very, exceedingly rather than most in comparison with others. It may be noted, though, that the context the in which the word is found does seem to imply a certain amount of comparison: In LotR, ancalima occurs as part of Frodos speaking in tongues in Shelobs lair (volume 2, Book Four, chapter IX): Aiya Erendil a Elenion Ancalima. No translation is given in the LotR itself, but Tolkien later stated that this means hail Erendil brightest of stars (Letters:385). a In Tolkiens mythology, Erendil carrying the shining Silmaril was set in the a heavens as the brightest of the stars. So here, the meaning does seem to be that of a genuine superlative, brightest in the full sense of brighter than all the others. In any case, no other information on how to form the superlative appears in published writings, so we have little choice but to use this formation. We must however be prepared that future publications may provide more information about this, involving alternative superlative formations. The prex an- in this form cannot be mechanically prexed to any Quenya adjective, or consonant clusters that Quenya does not allow would sometimes result. An- can be prexed as is to adjectives beginning in a vowel or in c-, n-, qu-, t-, v-, w-, and y-: an an ple!) an an an an an + alta great (in size) = analta greatest + calima bright = ancalima brightest (our sole attested exam+ + + + + norna tough = annorna toughest quanta full = anquanta fullest vanya beautiful = anvanya most beautiful wenya green = anwenya greenest yra old = anyra oldest a a
Perhaps we can also include adjectives in f - and h- (no examples): an + fna white = ?anfna whitest a a an + halla tall = ?anhalla tallest What would happen in other cases we cannot say for certain. Either an extra vowel (likely e or a) would be inserted between the prex and the adjective to break up what would otherwise be an impossible cluster, or the nal -n of the prex would change, becoming more similar (or wholly similar) to the rst consonant of the adjective. Such assimilation is observed elsewhere in our corpus, so this has to be our favourite theory regarding the behavior of an- as well. Before the consonant p-, the n of an would likely be pronounced with the lips closed because the pronunciation of p involves such a closure; hence n would turn into m. (Compare English input often being pronounced imput.) From pitya small we would thus have ampitya for smallest, this being the impossible word anpitya reworked
101
into a permissible form (Quenya does not have np, but the cluster mp is frequent even in unitary words). Before the consonants l-, r-, s-, and m-, the nal n of an- would probably be fully assimilated, that is, it becomes identical to the following consonant: an an an an + + + + lauca warm = allauca warmest ringa cold = arringa coldest sarda hard = assarda hardest moina dear = ammoina dearest
Cf. such attested assimilations as nl becoming ll in the compound N melu lt Flower of the West (UT:227, transparently a compound of the wello e known words n men west and lt ower). As for the group nm beu o e coming mm, this development is seen in the name of the Vanyarin Elf Elemm e mentioned in the Silmarillion: his (her?) name apparently means r Star-jewel (elen star + m e jewel). r
102
Vocabulary
canta four Nauco Dwarf parma book tiuca thick, fat mapa- verb grasp, seize tir- verb watch, guard lala- verb laugh (so according to a late source, PM:359; in earlier material the verb
lala-, of a quite dierent derivation, has the meaning deny: See the entry la in the Etymologies. We neednt discuss whether one obsoletes the other; here we will use lala- for laugh only.)
caita- verb lie (lie horizontally, not tell a lie) tulta- verb summon
e linda- verb sing (cf. the word Ainulindal or Music [lit. Singing] of the Ainur)
Exercises
1. Translate into English: A. I n lla. s a e B. I antiuca Nauco mta. a C. I tri t i aran. a ra D. I analta oron n taura. a E. I nr tulta i anvanya vend. e e e F. I aiw linda. e e G. I Naucor mpar i canta Eldar. a e H. I antaura aran n saila. a 2. Translate into Quenya: I. The woman is watching the greatest (/biggest) ship. J. The most evil (/evilest) men are dead. K. The Elf is seizing the book. L. Four men are lying under a tree. M. The wisest Elf is reading a book (careful : what probably happens to
the superlative prex when it is added to a word like saila wise?)
103
N. The king and the queen are reading the book. O. The birds are singing. P. The four Dwarves are watching a bird.
104
Lesson 6
Past tense
6.1
The previous lesson discussed the Quenya present tense, which is typically used to describe an on-going present action. However, Quenya has dierent tenses covering the entire trinity of past, present and future, and when recounting past events one will normally use the past tense. In English, very many past tenses are formed by means of the ending -ed, e.g. lled as the past tense of the verb ll. In Quenya, most past tense forms are likewise formed by means of an ending added to the verbal stem. As far as we know, all past tense verbs end in the vowel - (though further endings, e such as the plural ending -r that is used in the case of a plural subject, may of course be added after this vowel). In many cases, this vowel - is part e of the ending -n, that seems to be the most general past tense ending in e Quenya. As discussed in the previous lesson, most Quenya verbs are A-stems, meaning that they end in the vowel -a. The past tenses of these verbs are typically formed simply by adding the ending -n. For instance, the Etymologies mene tion a verb orta- raise (see the entry oro), and in Namri in LotR its a e past tense is seen to be ortan. (The simplest translation of ortan is of e e course raised; the somewhat free rendering in LotR employs the translation has uplifted instead, but Tolkiens interlinear translation in RGEO:67 reads lifted up which is merely an alternative wording of raised.) Other examples from Tolkiens notes: ora- urge, past tense oran urged (VT41:13, 18) e hehta- exclude, past tense hehtan excluded (WJ:365) e ulya- pour, past tense ulyan poured (Etym, entry ulu) e sinta- fade, past tense sintan faded (Etym, entry thin) e We may add the verb ahyan changed (or did change), only attested e like this in the past tense, as part of the question manen lamb Quendion e ahyan[?] how did the language of the Elves change? (PM:395). The verb e
105
change would seem to be ahya-. Regarding the verb ava- (apparently meaning refuse, forbid), Tolkien noted that its past tense avan revealed that it was not in origin a strong e or basic verbal stem. The latter seems to be more or less the same as a primary verb. He called avan a weak past tense form (WJ:370). That e probably goes for all the past tenses so far discussed. (What Tolkien would call a strong past tense is not quite clear. Perhaps he would use this term of the past tenses formed by means of nasal-inxion see below.) We must also consider the basic or primary, ending-less verbs, verbs that unlike the A-stems do not have a nal vowel: verbs like sil- to shine, tir- to watch, mat- to eat. It seems that the ending -n can be used to form the past tense of some e primary verbs as well. Tolkien mentioned tirn as the past tense of the verb e tir- to watch (Etym, entry tir), and he also quoted tamn as the past e tense of the verb tam- to tap (Etym, entry tam). In these cases, adding -n to the verbal stems in question does not produce impossible consonant e clusters: Both rn and mn are permitted by Quenya phonology. For this reason, the ending -n can probably also be added to verbal stems ending in e -n, since double nn is likewise a wholly acceptable combination in Quenya. For instance, the past tense of the verb cen- to see is presumably cenn e saw, though we have no attested example of the past tense of a verb of this shape. But whenever the stem of a basic verb ends in any consonant other than just -m, -n, or -r, simply adding the ending -n would produce consonant e clusters that Quenya cannot have. The past tense forms of verbs like mateat, top- cover or tac- fasten cannot be **matn, **topn, **tacn, e e e for clusters like tn, pn, cn are not found in the language. So what happens? The dicult way of describing what occurs is to say that the n of the ending -n is replaced by nasal-inxion intruding before the last consonant e of the verbal stem. What is inxion? We have already mentioned suxes, elements added at the end of a word (like the plural ending -r, added to the noun Elda in its plural form Eldar), and prexes, elements added at the beginning of a word (like the superlative prex an-, added to the adjective calima bright in its superlative form ancalima brightest). If you want to add something to a word, there are only so many places you can t it in; if it is not to be prexed or suxed, the nal option is to inx it, that is, jam it into the word. For instance, the verb mat- to eat has the past tense mant ate (VT39:7), an inxed n turning up before the nal consonant of e the verbal stem (t becoming nt). Similarly, the verb hat- break asunder has the past tense hant (Etym, entry skat). e Before the consonant p, the inx takes the form m rather than n, so that the past tense of top- to cover is tomp (Etym, entry top). Before c, the e inx appears as n (or actually n, see below), so that the past tense of tac to fasten is tanc (Etym, entry tak). The various forms of the inx n, e 106
m or n, depending on the environment are all nasals, sounds pronounced by making the stream of air from the lungs go out through the nose rather than the mouth. Hence nasal-inxion is a tting term for this phonological process. As I said, that was the dicult way of stating what happens. Put more simply: if adding the past tense ending -n to a primary verb would result e in any of the impossible clusters tn, cn, pn, the n and the consonant before it switch places. Tn and cn simply become nt and nc; what would be np changes to mp to ease pronunciation. (Actually what would be nc similarly changes to nc, using n for ng as in king as Tolkien sometimes did but ac cording to the spelling conventions here employed, nc is represented simply as nc.) Hence: mat- eat, past tense (**matn >) mant ate e e top- cover, past tense (**topn > **tonp >) tomp covered e e e tac- fasten, past tense (**tacn >) tanc fastened e e This, at least, is an easy way to imagine it for pedagogical purposes. We cannot know for certain whether Tolkien imagined this to be the actual development a form something like matn actually occurring at an e earlier stage, but later becoming mant by swapping around the consoe nants t and n. The linguistic term for such transposing of two sounds is metathesis, and there are other examples of metathesized consonants in the imaginary evolution of Tolkiens languages (see for instance the Etymologies, entry kel-). However, some clues suggest that Tolkien imagined these past tenses to reect genuine nasal-inxion occurring already in primitive Elvish, not merely a later transposition of consonants. After all, he had one of his characters observe that nasal-inxion is of considerable importance in Avallonian (SD:433; Avallonian is another term for Quenya). But this is an academic question. Primary verbs with -l as their nal consonant must be given special attention. The verb vil- to y is said to have the past tense vill (Etym, e entry wil). This double ll probably represents some combination of l and n. Perhaps vill is meant to represent older wiln with the normal past tense e e ending (notice that in this case, v comes from older w: root wil), the group ln turning into ll in Quenya. However, other examples suggest that older ln would rather produce Quenya ld. It may well be that vill is meant to e represent older winl, that is, a nasal-inxed variant of the verb wil- (since e nl also became ll in Quenya; for instance, the noun nell brook is said to e come from older nen-le: Etym, entry nen). Whatever development Tolkien may have imagined, primary verbs with l as their nal consonant seem to form their past tense form by adding -l. e
NOTE: In Telerin, the sister language of Quenya in the Blessed Realm, a verb formed from a root del (go) is said to have the past tense delle: WJ:364. As pointed out by Ales Bican, this form probably descends from older denle (with nasal-inxion). If it descended
107
from delne, it would likely have remained unchanged in Telerin, since the cluster ln is permitted in this language (cf. a Telerin word like elni stars, WJ:362). This observation supports the view that past tenses with nasal-inxion did occur already in Primitive Elvish.
The system set out above is what I shall consider the regular way of forming the past tense of a verb in Quenya. That is, as long as a verb conforms with this system, I will not explicitly list its past tense when I rst mention it. All the past tenses in the exercises below are constructed according to this system, so your task this time is to internalize the rules above. Some irregular forms will be discussed in later lessons, but even so, we will here survey certain alternative past tense formations (contrasting them with the more regular forms may actually be helpful in memorizing the normal system but the student is not expected to memorize this survey as such). So do skim through as much as you can take of the stu below, and proceed to the exercises when youve had enough. The past tense of primary verbs with -r as their nal consonant is relatively well-attested: Attested examples include car- make, do, pa.t. carn e (Etym, entry kar), tir- watch, pa.t. tirn (Etym, entry tir) and ture govern, pa.t. turn (Etym, entry tur). So above we set out the rule that e verbs of this shape have past tense forms that are constructed by adding the sux -n. But a few verbs behave quite dierently. The past tense of the e verb rer- sow is not **rern as we might expect, but rend: See Etym, e e entry red. The reason for this is precisely the fact that the original rootword was red rather than **rer. Thus the verb rer- appeared as red- at an earlier stage, and then the past tense rend is actually regular enough: e it is simply formed from red- by means of nasal-inxion + the ending - e (just like such a regular verb as quet- say has the pa.t. quent). What e slightly complicates matters is that in Quenya, original d only survived as part of the clusters ld, nd, and rd; in all other positions it was changed, and following a vowel it normally became r. Hence red- turned into rer-, while the past tense rend remained unscathed by the phonological changes. In e this perspective, the verb is strictly speaking not irregular at all; it just behaves dierently because it has a special history and this goes for very many of the irregularities in Quenya: As observed by his son, Tolkiens linguistic creations imagine language not as pure structure, without before or after, but as growth, in time (LR:342). Tolkien clearly liked leaving in various testimonials to this imaginary age-long growth. We dont know very many verbs in -r that should have past tenses in -nd e because of their special history. From the Etymologies we must presumably include the verbs hyar- cleave and ser- rest (since these come from roots syad and sed, see the relevant entries in Etym but Tolkien did not actually mention the past tense forms hyand and send). In a post-LotR e e source we have a verb nir- press, thrust, force; again no past tense form has been published, but since the stem is given as nid it should presumably 108
be nind rather than nirn (VT41:17). More attested examples could be e e quoted from early Qenya material, but these writings do not have full authority as regards LotR-style Quenya. For instance, the 1915 Qenya Lexicon seems to include the verb nyar- tell, relate in this category (past tense nyand, QL:68). But in later material, Tolkien derived this verb from a root e nar (entry nar2 in Etym) rather than nad, so now its past tense would presumably be regular (nyarn). e Some primary verbs are also seen to use a past tense formation that dispenses with any nasal sounds. The verb does receive the ending -, the vowel e displayed by all past-tense forms, but instead of adding a nasal sound (inxed or as part of the ending -n), the stem-vowel of the verb is lengthened. e For instance, the past tense of the verb lav- lick is seen to be lv (ata e tested in Namri as part of the verb undulv down-licked, that is, a e a e covered). Likewise, the past tense of the negative verb um- not do or not be is said to be um (Etym, entry ugu/umu; we will return to this e peculiar verb in Lesson Nine). This past tense formation is quite common in the early Qenya Lexicon, and it also turns up in relatively late (but still pre-LotR) sources. F riels Song of ca. 1936 agrees with the 1915 Lexicon that the past tense of the verb car- make, do is cr (QL:45, LR:72; the a e spelling used in the sources is kre). However, according to the Etymologies a (entry kar), the past tense is carn and that is the form we will use here: e The Etymologies is, at least in part, slightly younger than Firiels Song. Following the pattern of cr, some pre-LotR sources give t l as the past a e ue tense of the verb tul- come (LR:47, SD:246), but vill as the past tense e of vil- in the Etymologies suggests that the past tense came could just as well be tull (representing older tulne or tunle) instead. e It might seem that Tolkien eventually decided to limit the use of the past tense formation represented by t l and cr, though it was never wholly ue a e abandoned, as the form undulv in Namri in LotR demonstrates. We a e a e might actually have expected the past tense of lav- lick to be **lamb e rather than lv. A past tense form lamb would be constructed by nasala e e inxion of the original root-word lab (itself listed in Etym): In Quenya, original b normally became v following a vowel, but b persisted unchanged in the group mb. The Qenya Lexicon actually lists amb as the past tense of e a verb av- depart (QL:33); this may be an example of this phenomenon. However, **lamb as the past tense of lav- would clash with the noun e lamb tongue, language; perhaps this is why Tolkien decided to go for e the irregular formation lv instead. Or should we generalize from lav- and a e let all Quenya primary verbs in -v form their past tense after the pattern of lv? a e Luckily, these verbs are not very numerous. There is a distinct verb lavmeaning yield, allow, grant (root dab, see Etym), possibly a verb tuvnd (verbal stem isolated from a longer form), plus tyav- as the verb taste (see entry kyab in Etym). Should the past tense tasted be tyamb e 109
or tyv? The latter past tense form is actually attested in the Qenya a e Lexicon (p. 49), but since the QL is seen to use this formation quite liberally compared to later Quenya, we cannot be sure that the information is valid for the later stages of Tolkiens conception. (Tyv is attested in a posta e LotR source as a noun taste; whether this argues against the same form being used as a past tense tasted is unclear. In the 1915 Lexicon, Tolkien did have similar-sounding nouns and verbal tenses coexisting; see QL:49, entry kumu.) There are some curious cases where even longer, derived verbs (A-stems) drop their ending and have lv-style past tenses derived directly from the a e ending-less root. One early example is the verb serta- tie, past tense sr e e (QL:83) rather than **sertan as we might expect. These formations are far e from uncommon in the 1915 Lexicon, but the idea was not wholly obsolete in later Quenya either: The Etymologies of the mid-thirties records that the verb onta- beget, create has two possible past tenses: beside the regular form ontan we also have the irregular form n (Etym, entry ono). e o e The simplest A-stems, those that add the short ending -a to the root (and not a longer ending like -ta or -ya), may also drop this ending in some past tense formations. Above we quoted the QL form tyv as an a e attested past tense of the verb tyav- taste, but in the 1915 Lexicon, the verb taste is actually given as an A-stem tyava-: It is not a primary verb tyav- as it becomes in later sources (QL:49 vs. Etym, entry kyab). Within the later system, we would expect an A-stem tyava- to have the past tense tyavan, but the validity of either form in LotR-style Quenya is e highly questionable. More commonly, the simplest A-stem verbs have past tenses that are regular enough if you pretend that the nal -a does not exist! Above we quoted oran as an example of the regular past tense of e a simple A-stem verb (ora- urge), but immediately after writing oran, e Tolkien actually added orn as a parenthetic alternative (VT41:13). Of e course, orn would be a perfectly regular form if it were the past tense of a e primary verb **or- (cf. for instance tur- govern, pa.t. turn). In eect, e ora- may behave as a primary verb in the past tense, discarding its ending and jumping over into another class. The earliest material has examples of the same phenomenon: In the QL, the past tense forms of the verbs papa- tremble and pata- rap, tap are given as pamp, pant (p. 72), e e not **papan, **patan as we would expect according to the regular e e system. The nasal-inxed past tense forms would be perfectly regular if we assume that in the past tense, the simple A-stem verbs papa- and pataare masquerading as primary verbs **pap-, **pat-. Thus we cannot be certain whether the past tense of the verb mapa- grasp, seize should be mapan or mamp; writers have used both. Since Tolkien seems to imply e e that the past tense of ora- can be both oran and orn, perhaps both are e e permissible.
NOTE: In QL:59, Tolkien actually listed the past tense of mapa- as namp (sic!) In e
110
the 1915 scenario, there were two variant roots, mapa and napa, that shared the past tense namp. Do we dare to assume that this idea was still valid decades later? The verb e mapa- is listed in the Etymologies, but if Tolkien had still imagined its past tense to be as irregular as namp, I tend to think that it would have been explicitly mentioned in e Etym as well. Furthermore, in Etym there is no trace of the alternative root napa; we only nd map (LR:371) corresponding to mapa in the QL. But on the other hand, the form namp is attested, so if you like it better than the unattested forms mapan or e e mamp, feel free to use it. e
The verb lala- laugh is another example of one of the simplest A-stems. It may have the past tense lalan, but it is also possible that it should behave e as a primary verb in the past tense. But if so, we must take into account the fact that lala- is to be derived from older g-lada- (PM:359); this is one of the cases where an original d following a vowel turned into l rather than r (inuenced by the l earlier in the word). So if lala- has a short past tense, it should probably not be lall, but rather land derived from e e a nasal-inxed form of the original word g-lada-. On the other hand, the similar but distinct verb lala- deny found in the Etymologies (LR:367) never contained a d, so its past tense may well be lall (unless it is lalan, e e and I think I lean toward that form). The Etymologies actually provides a few examples of even more complex A-stems that also drop their ending and in eect transform themselves into primary verbs in the past tense. The verb farya- to suce is said to have the past tense farn (Etym, entry phar); here the whole ending -ya drops e out in the past tense, which is formed as if this were a primary verb **fer. Based on such a regular example as the one we quoted above namely ulya- pour, past tense ulyan we would of course expect the past tense e of ferya- to be **feryan. But actually even our regular example ulyae also has an alternative past tense form ull (Etym, entry ulu), and this e is a particularly interesting example, for Tolkien indicated that the two past tenses ulyan and ull were not interchangeable. They had somewhat e e dierent meanings. There will be a fuller discussion of this in Lesson Ten; for now it suces to say that I think most verbs in -ya would retain this ending when the past tense sux -n is added. (But ull as one past tense e e of ulya-, formed directly from ul- rather than the full form of the verb, would seem to conrm that primary verbs in -l normally have past tenses in -l. Except for ull, we only have the example vil- y, pa.t. vill to go e e e on so an extra, if indirect, conrmation of this pattern is very welcome!) Finally we will discuss a strange past tense formation that may occur in the case of verbs in -ta. Perhaps it should not be seen as irregular, for Tolkien actually described one such past tense as regular . . . for a -ta verb of this class (WJ:366). Nonetheless, its formation is less than straightforward. It is exemplied already in the earliest material: The 1915 Lexicon contains a verb lahta- (QL:50; the verb is not clearly glossed), but its past tense is not **lahtan as we might expect: Instead we nd lahant. In other words, the e e 111
verb lahta- is reworked into lahat- (the stem-vowel being repeated between the second and the third consonant, breaking up the consonant cluster, whereas the nal -a is dropped), and the past tense lahant is then formed e from this lahat- by means of nasal-inxion and an added -, in itself a quite e regular process familiar from primary verbs. A much later example can be found in the Etymologies, where the verb orta- rise, raise is assigned a past tense form oront (Etym, entry oro), e though oront is not there clearly presented as a Quenya form: In Etym, it e is actually quite unclear what language it is meant to belong to. However, in some of Tolkiens earlier drafts for Namri, the past tense of orta- did a e appear as oront, not regular ortan as it became in the nal version. e e So what is going on here? Our only real clue is what Tolkien wrote in WJ:366, where he somewhat surprisingly declared the form oant the past tense of auta- go away, e leave to be quite regular for a -ta verb of this class. According to the regular system we have tried to make out, oant instead of **autan ine e evitably seems highly irregular. Tolkien derived the verb auta- from a root awa (WJ:365), so its form in the primitive language is probably meant to be awat (my reconstruction). As primitive Elvish evolved towards Quenya as a we know it, the second of two identical short vowels in concomitant syllables was often lost; hence awat would have been shortened to awt = aut, and a a a this in turn is the direct ancestor of Quenya auta-. But it seems that the old past tense of such a verb as awat, with a vowel immediately preceding a the ending -t, was formed by nasal-inxion: Tolkien explicitly gave the past a tense of the primitive verb as awant (WJ:366; the spelling there used is ace tually awa-n-t, the hyphens before and after the n apparently emphasizing e that it is an inx whereas the accent on the initial a here only means that it is stressed, not that the vowel is long). In the case of a word like awant, the rule that the second of two identical e short vowels is lost could not apply (no **awnt ), for such loss does not e occur immediately in front of a consonant cluster and the nasal-inxion has here produced a cluster nt. The nal Quenya form of awant, namely e oant, is somewhat obscured because the group awa later became oa in e Quenya but this change has nothing to do with the past tense formation. Now we can explain a form like oront as the past tense of orta-: e In the Etymologies, the original root is given as oro (LR:379), so Tolkien probably meant the verb orta- to be descended from older orot - after the a regular loss of the second vowel. But the past tense of this orot - was the a nasal-inxed form oront (both are my reconstructions), and this produced e Quenya oront, the second vowel here being preserved because of the fole lowing cluster nt (no one wants to say **ornt!) e When Tolkien apparently changed his mind and altered the past tense of orta- from oront to ortan (a regular form according to the syse e tem we have set out), this would seem to suggest that he had now decided 112
that the primitive forms were instead ort - with past tense orta-n : There a e was never any vowel immediately in front of the ending -t after all, and a therefore the past tense was not formed by nasal-inxion, but by the independent ending -n (> Quenya -n). This is not the only example of Tolkien e e apparently changing his mind about which verbs actually belong to this exclusive class. The Etymologies lists a verb atalta- collapse, fall in (entry talat); no past tense is there mentioned, but in one text we have atalant e (LR:56, translated down-fell). This would seem to presuppose that the primitive forms were atalat - with past tense atalant (my reconstructions, a e but cf. WJ:319 regarding atalat as a derivative form of the root talat). Yet in Tolkiens later texts the past tense of atalta- becomes ataltan e (LR:47 and SD:247), simply formed by adding the normal ending -n. So e now Tolkien had presumably come to envision the primitive forms as atalt -, a past tense atalta-n (my reconstructions). e If the apparent revisions oront > ortan and atalant > ataltan do e e e e not reect changes in his ideas about the primitive Elvish forms, it may be that he imagined a development whereby the Eldar replaced the more complex past tense formations with simpler, analogical forms. For instance, oront as past tense of orta- could have been replaced by ortan because e e of analogy with such straightforward past tense formations as hehta-, pa.t. hehtan (WJ:365). In the Etymologies, the form oront is indeed marked e e with a symbol that indicates that it is poetic or archaic (cf. LR:347); is this to suggest that it was ordinarily replaced by the non-archaic form ortan? e Especially considering how Tolkien later came to envision the history of the Quenya tongue that it was used as a ceremonial language in Middle-earth, but was no longer anyones mother-tongue we could very plausibly assume that its grammar was somewhat simplied, more complex formations being suppressed and replaced by simpler analogical ones. Indeed oant rather e than **autan as the past tense of auta- to leave is the only verb I can e think of where we must use this special past tense formation, unless we are to accept some of the earliest Qenya material with no reservations (and I have plenty). With this we conclude our survey of various strange or irregular ways of forming the past tense; as I said above, the exercises below are meant to reect the regular system instead. Remember that just like present-tense verbs, a past tense form receives the ending -r if it has a plural subject (or multiple subjects). For instance, the simplest past tense of the verb lanta- fall is lantan, but with a plural e subject it becomes lantaner (SD:246). Naturally, he diaeresis over the nal - disappears, since the vowel is no longer nal when the plural ending -r is e added after it.
113
Vocabulary
lemp ve e elen star harma treasure (noun) sil- verb shine (with white or silver light, like star-shine or moon-shine) hir- verb nd cap- verb jump tec- verb write quet- verb speak, say mel- verb love (as friend; no Quenya word referring to erotic love between the sexes
has been published)
cen- verb see (related to cenda- read, which word is derived from a strengthened
form of the same stem and meaning, basically, to watch closely).
orta- verb rise, also used = raise, lift up. harya- verb possess; have (related to the noun harma treasure, basically referring to a possession)
Exercises
1. Translate into English (and practice your vocabulary at the same time; most of the words employed in exercises AH were introduced in earlier lessons): A. I nr cendan i parma. e e B. I Naucor manter. C. I aran tultan i tri. e a 114
D. N lindan. s e E. I vendi tirner i Elda. F. I lemp roccor caitaner nu i alta tasar. e G. I eleni siller. H. I Nauco cenn rocco. e 2. Translate into Quenya: I. A Dwarf found the treasure. J. The Elf spoke. K. The horse jumped. L. The king loved the Elves. M. A man wrote ve books. N. The queen rose. O. The kings possessed great treasures. P. The king and the queen summoned four Elves and ve Dwarves.
115
Lesson 7
Future tense and Aorist
7.1
In this lesson we will introduce two new tenses of the verb, the future and the aorist. We shall have to spend quite a few paragraphs trying to dene the function of the latter, but the function of the future tense is easy enough to grasp: This tense is used with reference to future actions. English (unlike, say, French) has no distinct future tense. Instead of a unitary, inectional form of the verb that only refers to future actions, English may fall back on longer phrases involving extra verbs like shall or will: A past tense form like came has no one-word counterpart with future reference that would exemplify a true future tense we only nd circumlocutions like shall come or will come (or even is going to come). It is even possible to use the present tense with future reference: He comes tomorrow. For this reason, linguists may refer to the English present tense as a non-past tense instead: It actually covers both present and future. These somewhat asymmetric features of English are avoided in Tolkiens Elvish. Languages like Quenya and Sindarin do possess true future tense forms of the verb. For instance, the future tense of the verb hir- nd appears near the end of Namri, in the sentence nai ely hiruva, maybe a e e thou shalt nd [it]. The example hiruva shall (shalt) nd includes what seems to be the normal possibly universal Quenya future tense marker: the ending -uva. This pattern is conrmed by the Markirya poem, that includes the examples cenuva shall heed, tiruva shall watch and hlaruva shall hear (verbs cen- see, behold, heed, tir- watch, hlar- hear). In LR:63, Tolkien translates the verb queluva as faileth, but this is only an example of the English present or non-past tense embracing the future as well. The context clearly indicates that the verbal action in question belongs to the future: Man tr antva nin Il vatar, Il vatar, enyr tar i a e a u u a e tyel e Anarinya queluva? What will Ilvatar, O Ilvatar, give me in r u u that day beyond the end, when my Sun faileth [literally: shall fail]? The examples listed so far exemplify the future tense of primary or 116
ending-less verbs only. It seems that the ending -uva is also used in the case of the more numerous A-stem verbs, which however lose their nal -a before the future tense ending is added (one exception, see note below). In a postLotR source, the future tense of the verb linda- sing appears as linduva (attested with a secondary ending here removed; see Taum Santoskis article in the October 1985 issue of the newsletter Beyond Bree). Also, what must be the future tense of the A-stem verb ora- to urge, impel is apparently given as oruva in another post-LotR source (VT41:13, 18; Tolkien actually wrote oruv, but the editor points out that the dot may be an inadvertently incomplete a: No Quenya word can end in -v.)
NOTE: Notice, however, that a nal -a does not drop out before the ending -uva when this -a is also the only vowel of the verbal stem. Thus, the future tense form of the copulas derived from the stem na to be (cf. n is) is not **nuva, but nauva: This word for a will be is attested in VT42:34.
It may be that Tolkien at one point imagined a somewhat more complicated system regarding the A-stems. Above we quoted a line from the pre-LotR Quenya text usually called F riels Song, including antva as the a future tense of anta- give (LR:63, 72). Here Tolkien seems to be using a system whereby A-stem verbs form their future tense by lengthening the nal -a to - and adding the ending -va (shorter variant of -uva?) However, a in light of the later examples linduva and oruva (instead of **lindva, a **orva), we may conclude that Tolkien eventually decided to make -uva a the more or less universal future tense marker: This ending simply causes the nal -a of A-stems to drop out. My best guess is that in LotR-style Quenya, the future tense of anta- should be antuva rather than antva, a since Tolkien may seem to have simplied the system. However, there is one possible complication in LotR-style Quenya as well, regarding the primary verbs. In Namri in LotR occurs the future tense a e form enquantuva, shall rell. Removing the prex en- re-, we have quantuva for shall ll. This used to be taken as the future tense of a verb quanta- to ll, related to the adjective quanta full. Tolkiens earliest Qenya wordlist indeed lists such a verb (QL:78, there spelt qanta-). However, about half a decade after publishing LotR, Tolkien in the essay Quendi and Eldar seemingly cited the Quenya verb to ll as quat- (WJ:392). This would seem to be a primary verb, past tense presumably quant (the pa.t. e qante is actually given in QL:78, but there it is evidently only meant as a permissible shortening of the full form qantane; the regular past tense of a verb quanta- would be quantan in later Quenya as well). If Tolkien had e decided that the Quenya verb to ll is actually quat-, and its future tense is quantuva as Namri would seem to indicate, should we conclude that a e the same verbs that form their past tense with nasal-inxion + the ending - similarly form their future tense with nasal-inxion + the ending -uva? e For instance, should the future tense of verbs like mat- eat, top- cover and tac- fasten be mantuva shall eat, tompuva shall cover, tan117
cuva shall fasten? (Compare the nasal-inxion in the past tense forms: mant, tomp, tanc.) Or should we just add the ending -uva to the e e e verbal stem without any further manipulations, hence matuva, topuva, tacuva instead? General principles would perhaps suggest the latter, but there remains the curious example of quantuva next to quat-. If there is to be no nasal-inxion in the future tense forms, we would have to accept that the verb ll can be both quanta- and quat-, with separate future tenses quantuva and quatuva. I have used future tense forms with nasal-inxion in certain compositions of my own (and so have some people who put greater trust in my so-called expert opinion than they possibly should). But it may well be that Tolkien, mentioning the form quat- in WJ:392, actually intended this to be simply the way the underlying root kwata manifests in Quenya. The exact wording in the source involves a reference to the verb stem *kwata, Q quat- ll. If quat- is merely the way the ancient stem kwata appears in Quenya, the actual verb ll could still be quanta- with future tense quantuva. (Compare for instance the entry pat in the Etymologies, this root pat producing the Quenya verb panta- open. There is also an adjective panta open, exactly parallelling quanta full next to the verb quanta- ll; perhaps the verb is derived from the adjective in both cases.) Alternatively, quat- really is the verb to ll and not just an underlying root-form, but the future tense quantuva still presupposes a longer Astem quanta-. Perhaps Tolkien had just plain forgotten that he had already published a form of the A-stem verb quanta- ll, so that he was no longer free to change it to a primary verb quat-. (See PM:367-371 for an example of Tolkien working out some elaborate linguistic explanations that he had to scrap because he discovered that they conicted with something he had already published in LotR a fatal footnote in an Appendix forcing him to reject his nice new ideas!) Thus, material presently available does not allow any certain conclusion in this matter. Writers can equally plausibly let verbs that show nasal inxion in the past tense do so also in the future tense (arguing from the quat/quantuva pair that this is how the language works) or choose to explain quat- dierently and form the future tense of any primary verb simply by adding the ending -uva (as in hir-/hiruva). As users of Quenya we can probably well aord to live with slightly dierent dialects regarding this detail, until future publications hopefully allow us to pick the right explanation. It must be assumed that the future tense, like all other tenses, receives the ending -r where it occurs with a plural subject (e.g. elen siluva a star will shine, but plural eleni siluvar stars will shine).
118
7.2
The Aorist
We have now discussed all the three tenses corresponding to the basic trinity of past, present, and future. Yet the Quenya verb has other tenses as well. One is called the aorist. The use of this term with reference to Quenya grammar was long disputed by some, but a Tolkien text that nally became available in July 2000 demonstrates that he had indeed invented a Quenya tense he called aorist (VT41:17). While even people with no linguistic training readily understand what the past, present and future tenses are for, it is hardly equally obvious what function the aorist tense has. (Some linguists would say that the aorist is strictly not a tense at all, according to certain denitions of that term; however, Tolkien did use the phrase aorist tense in VT41:17. We will not discuss this question here, wholly academic as it is.) So what, really, is an aorist? To start with the word itself, it comes from Greek and literally means something like unlimited or undetermined. The word was originally coined to describe a certain Greek form of the verb. In Greek this form contrasts with the past tense or imperfect, the latter being used of a past action that was being done over a period of time (not just a momentary action). The aorist, on the other hand, has no such implications regarding the duration of the action. It just denotes a past action, period, with no further distinctions. When contrasted with the imperfect, the Greek aorist can be used for a momentary or clearly nished (not on-going) action. Another use of the Greek aorist is not especially associated with the past: the aorist could be used to express general truths that are not limited to any specic time, like sheep eat grass. But this was the Greek aorist; the Quenya aorist is not used in quite the same way. Yet their functions do overlap in some respects, which must be the reason why Tolkien decided to employ this term from Greek grammar in the rst place. We will try to determine the function of the Quenya aorist before we discuss how it is actually formed. For now, just take my word that the verbs in the examples I cite are aorists. The Quenya aorist, like the Greek one, can be used to express general truths. Our best example is a sentence occurring in WJ:391, where Elves are described as i carir quettar mainen, those who make words o with voices. The aorist verb carir make here denotes a general habit of the Elves, covering past, present and future, for the Elves were making words throughout their history. The sentence polin quet I can speak e (VT41:6) includes another aorist verb, and again a general truth is presented, though in this case it relates only to the speaker: The meaning is of course I can (always) speak, presenting a general ability, not just something that applies only to the present time (as if the speaker was dumb yesterday and may go dumb again tomorrow). So one imporant function of 119
the Quenya aorist is that it is used, or rather can be used, of verbal actions that transcend the here and now rather describing some timeless truth or general situation. In Namri in LotR, Galadriel describes the gloomy a e state of Middle-earth using an aorist verb: sindanriello caita morni o e out of a grey country darkness lies (not present tense caita = is lying, e as if this were merely a strictly present phenomenon, soon to pass). The rst words of Namri also include an aorist: lauri lantar lassi, like gold fall a e e the leaves but this is not just a here-and-now description of leaves that are falling (which would presumably be lantar, present tense): The following e lines indicate that Galadriel describes the general situation in Middle-earth, the ever-recurring autumnal decay as she has been observing it throughout yni untim, long years uncountable. So our example sheep eat grass e o e is probably best rendered into Quenya using an aorist verb: mmar matir a salqu (singular sheep = mma, grass = salqu). As the example e a e polin quet I can speak demonstrates, the aorist can also be used to e describe the abilities or habits of a single individual (i mma mat salqu a e e = the sheep eats grass). It seems, however, that the Quenya aorist is not only used to describe timeless truths. In some cases Tolkien himself seems to waver in the choice between the aorist and the present tense, the latter more typically describing an ongoing here-and-now situation. This hesitation on Tolkiens part suggest that these tenses are to some extent interchangeable. We have an aorist in the sentence renya quet nin my heart tells me (VT41:11), which is o e apparently quite synonymous with the alternative wording renya quta o e nin (VT41:13) employing a present tense form instead of an aorist. In the famous greeting elen s l menn omentielvo, a star shines [or rather la u is shining] on the hour of our meeting, Tolkien nally decided to use a present tense form but in earlier drafts, he used an aorist sil instead e (RS:324). This greeting, having relevance for our meeting only, obviously cannot describe any general truth transcending time. Yet it is apparently permissible to use an aorist form even in such a context (though Tolkien decided that it was better to use the present tense). It should be noted that the Quenya aorist is generally associated with the present, not with the past as in Greek. As Jerry Caveney wrote about Tolkien on the Eling list (August 3, 2000): In what seems to me typical of his creativeness and fun in creating languages, he took the idea of the aorist aspect, and said, in eect, What if a language used the aorist to contrast present general (unlimited) actions to present continuative actions instead of using it to contrast past general actions to present continuative [as in classical Greek]? The result is Tolkiens present aorist. :) He thus created a language that could distinguish continuative from general present actions simply, something classi120
cal Greek could not readily do, and which modern English and French, for example, can only do with extra words (I walk, I am walking; je marche, je suis en train de marcher ). I suspect Tolkien enjoyed the elegance of this basic grammatical distinction, which I am not aware that any living language has. On the other hand, Carl F. Hostetter thinks the Quenya aorist is used to describe an action that is punctual, habitual, or otherwise durationless (VT41:15). This is probably correct in most cases, describing the typical function of the aorist. Yet some examples suggest that it may be better to say that whereas the present tense explicitly identies an ongoing action, the Quenya aorist is simply unmarked as far as duration is concerned. It does not necessarily contrast with the continuative present tense; an aorist as such does not signal that a verbal action must be non-continuative or durationless. Rather, as Caveney says, it is a general form, an all-purpose present tense that simply doesnt address the question of whether the action denoted is continuative, habitual or momentary. As Luk Novk obas a served on the Eling list (August 1, 2000): It seems that the aorist is so aoristos [Greek: unlimited] that it can express almost everything. In the exclamation auta i lm! the night is passing (Silmarillion ch. o e 20), the form auta would seem to be an aorist (contrasting with the present tense, which is probably auta) yet Tolkien employs the translation is e passing rather than passes. So it would seem that the aorist can also be used for an ongoing action; it just isnt explicitly marked as such, grammatically speaking. If this is correct, it would be dicult to pin down any case where it is palpably wrong to replace the present tense with an aorist. Using the aorist would be simply a rather neutral way of talking about present actions whether such action is actually ongoing, habitual, or merely an expression of general truths. (Hence mmar matir salqu = sheep a e eat grass could also be understood as sheep are eating grass, though for this meaning it is probably better but hardly mandatory to use the present tense: mtar.) In choosing between the aorist and the present tense, a the only hard-and-fast rule one has to go on seems to be that the present tense should not be used with reference to entirely duration-less actions: The Quenya present tense is always used about some kind of continuous action. (Indeed some students would dispense with the term present tense and rather speak of the continuative form.) Beyond this one restriction, it seems that writers can choose quite freely between the aorist and the present tense. Generally, however, it seems that the Quenya aorist corresponds to the English simple present (that shows either the ending -s or no ending at all, depending on the grammatical context). So Tolkien often translated Quenya aorists: e.g. top covers (LR:394), mac hews (VT39:11), tirin e e I watch (LR:394). The Quenya present tense, on the other hand, is often 121
best translated using the English is . . .-ing construction: tpa is covo ering, mca is hewing, t a ran I am watching. (The ending -n in the examples tirin/t ran, as well as in the form polin I can cited above, signies I: This sux will be discussed in the next lesson.) In Lesson Five we pointed out that the present tense form quta denotes is saying rather e than just says; conversely, the aorist quet is usually says rather than e is saying. If the Quenya aorist is used somewhat like the English simple present tense, the aorist can be used to describe actions that are perceived as duration-less or habitual. For instance, an aorist like cap jumps may dee scribe an action that is momentary (he jumps) or habitual/characteristic (any frog jumps). Yet we also seem to have examples of Tolkien using the Quenya present/ continuous tense instead of the aorist where English would still translate the verb in question as a simple present tense form, not as an is . . .-ing construction. Consider this line from Namri : h e untpa Calaciryo a e si u m mist covers the jewels of Calacirya. The present tense form unt pa ri u describes a continuous action, more literally is covering, but here Tolkien wrote covers instead. Presumably it would in no way have been wrong to use an aorist instead. After all, the mist covering the jewels of Calacirya is evidently perceived as a rather general state of things, not merely as an ongoing meteorological phenomenon that will soon pass! (The aorist would presumably be untup perhaps this form, stressed on the rst rather e than the penultimate syllable, just didnt t the meter of Tolkiens poem? Anyhow, the latter element of this verb untup- seems to be a variant of top- in the Etymologies, both verbs meaning cover.) Another example of a present tense where we might expect to see an aorist can be found in Cirions Oath (UT:305, 317), in the sentence i hrar a mahalmassen mi N men = those who sit on thrones in the West. This u refers to the Valar, and their being enthroned in the West must be considered a general truth, just like it is a general truth that Elves make (aorist carir) words with voices. Yet Tolkien used what seems to be a present tense instead of an aorist: hra, here plural hrar, apparently suggesting a primary verb a a har- to sit. The plural aorist would probably be harir instead. It may be noted that while Tolkien translated hrar as sit in the running English a translation in UT:305, he employed the more literal translation are sitting in his linguistic discussion in UT:317. Yet this seems to demonstrate that in Quenya, one can use the present tense as well as the aorist to describe also a general state of things. After all, the Valars agelong state of being enthroned is also after a fashion continuous. Cf. also the sentence yonya iny tye-mla, I too, my son, I love thee (LR:61), where Tolkien uses a e e present tense instead of an aorist: Literally iny tye-mla would seem to e e mean I am loving you, but the reference must be to a quite permanent emotional state. If anyone else that Tolkien had written this, I would strongly advice the writer to use an aorist (mel) instead of mla actually I still e e 122
think the aorist would be better in this context, even though it was Tolkien who wrote this! But this example conrms that the present tense can also be used to describe general truths or more or less permanent situations, though this is more typically the domain of the aorist. I can well imagine that after this discussion, the student wonders if there is any point in maintaining the aorist and the present as distinct tenses, since their functions seem to overlap to such an extent the only concrete rule being that if some kind of present action cannot in any way be seen as continuous, but is entirely duration-less, one must use the aorist. In just about all other contexts, either tense will apparently do, and the use of the aorist may not necessarily imply that an action has to be duration-less: For instance, it could also describe a general truth, or indeed an ongoing action (as in auta = is passing). The context must be taken into consideration. I can only say that I didnt make this language (another guy did. . . ) Perhaps future publications will throw more light on whatever subtle distinctions Tolkien had in mind. But in the exercises I made for this course, I have consistently used aorists for the English simple present, whereas I use the Quenya present tense for the English is . . .-ing construction. I do think writers transposing English usage to Quenya using this formula would get it right (or rather, wouldnt make palpable mistakes!) most of the time. That was the function of the aorist, dicult though it is to pin down. Now we must discuss how the Quenya aorist is actually formed. It seems that in Primitive Elvish, the rules for how the aorist is constructed were quite simple: In the case of a derived or A-stem verb, the aorist tense is simply identical to the verbal stem itself (irrespective of the fact that the aorist can of course receive such secondary endings as the plural marker -r, where such is required). No explicit tense-marker had to be present. Regarding the A-stems, this system persists in Quenya. The aorist of a verb like lanta- to fall is simply lanta falls (occurring in Namri, there a e with the plural ending -r to agree with its plural subject leaves: lauri e lantar lassi, golden fall [the] leaves). In the case of the primary or ending-less verbs like mat- to eat, they originally (in Primitive Elvish) formed their aorist tense by adding the ending -i : Eats apparently used to be mati. It is somewhat arguable whether the ending -i is here strictly an aorist tense-marker. If so, we might have expected to see it in the formation of A-stem aorists as well. Perhaps the rule for aorist formation in Primitive Elvish should rather be stated like this: The aorist is normally identical to the verbal stem, but in the case of primary or ending-less verbal stems, they receive the ending -i as a kind of stopgap to make up for the absence of any other ending. (I should add that this simplied view is not wholly unproblematic, but it works most of the time.) This system essentially persists in Quenya, but the phonological development occurring since Primitive Elvish has added one minor complication: Where nal, the short -i of Primitive Elvish was 123
at some point changed to -. (For instance, the Quenya noun rinc quick e e shake is said to descend from primitive rinki : See the entry rik(h) in the Etymologies. Where Quenya has nal -i, it is normally shortened from long - in the primitive language.) Hence the old form mati eats had turned into mat in Quenya. But since this change only occurred where -i was nal, e we still see mati- if the aorist form is to receive any ending, such as -r in the case of a plural subject. Hence Nauco mat a Dwarf eats, but with e a plural subject Naucor matir Dwarves eat. The ending shielded the nal -i so that it was not really nal at all, and therefore it did not change to -. e
NOTE 1: There are a few examples of what seems to be aorist forms where the ending - e persists in the form -e- even if the aorist receives an ending. For instance, what must be the plural aorist of the verb ettul- come forth appears as ettuler (instead of the expected form ettulir) in SD:290. Perhaps Tolkien at one stage imagined that the primitive ending -i had become -e in all positions, even where it was not nal like ettulir being altered to ettuler on analogy with the ending-less form ettul. But this seems to have been just a e passing phase in Tolkiens evolution of Quenya: In our best late source, the essay Quendi and Eldar of about 1960, the plural aorist of car- do, make appears as carir, not **carer (WJ:391). Hence Tolkien had reestablished the system he had also used a quarter of a century earlier, in the Etymologies. The form ettuler is (apparently) translated are at hand in SD:290; a more literal translation would presumably be are coming forth. This would conrm that it may be permissible to use the aorist also for ongoing actions; this tense is simply unmarked regarding the duration of the action, whereas the present or continuous/continuative tense explicitly identies an action as ongoing. In our exercises, we will nonetheless use the aorist in the most typical way (to denote actions that are momentary or habitual/timeless). NOTE 2: In the case of primary verbs, the aorist and the present tense dier not only regarding the ending. In the present tense, the stem-vowel is lengthened (mta is a eating), whereas in the aorist, it stays short (mat eats). Yet there are a very few e strange forms in our corpus that look like aorists by their ending, but still show a long stem-vowel, e. g. tp stops, blocks (Etym, entry tap). We would expect tap with a e e a short vowel (it is tempting to believe that the accent above a is just an ink-smear in Tolkiens manuscript. . . ) It may also be noted that a few derived verbs (A-stems) include an intrinsically long vowel, e. g. c na- bend, s ya- breathe or mta- labour, toil. u u o To use the latter verb as an example, its aorist would presumably be mta, even though o this may look like the present tense of a non-existing primary verb **mot-. (We must assume that the actual present tense of mta would be mta.) o o e
124
Quenya also has a tense termed aorist, which diers from the present tense in that the latter explicitly describes an on-going action. The aorist says nothing about the duration of the action, and while the use of an aorist form does not preclude that the action denoted is drawn-out or on-going, it seems that this tense is more typically used to describe duration-less, punctual, habitual, characteristic or altogether timeless actions. An example of an aorist is quet = speaks, as opposed to present tense quta is e e speaking. It may be that the Quenya aorist corresponds quite well to the English simple present tense (speaks), whereas the Quenya present tense rather corresponds to the English is . . .-ing construction (is speaking). In the case of A-stem verbs, the aorist is identical to the verbal stem itself (irrespective of any secondary endings the aorist verb may receive). In the case of primary verbs, the aorist tense is formed by means of the ending -i, which however changes to - if no secondary ending (e.g. -r for plural) e is to follow. Hence the aorist of mat- to eat is mat eats if there is e no further ending added to the word, but otherwise we see mati- + ending (e.g. matir eat in the case of a plural subject).
Vocabulary
enqu six e ilya, noun/adjective all, every (every before a singular noun, e. g. ilya Elda
every Elf, but ilya occurring by itself would rather mean all). Note that before a plural noun, this word also signies all and is inected for plural as a common adjective, hence becoming ily for older ilyai (cf. ily tier all paths in Namri e e a e and ily mahalmar all thrones in Cirions Oath) e
Atan Man (not sentient male, which is nr, but Mortal Man as opposed to Immore
tal Elf, or Dwarf. Within Tolkiens mythos, this word came to be used especially of the Elf-friends of Beleriand and their descendants, the ones called Edain or Dnedain in Sindarin. But even within the mythos, the word was originally used u simply of humans as opposed to Elves, and so do we use it here. Cf. Ilvatars words u in the Silmarillion, chapter 1: Behold the Earth, which shall be a mansion for the Quendi and the Atani [Elves and Men]!)
ohtar warrior r (rv-) lion a a Ambar the world (the Quenya word probably does not require the article i; it is
capitalized and apparently treated as a proper name)
r-, verb die, expire (cf. the adjective rin dead) tur-, verb govern, control, wield or, preposition over, above
Exercises
1. Translate into English: A. Rimb Naucor haryar harmar. e B. Anar ortuva ar i aiwi linduvar. C. Enqu neri tiruvar i ando. e D. Ilya Atan ruva. E. Ily Atani rir. e F. Saila nr cenda rimb parmar. e e G. Ilya elen sil or Ambar. e H. I Elda mapa i Nauco. 2. Translate into Quenya: I. Every Elf and every Man. J. The Elf will nd the Dwarf. K. The horse jumps over the Dwarf. L. The king controls many warriors and will control (/rule) all the world. M. The king and the queen will read the book. N. The warrior wields a sword. O. All lions eat esh. P. Six lions are eating esh.
126
Lesson 8
Perfect tense. Pronominal endings -n(y), -l(y), -s. e e
8.1
Tolkien certainly imagined the Quenya verb to have more tenses than the ones that appear in published material, but only one of these known tenses now remains to be discussed. The last known Quenya tense is the perfect. (There are still other forms of the verb that we shall have to discuss later, such as the innitive, the gerund and the imperative, but these dont count as tenses.) Linguistically speaking, English has no perfect tense, just as English has no future tense. However, just as the language quite regularly expresses the idea of futurity by involving extra verbs like shall or will, so the meaning of a true perfect tense is typically achieved by means of a circumlocution involving the verb have. For instance, some typical English constructions doing the job of a perfect tense are seen in these sentences: Peter has left, the guests have eaten (as opposed to a mere past tense: Peter left, the guests ate). The perfect tense thus describes an action that itself is past, but by using the perfect tense one emphasizes that this past action is somehow still directly relevant for the present moment: Peter has left [and he is still gone], the guests have eaten [and they are hopefully still satiated as we speak], etc. In English at least, such constructions may also be used to describe an action that started in the past and still goes on in the present moment: The king has ruled (or, has been ruling) for many years. Quenya, unlike English, does have a true perfect tense a unitary form of the verb that expresses this meaning, without circumlocutions and extra verbs. Several examples of this perfect tense occurs in LotR. Two of them are found in the chapter The Steward and the King in Volume 3. The rst example is from Elendils Declaration, repeated by Aragorn during his coronation. It goes, in part: Et Erello Endorenna utlien = Out of the a u Great Sea to Middle-earth I am come [or: I have come]. Removing the ending -n meaning I, we nd that the naked present tense have/has come 127
is ut li (according to the spelling conventions here employed, we must add u e a diaeresis to -e when it becomes nal). Later in the same chapter, Aragorn nds the sapling of the White Tree, and exclaims: Y! ut vienyes! I have e u found it! (The word y is not translated; it is apparently simply an exclamae tion Yes! or Yeah!) Ut vienyes can be broken down as ut vie-nye-s u u have found-I-it. We are thus left with ut vi as the perfect tense of a verb u e tuv- nd. (This verb is not otherwise attested, unless it can be equated with a verb tuvu- receive found in very early [1917] material; see GL:71. Whether this tuv- somehow diers in meaning from hir-, we cannot know. In the exercises of this course, I always use hir- for nd.) A post-LotR example of a Quenya perfect tense is found in VT39:9, Tolkien mentioning a form ir e has twisted evidently the perfect tense ci of a primary verb ric- twist (not otherwise attested, but the Etymologies lists a primitive root rik(h)- jerk, sudden move). As stated above, the form ut vi has found seems to presuppose a verb tuv- nd, and u e ut li has come is the perfect tense of a verb tul- come that is atu e tested in the Etymologies (entry tul-). From these examples it is clear that the perfect tense is formed with the ending -i, but the stem of the verb is e also manipulated in other ways. In the case of primary verbs at least, the u e ci stem-vowel is lengthened : ut vi, ut li, ir e. u e The ardent student will remember that a similar lengthening occurs in the present tense (we would have t va is nding, t la is coming, r u u ca is twisting), but the perfect tense formation diers from the present tense not only in the fact that the former receives the ending -i instead of -a. e The perfect, alone of all known Quenya tenses, also receives a kind of prex. This prex is variable in form, for it is always the same as the stem-vowel (but short). Hence the verbs tuv- nd and tul- come become ut vi u e and ut li in the perfect (I underline the prex), since their stem-vowel is u e u. On the other hand, the verb ric- twist, with the stem-vowel i, turns into ir e in the perfect tense. Further examples (constructed by me, with ci underlining of stem-vowel and prex): Stem-vowel Stem-vowel Stem-vowel Stem-vowel Stem-vowel A: mat- eat vs. amti has eaten a e E : cen- see vs. ecni has seen e e I : tir- watch vs. it e has watched ri O: not- reckon vs. onti has reckoned o e U : tur- govern vs. ut ri has governed u e
The prex seen in the perfect tense is usually referred to as the augment. It may also be noted that the process of copying or repeating a part of a word, like the prexing of stem-vowels seen here, is by a linguistic term called reduplication. So to use as many fancy words as possible, one feature of the Quenya perfect tense is that it includes a reduplicated stem-vowel that is prexed as an augment. So far we have only used examples involving primary verbs. The evidence 128
is actually extremely scarce regarding derived (A-stem) verbs. General principles suggest that they would drop the nal -a before the ending -i is e added. For instance, the perfect tense of lala- laugh or mapa- seize is presumably alli has laughed, ampi has seized. (Where such a verb a e a e has a long stem-vowel, it presumably just stays long in the perfect, where it would have been lengthened anyway. The augment should probably always be a short vowel, though; hence a verb like mta- toil may have the o perfect tense omti has toiled.) o e However, very many A-stems have a consonant cluster following the stemvowel, e.g. rn following the rst A in a verb like harna- wound. Since Quenya isnt fond of long vowels immediately in front of consonant clusters, we must assume that the lengthening of the stem-vowels simply does not occur in verbs of this shape. Otherwise the perfect tense would be formed according to the normal rules: reduplicate the stem-vowel as an augment and replace nal -a with the ending -i (so has wounded would be aharni, not e e **ahrni). We may have some attested examples of augment-less perfects a e that are seen to skip the lengthening of the stem-vowel where there is a consonant cluster following it (see below). The numerous A-stems that end in -ya may be somewhat special. Take a verb like hanya- understand. According to the rules so far given, the perfect has understood should be **ahanyi (or even **ahnyi with e a e a lengthened vowel, for it is rather unclear whether ny here counts as a consonant cluster or a unitary consonant palatalized n like Spanish n). However, such a form is impossible, for the combination yi does not occur in Quenya. We may have one example to guide us: In Namri, there occurs a perfect a e tense avni has passed (actually it appears in the plural: yni avnier a e e a ve lint yuldar lisse-miruvreva = years have passed like swift draughts e o of the sweet mead notice that the perfect, like other tenses, receives the ending -r when it occurs with a plural subject). In the essay Quendi and Eldar of ca. 1960, Tolkien explained avni (or vni without the augment) a e a e as being the perfect tense of the highly irregular verb auta- (WJ:366). But a quarter of a century earlier, in the Etymologies, he had listed a verb vanyago, depart, disappear (see the entry wan). It is eminently possibly that when he actually wrote Namri in the forties, he still thought of (a)vni a e a e as the perfect tense of this verb vanya-, though he would later come up with another explanation (perhaps he wanted to eliminate the clash with the adjective vanya fair, though the words would not be dicult to distinguish in context?) If so, Tolkien gave away how to treat verbs in -ya: In the perfect tense, the whole ending -ya is dropped before -i is added, and e what remains of the verb is treated just as if it were a primary verb. The perfect tense would therefore show both augment and lengthening of the stem-vowel, something like this:
129
hanya- understand, perfect ahni has understood a e hilya- follow, perfect ih e has followed li telya- nish, perfect etli has nished e e tulya- lead, perfect ut li has led u e Of course, from the perfect forms you cannot determine with certainty what the original verb stem looks like. For instance, ih e could also be li the perfect of a primary verb **hil- or a short A-stem **hila-. In this case, no such verb is known to exist, but ut li would be the perfect not only u e of tulya- lead, but also of the distinct primary verb tul- come. So one must apparently depend on the context to nd out whether the perfect ut li is formed from tulya- (so that it means has led) or from tul- (so u e that it means has come). Same with the perfect ahri: this form would a e mean has possessed if it is formed from harya, but has sat, has been sitting if it is the perfect of har- (apparently a primary verb sit; only the plural present tense hrar are sitting is attested: UT:305, 317). a Verbs including diphthongs: In some cases it may be somewhat dicult to determine what the stem-vowel is. Where a verb contains a diphthong in -i or -u, it is probably the rst vowel of this diphthong that functions as an augment in the perfect tense. For instance, the perfect tense of verbs like taita- prolong or roita- pursue would probably be ataiti, oroiti, and e e the perfect tense of hauta- cease, take a rest is presumably ahauti. (The e stem-vowel can hardly be lengthened when its part of a diphthong, so we wouldnt expect to see **atiti, **oriti, **ahuti.) The original roots a e o e a e of these verbs are given in the Etymologies as tay, roy, khaw, respectively; thus the proper stem-vowels of these verbs are seen to be a, o, a (again respectively). The nal -i or -u seen in the Quenya diphthongs arise from original consonants -y and -w, so they cannot count as stem-vowels. Unaugmented perfects: The material contains some examples of perfecttense verbs that are constructed according to the rules set forth about, except that they do not have any augment prexed. MR:250 (reproducing a post-LotR source) mentions a form f e has breathed forth or in later ri usage has died; the augment is missing, though there is no reason to assume that the full form if e would be wrong. (The actual translation of ri f e given in MR:350 is she hath breathed forth, but no element meaning ri she can be identied; it is evidently understood.) The verb avnier have a passed occurring in Namri was actually vnier with no augment in the a e a rst edition of LotR; Tolkien supplied the augment in the second edition (1966). Before this, in the essay Quendi and Eldar of about 1960, he explained the unaugmented variant as being simply a variant form appearing in verse (WJ:366). Adding a syllable, as Tolkien did when introducing the full form avnier into the poem in 1966, actually doesnt t the meter of a Namri very well but he evidently decided to let grammatical accuracy a e take priority. 130
In the other perfects occurring in LotR (ut lien, ut vienyes), the augu u ment was present also in the rst edition of 195455. Nonetheless, it seems that the whole idea of augmenting perfect-tense verbs appeared relatively late in Tolkiens evolution of Quenya. In early sources, the augment is missing. For instance, the phrase the Eldar have come appears as iEldar tulier in Tolkiens earliest Qenya (LT1:114, 270). The perfect of tulhere appearing features the same ending -ie- as in LotR-style Quenya, but the augment, as well as the lengthening of the stem-vowel, still have not been introduced into the language. Updating this sentence to LotR-style Quenya by implementing Tolkiens later revisions would probably produce Eldar ut lier (with an augmented perfect and no article before Eldar when it u refers to the entire Elvish race). In much later, but still pre-LotR, material, we nd lanti (with a plural e subject lantier) as a form of the verb lanta- fall: LR:56. These forms would also seem to be unaugmented perfects, showing the ending -i chare acteristic of this tense. True, Tolkien translated these forms as fell (lanti e nu huin fell under shadow, ari lantier seas fell) as if they represent e e some kind of past tense form not perfect has/have fallen. However, he later noted that the forms of past and perfect became progressively more closely associated in Quenya (WJ:366). If this is to mean that Quenya might sometimes use the perfect where English would rather have a past tense, we can explain fell rather than has/have fallen as a possible translation of lanti/lantier. In SD:310, where Christopher Tolkien discusses a later e version of the text in question, he records how his father changed lantier to lantaner apparently substituting a true past tense form for a perfectused-as-past. If lantier, sg. lanti, can indeed be considered a perfect tense form, it e would conrm that the stem-vowel cannot be lengthened before a consonant cluster (not **lnti). Around this stage, Tolkien had certainly introduced a e such lengthening of the stem-vowel in the perfect; F riels Song has crier a for made (or they made, since the plural ending -r is included). This form of the verb car- make, do would seem to be another perfect-usedas-past, judging from the translation. Since the stem-vowel is lengthened in crier, we must assume that it stays short in lantier for purely phonological a reasons: no long vowels are allowed before a consonant cluster. It may be that the absence of the augment in some early sources is simply due to the fact that Tolkien had not invented it yet; in LotR-style Quenya I would recommend alanti as the perfect tense of lanta- and acri as the perfect e a e of car-. Nonetheless, the above-cited example f e has breathed forth, has exri pired from a post-LotR source (MR:250) would seem to indicate that even in LotR-style Quenya, it is permissible to leave out the augment, constructing the perfect simply by means of the ending -i + lengthening of the steme vowel if there is no consonant cluster following it. Possibly unaugmented 131
perfects are meant to be more common in spoken or informal language, and in poetry one can leave out the augment if the extra syllable would spoil the meter (hence vnier for avnier in Namri, though Tolkien changed a a a e his mind in 1966 and introduced the full form). However, in the exercises I made for this course, all perfect-tense forms do include the augment. Verbs beginning in vowels: Verbs beginning in a vowel pose a problem. Where a verb has a prex beginning in a vowel, the augment may slip in between the prex and the most basic verbal stem. For instance, the verb enyal- recall, remember is quite literally en-yal- re-call, where yal- and not en- is the basic verbal stem incorporating the stem-vowel; in such a case I would expect the perfect to be enayli. But some verbs begin in a vowel a e even without any prexed element, e.g. anta- give. In such a case the rst vowel is also the stem-vowel, here occurring without any consonant in front of it. A verb may also include a prex that happens to be identical to the stem-vowel, e.g. onot- count up (formed from not- reckon with a prex o- meaning together, hence onot- is literally reckon together). Other verbal stems already prex the stem-vowel as a kind of intensication, e.g. atalta- collapse, fall in (vs. the verb talta- with a somewhat less harsh meaning: slope, slip, slide down). In all of these cases, it is dicult to prex the stem-vowel as an augment in the perfect tense. We cannot well have aanti for has given, oonti for has counted up, aatalti for e o e e has collapsed. So what do we get instead? One popular assumption has been that in such cases, the entire rst syllable is reduplicated as an augment: Hence the perfect tense of a anta- give would be ananti (antanti?), and so on. With the publication of Vinyar e e Tengwar #41 in July 2000, this theory was almost conrmed. It turns out that in a late source, Tolkien listed orri as the perfect tense of the verb o e ora- urge (VT41:13, 18; actually this form is not explicitly identied as the perfect tense, but it can hardly be anything else). Notice that the entire rst syllable (or-) is reduplicated in the perfect: By reduplicating the consonant following the stem-vowel as well as the stem-vowel itself, the awkward form **ori is avoided; in orri the reduplicated consonant r keeps the o e o e augment and the initial vowel of the verbal stem comfortably apart. Well and good the only problem is that after writing down the form orri, o e Tolkien struck it out! Whether this means that we are back on square one, or whether Tolkien struck out the form orri not because he invalidated o e it but simply because he didnt feel like discussing the perfect tense of orathere and then, none can say. Since it is rather unclear how we should add the augment to most verbs beginning in a vowel, I have simply avoided the perfect tense of such verbs in the exercises I made for this course. But since augmentless perfects seem to be permissible, the easiest solution must be to simply omit the augment in the case of such verbs: anta- give becoming anti has given, onote count up becoming onti has counted up (though this is also the o e 132
perfect of not- reckon!), and so on. After the rejected form orri, Tolkien o e actually wrote ori. Was this a replacement perfect tense, with no augment? e I would expect ri with a lengthened stem-vowel; ori looks more like a o e e quite dierent form of the verb (a gerund, to be discussed in later lessons). This word may be worth noticing, all the same. Before leaving the perfect tense, I should briey comment on a somewhat strange form occurring in the Silmarillion, chapter 20. Here we have the exclamation ut lien aur, translated the day has come. Ut lie (ut li) u e u u e is clearly the perfect tense of tul- come, as conrmed by the translation has come. However, the added n is something of a mystery. What is this extra consonant doing there? The form ut lien is reminiscent of ut lien u u I am come in Elendils Declaration in LotR, but here -n is a pronominal ending I (see the next section). No such ending can be present in ut lien, u given Tolkiens translation. The apostrophe inserted before this latter n probably indicates a dierent pronunciation as well; in ut lien the nal u consonant is perhaps meant to be sounded as a separate syllable. It may be that this n is added simply for the sake of euphony, preventing three vowels in sequence (since the next word also begins in a vowel; if you count the diphthong au in aur as two vowels, there would even be four sequential e vowels). If a perfect tense appears with no secondary ending added to i, and the next word begins in a vowel, should we always insert n to e avoid too many vowels in hiatus? I have used such a system in at least one composition of my own, but this conclusion is extremely tentative: In the exercises below I have never used this extra n, since no one really knows its function. Some even think it represents an alternative incarnation of the article (which normally appears as i): After all, Tolkien did employ the translation the day has come. Hence ut lien aur = ?ut li en aur u e u e e or ?ut li in aur has-come the day??? (For a possible attestation of in u e e as a Quenya article, see PM:395.) We can only hope that future publications will shed some more light on this. It may be noted that Christopher Gilson, who has access to unpublished Tolkien material, advocates the n = the interpretation.
8.2
Pronouns
It is time to introduce one of really economic devices of Language, the pronouns. (If you know perfectly well what a pronoun is, and you also know about the three dierent persons that personal pronouns are divided into, please scroll down until you see the word Quenya in red. Im not trying to waste anyones time here!) The word pronoun is a giveaway; it simply means for (instead of) a noun. Pronouns are words (or endings) that can replace a noun, often referring back to a noun that has already been mentioned. Thus you dont
133
have to repeat the noun itself all the time. Pronouns provide a kind of spoken shorthand, saving the language from utter tedium. Thanks to pronouns, speakers of English can keep up a conversation with another people without having to endlessly repeat the other partys name every time they are being addressed; instead the pronoun you is substituted. Instead of having to say the group just referred to or the people presently being discussed speakers of English have at their disposal the short, snappy word they. And try to imagine how you would go about referring to yourself without the pronoun I. Phrases like this person or the one who is talking now get tedious really fast. There are several kinds of pronouns (even interrogative ones like who), but the ones most frequently found are the personal pronouns, which we will focus on in this introduction. Customarily, they are divided into three dierent persons (not that the pronouns involved only refer to sentient beings; in this context, person is just an established term for a pronoun class). In English, this traditional tripartite classication produces a table something like this: First person (referring to oneself or ones own group): Singular I, as object me, of ownership my and mine; plural we, as object us, of ownership our and ours. Second person (directly addressing another person or another group): Singular and plural both you, which is also the object form; of ownership your and yours. Archaic English also has distinct singular pronouns: Thou, as object thee, of ownership thy and thine. Third person (referring to another person or group): Singular he, she, or it depending on the gender and/or nature of what is being referred to; as object him, her, or it (the latter being the same as the subject form), of ownership his, her (the latter happens to coincide with the object form, but there is also hers) and its. In the plural we have they, as object them, of ownership their and theirs. While the concept of these three persons as such is near-universal in the languages of the world, it is quite arbitrary what other distinctions languages build into their pronoun tables. The system is not necessarily symmetric, either certainly not so in English. English pronouns normally maintain a distinction of singular vs. plural, e.g. singular I vs. plural we, but this distinction is suddenly abandoned in the second person, where you is used no matter whether the speaker addresses one person or several people. On the other hand, English suddenly becomes very pedantic in the third person singular. Here you have to use he if you are referring to a male, she if you are referring to a female (or a ship!), and it if you are referring to something inanimate or abstract, or to an animal (unless it is a pet and you feel
134
you know it so intimately that you must consider saying he or she instead!) Such hair-splitting distinctions are not made anywhere else in the English pronominal system, and certain other languages dispense with them. Finnish, ever relevant for this study since it was Tolkiens foremost inspiration for Quenya, only has a single word (hn) that covers both he and a she: The Finns get happily along without making this distinction. On the other hand, other languages may go even further than English. For instance, the Hebrews apparently thought the masculine/feminine distinction was so interesting that it wasnt enough to have separate words for he and she. Hebrew also has separate words for you (atta when speaking to a man, att when addressing a woman); the language even carries the principle into the plural by having separate words for they (hem when referring to a group of men, but with reference to an all-female group, they is henna. . . as I understand it, a mixed group is referred to by the masculine term hem, and then we are left to wonder if a group of 10,000 women and 1 man is still hem rather than henna). What, then, about Quenya? What pronominal distinctions did Tolkien have his Elves make? It is somewhat dicult to say anything very denite about the Quenya pronominal system. Even now, with enormous amounts of material still unavailable to scholarship, it is already safe to say that the pronouns of Tolkiens Elvish languages were rather unstable probably even more so than many other aspects of his ever-uid linguistic constructs. The pronoun tables seem to have undergone countless revisions, and some think Tolkien never quite managed to sort out every detail. (Personally I think he did the problem is rather that he did it so often!) We know that the Quenya pronominal system, as Tolkien envisioned it in his later years, makes some distinctions that are not regularly expressed in English. For one thing, just as Quenya has a dual form of the noun in addition to the singular and plural forms, so there are also at least some dual pronouns. So in the First Person we dont nd singular I and plural we only, but also a distinct dual pronoun meaning you (sg.) and I or the two of us. Another subtle distinction is made in the words for we: In Quenya, there are separate words or endings for we, depending on whether or not the party that is addressed is included in we or not. On the other hand, it seems that Quenya does not always maintain the distinction between he, she and even it; all of these may be covered by a single pronoun. As this course proceeds, we will discuss various parts of the pronoun table and their associated obscurities, and also return to the special pronominal distinctions made in Quenya. However, let us introduce a few pronouns right away. One thing should be understood: in Quenya, pronouns typically appear as endings, not so often as independent words. (Where a Quenya pronoun 135
does appear as a separate word, it is often emphatic producing much the same eect as putting an English pronoun in italics: You [and no one else] did it. We will return to the independent pronouns later.) In the nal lines of Namri we nd the word hiruvaly, translated thou shalt nd by a e e Tolkien. If you have worked your way through all the exercises, you will remember the form hiruva, future tense of hir- nd. This hiruva shall nd here appears with the pronominal ending -ly attached, denoting the e subject of the verb. This ending belongs to the Second Person and signies thou or using a less archaic translation, you. Hence hiruvaly = e thou shalt nd, or you will nd. The sux -ly can be attached to any e verb to indicate that its subject is you, thou. Having mentioned this pronoun we however run into Instant Obscurity, which is a situation we shall often nd ourselves in while discussing Quenya pronouns. It is unclear whether or not this ending -ly covers both singular e and plural you; in Namri it is singular, as demonstrated by the transa e lation thou. In one of Tolkiens draft texts for the LotR Appendices, he actually wrote that the Elvish languages did not distinguish between singular and plural you (no more than English does): All these languages . . . had, or originally had, no distinction between the singular and plural of the second person pronouns; but they had a marked distinction between the familiar forms and the courteous (PM:4243). The ending -ly, used e by Galadriel to address a relative stranger like Frodo, would seem to be a polite or courteous you. In Namri it is thus used as a singular thou, a e only one person being addressed, but according to PM:4243 just quoted it could equally well be plural you (so if all the members of the Fellowship had understood Quenya, they still couldnt be certain whether Galadriel was addressing them all, or Frodo alone). However, in the essay Quendi and Eldar written about half a decade after the publication of LotR, Tolkien did imply the existence of pronominal endings that make a distinction between singular and plural you (WJ:364). Here he referred to reduced pronominal axes of the 2nd person, implied to be -t in the singular and -l in the plural. This -l could well be a reduced form of -ly, which would then be a plural you. Even so, Tolkien e indisputably used this ending for a singular you in Namri, since he a e translated it as thou in the text in LotR. This shorter ending -l is also attested as part of the verb hamil you judge (VT42:33), and this may also be taken as a singular you, though the context is not conclusive either way. It would seem that in the second half of the fties, Tolkien had been rethinking the pronominal system. The statement made in the draft text for the LotR Appendices, to the eect that Elvish did not distinguish singular and plural you, had not actually made it into the published LotR. Therefore he would not be bound by it. (Whenever we are dealing with Tolkien material that has been published only posthumously, we can never be certain that the information provided is entirely canonical: The author 136
could always change his mind, and so he often did, especially regarding the languages.) Tolkien had apparently discovered that Quenya possesses distinct pronouns for singular and plural you after all. Perhaps the new (ca. 1960) idea goes something like this: -ly and the shorter variant -l would properly e be a plural you, but it is also used as a polite singular, hence the translation thou in Namri. The idea of addressing a single person as if (s)he a e were several people could be a way of showing respect, emphasizing the importance of the other. Parallels are found in languages of our own world. (A former British PM on occasion carried this system over from the Second Person to the First, referring to herself as a plural we rather than a singular I, apparently to emphasize her own importance. Of course, royals have been using this linguistic device for centuries and for that matter, even the author of this course sometimes refers to himself as we! But actually I tend to include the reader in this we, so that you quite undeservedly receive some of the credit for the gradual unraveling of Quenya grammar that we are undertaking here. . . ) As for English thou vs. you, Matthew Skala wrote on the Eling list (January 4, 2001): Thou is second person singular, and you is second person plural, with the added rule that in formal contexts its polite to use plural even when talking to an individual. Much like French tu (singular/informal) and vous (plural/formal). In English it has become standard to use you for both singular and plural regardless of politeness, but that is a recent innovation; until about 100200 years ago, English speakers routinely used thou in informal contexts. The bizarre situation today is that because of this historical change, most of us only ever encounter thou in historical and formal contexts, such as the Bible. . . and so now, if you use it you sound like youre being especially formal and polite. The formality/politeness charge has been ipped backwards. It may be, then, that Quenya -ly or -l corresponds to the original use of e English you, before the historical change Skala describes but because of that change, -l(y) used as a polite singular may now be rendered thou, e as Tolkien translated it in LotR. To summarize: the ending -l(y) can certainly be used as a singular you, e and it is probably a polite/courteous form rather than a familiar/intimate form. It may be that -l(y) also covers plural you, this may even be e its proper meaning, but this is where things get somewhat obscure. Tolkien probably changed his mind back and forth about the details. In the exercises below, I have simply used the neutral word you as the equivalent of -l(y). e Then it is impossible to go wrong. 137
But we seem to have plunged right into the Second Person; let us return to the First. In the First Person singular, things are luckily crystal clear (well, very nearly so). The pronoun I is most often represented by the ending -n. (Linguists have noted that in the languages of the world, the term for I, me remarkably often includes some nasal sound like N or M. Whatever subtle features of human psychology underlie this phenomenon, Tolkien seems to have liked this association, and worked it into several of his languages. Cf. Sindarin im = I.) Notice how the ending -n is added to the verbs ut li (perfect tense of tul- come) and maruva (future tense u e of mar- abide, dwell) in Elendils Declaration: Et Erello Endorenna utlien = out of the Great Sea to Middle-earth a u I am come. Sinom maruvan = in this place will I abide. e However, the ending -n for I also occurs as a longer variant, -ny. (As e noted above, the ending -ly for you has a shorter variant -l; the variation e -ny vs. -n for you would parallel this.) This longer variant is seen in a e word we have already touched on in this lesson, the form ut vienyes! I u have found it! Aragorns exclamation when he found the discovered the sapling of the White Tree. The word ut vi, apparently the perfect tense u e of a verb tuv- nd, here occurs with two pronominal endings. The rst of them, -ny or I, denotes the subject of the verb: Ut vie+ny have e u e found+I = I have found. However, following -ny we have yet another e pronominal ending, the Third Person Singular sux -s, meaning it. Thus an entire sentence of verb, subject and object has been telescoped into a single word: utvienyes = I have found it. u
NOTE: Notice that according to the spelling conventions here employed, nal - loses e its diaeresis whenever an ending is added so that it is not nal anymore: ut vi + -ny = u e e ut vieny and not ut viny; adding -s to ut vieny likewise produces ut vienyes u e u e e u e u and not ut vienys. This is solely a matter of orthography. u e
We can abstract this grammatical rule: if a verb is to receive two pronominal endings, one denoting the subject of the verb and the second the object, the subject ending is attached rst and the object ending next. In published material, there are two or three other examples of this, beside ut vienyes. u It is then obvious why the long form -nye- is preferred here. While ut vu ien would do nicely for I have found, the object ending -s it could not have been added to the short ending -n, since **ut viens is not a possible u Quenya word. So we can formulate another rule: The long form -ny (-nye-), e NOT short -n, must be used for I if another pronominal ending is to follow it. (Similarly, for you one must use the long ending -ly [-lye-], not the e shorter form -l, if a second pronominal ending is to be added: You have found could be either ut viel or ut viely, but you have found it must u u e be ut vielyes, since **ut viels would be impossible.) u u The long ending -ny I may however occur even if there is no obe 138
ject pronoun following it (as can the long form -ly for you, thou, cf. e hiruvaly thou shalt nd in Namri ). The form linduvany I shall e a e e sing occurs on the frontispiece of the 1975 French bilingual edition of The Adventures of Tom Bombadil (ISBN 2-264-00913-6). The frontispiece reproduces a manuscript page by Tolkien, including some brief linguistic notes. (Taum Santoski, analyzing these notes in the newsletter Beyond Bree, October 1985, read this form as linduvanya but as pointed out by Carl F. Hostetter, Tolkien probably intended linduvanye instead. Tolkien was capable of wonderful calligraphy, but his normal handwriting is often a challenge to transcribers!) As long as no second pronominal ending is to follow, it is apparently wholly optional whether one uses the long ending -ny or e the short ending -n for I. We have the long ending in linduvany I shall e sing, but Elendils Declaration uses the short ending in maruvan I will abide. Certainly these examples could be scrambled to produce linduvan, maruvany of exactly the same meaning. e It seems, however, that the short ending -n is much more common than the longer sux -ny. We have already encountered this -n attached to e several verbs, such as polin I can, tirin I watch in the previous lesson. Tolkien very often cites primary verbs like this, listing them as they appear in the 1st person aorist (with the ending -i- intact because it is followed by an ending and hence not nal, so that it would become -). Tirin is an e example actually found in the Etymologies (entry tir), but by the standards of this eld, examples truly abound: carin I make, build (entry kar), lirin I sing (glir) or I chant (lir1 ), nutin I tie (nut), nyarin I tell (nar2 ), rerin I sow (red), serin I rest (sed), sucin I drink (suk), tamin I tap (tam), tucin I draw (tuk), tulin I come (tul), turin I wield (tur), tyavin I taste (kyap), vilin I y (wil), umin I do not (ugu/umu). The form polin I can (VT41:6) is one of several examples from post-LotR sources. Presumably it would in no way be wrong to use the long ending -ny instead (e.g. poliny), but -n is the commonest e e ending in the published corpus. But especially for the purpose of poetry, it is often practical to be able to choose between a long and a short pronominal ending, so that one can include or get rid of a syllable if the meter demands this. Also notice that the ending -ny, as well as -ly for you, cause the e e accent to fall on the syllable preceding the ending because ny and ly here count as consonant clusters. Cf. the stress rules set out in Lesson One. If hiruvany I will nd (with the accent on a) doesnt sound good in your e poem, you can always use the short form hiruvan and have the accent land on i in the rst syllable instead. (Again, we may have the same system in the Second Person: It is entirely possible that in Namri, Tolkien wrote a e hiruvaly rather than the shorter form hiruval simply because the former e variant t his poetic meter better.) As for the ending -s meaning it, occurring as an object pronoun in 139
ut vienyes I have found it, it seems that it may also be used as a subject. u For instance, if polin is I can, we must assume that it can would be polis. However, the ending -s brings us into the Third Person with its own set of obscurities, which we will save for later (Lesson 15). In the exercises below, -s is used in the same way as in the example ut vienyes: attached u to another pronominal ending to denote the object of the verb (just as the rst ending added to the verb denotes its subject).
Vocabulary
otso seven seldo boy (actually Tolkien didnt provide an explicit gloss, but the word is cited in
a context where he is discussing Quenya words for child, and seldo seems to be a masculine form. See the entry SEL-D- in the Etymologies.)
ml thrall, slave o an for (or since, because, introducing a sentence giving a reason, as in I rely on
him, for he has often been of help to me.)
tul- verb come lanta- verb fall nurta- verb hide (cf. the Nurtal Valinreva or Hiding of Valinor referred to e o
in the Silmarillion)
lerya- verb release, (set) free, let go metya- verb end = put an end to roita- verb pursue 140
Exercises
1. Translate into English (and practice your vocabulary; except for the numeral otso and the pronominal endings, exercises AH only employ words you are supposed to have memorized in earlier lessons): A. I nr ih e i harma. e ri B. I rvi amtier i hrv. a a a e C. I aran utulti i tri. e a D. I nissi ecendier i parma. E. I uma tri ampi i otso Naucor. e a a e F. Etciely otso parmar. e e G. Equtien. e H. Ecnielyes. e 2. Translate into Quenya: I. The man has come. J. The seven Dwarves have eaten. K. The boys have seen a lion between the trees. L. The six Elves have pursued the seven Dwarves. M. The Dwarf has hidden a treasure. N. I have praised the king, for the king has released all thralls. O. You have fallen, and I have seen it. P. I have put an end to it [/I have ended it].
141
Lesson 9
The innitive. The negative verb. Active participles.
9.1
The innitive
All of the forms of the verb that we have discussed so far, all the tenses, are what a linguist would call nite verb forms. The denition of a nite verb is that it is capable of functioning as the predicate of a sentence, the part of the sentence that tells us what the subject does (or is in Lesson Four we pointed out that a phrase made up of copula + noun or adjective also counts as a predicate, e.g. gold is beautiful, but here we will deal with more normal verbs instead). In a sentence like i Elda mta massa the a Elf is eating bread, linguists can readily label the roles of all parts of the sentence: just like i Elda the Elf is the subject and massa bread is the object, so the verb mta is eating is the predicate of the sentence. And a precisely because the form mta, present tense of mat- eat, is able to a function as a predicate here, we can tell that mta is a nite form of the a verb. The innitive is another story. It is, as the name suggests, in-nite non-nite. It is not inected for time, as are the tenses. It does not receive the ending -r, even if the subject of the sentence is plural. So by itself, an innitive is not capable of functioning as the predicate of a sentence. An innitive cannot be directly teamed up with a subject. What, then, is its use? English innitives have various uses, but an important function of the innitive is that it allows several verbs to be combined in one sentence. In a sentence like the Dwarves wanted to eat, the verb wanted is a nite form, appearing in one specic tense (in this case past tense). But the verb eat appears as an innitive, to eat, complementing the nite verb to form a longer verbal phrase wanted to eat. In English, innitive verbs are very often marked by inserting to before the verb proper, but this to is not always included. In a sentence like I let him go, the verb go counts as an innitive even though there is no to before it. (Contrast I allowed him to go.) Neither is to included before an innitive following certain 142
verbs like can or must (e.g. I must go, not **I must to go). In Quenya, there seems to be no independent innitive marker like English to, so we dont have to worry about where to include or omit it. Attested examples of Quenya innitives most certainly do not abound, but there is the sentence polin quet I can speak (VT41:6). Here the verb polin I e can is a nite form, the aorist of the primary verb pol- appearing with the pronominal ending -n I attached but the word quet must be analyzed e as an innitive. Of course, quet is similar in form to an aorist speaks, e but as indicated by the translation speak as well as the context, the form quet is innitival here. We can tell, then, that primary verbs like quete have innitives in - (undoubtedly representing Primitive Elvish -i ). The e ending may be analyzed simply as a kind of stopgap that is supplied to make up for the absence of any other ending, or quet may be seen as representing e an uninected primitive I-stem kweti. No matter how we imagine the ultimate derivation and the meaning of the ending -, we probably know e enough to actually start using the innitive form of primary verbs. Here are some (home-made) examples combining innitives with various nite forms (tenses) of the verbs mer- wish, want and pol- can. Finite verbs in red, innitives in blue: I Elda pol cen i Nauco the Elf can see the Dwarf (notice that though e e
the verbs pol- want and cen- see, behold receive the same ending - here, the former e is an aorist and the latter is an innitive: The context must decide whether the form cen e is to be understood as an aorist sees or an innitive [to] see)
I Naucor merner mat the Dwarves wanted to eat (nite verb merner e
wanted, inected for past tense and plural, + innitive verb mat to eat) e
I seldo poll hlar ilya quetta the boy could hear every word e e Polily car ilqua you can do everything e e I nissi meruvar tul the women will want to come e What, then, about A-stems? In the Etymologies, Tolkien often glossed Astem verbs as if they were innitives, e.g. anta- to present, give, varyato protect or yelta- to loathe (entries ana1 , bar, dyel). This is not by itself conclusive evidence that a form like anta could actually be used as an innitive to give in a Quenya text, for in the tradition of Western linguistics, the innitive is commonly the form used to name, list or gloss a verb in wordlists. Sometimes this system is carried through even where such a gloss is strictly wrong: A Hebrew-English wordlist may insist that nathan means to give, though it actually means he gave this being the simplest and most basic form of this verb, the logical form to be listed in a dictionary. However, a form like anta- is simply an uninected A-stem, and Tolkien did refer to certain grammatical circumstances when the bare stem of the verb is used . . . as innitive (MC:223). The general system also seems to suggest that A-stems with no additions can function as innitives. (Notice that the innitives of both primary verbs and A-stems seem to be 143
similar in form to ending-less aorists.) So I guess we can have sentences like the following (and let me just underline the innitives to avoid too much fancy coloring): I vendi merner linda the maidens wanted to sing I norsa pol orta i alta ondo the giant can lift the big rock e Merin cenda i parma I want to read the book In some cases, English may prefer a form in -ing to a regular innitive, for instance after the verbs start and stop. I think it is a fair guess that Quenya would use the normal innitive in such cases as well: I nissi pustaner linda the women stopped singing (or, . . . ceased to sing) Several innitives can probably be juxtaposed by means of ar and: I neri merir cenda ar tec rimb parmar the men want to read and e e (to) write many books The discussion above certainly does not cover all there is to say about Quenya innitives. Some more details are known and will be lled in later in this course, but there are many obscure points. In some very late (ca. 1969) notes, Tolkien refers to the general (aorist) innitive formed by added -i (VT41:17), but since only brief quotes from this material have been published, we cannot be certain what he means. Is there a specic aorist innitive? We have earlier discussed the distinction made between such forms as mta is eating (present/continuative tense) and mat eats a e (aorist). Does Quenya carry these distinctions over into the innitive, so that one can somehow distinguish to eat (aorist innitive) from to be eating (continuative innitive)? Moreover, what does Tolkien refer to by added -i ? Obviously there is an innitive that is formed by adding -i to the verbal stem (of primary verbs at least). But is this ending a contemporary Quenya sux, or does it represent a Primitive Elvish form? As mentioned above, the attested innitive quet e (to) say may be meant to represent a primitive form kweti, which would indeed be the root kwet- with added -i . But if this -i is a contemporary Quenya sux, there would be an alternative innitive queti to say. How it is used, and whether it is interchangeable with the attested form quet, e we cannot even begin to guess. In the essay Quendi and Eldar, Tolkien did mention a few verb forms that may seem to exemplify an innitive in -i, namely auciri and hciri, both meaning cut o (in two dierent senses, o see WJ:365366). But later in the essay, he quoted the same forms with a hyphen attached (auciri-, hciri-), as if these are verbal stems rather than o independent innitive forms (WJ:368). So we cannot be sure of anything, and must await the publication of more material. As noted above, the innitive is traditionally used to name or list verbs, or to give their meaning as a general gloss. From now on we will often 144
dene verbs in such a way, e.g. glossing a verbal stem like tul- as to come and lanta- as to fall (rather than simply come, fall). It must still be understood that the mere stem of primary verb like tul- cannot function as an actual innitive (tul) in a Quenya text (it must be tul instead). It is e simply customary and convenient to give the meaning of a verb by quoting its gloss in the innitive. In the Vocabulary listings of Lessons 5 through 8, I had to write verb in front of the gloss of every new verb to make it crystal clear what part of speech the new word belonged to. Sometimes this was actually necessary: If I dened lanta- simply as fall, some student would surely manage to overlook the nal hyphen of lanta- that is meant to suggest that this is a verbal stem, and conclude that fall is here a noun autumn, or something! Finally having introduced the innitive, I will use this form instead when glossing verbs like to fall in this case.
NOTE 1: In English, innitives introduced by to (or in order to) are often used to describe an intention: I came to see you. In such a context, it seems that Quenya does not use the forms discussed so far, but a quite dierent construction (gerund in dative, to be discussed in a later lesson).
9.2
This may be a good place to introduce a somewhat peculiar Quenya verb. Earlier we have mentioned the copula n is, which we can now refer to as a a tense of the verb to be. (Dont ask me if n is the present tense or the a aorist, and the other tenses of this verb are unfortunately even more obscure: The verb to be is notoriously irregular in the languages of the world, and Tolkien may well have invented some nice irregularities for Quenya as well.) Anyhow, Quenya also has a unitary verb meaning not to be; you can express this meaning without combining some form of n with a separate a word for not (though Quenya does have such a negation as well). This verb is listed in the Etymologies, entry ugu/umu, where it appears as umin I am not (another example of Tolkiens frequent habit of listing primary verbs in the 1st person aorist). The past tense is also listed, somewhat irregular: it is um, not **umn as it would have to be according to the simplest e e e as the past tense of a primary verb um- would regular pattern. Um seem to belong to the same pattern as lv, pa.t. of lav- to lick (cf. a e undulv down-licked = covered in Namri in LotR). One must take a e a e care not to confuse the past tense form um was not with the ending-less e aorist um is not. e As the future tense of this verb, we might expect umuva, and this unattested form may well be permissible but actually a shorter form uva occurs in F riels Song. Here we have the phrase uva . . . fara, will not e be enough (fara = adjective enough, sucient). Possibly, this uva is e actually the future tense of another verb: Besides umin I am not from the
145
root umu Tolkien also listed a form uin of the same meaning apparently derived from the root ugu. Perhaps uva is strictly the future tense of the latter verb. It could represent a primitive form something like uguba, whereas uin is to be derived from ugin (or ugi-ni at an even older stage). Between vowels, g was lost in Quenya, so that the two us of uguba merged into one long u in uva, whereas the u and the i of ugin merged into a diphthong ui (as in uin) when the disappearance of g brought the two vowels into direct contact. Whatever development Tolkien may have imagined, we will here use uva as the future tense of um- not to be, avoiding the unattested (and perhaps somewhat awkward) form umuva. Like n, this negative copula can presumably be used to connect a a subject with a noun or an adjective: I Nauco um aran the Dwarf is not a king e I nissi umir tiuc the women are not fat e I rocco um mor the horse was not black e e I neri umer sail the men were not wise e Elda uva uma an Elf will not be evil e Nissi uvar ohtari women will not be warriors (sorry, Eowyn!) Or, using pronominal endings instead of an independent subject: Umin Elda I am not an Elf Umen saila I was not wise Uvaly ohtar you will not be a warrior e But above I said that this was a good place to introduce the negative verb. This is because it can probably be combined with innitives as well. We lack actual examples, but in the entry ugu/umu in Etym, Tolkien indicated that umin does not always signify I am not. It can just as well mean I do not. By combining such a verb with an innitive, one can probably negate the verbs in question. Home-made examples involving various tenses of the negative verb: Umin tur macil I do not wield a sword e Mma um mat hrv a sheep does not eat esh a e e a e I Nauco um tul the Dwarf did not come e e I neri umer hir i harma the men did not nd the treasure e I n uva linda the woman will not sing s I neri uvar cen i Elda the men will not see the Elf e We must assume that following the negative verb, as well as in other contexts, several innitives may sometimes be combined, like mer and e cen in this sentence (the nite verb in red, the two innitives in blue and e pink): I Elda um mer cen i Nauco. The Elf did not want to see the e e e Dwarf.
146
Or again, with the innitives mer and cenda: e I Nauco uva mer cenda i parma. The Dwarf will not want to read e the book. Presumably the present/continuous tense of the negative verb, which would have to be uma, can be used to deny the existence of an ongoing action: I Nauco uma linda the Dwarf is not singing (just now) Contrast the aorist: I Nauco um linda the Dwarf does not sing. The e latter would often (but not necessarily) have a wider application, like the Dwarf is not a singer. Anyhow, we will stick to the aorist in the exercises below.
9.3
Active participles
The various parts of speech, such as nouns, verbs, and adjectives, remain relatively distinct categories most of the time. However, some words fuse the properties of several parts of speech. The participles are words with a basically adjectival function, but they are directly derived from verbs, and in the case of active participles, they are still able to take an object. The participles are subdivided into two categories, often called present participles and past participles. These terms are somewhat misleading, for the most important distinction between has nothing to do with tenses. The alternative terms active participles and passive participles are better, and I will try to use them consistently here. We will save the past or passive participle for the next lesson and focus on the present or active participles here. In English, this form is derived by means of the ending -ing. For instance, the verb follow has the active participle following. This verbal adjective describes the state of something or someone that carries out the action of the corresponding verb: The day that follows can be described as the following day. If the verb is able to take an object, so is its corresponding participle. A person who loves Elves can be described as a person loving Elves. In English, the form derived from verbs by adding -ing is somewhat ambiguous. It can also function as a noun. The active participle of a verb like kill is of course killing, as it is clearly adjectival in a phrase like a killing experience, but in a sentence like the killing must stop, it is equally clear that it is used as a noun. But in the latter sentence, killing is a verbal noun, an abstract noun denoting the action of killing. Here we are only interested in verbal adjectives = participles. In Quenya, the two do not coincide in form. The Quenya ending corresponding to English -ing (when used to form participles) is -la. There are many examples of active participles in the 147
Markirya poem. For instance, Tolkien in his annotation indicated that ilkala [is the] participle of ilka gleam (white) (MC:223). The participle ilcala (as we would spell it here) thus means gleaming, and so it is used in the poem, in a phrase translated in the moon gleaming (MC:215). It seems that in a Quenya active participle, the stem-vowel is lengthened if possible. In ilcala the i cannot become long because there is a conso nant cluster following it. However, Tolkien in MC:223 also mentioned a verb hlapu- to y or stream in the wind (one of the rare U-stems, a rather obscure category of verbs). Its participle appears as hlpula on the previous a page: Winga hlpula, translated foam blowing (cf. MC:214). We must a assume, then, that the participle of a verb like lala- to laugh is llala a (!) laughing: The stem-vowel is lengthened. If the verbal stem includes a vowel that is long already, it simply stays long in the participle: The participles of p ca- to lessen, dwindle and r ma- to shift, move, heave appear u as p cala and r mala in the Markirya poem. u In the case of longer verbal stems where the stem-vowel occurs twice, as in falasta- to foam (root evidently phalas), it seems that it is the second occurrence of the stem-vowel that is to be lengthened if possible. In this case it cannot be lengthened, since it is followed by a consonant cluster; the participle foaming is attested (in Markirya) as falastala. The rst occurrence of the stem-vowel could have been lengthened as far as phonology is concerned (**flastala), but this rst vowel evidently does not count a for the purpose of lengthening. (Presumably it is not lengthened in the present tense, either: falasta is foaming, hardly ?flasta, much less e a e **falsta.) a e The primary verbs are a problem. Adding the ending -la to their stems would usually result in consonant clusters not permitted in Quenya. For instance, the participle of the verb tir- to watch cannot be **tirla (let alone **t rla), a quite impossible Quenya word. It has been assumed that in such cases, one may start by constructing the continuative stem (similar to the present tense) by lengthening the stem-vowel and adding -a, e.g. t is ra watching, and then derive the participle by adding the participial ending -la to this form: t rala watching. Markirya has hcala as a participle a yawning; unfortunately the underlying verb to yawn is not attested, but if it is a primary verb hac-, the attested participial form would conrm such a theory. But of course, the verb underlying the participle hcala could a just as well be an A-stem haca- or hca- (cf. hlpula blowing, streaming a a from hlapu- and p cala dwindling, waning from p ca-). With the publication of The Peoples of Middle-earth in 1996, a form that may seem to be the participle of a primary verb became available: PM:363 refers to the root it [as in] itila twinkling, glinting, and a ash, itata verb to sparkle. But is itila really the participle of a primary verb it-? Tolkien refers to it- as a stem or root (cf. PM:346), not as a Quenya verb. The actual Quenya verb in question is listed as ita-, a short A-stem meaning 148
to sparkle. Its participle would presumably be tala, not itila. If the latter is a participle at all, it is a peculiar one: it shows no lengthening of the stemvowel (not ** tila), and a connecting vowel -i- is inserted before the ending -la. Since the aorist of a verb it- would be iti- (becoming it only in the e absence of any endings), one may wonder if itila is an aorist participle. This would mean that Quenya is able to carry the distinction of aorist/present tense over into the participle, so that there are dierent forms for doing (habitually or momentarily) and doing (continuously): perhaps something like carila and crala, respectively (from the verb car- to do). But this a is speculative, and I cannot recommend such a system to writers; we must await the publication of more material. It may be that itila is simply an old adjectival formation that no longer counts as an adjective in Quenya. The ending -la occurs in adjectives as well, e.g. saila wise; undoubtedly -la is in origin simply an adjectival ending that came to be favoured as the sux used to derive verbal adjectives = participles. Even so, Quenya participles seem to have established themselves as formations quite distinct from adjectives, for in one respect their behaviour diers: Unlike adjectives, the active participles apparently do not agree in number. For instance, Markirya has rmar sis a lala for wings shining (the second word being the participle of the verb sis la-, a longer variant of the verb sil- shine white). As we remember, normal adjectives in -a have plural forms in - (representing archaic Quenya -ai). So if sis e lala were to be agree in number with the noun it describes, we would have expected **rmar sis e. Perhaps Tolkien did not want participles in -la to agree a lal in number precisely because the plural form of the participial ending would have to be -l: This ending could easily be confused with the prominent e abstract ending -l, which is added to verbal stems to derive verbal nouns e e.g. lindal singing from linda- to sing (as in Ainulindal Ainue e singing, free rendering Music of the Ainur). While lindala and lindal e both translate as singing in English, the latter is a noun (a singing), whereas the former is singing in the adjectival sense. English very often employs the active participle to express the meaning of a continuative tense, combining the participle with a copula like is or was, e.g. the boy is laughing. But regarding present actions at least, Quenya would rather express this meaning by using the genuine present/continuative tense: I seldo lla. None can say whether the Engae lish-style wording i seldo n llala is a valid Quenya sentence; one suspects a a that while it would be intelligible, the Eldar (/Tolkien) would not think of it as good Quenya. While we have no attested example of an active participle taking an object, we must assume that it is possible, e.g. Nauco t rala Elda, a Dwarf watching an Elf.
149
Vocabulary
tolto eight pol- to be (physically) able to, normally translated can (where this refers
to some physical ability not can in the sense know how to, referring to intellectual skill, or can in the sense may = is permitted to, referring to freedom from prohibitions. For the two latter meanings, Quenya uses distinct words.)
um- negative verb not to do or not to be, past tense um, future tense e uva mer- to wish, want hlar- to hear(related to Sindarin lhaw as in Amon Lhaw, the Hill of Hearing mentioned in LotR)
verya- to dare (from the same root as the Sindarin name Beren, meaning bold or
daring one)
lelya- to go, proceed, travel, past tense lend, perfect [e]lendi (more e e
about this irregular verb in the next lesson)
pusta- to stop ruhta- to terrify, to scare (ultimately related to Urco or Orco, the Quenya words
for bogey, Orc)
150
coa house (building only, not house = family) mir preposition into ter preposition through (a longer variant ter also exists, but I have used ter in e
the exercises below)
Exercises
1. Translate into English: A. S lala Isil orta or Ambar. e B. I cpala Nauco lantan ter i talan. a e C. Polin hlar lindala vend. e e D. Min nr t rala min macil uva ruhta i tolto taur ohtari. e e u e e E. Ml mpala taura nr um saila. o a e e F. I tolto rvi caitala nu i aldar ortaner, an i rvi merner a a mat i neri. e G. R um pol pusta mat hrv. a e e e a e H. I ruhtala ohtar pustan tir i li, an i ohtar um saila. e e e e 2. Translate into Quenya: I. The man pursuing the Dwarf is a warrior. J. The king wanted to go. K. The maiden did not dare to see the queen. L. The laughing women went into the house. M. The eight traveling Dwarves can nd many treasures. N. You did not praise the Elf, you do not praise the Man [Atan], and you will not praise the Dwarf. O. I want to travel through the world and free all peoples. P. A daring man went through the gate and into the mountain.
151
Lesson 10
Adverbs. The pronominal endings -nt and -t. Innitives e with object pronouns. The past tense of intransitive verbs in -ya. Passive participles.
10.1
Adverbs
Adverbs form a part of speech that is used to provide extra information in a sentence. A typical sentence provides information about who does what (to whom), involving a subject, a predicate and if necessary an object. But you may also want to slip in information about when, where or in what manner the verbal action occurs. This is where adverbs enter the linguistic stage. In many cases, adverbs are to verbs what adjectives are to nouns. Like an adjective may describe a noun, an adverb may describe the nature of the verbal action of the sentence. In a sentence like they left swiftly, the last word is an adverb describing how or in what manner they left. If we say she is singing now, the word now is an adverb answering the question of when the verbal action is taking place. And if we say they did it here, the word here is an adverb telling us where the verbal action took place. Some adverbs may be called basic, since they are not derived from another part of speech. Just consider such an adverb of time as English now and its Quenya equivalent s neither can be further analyzed. But very ; many English adverbs are not basic in this way. They are transparently derived from adjectives, as in one of the examples we just used: The adverb swiftly is obviously based on the adjective swift. The Great English Adverb-Former is the ending -ly, which can in principle be added to any adjective, turning it into an adverb (producing such pairs as deep/deeply, nal/nally, great/greatly, high/highly, swift/swiftly and countless others. . . but preferably not good/goodly, since the place of goodly is already occupied by the basic adverb well !) Since we have only a handful of words that Tolkien explicitly identied as adverbs, but plenty of adjectives, it would be nice if we could pin down a Quenya adverb-former like the English ending -ly. Then we could derive our own Quenya adverbs. We may have such a Quenya ending. It occurs in LotR, as part of the Cor152
mallen Praise (volume 3, Book Six, chapter IV: The Field of Cormallen). As part of the praise received by the Ringbearers we have the two words andav laituvalmet, translated long we will praise them in Letters:308. e Here we have the adverb andav, long (here meaning for a long time). e We know that the Quenya adjective long is anda (cf. Sindarin and as in And +duin = Anduin, Long River). It would seem, then, that this adjective has been turned into an adverb by supplying the ending -v (probably ree lated to the Quenya preposition ve as, like). In the case of anda/andav, e the English translation is long in either case, but normally the ending -v e would correspond to English -ly. So if alta is great, can we use altav e for greatly? Since tulca is a word for rm, would rmly be tulcav? e Knowing that saila means wise, can we assume that sailav an accepte able word for wisely? By and large, I think such formations are plausible, though the potential application of the ending -v may not be literally lime itless. The Quenya adjective good is mra; one wonders if mrav for a a e well would sound just as weird as goodly in English! (A basic adverb vand well occurs in Tolkiens earliest Qenya wordlist [QL:99]; whether e this was still a valid word in LotR-style Quenya some forty years later, none can say.) Like anda long, the vast majority of Quenya adjectives end in -a. The less frequent adjectives in - in practically all cases descend from Primitive e Elvish forms in -i, which vowel would be preserved unchanged before an ending or in compounds: Compare mor dark, black with the compound e Moriquendi Dark Elves. We must assume that the original quality of the vowel would also be preferred before the adverbial ending -v so if we e try to derive an adverb darkly from mor, it should probably be moriv e e rather than morev. Actually, very few of the adjectives in - are likely e e to have any corresponding adverbs; they mostly denote colours. Perhaps we can have muss/mussiv soft/softly, nind/nindiv thin/thinly e e e e and ring/ringiv cold/coldly (but in a later source, the word for cold e e appears as ringa rather than ring, and then the adverb would simply be e ringav). e How the ending -v would be added to the few adjectives in -n is quite e unclear. The adjective melin dear (not to be confused with the similarsounding 1st person aorist I love) could have a corresponding adjective melinv dearly, for while nv does not occur in unitary words, it is a e possible Quenya combination (cf. Aragorns title Envinyatar Renewer, where en- = re-). On the other hand, if the ending -v is related to the e preposition ve as, like, both probably descend from be in Primitive Elvish. We could then argue that the original melin-be would rather come out as melimb in Quenya. On the yet other hand (if we can postulate even more e hands), adjectives in -in seem to be shortened from longer forms in -ina, and then one could argue that this a would be preserved before an ending. Thus, dearly could be melinav. (Id say, forget about melin and start e 153
from melda or moina instead, which adjectives also mean dear. Then we can simply have meldav or moinav!) e e In English at least, an adverb does not necessarily describe a verbal action. It can also be used to modify the meaning of an adjective (or even another adverb). This is a kind of meta-description, one descriptive word describing another. Whether Quenya adverbs (or specically the ones in -v) can be e used in such a way, nobody knows. For instance: Knowing that valaina is the Quenya adjective divine, can we feel free to use valainav vanya e for divinely beautiful? Tolkien provided aqua as the adverb fully, completely, altogether, wholly (WJ:392 this is a basic adverb not derived from an adjective, unlike the English glosses in -ly that are derived from the adjectives full, complete, whole). It does seem very likely that this aqua can modify an adjective, e.g. aqua mor completely dark. If this is not e so, Tolkien ought to have told us. . . ! It may be noted that in some early sources, Tolkien uses adverbs in -o rather than -v. The one attestation of the latter is, as I have pointed e out, andav vs. the adjective anda long. However, there exists an early e Qenya sentence which translates as the Elves were lying long asleep at Kovienni [later: Cuivinen]; see Vinyar Tengwar #27. In this sentence, the e e adverb long appears as ando, not andav. Further examples of adverbs e in -o include ento next and rato soon (from an Arctic sentence quoted in Father Christmas Letters obviously a form of Qenya, though appearing in a context that has nothing to do with Tolkiens serious literary output). We may even include the adverb voro ever, continually from such a relatively late source as the Etymologies (entry bor), though in this word, the nal -o may be simply the stem-vowel reduplicated and suxed. The example ando long (not to be confused with the noun gate), which is obviously derived from the adjective anda, would seem to indicate that the ending -o can be used to derive adverbs from adjectives. May we then have (say) tulco rmly from tulca rm, as an alternative to tulcav? Or are we to understand that Tolkien, by the LotR period, had e dropped -o as an adverbial ending? If so he introduced -v as a replacement, e not an alternative (changing ando to andav). e We cannot know whether -o is still a valid adverbial ending in LotR-style Quenya. But when deriving adverbs from adjectives, I would recommend using the safe (or at least safer) ending -v instead. In the exercises below, e I have not used the ending -o, but only -v. On the other hand, at this stage e I would not tamper with attested adverbs like ento, rato, voro (changing them to ?entav etc.) e Do adverbs, like adjectives, agree in number? It has been suggested that andav is actually a plural adverb, agreeing with a plural verb (andav e e laituvalmet long we will praise them notice the plural subject ending attached to the verb). If so, -v could be the plural form of a singular e adverbial ending -va, completely unattested. According to this system, we 154
would have such variation as i nr lend andava the man traveled long e e (singular adverb corresponding to a singular verb) vs. i neri lender andav e the men traveled long (plural adverb to go with a plural verb). But this is 100 % hypothetical. While nothing can be ruled out at this stage, I tend to believe that this there is no such variation. More likely, the adverbial ending -v is invariable in form, related to the preposition ve as, like as suggested e above. In closing, I should mention that some Quenya adverbs are derived from other parts of speech than adjectives. In Namri we have oial as the ada e e verb for ever (or everlastingly, as the interlinear translation in RGEO:67 goes). But the Etymologies, entry oy, indicates that oial is properly or in e its origin a noun meaning everlasting age. Apparently this noun is used as an adverb in Namri. a e Phrases involving prepositions very often have an adverbial function to begin with, and sometimes unitary adverbs may evolve from them: In Cirions Oath we have tennoio as another Quenya word meaning for ever, but in UT:317, Tolkien explains that this form is simply a contraction of two originally distinct words: the preposition tenna up to, as far as + oio an endless period. Finally we have what I have already called basic adverbs, not derived from any other part of speech. Aqua completely and s now men tioned above are just two examples; we may also include words like amba up(wards), hya far o (read perhaps haiya as the Third Age form), a oi ever, and others.
10.2
In Lesson Eight, we introduced three pronominal endings: -n or longer -ny e for I, -l or longer -ly for you, and -s for it. But obviously there are e more pronouns, and we will now attempt to identify the pronominal endings of the Third Person Plural: as subject they, as object them. Cirions Oath in UT:305 includes the word tiruvantes, in UT:317 translated they will guard it. The verb tir- watch, guard, the future-tense ending -uva shall, will and the pronominal ending -s it ought to be familiar to the student by now. We are left with -nte- as the element translated they. UT:317 explicitly conrms that -nt is the inection of 3[rd e person] plural where no subject is previously mentioned. Like most brief linguistic notes of Tolkiens, this one does require some exegesis. I shall here assume that Tolkiens intention is this: If a sentence has a plural subject that has been previously mentioned, occurring before the verb, the verb would only receive the normal plural ending -r (e.g. i neri matir apsa the men eat meat). But if there is no subject previously mentioned, the ending -r is replaced by -nt, meaning they: Matint apsa, they eat meat. e e
155
Apparently, this ending would still be used if the subject is identied later in the sentence; perhaps we can have such a sentence as matint apsa i neri e they eat meat(,) the men (do). Cirions Oath also identies the subject a u later in the sentence (nai tiruvantes i hrar mahalmassen mi Nmen be it that they will guard it, the ones who sit on thrones in the west. . . ) Cirions Oath occurs in post-LotR material, so the information provided in UT:305, 317 was certainly intended to be LotR-compatible. However, a quite dierent pronominal ending for they occurs in Tolkiens early material. In LT1:114, we nd the Qenya form tulielto they have come, including the ending -lto for they. This ending was current as late as when Tolkien wrote F riels Song, which includes the forms crielto they made and a antalto they gave (LR:72). Whether it is also valid in LotR-style Quenya is another matter. The ending -lto looks somewhat strange compared to the other known pronominal endings. Of the pronominal endings attested in LotR or during the post-LotR period, all the subject endings that constitute a separate syllable end in the vowel - (six endings in all, if we include -nt e e discussed above). A sux -lto ending in -o doesnt seem to t in very well (so some would alter -lto to -lt in LotR-style Quenya, though there is e no evidence for such an ending). I tend to assume that Tolkien eventually scrapped this ending completely, replacing it with -nt. e The opinion has been voiced that -lto is valid all the same. Some would interpret Tolkiens note about -nt being used where no subject is previously e mentioned in an absolute sense: It wouldnt be enough that the subject has not been previously mentioned in the same sentence, as I assumed above. Of course, when the word they is used in English, it usually refers back to some group mentioned earlier in the text or conversation. According to the strict interpretation of Tolkiens note about -nt, this pronominal ending e cannot be used for any they that refers back to some group mentioned earlier, even if it was in a quite dierent sentence. The ending -nt would e only point forward, to some group that is to be identied later in the text or sentence (as is the case in Cirions Oath). They referring back to some other group (already mentioned in another sentence) would require a quite dierent ending, perhaps the -lto attested in earlier sources. I cant claim that this isnt a possible interpretation of Tolkiens words or the available examples. However, I still have a bad feeling about using the ending -lto in LotR-style Quenya. In the exercises I have made for this course, I have ignored -lto, assuming that -nt can be used as a pronominal e ending signifying they in a general sense. When Tolkien speaks of -nt e being used only for a subject that has not been previously mentioned, I assume that he means not previously mentioned in the same sentence (for if a plural subject had already occurred, the verb would receive only the normal plural marker -r). Hence we can presumably have forms like these, with -nt attached to the various tenses of pusta- to stop: e
156
Aorist pustant they stop e Present pustant they are stopping e e Past pustanent they stopped e Future pustuvant they will stop e Perfect upustient they have stopped e As indicated by the attested example tiruvantes = they will guard it, a second pronominal ending can be attached following -nt (-nte-), denoting e the object of the sentence. This brings us over to another question: If -nt is e the subject ending they, what is the corresponding object ending them? Discussing adverbs above, we have already quoted the sentence andav e laituvalmet long we will praise them from LotR. Knowing that laituvalmet means we will praise them, we can easily isolate the nal -t as the element translated them. (The cunning student will also be able to isolate the pronominal ending signifying we, but we will save that one for later: Actually Quenya has several endings for we, with dierent shades of meaning.) As usual, things are not quite crystal clear. The ones being praised here are Frodo and Sam, two persons. Some have therefore assumed that this -t is a dual them, even suggesting that laituvalmet may be rendered we will praise both [of them]. Those adhering to this theory have been encouraged by the fact that there is also a dual ending -t (as in ciryat 2 ships; look up Lesson Three again). Nothing can be denitely ruled out at this time, but the ending -t them would seem to match -nt they quite well. I e dont think -t is exclusively dual, but in any case, this is one ending that can be translated them. Hence, forms like the following must be possible: Tirnenyet = I watched them Melilyet = you love them Hiruvanyet = I will nd them and even: Pustanentet = they stopped them Likely, this would refer to two dierent groups. They stopped themselves is probably expressed in another way (unfortunately we dont really know how).
10.3
So far, we have identied two pronominal endings that can be used as the object of the sentence, -s for it and -t for they. As is evident from attested examples (tiruvantes they will guard it, laituvalmet we will praise them), these object endings may be attached to a nite verb following
157
another pronominal ending denoting the subject. But what about a longer verbal phrase involving an innitive? Let us start with a sentence like i ml veryan cen i aran ar i tri, o e e a the thrall dared to see the king and the queen. Here we have a nite verb veryan dared + an innitive cen to see. Now we want to get e e rid of the whole phrase the king and the queen, replacing it with the object pronoun them, hence the thrall dared to see them. (Notice that I deliberately construct an example that will be compatible with the theory of -t them being dual only, even though I dont believe this to be the case. . . unnecessary risks are just that, unnecessary!) Well, where do we put the ending -t? Quite obviously, it must be attached to the innitive cen to e see. Cenet, then? Or, since the innitive cen seems to represent Primitive e Elvish keni and primitive -i changes to - only when nal, one might think e o that cenit is a better choice. So the thrall dared to see them = i ml veryan cenit, right? e Wrong! In Vinyar Tengwar #41, July 2000, it was revealed that the innitive of primary verbs is formed with the ending -ita if any pronominal endings are to be added (actually the sux is only -ta-, which added to an innitive like cen = ceni- produces cenita-). Tolkien in some of his late e (ca. 1969) notes refers to the general (aorist) innitive formed by adding -i (not as such capable of any further suxion; with pronominal axes it was the stem of the aorist tense); the particular innitive with -ita diering in use from the preceding mainly in being able to receive pronominal object axes (VT41:17). He went on to quote the example caritas, doing it (or probably just as well to do it) an innitive of the verb car- do with the object ending -s it attached. As I pointed out in the previous lesson, it is unclear whether the reference to an innitive constructed by adding -i implies that there is a contemporary Quenya innitive that shows the ending -i. Tolkien may simply refer to the original form of the innitive ending, e.g. Primitive Elvish kweti as the form underlying the contemporary Quenya form quet (to) speak (ate tested in the sentence polin quet I can speak). Anyhow, this innitive e was not as such capable of any further suxation, apparently to avoid confusion with the stem of the aorist tense. The innitive of car- make, do would be car (cari-), but if we tried to add an ending like -s it die rectly to it in order to express to do it, the resulting form **caris would look just like the aorist it does or it makes. The actual form caritas is not ambiguous. In the case of they make vs. to make them, there would be a distinction even without the extra -ta-, since the subject ending for they (-nt) diers from the object ending them (-t). Even so, Tolkien appare ently decided to eliminate any possible confusion between aorist forms with subject endings and innitives with object endings: The innitives insert -ta- between the innitive proper and the pronominal suxes. Therefore, 158
the innitive to see is expanded from cen to cenita- when it is to receive e o any object ending. The thrall dared to see them must actually be i ml veryan cenitat, the extra -ta- intruding between the innitive and the e object ending. It is unclear whether A-stem verbs behave in the same way. Vinyar Tengwar #41 published only a very brief quote from Tolkiens 1969 notes (the editor apparently needed the space for more important things, like an indepth article about the optimal Bulgarian translation of the Ring Poem). The quote, reproduced above, apparently only deals with the innitive form of primary verbs the ones that have aorists in - or with endings -i-. Some e writers have assumed that A-stem verbs functioning as innitives would similarly add -ta before any object pronominal endings are suxed. So with verbs like metya- to end, to put an end to and mapa- to seize, it would work something like this: Merint metyatas they want to end it e I ohtari uvar mapatat the warriors will not seize them Perhaps such sentences are quite OK, perhaps not. Presently there is no way of telling. One may doubt that the ending -ta would be added to the stem of a verb that ends in -ta already, like orta- to lift up, to raise. Should I can lift it really be polin ortatas? Generally, Quenya is not very fond of two adjacent similar-sounding syllables, like the two tas here. Luckily, we can work around this uncertainty. We can simply avoid attaching object pronominal endings to the innitives of A-stem verbs, since we know at least some independent object pronouns (e.g. te them instead of the ending -t so for, say, you wanted to seize them we can have mernely mapa te instead of the uncertain construction ?mernely mae e patat). We will discuss the independent pronouns in a later lesson. In the exercises below, the innitives in -ita + object sux only involve primary verbs. It is interesting to notice that Tolkien translated caritas as doing it (VT41:17). This may suggest that such innitives can also function as the subject of a sentence, e.g. cenitas farya nin seeing it is enough for me (farya- verb to suce, to be enough; nin to/for me).
10.4
In Lesson Six, we set out some rules for regular past tense formation, but we also touched on various irregular forms (that is, past tense formations that dont readily fall into the most common patterns). Some of these may actually form sub-groups that are regular enough according to their own special rules. Let me rst introduce a couple of terms that will facilitate the following discussion: transitive and intransitive. In linguist terminology, a verb is said 159
to be transitive if it can have an object. Most verbs readily can, but not all. A verb like to fall is not transitive (= intransitive). The subject itself may fall, but the subject cannot fall something else; there can be no object. A typical intransitive verb only describes an action which the subject itself performs, not an action that is, or can be, done to someone or something. (I say typical, for Quenya actually has some verbs that cannot even have a subject, the so-called impersonal verbs to be discussed in Lesson Eighteen.) Some verbs form pairs where one verb is transitive, the other intransitive. The subject may raise an object (transitively), but the subject by itself can only rise (intransitively) not involving any object at all. Other examples of such pairs include transitive to fell vs. intransitive to fall, or transitive to lay vs. intransitive to lie. But in many cases, English uses the same verb form both intransitively and transitively, e.g. to sink. A subject may sink an object (e.g. the torpedo sank the ship, transitive verb with both subject and object), or the subject just sinks all by itself, so to speak (e.g. the ship sank, intransitive verb with subject only obviously sank is used with two quite dierent meanings here). Such ambiguity may also occur in Quenya; for instance, orta- covers both to raise and to rise, and the context must be taken into account to determine which meaning is relevant. (To be more concrete: check if the sentence includes an object or not! E.g. i aran orta = the king rises, but i aran orta ranco = the king raises an arm.) Let us then consider some irregular Quenya verbs. The verb farya- to suce, to be enough is said to have the past tense farn, irregular in the e sense that the ending -ya of the verbal stem drops out before the past tense ending -n: We might have expected **faryan, but the Etymologies lists e e a few more verbs that exemplify the same phenomenon: Vanya- to go, depart, disappear has the past tense vann. (Likely, Tolkien later replaced e the verb vanya- with auta- of similar meaning, but we may still consider it here.) To these examples from the Etymologies (see entries phar, wan) we may add a verb the student is supposed to have memorized as part of the previous lesson: lelya- to go, proceed, travel from WJ:363. Its past tense is not **lelyan, but lend, seemingly a quite irregular form (though not e e as wildly irregular as English to go vs. its past tense went!) The sudden appearance of the cluster nd is no big mystery; it arises by nasal-inxion of the original root led. (This root is listed in the Etymologies, though according to a later source, led is reworked from even more primitive del. Lelya- is meant to descend from primitive ledy - [ledj -], since dj became a a ly medially in Quenya [WJ:363]. The past tense lend would come from e lend, not so dissimilar from the verb ledy - as these forms later became.) e a The real mystery here is this: Why do the verbs farya-, vanya-, and lelyasurrender the ending -ya in the past tense? It may be noted that by their meaning, all three verbs are distinctly intransitive: To suce, to disappear, to go. This could be just a coincidence, 160
of course, but the Etymologies provides us with another highly interesting example. In the entry ulu, a verb ulya- to pour is listed. Tolkien indicated that it has a double past tense. If the verb is used in a transitive sense, as in the servant poured water into a cup, the past tense poured is ulyan. This would be an entirely regular form. However, if the verb e is used intransitively, the past tense of ulya- is ull instead (presumably e representing older unl, formed by nasal-inxion from ul- without the ending e -ya; cf. vill as the past tense of vil- to y, though in the latter case no e ending -ya appears in any form of the verb). So if you want to translate the river poured into a gorge, the form to use is ull, not ulyan. e e It seems, then, that we can discern a pattern here: Intransitive verbs in -ya drop this ending in the past tense; the past tense is formed from the ending-less root, as in the case of primary verbs. Or put dierently: In the past tense, intransitive verbs in -ya surrender this ending to masquerade as primary verbs. In the rare cases where a verb can be both transitive and intransitive, the ending -ya is retained when it is used in a transitive sense (as in the pa.t. form ulyan), but dropped when the verb is used in an e intransitive sense (ull). e Why this should be so is of course entirely obscure. In other tenses than the past, the verb ulya- to pour would seem to appear in the same form no matter whether it is transitive or intransitive (aorist ulya pours, present tense ulya is pouring, future ulyuva shall pour etc.) But it was never e Tolkiens intention to make a new Esperanto, a language aiming to be 100 % regular and logical. Within his mythos, Quenya is supposed to be an ordinary spoken tongue, developed over thousands of years. Hence, Tolkien may deliberately have included what you will nd in any natural language: certain features that dont necessarily make immediate sense. Most verbs in -ya are transitive, and would presumably retain their ending in the past tense, before the pa.t. sux -n is added (as in the attested e example ulyan). Here are most of the remaining intransitive verbs in -ya, e though Tolkien did not actually mention any past tense forms in their case: hwinya- to swirl, to gyrate (past tense hwinn?), mirilya- to glitter e (pa.t. mirill? cf. ulya-, pa.t. ull), ranya- to stray (pa.t. rann?), e e e s ya- to breathe (pa.t. s n?), tiuya- to swell, grow fat (pa.t. tiun?) u u e e The verb yerya- can be both transitive to wear (out) and intransitive to get old. Perhaps the past tense is yeryan in the former sense and e yern in the latter sense, just like we have transitive ulyan coexisting e e with intransitive ull as the past tense poured? e I should add that all of this is somewhat hypothetical, since Tolkien did not actually mention the past tense of very many intransitive verbs in -ya. But the student should at least notice the attested irregular past tenses, including the double pa.t. of ulya- to pour and especially lend went e as the rather unexpected past tense form of lelya- to go, travel, proceed.
NOTE: The perfect tense of this verb appears as lendi in some texts. SD:56 indicates e
161
that in one draft, Tolkien used lendien rather than ut lien for I have come in Elendils u Declaration (out of the Great Sea to Middle-earth have I come). Lendien would mean, literally, I have gone/went/traveled or something similar. This perfect form is not augmented, possibly simply because Tolkien had not yet invented the augment that is usually prexed in the perfect tense. I would normally supply it, using elendi as the perfect of e lelya-. I have used this perfect in (the key to) one of the exercises below.
10.5
Passive participles
Then we will return to the participles. The logical counterpart of the active participles discussed in the previous lesson is obviously the passive participles. They are often called past participles instead (just like the active participles are frequently referred to as present participles). However, the term passive participle is very tting. This participle is an adjectival form derived from the stem of a verb, and it describes the state that something or someone is left in by being exposed to the corresponding verbal action. For instance: If you hide something, it is hidden. Therefore, hidden is the passive participle of the verb to hide. The word hidden can be used as an adjective, both predicatively (the treasure is hidden) and attributively (hidden treasure). The passive participle hidden contrasts with the active participle hiding: The latter describes the state of the subject, the acting party, whereas the passive participle describes the state of the object, the one passively exposed to the verbal action. In the case of intransitive verbs, where no object can be involved, this participle describes the state of the subject itself after carrying out the verbal action in question: If you fall, you will thereafter be fallen; if you go, you will thereafter be gone. Here the often-used term past participle makes sense; participles like fallen or gone describe the condition of the subject after carrying out some past action. They are seen to contrast with the present participles (active participles) falling and going, which describe the condition of the subject while the verbal action is still present or on-going. But as long as we are dealing with transitive verbs and most verbs are transitive I still think it is better to speak of active participles vs. passive participles. In English, quite a few passive participles have the ending -en, as in the examples hidden, fallen above. But in very many cases, English passive participles are similar in form to the past tense of verbs, though the words have wildly dierent functions (a form like tormented is a past tense verb in a sentence like they tormented the Dwarf, but a passive participle in a sentence like the Dwarf was tormented). So what do the corresponding Quenya forms look like? The vast majority of Quenya participles seem to be formed by means of the ending -na or its longer variant -ina. Some attested A-stem participles
162
are seen to include the longer ending, the nal -a of the verbal stem and the i of the sux -ina merging into a diphthong -ai- (which receives the stress, like any diphthong in the second-to-last syllable). An example is provided by the phrase Arda Hastaina, Arda Marred, an Elvish term for the world as it is, tainted by the evil of Morgoth (MR:254). This hastaina marred would seem to be the passive participle of a verb hasta- to mar, not otherwise attested. However, the verb hosta- to gather, collect, assemble is attested both in the Etymologies (entry khoth) and in the Markirya poem (MC:222223). Its passive participle turns up in F riels Song, where it is implied to be hostaina (attested in the form hostainiva will be e gathered; the sux -iva will be is hardly valid in LotR-style Quenya, e but the underlying participle certainly is). We can probably conclude that A-stems in -ta nearly always have passive participles in -taina. Since antameans give, the participle given would be antaina. Since orta- means raise (or used intransitively, rise), the word for raised (and risen) would seem to be ortaina. Perhaps the ending -ina can be added to nearly all A-stems? From a verb like mapa- grasp, seize, I think we may well derive mapaina as the participle grasped, seized. (Indirect support for this: The ending ina is also used to derive adjectives, as in valaina divine obviously an adjectival formation based on Vala, which noun is analogous in form to a simple A-stem like mapa-. Indeed it is hinted that the noun Vala is originally derived from a simple A-stem verb vala- to order, to have power: WJ:4034. If it had remained a verb only, valaina could have meant ordered instead.) The behaviour of A-stems in -ya is slightly obscure. In the Etymologies, Tolkien listed a root per divide in middle, halve (cf. Sindarin Perian haling, Hobbit). He then mentioned a Quenya word perya, evidently a verb preserving the root meaning. Immediately after perya, he listed an undened word perina. Is this the passive participle halved? I think this is almost certainly the meaning of this word, but perhaps we should see it as an independent adjectival formation derived directly from the root, not as the passive participle of the verb perya-. (We might have excepted prina e with a long if it were a passive participle; see below regarding the rcina e a pattern.) Elsewhere in the Etymologies, in the entry gyer, we have a verb yeryato wear (out), get old. The same entry also mentions a word yerna worn. As far as the English glosses are concerned, yerna could be the passive participle of the verb yerya-. Should we conclude, then, that verbs in -ya form their passive participles by replacing this ending with -na? Again I think yerna is not actually the participle of yerya-, but rather an independent adjectival formation. The following facts support this: 1) Tolkien traced yerna all the way back to Primitive Elvish gyern, so it was not derived from the a verb later; 2) Tolkien actually listed the form yerna before he mentioned 163
the verb yerya-, again suggesting that the former is not to be derived from the latter, 3) yerna is glossed old as well as worn, and the rst gloss suggests that yerna is to be considered an independent adjective, not a participle. Same story as with perina above, then. This would also go for a pair like halya- veil vs. halda veiled, hidden (entry skal1 ): The latter form Tolkien referred to Primitive Elvish skaln (initial sk - becoming h- and ln a becoming ld in Quenya). It may well be that in Primitive Elvish, skaln did a count as the passive participle of the verbal root skal- screen, hide, but its Quenya descendant halda has developed into an independent adjective (one of Tolkiens glosses for this word, shady, is also an adjective). Thus, halda is not necessarily the passive participle of the verb halya- derived from the same root, though it has somewhat the same meaning as the actual participle would have. So how, really, are we to treat verbs in -ya? I think a highly interesting clue is provided in MR:326 (cf. MR:315), where Christopher Tolkien tells us that in a post-LotR text, Tolkien used Mirruyainar or Mirroyainar for the Incarnate (plural). This may seem to be passive participles inected as nouns: incarnated ones. Removing the plural ending -r, we are left with mirruyaina/mirroyaina as a possible participle incarnated and if we peel away the presumed participial ending as well, the verb to incarnate would seem to be mirruya- or mirroya-. Tolkien later changed the word Mirruyainar/Mirroyainar to Mirroanwi, not involving any -ya- at all, but the rejected forms may still give away what the passive participle of a verb in -ya should look like. Such verbs would seem to have participles in -yaina, just like verbs in -ta have participles in -taina. So given that lanya- is the verb to weave, the word for woven may well be lanyaina. The regular passive participles of the verbs perya- to halve, yerya- to wear (out) and halya- to veil would similarly be peryaina, yeryaina, halyaina (meaning much the same as the related adjectives perina, yerna, halda, of course, but the latter may not so clearly imply that the described states are inicted see below regarding harna- vs. harnaina). We can probably conclude that nearly all A-stem verbs form their passive participles by adding -ina. The only exception occurring in the published corpus is the form envinyanta healed or more literally renewed (MR:405). It would seem to be the passive participle of a verb envinyata- renew (not attested by itself, but cf. Aragorns title Envinyatar Renewer). This participle is formed by means of nasal-inxion intruding before the ending -ta. We cannot know whether the more regular formation envinyataina, itself unattested, would be a valid form. However, the ending -ina is not only used in the case of A-stems; primary verbs with c or t as their nal consonant also form their passive participles by means of this ending. The Markirya poem includes a form rcina broken (man tiruva rcina cirya[?] who shall see [/watch] a a a broken ship?, MR:222). Tolkien explicitly identied rcina as the passive a 164
(or past) participle of the verb rac- to break (MC:223). The verb to reckon, to count is not-, and in F riels Song we have ntina as the passive o participle counted. It seems, then, that primary verbs ending in unvoiced stops like c or t form their passive participles by lengthening the stem-vowel and adding the long ending -ina. We dont seem to have any attested example of the participle of a primary verb ending in -p (another unvoiced stop), but it would in all likelihood slip into the same pattern: The verb top- to cover would have the passive participle tpina covered. (The verb topo is listed in the Etymologies; the poem Namri in LotR may suggest that a e Tolkien later changed it to tup-. If so, the participle would of course be t pina instead.) Perhaps primary verbs in -v also form their passive paru ticiples according to this pattern, e.g. lvina allowed, granted from the a verb lav- allow, grant (not to be confused with a similar-sounding verb meaning lick). We lack examples, though. Attested examples dont exactly abound for other primary verbs either, but most of them probably prefer the short ending -na to -ina. MR:408 (cf. MR:405) indicates that Tolkien used vincarna for healed; the more literal meaning is transparently renewed or wholly literally newly-made: Vinis the stem of the Quenya adjective vinya new, and carna made can only be the passive participle of the verb car- make. So primary verbs ending in -r have passive participles in -rna (and because of the consonant cluster here arising, the stem-vowel preceding it obviously can not be lengthened as in the rcina class discussed above). Given that mer- is the Quenya a verb to want, the Wanted posters of the Quenya Wild West would evidently read Merna. Perhaps mrina, crina (following rcina) would be e a a possible alternative passive participles of mer-, car-, perhaps not. I think it is best to let the attested example carna guide us here. For primary verbs in -m and -n, we only have what may be called indirect examples of their passive participles, but they are probably good enough. The verb nam- to judge (namin I judge, VT41:13) seems to have the passive participle namna. This form is attested as a noun meaning statute (as in Namna Finw M e riello, the Statute of Finw e and M riel, MR:258). Apparently the participle namna, basically meaning judged, is also used as a noun judgement, juridical decision and then statute. As for primary verbs in -n, we may consider such nouns as anna gift and onna creature vs. the verbs anta- to give and onta- to create (see the entries ana1 , ono in Etym). These are not primary verbs, of course (and in Quenya we would expect them to have the participles antaina, ontaina) but the nouns anna, onna may descend from primitive participial formations based on the naked root-word, before -ta was added to produce the verbs as they appear in Quenya. So anna may come from a primitive participle given, only later used as a noun something that is given = gift. Onna might likewise represent an original passive participle created, later used as a noun created one = creature. I tend to 165
think, therefore, that the ending -na can be added to the stems of Quenya primary verbs ending in -n. For instance, since cen- is the verb to see, cenna may well be the passive participle seen. But again, cnina may be e a permissible alternative formation (perhaps we can also have nmina for a judged, for all I know). What about primary verbs in -l, such as mel- love? If we dont resort to the pattern of rcina once again, using mlina for loved, the ending a e -na would have to be added directly to the verbal stem. But since **melna is not a possible Quenya word, ln would become ld, just as in one example discussed above (Quenya halda descending from Primitive Elvish skaln ). a The Etymologies actually lists a word melda, glossed beloved, dear. These glosses are adjectives, but by their meaning they are of course very close to the participle loved. So are we once again looking at an original participle that has developed into an independent adjective? Would the actual participle of mel- dier in form, precisely to distinguish it from this adjective? If so we might consider mlina again. Or is melda both the adjective dear e and the participle loved? One may well ask whether there is any point in even trying to distinguish between them, since their meanings would be virtually the same. Another example may also be considered: The Quenya verb to bear, to wear, to carry seems to be col-, though it has never been independently attested: Only various derivations are found in our corpus. One of them appears in MR:385: colla = borne, worn (also used as a noun vestment, cloak, considered as something that is worn). Is this an example of the past participle of a primary verb ending in -l? Can we use mella for loved, then? I tend to think that colla is rather an adjectival derivative perhaps representing primitive konl with nasal-inxion of the root kol (not a in Etym). By its original derivation it would then parallel such a Quenya adjective as panta open (which Tolkien referred to Primitive Elvish pant, a derived from the root pat listed in Etym). Im afraid no quite certain conclusion can be reached regarding the passive participles of primary verbs in -l, but I think the safest would be to either use the ending -da (representing earlier -na), or the longer ending -ina combined with lengthening of the stem-vowel. Should passive participles agree in number, like normal adjectives do? In other words, should the nal -a turn into - (for older -ai) if the participle e describes a plural noun? As far as I can see, the corpus provides no example that could guide us. We recall that active participles (ending -la) do not agree in number. However, I tend to think that passive participles do behave like normal adjectives in this regard. We have just seen that in many cases it is dicult to even determine whether a form is to be considered a passive participle or an adjective, since adjectives may be derived with the same endings. (For that matter, this goes for English as well: An adjective like naked could well have been a passive participle by its form; however, 166
there is no corresponding verb **nake denude, so we cant set up a pair **nake/naked like we have love/loved.) Since adjectives like valaina divine and yerna old must be assumed to agree in number, it is dicult to imagine that participles like hastaina marred or carna made would not show such agreement. So I would change the nal -a to - where the e participle describes a plural noun (or several nouns). In English, past/passive participles are used as part of the circumlocutions which simulate the function of a true perfect tense: The Dwarf has seen the Elf; the woman is (or, has) fallen. But here Quenya would simply use the real perfect tense instead: I Nauco ecni i Elda; i n alanti. e e s e Perhaps n lantaina is also permissible for is fallen, but rendering the a Dwarf has seen the Elf as **i Nauco harya cenna i Elda (copying the English wording directly) only results in nonsense. A nal note: In some cases, forms in -na that were originally participial or adjectival have themselves become A-stem verbs. The primitive word skarn, a listed in the entry skar in the Etymologies, was perhaps originally a passive participle torn, rent (since the root skar itself is said to mean tear, rend). In Quenya, skarn turned into harna wounded, probably felt to a be an adjective rather than a participle. The funny thing is that harna- also came to be used as a verb to wound, and if this verb has its own passive participle harnaina, we would have come full circle! In English, both harna and harnaina must be translated wounded, but whereas harna would merely describe the state of being wounded, harnaina clearly implies that the wounds were inicted. Cf. the English adjective full (merely describing a state) vs. the passive participle lled (implying that the state in question results from the act of lling).
167
someone or something by the corresponding verbal action: what you hide (verb) becomes hidden (passive participle). A-stem verbs seem to form their passive participles in -ina (e.g. hastaina marred from hasta- to mar). This ending is also used in the case of primary verbs ending in -t and -c, probably also -p and possibly even -v; in this class of verbs, the ending is combined with lengthening of the stem-vowel (e.g. rcina broken from a rac- to break). It may be that the same pattern can be applied to all primary verbs, but verbs in -r are seen to take the simple ending -na instead, with no lengthening of the stem-vowel (carna made from car- to make). Primary verbs in -m, and probably also -n, would similarly take the simple ending -na (e.g. namna judged from nam- to judge, cenna seen from cen- to see). It is somewhat uncertain how we should treat primary verbs in -l; if we are to use the simple ending -na, it would turn into -da for phonological reasons (e.g. melna > melda loved as the passive participle of mel- to love; melda is attested as an adjective beloved, dear). Passive participles probably agree in number in the same way as adjectives, changing -a to - if they describe a plural noun or several nouns. e
Vocabulary
nert nine e n ra deep u anwa real, actual, true nulda secret telda nal (adjective derived from the same root as the name of the Teleri, the Third
Clan of the Eldar, so called because they were always the last or hindmost during the March from Cuivinen far behind the Vanyar and the Noldor, who were more e eager to reach the Blessed Realm)
linta swift (pl. lint in Namri, which poem refers to lint yuldar = swift e a e e
draughts)
hosta- to assemble, gather nr land (a land associated with a particular people,WJ:413) o e lamb tongue = language (not tongue as a body part) e car- to make, to do farya- to suce, to be enough, pa.t. farn (NOT **faryan because the e e
verb is intransitive?)
168
Exercises
1. Translate into English: A. Melinyet n rav. u e B. Lindant vanyav, ve Eldar. e e C. Ily nert andor nar tirn. e e e D. Merint hiritas lintav. e e E. Haryaly atta parmar, ar teldav ecendielyet. e e F. Anwav ecnien Elda. e e G. I nurtaina harma uva hirna. H. Ument mer caritas, an cenitas farn. e e e 2. Translate into Quenya: I. They have traveled [/gone] secretly through the land. J. The assembled Elves wanted to see it. K. Written language is not like spoken language. L. Five ships were not enough [/did not suce]; nine suced. M. I will really stop doing it [/truly cease to do it]. N. They swiftly gathered the nine terried Dwarves. O. Finally you will see them as you have wanted to see them. P. They dont want to hear it.
169
Lesson 11
The concept of cases. The Genitive case.
11.1
Cases
Lessons 110 have mainly been concerned with adjectives and verbs. As for nouns, we have only discussed how their plural and dual forms are constructed. There is, however, much more to say about the inection the Quenya noun. The second half of this course will predominantly be concerned with the elaborate case system of Quenya, which is indeed the most characteristic feature of the language. It is in the treatment of nouns that the grammatical structure of Quenya most clearly reects two of Tolkiens inspirations, Finnish and Latin. What, linguistically speaking, are cases? A noun may have many functions in a sentence. English may indicate what function a noun has by means of word order alone. In a sentence like the man loves the woman, it is merely the word order that gives away the fact that the man is the subject and the woman is the object. The rule that very early slips into the subconscious mind of children exposed to English goes something like this: The noun in front of the predicate verb is its subject, while the noun that comes after it is normally its object. Where word order is not enough, English may slip in clarifying prepositions in front of a noun, e.g. to in a sentence like the Elf gives a gift to the Dwarf. There are languages that wouldnt need to have a to here; instead the noun Dwarf would occur in a special, inectional form. Of course, Quenya also has prepositions, and the student will already have encountered several: nu under, or over, imb between, ve as, like, e mir into (which word, by the way, is formed from the simpler preposition mi in). But it is a characteristic of Quenya that where English would often place a preposition in front of a noun, or rely on word order alone to indicate what the function of a noun is, Quenya would have a special form of the noun which by itself indicates its function. These various, specialized noun-forms are called cases. For instance, our example above the Elf gives a gift to the Dwarf would translate into Quenya something like i Elda 170
anta anna i Naucon, where the case ending -n added to Nauco Dwarf corresponds to the English preposition to. (This particular case is called the dative, to be fully discussed in Lesson 13.) Certain prepositions may also demand that the word (noun or pronoun) following them appears inected for some case sometimes quite irrespective of the normal, independent function of this case. The relevant preposition is then said to take (or govern) this or that case. The same phenomenon may be found in English, if one looks closely. While the case system is all but gone as far as English nouns are concerned, many English pronouns at least preserve a specic form that is used when the pronoun is the object and not the subject of a sentence. That is why Peter saw he is wrong; it must be Peter saw him, with the object form of this pronoun. (He is the subject form instead, and therefore quite proper in a sentence like he saw Peter.) But while the primary function of the form him is to function as the object of a sentence, very many prepositions also insist on being followed by this form. For instance, from he does not sound well; it must be from him, though him is not the object of a sentence here. The Quenya noun-forms so far discussed (whether singular, plural or dual) are examples of the nominative case. The most important grammatical function of the nominative is that this is the form a noun has when it functions as the subject of a verb. In Lesson Five, we very briey touched on another form of the noun the accusative case, which is the form a noun assumes when it is the object of a verb. Modern English does not preserve any distinction between nominative and accusative in nouns (though such a distinction persists in parts of the pronoun table, like nominative he vs. accusative him in our examples above). English nouns do not change their form dependent on whether the noun is the subject or the object of the sentence and neither do nouns in Third Age Quenya. Tolkien imagined an archaic form of Quenya, Book Quenya, that did have an accusative case distinct in form from the nominative. The noun ship would be cirya (pl. ciryar) if it was used as the subject of a sentence, but ciry (pl. ciryai) if it appeared a as the object: nominative vs. accusative. However, the distinct accusative disappeared from the language as spoken in Middle-earth; the forms cirya (pl. ciryar) came to be used both as subject and object. So either you can say that in Third Age Quenya, the nominative and the accusative cases have come to be identical in form, or you can say that the nominative has taken over the functions of the distinct accusative so that in eect, there is no accusative anymore. It boils down to exactly the same thing. But as far as we know, the accusative was the only Quenya case that was lost among the Exiles. The remaining cases, in addition to the nominative, are the genitive, the possessive, the dative, the allative, the ablative, the locative, and the instrumental. (I should add that learning the form and function of the cases is more important than learning their Latin names.) There is also a mysterious case which Tolkien listed in the Plotz Letter, but 171
without discussing its name or use so there is little I can say about it here. In Lessons 1116, we will work our way through the list of Quenya cases, discussing their functions and how they are formed. Precisely because we have the blessed Plotz Letter, we are now on somewhat more solid ground than we usually nd ourselves upon when discussing Quenya grammar. (Tolkien really should have sent Dick Plotz a list of pronouns and verb forms as well!)
11.2
The Genitive
We will start our discussion of the Quenya cases with the few Quenya noun forms that actually have a direct English equivalent (sort of). Where Quenya has nine or ten noun cases, English has only two: nominative and genitive. The nominative we have already discussed: In English as in Exilic Quenya, a noun appears in the nominative when it is the subject or the object of a noun. In both languages, the nominative singular may well be considered the simplest form of the noun. There is no special ending or other inectional element to signal that this is a nominative form; rather it is the absence of any such element that tells us what case the noun is. All the other cases or in English, the one other case do display special endings, though. The one noun case of modern English, except for the nominative, is the genitive. (Please notice the spelling; I am tired of seeing genetive on certain mailing lists.) In the singular, it is formed by adding the ending s to the noun, e.g. girls from girl. In the plural, this ending normally merges with the plural ending -s, but its theoretical presence is hinted at in writing by means of an apostrophe (girls for **girlss. . . Gollum would have loved the latter form). The grammatical function of this case ought to be familiar enough to anyone who is capable of reading this text; already in Lesson Two, we briey touched on this ownership form. As stated in my handy Oxford Advanced Learners Dictionary of Current English, the genitive case is used to indicate source or possession. In a combination like the girls doll, the genitive case is used to coordinate two nouns so as to indicate that the former is the owner or possessor of the latter. (This latter word which the genitive form connects with, like doll in our example, is sometimes said to be governed by the genitive. Conversely, the genitive form itself may be said to be dependent on this other word; this is Tolkiens wording in UT:317.) The English genitive does not necessarily connote ownership in the strictest sense, but may also be used to describe other kinds of belonging, such as family relationships e.g. the girls mother. As for the genitive suggesting source, we can think of such phrases as the architects drawings (the drawings made by the architect, not necessarily owned by him, but originating with him). The genitive noun may not even denote a sentient being, e.g. Britains
172
nest artists (the nest artists coming from/living in Britain). The latter example may also be termed genitive of location; Britains nest artists are the nest artists located in Britain. The noun a genitive form is dependent on may well be another genitive, which in turn refers to a third noun e.g. the queens sisters house. In principle we can string up an innite number of genitives (the kings fathers aunts brothers dogs. . . [etc. etc.]) though it should not come as a great shock to anyone that people who care about style and legibility normally wont push this too far. Somewhat like adjectives, genitives can be used both attributively and as predicates. All the examples above are examples of attributive genitives, directly teamed up with a noun which the genitive is then dependent on. A genitive would however function as a predicate in a sentence like the book is Peters. But rather than using genitives as predicates, English often resorts to circumlocutions (like the book belongs to Peter ). Quite often, English does not use a genitive, but instead employs a phrase involving a preposition predominantly of, e.g. the nest artists of Britain rather than Britains nest artists. In some contexts, of-constructions are actually preferred to genitives, e.g. the end of the road rather than the roads end. So what about Quenya? The functions of English genitives, as well as English of-constructions, are covered by two Quenya noun cases; we will discuss the other relevant case in the next lesson. The functions of the case normally referred to as the Quenya genitive are somewhat more limited than the functions of the English genitive. But rst of all, let us discuss how the Quenya genitive is formed. The basic Quenya genitive ending is -o. Starting from nouns that should be well known to the student by now, we can derive genitives like arano kings, trio queens, vendo maidens. If the noun ends in -o ala e ready, the genitive ending normally becomes invisible. In UT:8 we have ciryamo for mariners. This is our sole attestation of this noun, but there is no reason to doubt that its nominative form mariner is likewise ciryamo (this word is obviously derived from cirya ship, and the masculine/personal ending -mo [WJ:400] is well attested elsewhere: hence cirya-mo = ship-person). A name like Ulmo could be both nominative Ulmo and genitive Ulmos; the context must decide how the form is to be understood. (However, in the case of nouns in -o that have special stem-forms in -u, like curo, curu- skillful device, we would probably see curuo as the genitive form.) Nouns ending in -a lose this vowel when the genitive ending -o is added: Since Quenya phonology does not permit the combination ao, it is simplied to o. For instance, Namri demonstrates that the genitive Vardas is a e Vardo, not **Vardao. It follows, then, that a few otherwise distinct nouns coincide in the genitive; for instance, it would seem that anta face and 173
anto mouth both have the genitive form anto. The context must be taken into account to determine which noun is meant. In the plural, the genitive ending -o is expanded to -on (as we shall see later, the plural marker -n occurs in several of the Quenya case endings). This ending -on is added to the simplest (nominative) plural form of the noun, in -r or -i. Hence an r-plural like aldar trees has the genitive plural aldaron trees, of trees whereas an i-plural like eleni stars has the genitive form elenion stars, of stars. (The normal stress rules still apply, so while eleni is accented on the rst syllable, the stress must fall on -len- in the longer form elenion.) Both of these are attested in LotR: Namri has rmar aldaron for wings of trees (a poetic circumlocution a e a for leaves), and Frodo speaking in tongues in Cirith Ungol referred to Erendil as ancalima elenion, brightest of stars. a A prominent example of a genitive plural is the very title of the Silmarillion, formed from the nominative plural Silmarilli Silmarils. This title makes good sense considering that it is properly only one half of a longer genitive phrase, found on the title page following the Ainulindal and the e Valaquenta: Quenta Silmarillion, The History of the Silmarils. As is already evident, a Quenya genitive is often best rendered as an English of construction, not as an English genitive with the ending -s or -s : Stars brightest or the Silmarils History would not be good English. As for dual genitive, Tolkien indicated that its ending is -to, combining the dual ending -t with the basic genitive ending -o. In the Plotz letter, Tolkien used the example ciryato, of a couple of ships. There is one uncertainty here, not addressed in Plotz: Should the ending be -to also in the case of the nouns that have dual forms in -u rather than -t? Or would the u simply replace t here, so that such nouns have dual genitives in -uo instead? Concretely: if the nominative (the) Two Trees is Aldu, should the genitive of (the) Two Trees be Alduto or Alduo? A form like Alduto would have a double dual marker, both u and t, but then attested plural genitives likewise include double plural markers (elenion, aldaron). Even so, I am not ready to rule out the possibility that genitives in -u should have genitives in -uo, e.g. i cala Alduo for the light of (the) Two Trees. But since published material allows no certain conclusions in this matter, I have simply avoided the problem in the exercises below. The special stem forms of some nouns are relevant for the formation of genitives as well. From r (rv-) lion we would have the genitive rvo a a a lions; from n (niss-) woman we would have nisso womans. The s plural forms would be rvion lions, of lions and nission womens, of a women cf. the nominative plurals rvi, nissi. I am not quite sure about a the dual forms; perhaps we can have rveto, nisseto (an -e- intruding a before the ending -to so that impossible consonant clusters do not arise; see later lessons regarding attested examples of an extra -e- being slipped in like this). 174
So far the formation of the genitive; now we must return to its function. In English, the genitive very often indicates who owns what, as in the mans house. Indeed this is the main function of the English genitive. However, the Quenya genitive case is not normally used to describe simple ownership of things. Tolkien expressly noted that properly this case was not [used] as a possessive, or adjectivally to describe qualities (WJ:368). To understand its function it is often useful to bear in mind its ultimate derivation. Tolkien explained that the source of the most used genitive inection of Quenya was an ancient adverbial or prepositional element basically meaning from or from among. According to WJ:368, it originally had the form ho, or as an element added to nouns, -h. The latter was the o direct source of the Quenya case ending -o (plural -on). But according to the Etymologies, Quenya also had a regular preposition ho from, and in WJ:368 Tolkien mentions h- from, o as a verbal prex, e.g. in htulio o come away or literally from-come. Even the case ending -o may occasionally express from, the most basic meaning of the primitive element ho. In the prose Namri, we have the a e line Varda . . . ortan mryat Oiolosso, Varda . . . raised her hands e a e from Oioloss (essentially the same in the version in LotR, but with a more e complicated, poetic word order). The translation in LotR goes Varda . . . from Mount Everwhite has uplifted her hands Oioloss Ever-white e being a name of Taniquetil, the great mountain of the Blessed Realm where Manw and Varda dwell. e However, Oiolosso is our sole example of the Quenya genitive being e used with such a meaning. (For from, Quenya regularly uses another case the ablative, to be discussed in a later lesson.) Normally, the ending -o is seen to have acquired other, more abstract meanings. Nonetheless, one important function of the Quenya genitive still clearly reect the idea of something coming from something or someone else: The Quenya genitive can be used to describe the source, origin or former possessor of something so-called derivative genitives (WJ:369). Tolkien explained that rma Oromo Oroms horn refers to a horn coming from Orom, not a o e e e horn that Orom still has, or still had at the time that is being considered e (WJ:168). Likewise, lamb Eldaron could not be used for the language e of the Eldar, for this would mean the language coming from the Eldar; Tolkien added that such a wording would only be valid in a case where the whole language was adopted by another people (WJ:368369). In light of this, the genitive phrase Vardo tellumar Vardas domes in Namri a e may not necessarily imply that the heavenly domes were somehow owned by Varda, but rather that she made them, that they originated with her. Tolkien also listed from among as one of the meanings of the primitive element ho, and this meaning is discernable in Quenya examples of partitive genitive, the genitive indicating what something or someone is part of. In the phrase Erendil Elenion Ancalima Erendil brightest of stars a a 175
(Letters:385), the words elenion ancalima actually imply the brightest one among the stars: After his mythical transformation, Erendil carrying a the Silmaril is himself one of the stars, as indicated by the chapter The Mirror of Galadriel in Volume One of LotR (Erendil, the Evening Star, most a beloved of the Elves, shone clear above. . . ) It seems that a partitive genitive can denote what something is part of in a wholly physical sense as well: In a phrase translated the hands of the Powers, F riels Song uses the plural genitive Valion for of the Powers (sc. of the Valar as indicated by the Etymologies, entry bal, Vali is a valid alternative to Valar as the plural form of Vala). The hands of the Valar, whenever they are incarnated, are physically part of the Valar themselves. The relationship between a place and something located in that place can also be expressed by means of the genitive case (cf. our own example Britains nest artists). Namri has Calaciryo m for Calaciryas a e ri jewels = the jewels of Calacirya (Calacirya Light-cleft being a place in the Blessed Realm; notice that as in the case of Vardo Vardas, the genitive ending -o swallows up the nal -a). Perhaps this can also be analyzed as a partitive genitive, if something located in a place is somehow considered a part of that place. A more abstract, but perhaps basically similar construction is found in C rions Oath: Elennanro alcar the glory of the land o e of Elenna or literally (the) Elenna-lands glory. If we dont perceive the alcar or glory as being somehow located in Elenna (= Nmenor), we must u think of it as emanating from Elenna, so that the genitive denotes source. (See the next lesson concerning the comparable case alcar Oromo.) e Family relationships are denoted by the genitive case. In Treebeards Greeting to Celeborn and Galadriel occurs the genitive phrase vanimlion a nostari, parents of beautiful children (Letters:308) or more literally begetters of fair ones (SD:73) vanimli meaning fair ones (genitive pl. a vanimlion) and nostari meaning begetters. One could also argue that a this example shows that a noun denoting some kind of agent, and another noun denoting the one whom this agent does something to, can be coordinated by means of the genitive case (the fair ones were begotten by the begetters). Whatever the case, we have other examples of family relationships described by means of a genitive. In the Silmarillion Index, entry Children of Ilvatar, we learn that this is a translation of H Il vataro. u ni u Since Ilvatar (All-father) is a title of God, this example is somewhat prou found, but the genitive case would certainly also be used in such phrases as the kings sons (probably i arano yondor). As long as the genitive case describes parents relationship to their ospring, we could analyze the constructions as derivative genitives, parents being the physical origin of their children. But in the example Indis iCiryamo the Mariners Wife (UT:8), the genitive unquestionably describes a family relationship and nothing else, since the Mariner is not in any way the source or origin of his wife. 176
Perhaps we can generalize even further and say that relationships between people can be described by the Quenya genitive case. In WJ:369, Tolkien indicated that the genitive would be used in such a phrase as Elw, Aran e Sindaron Elw [= Thingol], King of the Sindar [Grey-elves]. Here the e relationship is that between a ruler and the ruled. The same construction could however be used with reference to the area that is ruled: King of Lestanr would be Aran Lestanro (Lestanr being the Quenya name oe o e oe of the land called Doriath in Sindarin). The genitive case may also refer to things that are ruled: In a booklet which accompanied an exhibition at the Marquette University Archives in September 1983, Catalogue of an Exhibit of the Manuscripts of JRRT, Taum Santoski presented Tolkiens Quenya translation of the title Lord of the Rings: Heru i Million, which is heru lord + i the + what is probably the plural genitive of a noun mill e ring, not otherwise attested. In the LotR itself, the Quenya word for ring is given as corma instead, Frodo and Sam being hailed as Cormacolindor or Ring-bearers (this word occurring in the Cormallen Praise). For Lord of the Rings we might therefore have expected Heru i Cormaron, but anyhow, the phrase Heru i Million conrms that the genitive case can be used to describe the relationship between a ruler and the ruled (people, area or thing). One of the most abstract meanings the genitive case may take on is of = about, concerning, as in Quenta Silmarillion the History of (= concerning) the Silmarils. Another attested example is quental Noldoron the e history of the Noldor (VT39:16). It may well be that the genitive can be used in this sense also in connection with verbs like nyar- tell, relate or quet- speak, e.g. nyarnen i Eldo I told about the Elf or i Naucor quetir alt harmaron the Dwarves speak of great treasures. We lack e attested examples, though. Sometimes the precise meaning of a genitive is dicult to clearly dene. In the famous greeting elen s l menn omentielvo, a star shines upon la u the hour of our meeting or literally . . . our meetings hour, the genitive simply coordinates the nouns meeting and hour to indicate that the meeting took place in the hour. In the phrase Heren Istarion Order of Wizards (UT:388), one may ask whether the genitive Istarion of Wizards implies that the order was founded by wizards, that it belongs to wizards, that it is made up of wizards, that it organizes or controls (or even is controlled by) wizards, etc. In all likelihood, several or all of these shades of meaning could be involved at the same time. Also consider this passage from LotR, in the chapter The Houses of Healing in the third volume: Thereupon the herb-master entered. Your lordship asked for kingsfoil, as the rustics name it, he said, or athelas in the noble tongue, or to those who know somewhat of the Valinorean. . . 177
I do so, said Aragorn, and I care not whether you say now asa e aranion or kingsfoil, so long as you have some. So asa aranion is the Quenya (or Valinorean) for kingsfoil, the herb e called athelas in Sindarin. The word asa refers to some kind of helpful or e benecial plant, but what precise meaning does the genitive plural aranion of kings express here? The kings didnt own or originate the kingsfoil; it was merely used by them for healing purposes. Unless this is comparable to a Calaciryo m ri-construction because the kingsfoil was physically with the kings when they used it for healing (life to the dying / In the kings hand lying!), we must conclude that the genitive can also be used to indicate rather ill-dened states of belonging, or mere association. Word order: In the prose version of Namri, Tolkien placed a genitive a e in front of the noun it is dependent on: Aldaron lassi = literally trees wings, maryo l o rinen = literally in her voices song, Calaciryo m ri = literally Calaciryas jewels cf. the interlinear translation in RGEO:66 67. (It should be noted that aldaron lassi was altered from lassi aldaron in the poetic version in LotR.) Above the entire prose version, Tolkien also placed the superscript Altariello naini, Altariels (= Gale adriels) lament. Cirions Oath displays the same word order: Nro alcar o e the glory of the land or literally (the) lands glory, Elendil vorondo voronw the faith of Elendil the Faithful or literally Elendil (the) Faithe fuls faith (the genitive ending being attached to the last word in the phrase Elendil voronda E. [the] Faithful; as usual, the ending displaces a nal -a). In LotR we also have elenion ancalima for stars brightest [one] = the brightest [one] of [the] stars. So in normal prose, should the genitive always precede, just like the English genitive in s? Not necessarily, it would seem. Most attested Quenya genitives follow the noun they are dependent on, with the same word order as an English of construction. In the case of most of these attestations, we have no reason to suppose the word order is particularly poetic: Quenta Silmarillion History of the Silmarils, Heru i Million Lord of the Rings, l menn u e e omentielvo on the hour of our meeting, asa aranion asa [helpful plant] of kings (kingsfoil; the two latter examples are from LotR), H ni Il vataro Children of Ilvatar (Silmarillion Index), mannar Valion u u into the hands of the Powers (F riels Song), Heren Istarion Order of Wizards (UT:388), Pelri Valion Fencing Heights of the Vali [Valar] o (MR:18), aran Sindaron King of the Sindar (WJ:369), Aran Lestau u nro King of Doriath (ibid.), i equessi R milo the sayings of Rmil o e (WJ:398), lamb Eldaron or lamb Quendion the language of the e e Elves (WJ:368/PM:395), Rithil-Anamo Ring of Doom (WJ:401). Where Tolkien rendered such a construction by employing an English genitive in -s, he must of necessity reverse the Quenya word order: Indis iCiryamo The Mariners Wife (UT:8). 178
One potential misunderstanding may be mentioned here: Occasionally people are seen to be completely seduced by English of -constructions, thinking that the genitive ending -o should appear at the same place in the phrase as the English preposition of does. Therefore they end up attaching the genitive ending to the wrong word in an hopeless attempt to copy the English order of all the elements in the phrase. Ask ten people to translate the glory of Aman into a Quenya genitive phrase, and it is a good bet that several of them will come up with something like i alcaro Aman, which actually means the glorys Aman or Aman of the glory! What we want is either Amano alcar (think Amans glory) or (i) alcar Amano. As for the word order employed when a preposition is used in conjunction with a genitive phrase, the prose Namri provides the strange example a e Vardo nu luini tellumar. Tolkien translated this as under Vardas blue domes. As we see, the Quenya wording is literally Vardas under blue domes, the preposition following the genitive noun a most unexpected order, especially considering that this is supposed to be normal prose. Yet the prose Namri also has ve aldaron rmar for like the wings of trees (or a e a literally like trees wings). Here the word order is exactly what we would expect, namely preposition + genitive + the noun it governs (not **aldaron ve rmar or whatever!) It is almost tempting to assume that Vardo nu a luini tellumar is simply an error for ?nu Vardo luini tellumar. At this stage at least, I would always use the English-style word-order exemplied by ve aldaron rmar. Perhaps Vardo nu luini tellumar is an example a of the exceedingly esoteric syntax preferred by the Eldar, whose thoughts are not like those of Mortal Men. . . or perhaps it is just a typo. We must await the publication of more material. The use of the article: A genitive determines the noun it is dependent upon, just like the denite article does: Indis iCiryamo does mean the Mariners Wife = the Wife of the Mariner. It cannot be interpreted a wife of the mariner in an indenite or undetermined sense, even though the denite article i is missing before the noun indis wife, bride. Same with lamb Quendion the language of the Elves (PM:395, emphasis e added); this cannot be interpreted a language of the Elves, for lamb is e determined by the genitive Quendion. Cf. English the Elves language = the [not a] language of the Elves, even though there is no the before language in a phrase like the Elves language. One must understand that while the rst noun of an English of -construction may or may not be denite and accordingly receives the appropriate article (the or a), a Quenya noun connecting with a following genitive is always determined, whether or not the article i is used. The system is actually the same as in English, with one minor complication added: whereas an English genitive always precedes the noun it is dependent on, a Quenya genitive may come after this noun as well. The latter word order inevitably makes one think of English of constructions, but they are strictly not comparable as far as grammar is 179
concerned even though Quenya genitive phrases are often best rendered as English of -constructions. Where the genitive follows the noun it is dependent upon, the use of the denite article before this noun is apparently optional. The noun is denite anyway, so including the article is in a way superuous; yet we have the examples i arani Eldaron the kings of the Eldar (WJ:369) and i equessi R milo the sayings of Rmil (WJ:398). Equessi R milo, u u u arani Eldaron without the article would have meant precisely the same thing. Conversely, the phrase indis i ciryamo the mariners wife could presumably have been expanded to read i indis i ciryamo the wife of the mariner, again without altering the meaning. No attested example of a preceding genitive is followed by an article. But if we can choose freely between i equessi R milo and just equessi u R milo, perhaps this principle would still apply if the genitive is moved to u the beginning of the phrase? R milo equessi Rmils sayings is certainly u u a valid wording, but what about R milo i equessi? Would this be equally u possible, or would it sound just as weird as Rmils the sayings in English? u I, for one, would avoid this uncertain and unattested construction. A few prepositions govern the genitive case. It is said that u without is normally followed by genitive, Tolkien mentioning the example u calo without light (VT39:14). This calo would seem to be the genitive form of a noun cala light (as in Calaquendi Light-elves or Calacirya Lightcleft).
180
= the brightest one of (/among) stars. The relationship between a place and something located in that place may also be expressed by means of a genitive: Calaciryo m the jewels of Calacirya. A genitive can also ri express of = about, concerning, as in Quenta Silmarillion the History of the Silmarils. Furthermore, the preposition u without normally takes the genitive case.
Vocabulary
cainen ten laman (lamn-) animal (the stem-form may also simply be laman-, but we will
use lamn- here)
yulma cup limp wine (within Tolkiens mythos, limp was some special drink of the Elves e e
or of the Valar but in the Etymologies, entry lip, Tolkien also provided the parenthetical gloss wine, and we will use the word in that sense here)
rass horn (especially on living animal, but also applied to mountains Etym., e
entry ras)
toron- (torn-) brother Menel the rmament, sky, heaven, the heavens (but the Quenya word is
singular. It is apparently not used in a religious sense, but refers to the physical heavens only. Cf. Meneltarma Pillar of Heaven as the name of the central mountain in Nmenor. The word Menel is capitalized and apparently treated as u a proper name, hence not requiring any article.)
ulya- to pour (transitive past tense ulyan, intransitive ull) e e s e river r cilya cleft, gorge (also cirya, as in Calacirya Pass of Light or Light-cleft,
which name actually appears as Calacilya in some texts but since cirya also means ship, we will use cilya here)
anto mouth (possibly representing earlier amat, amto; if so it likely comes from o
the same root as the verb mat- to eat)
Exercises
1. Translate into English: A. Hirnent i rin ohtaro macil. e B. Menelo eleni s lar. C. Tirnen i nisso hendu. 181
D. Cenuvant Aran Atanion ar ily nrion. e e o E. Coa u talamion um anwa coa. e F. I trio umi torni merir tur Ambaro lier. a e e G. I rassi i lamnion nar alt. e H. I cainen rvi lintav manter i rocco hrv. a e a e 2. Translate into Quenya: I. The birds of heaven will see ten warriors between the great rivers. J. The kings thrall poured wine into the biggest of the cups. (Biggest, greatest = analta. Time to repeat Lesson Five, where we discussed superlatives?)
K. The Elfs brother gathered (together) the ten books about stars. L. The great river of the land poured into a gorge. M. A man without a mouth cannot speak. N. I have seen the greatest of all mountains under the sky. O. I want to nd a land without great animals like lions. P. You will see an animal without horns (dual: a couple of horns)
182
Lesson 12
The Possessive-Adjectival case. Verbal or Abstract nouns and how they interact with the Genitive and Possessive cases.
This lesson is mainly devoted to a case that by its function in many ways complements the genitive case. But rst of all, let me say that there is no easy answer to the question of what this case should be called. Tolkien listed it in the Plotz Letter, but he did not name it. The case in -o or -on that we discussed in the previous lesson is referred to simply as the genitive in several sources. But in WJ:369, Tolkien refers to the forms in -o(n) as partitive-derivative genitives, whereas the other case that we will now discuss is called a possessive-adjectival [genitive]. On the previous page, he noted regarding the case with the ending -o(n) that properly it was used partitively, or to describe the source or origin, not as a possessive (emphasis added). The context indicates that the other case that he went on to describe is used as a possessive. So simply to have a some suitable designation of this case, I shall adopt the word possessive as its name. (Another plausible term is adjectival case, which is also used by some students.)
12.1
The Possessive
By its function, this case rather than the case in -o(n) which Tolkien normally terms the genitive when discussing Quenya grammar corresponds much better to the English genitive in -s. Even so, in certain contexts this case is also best translated using English of -constructions. The possessive case is formed by adding the ending -va, e.g. Eldava as the possessive form of Elda. In the case of a noun ending in a consonant, the ending probably takes the form -wa instead. The assumption that the ending -va appears in the variant form -wa after consonants is also supported by this fact: The sux -va is in origin a mere adjectival ending, found in some common adjectives as well, and in such cases it is seen to appear as -wa following a consonant e.g. anwa real, actual, true or helwa pale blue. In Primitive Elvish, the ending had the form -w, but in Quenya, w a 183
normally became v when intervocalic (occurring between vowels). Cf. another common adjective displaying this ending, tereva ne, acute, which word Tolkien noted had been terw in Primitive Elvish (see Etym., entry e a ter, teres). Since most Quenya nouns end in a vowel, the w of -w typia cally became intervocalic when this ending was added, and normally turned into v (e.g. Eld-w, Eldaw becoming Eldava, just like terw became a a a e a tereva). But if we combine this ending with a noun ending in a consonant, e.g. atar father (unchanged since Primitive Elvish), atar-w would prea sumably produce Quenya atarwa, original w remaining w because it is not here intervocalic. The Plotz Letter lists no dual forms of the possessive case, but I cant imagine why such forms should not exist. Even so, I wont construct any exercises involving these slightly hypothetical forms, but presumably the simple sux -va would be used after a dual form in -u e.g. Alduva as the possessive form of Aldu Two Trees. The more frequent dual forms in -t would likely have possessive forms in -twa, a dual like ciryat a couple of ships becoming ciryatwa (accented on the second-to-last syllable because of the consonant cluster tw). Just like the Plotz Letter lists no dual form of the possessive case, Tolkien mentioned no plural form either which fact led some investigators to conclude that this case has no plural at all! But other material does indicate that such a form exists (suggesting that we can also feel free to extrapolate a dual form as we tried to do above: the Plotz Letter does not necessarily include everything). In WJ:368 Tolkien indicates that the possessive has a plural form in -iva, combining the simple ending -va with the plural marker -i. In this case, this ending is used even if the possessive sux is added to words that would normally have nominative plurals in -r, like Eldar: The plural possessive is not **Eldarva or **Eldarwa or whatever, but Eldaiva, attested in the phrase lamb Eldaiva language of the Eldar e (WJ:369). The plural form -iva is said to be an innovation in Quenya, not a form inherited from older stages of Elvish. When the initial vowel of the ending -iva merges with the last vowel of the noun to produce a diphthong, like ai in Elda + iva = Eldaiva, this diphthong of course receives the stress (eld-AI-va). Most nouns in - would e at an older stage have behaved in a similar way, a diphthong ei arising; the plural possessive of lass leaf may at one point have been lasseiva (for e even older ?lasseiw, if such a form was ever in use). But the diphthong a ei eventually became long in Quenya, so perhaps the current form was lass with a long still attracting the stress. In the Plotz Letter, such a va long is observed in the plural form of another case: lass nen as the plural instrumental, to be discussed in a later lesson. It is not quite clear what would happen when the ending -iva is added to a noun already ending in -i, like tri queen, or a noun with a stem-form a in -i, like lm (lmi-) night (SD:415). Possibly the two is would merge o e o 184
into a long so that of queens or of nights is something like ?tr , a va, ?lm whereas the singular forms of a queen and of a night must o va be triva, lmiva. (The pronunciation would be markedly dierent: these a o singular forms are accented on the rst syllable, the third from the end, while the plural forms would be accented on the second-to-last syllable because of the long vowel that suddenly turns up there if the nal -i of the noun and the rst vowel of the ending -iva do indeed merge into a long But .) it is also possible that a form like triva has to do duty for both singular a and plural, so that one must rely on the context to distinguish of a queen from of queens. There are a few more things to say about the formation of the possessive case (see Various notes below), but we will now return to its function. This is the case you use to describe simple possession, the typical function of the English genitive. In the previous lesson, we have described how the Quenya genitive is rather used to indicate source or origin, not simple ownership. If the genitive describes the relationship between owners and the things they own, we are dealing with former rather than current ownership. Tolkien nicely explained this by contrasting the genitive and possessive cases, and we can well aord to quote him, recapitulating the function of the genitive in the process: Possession was indicated by the adjectival ending -va. . . Thus Oromes horn was rma Oromva (if it remained in his posseso e sion) . . . but [the genitive phrase] rma Oromo meant a horn o e coming from Orome, e.g. as a gift, in circumstances where the recipient, showing the gift with pride, might say this is Oromes horn. If he said this was Oromes horn, he would say Oromva. e Similarly [the genitive phrase] lambe Eldaron would not be used for the language of the Eldar (unless conceivably in a case where the whole language was adopted by another people), which is [rather] expressed . . . by . . . lambe Eldaiva. [WJ:368369] So the possessive case may indicate simple ownership at the time that is being considered (past or present whereas origin, or former possession, is indicated by the genitive case). An example from the Silmarillion is Mindon Eldaliva, the Tower of the Eldali [= Elf-people], meaning simply a e e tower owned by the Eldali. (Certainly they had also originated it, but they e were still its owners, so a genitive would be less appropriate.) We would also have such phrases as (i) coa i Eldava the Elfs house/the house of the Elf, i parmar i vend the books of the maidens, i m i va ri Naucoiva the jewels of the Dwarves. As for this word order, it should be observed that the noun which receives the possessive ending appears as the last word of the possessive phrase in all attested instances: The noun it governs (denoting the thing that is owned ) comes before it. It may well 185
be that one could reverse the word order and say (for instance) ?i Eldava coa with the same word order as in English: the Elfs house. However, I would avoid this construction until we have it attested in Tolkiens papers. The noun governed by the possessive form does not receive the article in most of our attested examples; it is already suciently determined: Rma Oromva is not indenite a horn of Oroms, as if it is rst ino e e troduced into the story, or it is implied that Orom had other horns as e well. (According to Tolkien, this meaning would be expressed by means of a loose compound, the words simply being juxtaposed without involving any case endings at all: Orom rma = an Orom horn.) Rma e o e o Oromva is Oroms horn = the horn of Orom, rma being detere e e o mined by Oromva. But we could certainly slip in an explicit article and e say i rma Oromva without changing the meaning; as demonstrated in o e the previous lesson, both constructions are equally valid in a phrase involving a genitive noun. An attested example involving the possessive case is the phrase i arani Eldaiv the Kings of the Eldar (WJ:369; this primarily e means those kings in a particular assembly who were Elvish, whereas i arani Eldaron with a genitive means those among the Eldar who were kings, or simply the kings ruling the Eldar). The article could probably be omitted without changing the meaning: Arani Eldaiv would still mean e the kings of the Eldar, the possessive form Eldaiv determining arani e anyway. (As for why the ending -iva here appears as -iv, see below; this e probably contradicts some evidence from LotR, so we may read Eldaiva instead.) The possessive case does not always indicate possession in the narrowest sense, but may also describe somebodys relationship to their more-or-less abstract attributes or properties. In such contexts, one can use the genitive as well. Tolkien mentioned that the splendour (glory) of Orom could be e expressed in two ways: One may use the possessive-adjectival case and say alcar Oromva, referring to Oroms alcar or splendour as a permanent e e attribute of his. But one could also use the genitive case; the wording alcar Oromo emphasizes that Orom is the source of the splendour. This could e e refer to his splendour as seen at the moment (proceeding from him) or at some point in a narrative focusing on the moment rather than on some permanent state (WJ:369). Cirions Oath uses the genitive in the phrase Elennanro alcar the glory of the land of Elenna. If one used the o e possessive instead, to produce the wording (i) alcar Elennanreva, it o would apparently put the emphasis on the glory of Elenna as a permanent attribute of the land. In Middle-earth time, Cirions Oath was spoken long after Elenna (Nmenor) had been destroyed and its glory proven to be u rather less than permanent, so perhaps this would be inappropriate. In our home-made example alcar Elennanreva, we added the posseso sive ending to a noun that does not denote a sentient being. This is hardly improper, for we have such attested examples as Taur Huinva Fore e 186
est of Gloom and Nurtal Valinreva Hiding of Valinor. Where no e o sentient is involved, the possessive case obviously takes on other shades of meaning; no ownership can be involved, since things or substances cant own anything. Cf. for instance the rst example of this case that was ever published, in Namri in LotR. Here we have yuldar . . . lisse-miruvreva a e o for draughts of [the] sweet mead (in the prose Namri in RGEO:68, the a e words are actually directly juxtaposed as yuldar lisse-miruvreva; in the o poetic version in LotR, a number of other words intrude between the two elements of this phrase). For decades, this was the sole available example of the case in -va. Here, this case ending implies (made) of: The yuldar or draughts consist of lisse-miruvr or sweet mead. Following this o e example, two nouns like ri crown and telp silver can evidently be e e combined as ri telpeva, crown of silver. It may be noted that in such e a case the possessive noun denoting a material the noun it governs is not necessarily be determined by it (not the crown of silver). Otherwise, yuldar lisse-miruvreva would have to mean **the draughts of sweet o mead, but Tolkien did not translate it in this way. Having only this one example from Namri to work from, early researchers thought the case in a e -va was what they called a compositive case denoting what something consists of (is composed of). This usage should be noted, but we now know that this is only one of the secondary functions of this case. Yet the fact remains that the ending -va is in origin simply adjectival, so this case may easily take on a descriptive function. Regarding the genitive case in -o, Tolkien noted that properly it was NOT used adjectivally to describe qualities (WJ:368): this is rather the function of the case in -va. The example Taur Huinva (Etym, entry phuy) apparently means e e Forest of Gloom; cf. the nouns taur forest and huin deep shadow, e e gloom. One may almost just as well treat huinva as a regular adjective e and translate Taur Huinva as Gloomy Forest or Shadowy Forest. e e The idea is that the forest is characterized by gloom, so the case in -va can describe what characterizes something or someone. This characteristic may also be an abstract or action: In early material (LT1:14) we nd the example Mar Vanwa Tyaliva Cottage of Lost Play the mar or e cottage being characterized by vanwa tyali, lost play (one must read e the earliest Silmarillion manuscripts as reproduced in LT1 and LT2 to understand precisely what this refers to). It should however be noted that the genitive case may also be used in such a context; in the late essay Quendi and Eldar we have Rithil-Anamo for Doom-ring or more literally Ring of Doom (WJ:401; the Old Quenya word rithil ring, circle would probably become risil in Exilic Quenya). Rithil-Anamo does not refer to Saurons Ring, but to the Mhanaxar, the circle where the Valar passed judgement. a The word anamo is not otherwise attested, but must be the genitive of either anama or anan (with stem anam-); it apparently means doom, judgement, judging the activity characterizing or going on in the Circle 187
(Rithil). Perhaps the possessive case could have been used instead (?Rithil Anamva or ?Rithil Ananwa) without changing the meaning. a In some instances one may indeed be in doubt which case to use, the genitive or the possessive; sometimes Tolkiens own choice is slightly surprising. He used the possessive in the phrase Noldo-quentasta Ingoldova Ingoldos History of the Noldor (VT39:16) the Elf Ingoldo being the author of this particular Noldo-quentasta or Noldo-history. Yet the emphasis is hardly on the fact that Ingoldo owns this Noldo-history (unless copyright was a big issue in Valinor). Ingoldo is just the author or originator, and for this meaning we might expect the genitive case to be used instead, since it frequently describes origin or source. Yet there may be certain conicting concerns here: Since the genitive case may also signify about, concerning (as in Quenta Silmarillion), perhaps Noldo-quentasta Ingoldo with a genitive instead could easily have been misunderstood as the Noldo-history about Ingoldo. Anyhow, in one attested example, Tolkiens choice of case certainly amounts to an outright contradiction of what he had written earlier, in the essay Quendi and Eldar : We have quoted his explanation of why it would normally be improper to use the genitive in a phrase like lamb Eldaron e the language of the Eldar this would imply the language coming from the Eldar, later taken over by others! One had to use the possessive case instead: lamb Eldaiva. Yet Tolkien himself used lamb Quendion for e e the language of Elves in a very late source (PM:395) and Quendion is unmistakably a plural genitive. The fact that Tolkien here uses another word for Elf (Quend instead of Elda) can hardly make any dierence: Ace cording to the system set out in Quendi and Eldar, we would expect lamb e Quend va, the possessive case being used of current ownership. Perhaps we can resolve the contradiction in internal terms, appealing to a linguistic development within the mythos: Tolkien noted that there was an increasing tendency to prefer the genitive case, people sometimes using it instead of the possessive case (WJ:369). So in late usage it would perhaps be more natural to say lamb Quendion rather than lamb Quend e e va, the former distinctions fading away. If one is in doubt which case to use, the genitive or the possessive, it is probably best to pick the former.
12.2
Various notes
lling in some details
NOTE #1: Vowel-lengthening in the syllable preceding the case ending: The observant student will have noted that sometimes, the last vowel of a noun is lengthened when the ending -va is added. For instance, Eldali + va e produces Eldaliva with a long (which must then receive the stress, ace e cording to the normal rules). Oromva and tyaliva as the possessive forms e e 188
of the nouns Orom and tyali are other examples. Notice that the words e e Eldali, Orom, tyali all end in two short syllables (containing neither e e e consonant clusters, diphthongs or long vowels). If the ending -va were added after them and no further changes were made, the extra syllable provided by this ending would make the stress move to what is now the third syllable from the end (cf. the stress rules set out in Lesson One). This would result in the rather awkward pronunciations **orOMeva, **eldaLIeva, **tyaLIeva. So where the ending -va is added to a noun ending in two short syllables, and there is no nal consonant, the vowel of the last of these syllables is apparently lengthened to make sure that it will receive the stress: oroM Eva, eldaliEva, tyaliEva. But if the noun ends in a consonant, there is never any need to lengthen the vowel, for where we are dealing with a noun of such a shape, the suxing of the case ending (probably appearing as -wa) will result in a consonant cluster which will make the stress move to the vowel before the new cluster anyhow. For instance, while a name like Menelmacar (the Quenya name of Orion) is naturally accented on the third syllable from the end because it ends in two short syllables, its possessive form Menelmacarwa would be accented on -arw - because of the cluster rw here arising: This cluster makes what is now the second-to-last syllable long, and therefore it receives the stress. It is unclear whether the system just sketched the nal vowel of a noun ending in two short syllables being lengthened before the ending -va would still be valid in the case of a word that only consists of these two short syllables. Such words are stressed on the rst syllable when they occur by themselves, e.g. li people. The example Eldaliva shows lengthening of e e the , but li is here compounded, and Eldali without the case ending is e e e accented on the a. It may be that if li occurred by itself, its possessive form e would be lieva (accented on the rst syllable) rather than liva (accented e on the ). We cannot know. In the case of a word like Vala, my gut feeling e is that its possessive form should be Valava rather than ?Valva. But since a there is no way of being certain, I will avoid these problems in the exercises below. Huinva (instead of **huineva) as the possessive form of huin shae e dow, gloom is however a puzzling example. For a while I actually thought nal - is always lengthened before the ending -va, but the Plotz Letter e indicates that the possessive form of lass leaf is lasseva (not **lassva). e e If the ui of huin is counted as two syllables (u-i), not as a diphthong, this e example would conform with the rule set out above: hu-i-n would have e its nal vowel lengthened when -va is added, producing huinva. But since e Tolkien explicitly stated that Quenya ui is a diphthong hence pronounced as one long syllable and not as two short ones this explanation is not satisfactory. Yet ui is supposed to be a diphthong in Sindarin as well, but in one Sindarin poem, ui occurs where the poetic meter demands two syllables. Perhaps ui, although a diphthong, is somehow overlong and sometimes 189
counts as two syllables, even though it is perceived as one syllable by the ear. Bottom line is, if the case ending -va is to be added to a noun with ui in its second-to-last syllable, the vowel in the nal syllable is apparently lengthened before -va is suxed. So the possessive form of nouns like cuil e life or tuima sprout should evidently be cuilva, tuimva. e a As for the genitive ending -o, there is no similar lengthening when the ending is to be added to a noun ending in two short syllables: The genitive form of Orom is attested as Oromo, not **Oromo (contrast possessive e e e Oromva). The form Oromo must be accented on -rom-. It seems likely, e e then, that nothing special happens when -o is added to a word like huin e either (genitive probably huino, hardly ?huino). However, I should like e e to see an attested example of what happens when the ending -o is added to a noun ending in two short vowels in hiatus most frequently -i, as in e Vali female Vala. ?Valio would have to be accented on i, which sounds e e rather awkward; the same goes for the plural form ?Valieron. I strongly suspect that in such a case, the vowel in the syllable before the genitive ending would be lengthened, thus attracting the stress: Valio, Valiron. e e But once again, there is no way of being certain; we must await further publications. NOTE #2: Special stem-forms of nouns: Where a noun has a special stem-form, it would always appear when the genitive ending -o is added. The genitive of n (niss-) woman or talan (talam-) oor would be nisso s womans and talamo oors. Yet the ending -va or -wa for possessive may sometimes produce more complex results. Adding -wa to a noun like talan, talam- would probably result in talanwa, not **talamwa, since mw regularly becomes nw in Quenya. Suxing -wa to lit (lic-) bird would result in licwa all right, but this we must spell liqua according to the normal conventions. I am not quite sure what the possessive form of n (niss-) woman should be. **Nisswa is certainly not a possible s Quenya word; perhaps we would see something like nisseva, an extra e turning up before the ending to break up the impossible consonant cluster (and following a vowel, we would regularly see -va instead of -wa). The stem-form of some nouns is simply a contraction, e.g. fern- as the stem of feren beech-tree. Surely the genitive would be ferno, but the possessive may well be ferenwa with no contraction, since other examples indicate that such contraction does not occur before a consonant cluster (**fernwa is not a possible Quenya word). Of course we could slip in an e here as well, producing ?ferneva, but I would certainly put my money on ferenwa. NOTE #3: A Tolkienian rule we can aord to ignore (!): In WJ:407, Tolkien states that the case derived by adding -va never lost it strong adjectival connotations; he actually says that it was and remained an adjective. As we remember from Lesson Four, adjectives in -a have plural forms in - e (for archaic -ai). According to what Tolkien says in WJ:407, a possessive noun (with ending -va) that governs a plural word would agree with it in 190
number just like any other adjective, the ending -va turning into -v. For this e reason, he used i arani Eldaiv for the Kings of the Eldar in WJ:369: e Eldaiva of the Eldar becomes Eldaiv (archaic Eldaivai) to agree in e number with the plural noun it is dependent on, namely arani kings. However, this may be one of the cases of Tolkien revising Elvish grammar without noticing that his new ideas contradicted something he had already published. For in Namri in LotR, we have yuldar . . . lisse-miruvreva a e o for draughts of sweet mead, and Tolkien later conrmed this construction in The Road Goes Ever On. Since yuldar draughts is a plural word, lissemiruvreva should have been lisse-miruvrev according to the system o o e Tolkien later set out in WJ:407. As I said, the likeliest external explanation is simply that Tolkien introduced a new rule without noticing that he had already published something that contradicted it. In internal terms, we may perhaps assume that the possessive form was still perceived as a kind of derived adjective in the older period, and therefore it also agreed in number like regular adjectives. But as the Ages went by in Middle-earth, the forms derived by means of the ending -va came to be perceived more strictly as a noun case only, and by the late Third Age when Galadriel composed her Lament, the adjective-style agreement in number had been abandoned. I do not use it in the exercises I have made for this course.
12.3
We have earlier dened nouns as words denoting things, whereas verbs are words that denote actions but we have also hinted that linguists would nd such denitions rather simplistic. Some nouns do denote actions, and they are appropriately called verbal nouns. Since such nouns may interact with the genitive and possessive cases in a way that should be noted, this is a good place to introduce them. A verbal noun is derived from the stem of a verb; in English, the relevant ending is -ing. (This is also the ending used to derive active participles, but they are adjectives, not nouns; the forms merely happen to coincide in English.) Singing is the verbal noun corresponding to the verb sing; in other words, singing is the action you perform when you sing. In Quenya, the stems of some primary verbs are the source of abstract formations in -m; some of them seem to have been verbal nouns in origin. For e instance, whereas the verb to love is mel-, the noun love (or loving) is melm. Some of these may take on more specialized meanings. Carm e e is used for art (UT:439), though this is basically simply a kind of verbal noun derived from the verb car- make, do hence literally making. (See below regarding oiencarm.) e Primary verbs may also receive the ending -i; the verb tyal- to play e 191
corresponds to the abstract formation tyali play, playing (as noun; cf. e the Mar Vanwa Tyaliva or Cottage of Lost Play mentioned above). e Added to an A-stem verb, the ending -i makes the nal -a drop out; cf. e naina- to lament producing the abstract noun naini a lament(ing). e Yet another frequent formation is to lengthen the stem-vowel of a primary verb and add the ending -. The verb ser- rest corresponds to the abstract e noun sr rest, repose, peace. Very often, the nouns so derived have taken e e on a somewhat more concrete meaning. From the verb sir- to ow we have s e, which would basically refer to a owing, but this noun is used r = river. The noun n t connects with the verb nut- to tie, but the u e noun has developed beyond the full abstract tying, binding and has come to signify knot instead. From lir- to sing, chant we have l e, used for r song rather than just singing, chanting. Yet the underlying idea of a verbal noun is often still discernible. The stems of some A-stem verbs, especially in -ta, can also be used as abstract nouns with no additions. Vanta- is the verb to walk, but vanta is also used as an abstract: a walk (that is, walking considered as a noun). Likewise, the verbs lanta- to fall corresponds to the noun lanta a fall(ing). However, the noun may also be lant, as in the name of the e song Noldolant or Fall of the Noldor mentioned in the Silmarillion. In e Quenya, abstract nouns indeed show a great preference for the ending -, e either alone or as part of a longer ending. One such ending is -l, which seems to one of the most versatile Quenya e abstract suxes. It may be that it can in principle be added to any Astem verb, and the resulting word is basically a verbal noun. The verb laita- to bless/praise occurs in LotR (in the Cormallen Praise), and the corresponding abstract noun laital praise or praising occurs in e UT:166, 436 (where reference is made to the Erulaital or Praise of Eru, e a Nmenrean festival). In earlier lessons we have used the verb nurta- to u o hide, which is actually only attested as a verbal noun nurtal hiding (see e o below concerning the phrase Nurtal Valinreva Hiding of Valinor). e Then let us return to the genitive and possessive cases. If you combine a verbal noun (or an abstract formation that still clearly connects with a verb) with a noun in the genitive case, it suggests that this noun is the subject of the corresponding verb. An attested example is Altariello naini for Gale adriels lament (RGEO:66; the Quenya form of Galadriels name is Altariel with stem Altariell-). The genitive Altariello Galadriels governing the abstract noun naini lament, lamenting indicates that Galadriel is the e one who does the lamenting: subject genitive. Perhaps the phrase i equessi R milo the sayings of Rmil (WJ:398) may also be analyzed in such a u u way: Rmil is the subject who originally said the sayings. An obvious u case is provided by the phrase Oiencarm Eruo the Ones [Erus, Gods] e perpetual production (MR:471). Eru is the one who does the perpetual production (oi-en-carm = probably ever-re-making), and this is indie 192
cated by the genitive form Eruo: subject genitive yet again. Way back in Lesson Two, I pointed out the error contained in the title of the fanzine Parma Eldalamberon; it should have been Parma Eldalambion instead. I must now take aair with the title of another (good!) Tolkien-linguistic journal, Tyali Tyelelliva. This was meant to signify e e Play of the Tyelelli (a folk of little Elves). But since the Tyelelli are the e e subject of the abstract noun play (the ones who do the playing), it would probably have been better to use the genitive case here: perhaps Tyali e Tyelellio. e So far subject genitive; what about object genitive? This kind of genitive is usually replaced by an of -construction in English: the discovery of America = the discovery which America was the object of. Subject and object genitive can even be combined in a phrase like Columbus discovery of America (Columbus is the subject who does the discovery, America is the object that is discovered). Our one-and-only attested example of a Quenya object genitive seems to indicate that for this meaning, Quenya uses the case in -va. This one example is found in the Silmarillion, near the end of Chapter 11: Nurtal e Valinreva, the Hiding of Valinor (Valinreva is formed from Valinr, o o o e an older variant of the name normally shortened as Valinor). The point is that the Valar hid Valinor, so Valinor is the object of the nurtal or e hiding. If one used the genitive case instead, saying Nurtal Valinro, e o e it might imply that this is a subject genitive Valinor doing the hiding instead of being its object. This would make little sense, since Valinor is not a person who can hide anything. Conversely, oiencarm Eruo cannot e be understood as perpetual production of the One even if some kind of sense could be made of this, for if Eru were the grammatical object that is produced, we would evidently see oiencarm Eruva instead. e Probably, the o-case could be used for subject genitive and the va-case for object genitive within the same phrase; if so it would probably be best to let the former genitive precede the verbal noun. Nurtal Valinreva or Hide o ing of Valinor could then be expanded to Valaron nurtal Valinreva, e o the Valars hiding of Valinor. Or, to use a wholly home-made example: e Eruo melm Ataniva = Gods love of Men and conversely: Atanion melm Eruva = Mens love of God e
added to nouns with dual forms in -u, whereas the case ending might appear as -wa when added to a dual form in -t.) If the ending -va is to be added to a noun ending in two short syllables, and there is no nal consonant, this nal vowel is lengthened before the case ending is added, so that it attracts the stress: the possessive form of Orom is therefore Oromva (not e e **Oromeva). For some reason, such lengthening also occur if the diphthong ui occurs in the second-to-last syllable of the noun; the possessive form of huin gloom is therefore huinva. A possessive phrase like X Yva e e (this is the only attested word order) may mean Ys X or Y of X referring to simple ownership, e.g. lamb Eldaiva the language of the Elves e or coa i Eldava the Elfs house. The pattern X Yva may also refer to a permanent attribute (e.g. alcar Oromva the glory of Orom), or e e to the prevalent characteristic of a place (e.g. Taur Huinva Forest of e e Gloom). Another use of this case is expressing X that consists of Y (e.g. yuldar lisse-miruvreva draughts of sweet mead). Verbal nouns, or o abstract nouns derived from verbs, denote an action viewed as a thing or process. Such nouns may be derived in a variety of ways; relevant endings include -m, -l, -i and -. Notice especially the ending -l, which (it e e e e e seems) may in principle be added to any A-stem verb, as when the verb linda- to sing produces lindal singing, music. When dependent on a e verbal noun or an abstract clearly associated with some verb, the genitive e case takes on the meaning of a subject genitive (as in Altariello naini Galadriels lament), whereas the possessive case is used for object genitive (Nurtal Valinreva Hiding of Valinor). e o
Vocabulary
minqu eleven e varya- to protect alya rich seler (sell-) sister malta gold (so according to Appendix E of LotR; the Etymologies, entry smal, gives
malda instead but post-LotR sources seem to indirectly conrm that malta was Tolkiens nal decision, as when PM:366 cites the Eldarin root yielding words for gold as malat.)
engw thing e muil secrecy (including one of the abstract endings mentioned above, -l; in this e e
case it is added directly to the root muy, here manifesting as mui-. Apparently this word is related to Sindarin muil as in one place-name occurring in LotR: Emyn Muil, possibly meaning something like Hills of Secrecy or Hidden Hills).
sr peace (in origin an abstract formation based on the verb ser- to rest, derived e e
from the same root sed which also produces the name of Est [from Esd /Ezd ], e e e the Vali or goddess of rest and sleep) e
194
ramba wall ondo stone (as material, though ondo is also used = a rock; the Sindarin equivalent gon, gond occurs in the names Gondor and Gondolin, the latter of which is adapted from Quenya Ondolind) e
osto city (according to late sources also used = fortress, but we will use it in the
sense of city here; the word seems to refer primarily to a fortied city, so there may not be much of a distinction anyway)
morni darkness (cf. mor black; the word morni is actually an abstract fore e e
mation based on another adjective derived from the same primitive root mor, namely morna = dark)
Exercises
These exercises involve both the genitive case and the possessive/adjectival case. Make sure to pick the right case in Exercises IP (though sometimes, either case will do). 1. Translate into English: A. I limp Eldaron vs. i limp Eldaiva (and since both phrases may e e
have the same English translation, explain what the dierence is)
B. Haryaly yulma maltava. e C. I rocco i Eldava alanti mir i n ra cilya. e u D. Neri sreva uvar ohtari. e E. Alt rambar ondova nurtaner i coar i cainen analy nere e iva i osto. F. I coa i arano selerwa n carn. a e G. Min i mlion ampi i macil i aranwa. e o a e H. I vendo toron hirn ily i harmar i minqu Naucoiva e e e e imb i canta rassi i ninqui orontion. e 2. Translate into Quenya: I. Rivers of wine poured into the mans mouth. J. The boys sister [/the sister of the boys] gathered (together) the things of the boys and went into the house of the queen. K. The secrecy of the women protected a great treasure of gold. L. The eleven warriors could not protect the peace of the city, for a great darkness fell. M. They will go through a land of great trees and many rocks, for they want to see the city of the mighty warrior. 195
N. A wall of secrecy protected the hidden gold of the city, and I did not nd it. O. The land of the Elves is a land of many beautiful things; a land without Elves is a land of darkness, for the Men (Atani) of the land do not hear the rich language of the Elves. P. The kings sisters gathering of books about Elves. (To make an
abstract noun gathering, try adding the ending -l -ing to the verb hostae to gather.)
196
Lesson 13
The Dative case. The Gerund. Pronominal endings -lm and e -mm. An indenite pronoun. e
13.1
Back in Lesson Five, we introduced the concept of grammatical objects, the target of the verbal action performed by the subject: I Elda mta massa a = the Elf is eating bread , i Nauco hirn harma = the Dwarf found a e treasure. So far in this course, all the objects that we have concerned ourselves with have been, more precisely, direct objects. These are objects directly aected by the verbal action. In archaic Quenya, such objects had their own grammatical case, the accusative but this case no longer had any distinct forms in Third Age Quenya. But there are also indirect objects, to which Quenya grammar likewise assigns special case forms. The case marking indirect objects, the dative, was still alive and well in Third Age Quenya. But before discussing how dative forms are constructed, let us have a closer look on indirect objects. An indirect object is, logically, an object indirectly aected by the verbal action of the sentence. Thus the indirect object is often the beneciary of the verbal action (though it may also denote a party that is adversely aected by this action). The archetypal example involves the verb to give, which must logically entail three parties: the subject that does the giving, the direct object which is the thing that is given, plus the indirect object that is the recipient of the gift and thus the beneciary of the verbal action: The man [subject] gave the boy [indirect object] a book [direct object]. Modern English (unlike, say, German) has no distinct dative case, so in the English example above, the noun boy receives no extra inectional elements to explicitly mark it as the indirect object of the sentence. In English, indirect objects may be indicated simply by word order; the indirect object is then jammed into the sentence in front of the direct object, just like the boy appears before a book in our example above. But just like
197
English often uses a phrase involving the preposition of instead of the genitive case, the absence of a distinct dative case is frequently compensated for by means of prepositional circumlocutions: The two prepositions most often used to simulate the function of the dative case are to and for. So instead of saying the man gave the boy a book (word order alone identifying the boy as the indirect object), one might say the man gave a book to the boy. Examples with for could be, say, we did it for the children, or the men work for the queen. In Quenya, the boy, the children and the queen of these examples would be considered indirect objects the parties indirectly aected by the verbal action and the corresponding nouns would be inected for the dative case. There would be no need to maintain a specic word order, or to use prepositions like to or for. Cf. the question occurring in the middle of Namri, where the pronoun ni I (related to the pronominal ending -n or a e -ny of similar meaning) appears in the dative case: e S man i yulma nin enquantuva? Now who will rell the cup for me? The element here translated for is not a separate word in Quenya; it is simply the nal -n of nin which -n is the Quenya dative ending. Hence nin = I-for, or in correct English: for me. In other contexts it could also have been translated to me or (where in English the indirect object is identied by word order alone) simply me: The verb to give being anta-, the Quenya equivalent of you will give me a book would in all likelihood go something like antuvaly nin parma. The dative pronoun nin turns up in e the last line of F riels Song, as part of a sentence which Tolkien translated what will the Father . . . give me. . . ? (LR:72). Of course, one might also translate . . . give to me. The Plotz Letter conrms that the ending -n for dative is not only applicable to pronouns; it can also be attached to common nouns. (Plotz lists ciryan as the dative form of cirya ship and points to lassen as the dative form of lass leaf.) Thus we can build sentences like carnelyes i Naucon e you did it for the Dwarf or i nr antuva i parma Eldan the man will e give the book to an Elf. In English, the latter sentence could also be translated the man will give an Elf the book, word order alone indicating that an Elf is to be understood as the indirect object of the English sentence. In Quenya, word order would be much freer (the main advantage of a highly inected language!), the case ending indicating that the noun in question is the indirect object no matter where the noun occurs in the sentence. This enables the speaker to move the indirect object around to express subtle nuances of emphasis. For instance, we may probably front the indirect object to put special focus on it: Eldan i nr antuva i parma, meaning e something like it is to an Elf [not to a Dwarf, etc.] the man will give the book. Whether the direct or the indirect object comes rst may not be very 198
material; while i nr antuva i parma Eldan means the man will give e the book to an Elf, i nr antuva Eldan i parma may be translated the e man will give an Elf the book. In the plural, dative forms end in -in. Nouns with nominative plurals in -i would in eect simply add the dative ending -n; the Plotz Letter points to lassin as the dative form of lassi leaves. The dative plural of a word like laman (lamn-) animal would therefore be lamnin = to animals or for animals. The ending -in is however also added to nouns that would have nominative plurals in -r; the plural marker -r does not appear in dative forms. From the rst line of F riels Song (translated the Father made the World for Elves), we know what the plural dative of Elda would be: Eldain. It should be noted that the ending -in merges with the last vowel of the noun to form a diphthong, like ai in this case. F riels Song also provides the dative plural of Mortals: F rimoin (nominative plural F rimor, LR:245; Tolkien later came to prefer the form F rimar, as in the published Silmarillion, chapter 12. The dative plural would then become F rimain, of course.) In the previous lesson I pointed out that it is not quite clear what would happen when the [plural possessive-adjectival] ending -iva is added to a noun already ending in -i, like tri queen, or a noun with a stem-form in -i, like a lm (lmi-) night (SD:415). The same problem arises in the case of the o e o plural dative forms. The dative singular to/for a queen would presumably be trin, but when trying to derive a plural dative by adding -in to tri, a a we are probably left with trin once again (the nal -i or the noun and the a initial i- of the ending simply merging). Conceivably the two vowels might merge into a long i, producing tr as the word for to/for queens, but a n this seems a somewhat unlikely word: Quenya rarely has a long vowel in a nal unaccented syllable (though there is the word palant It may well r). be that trin has to do duty for both singular and plural, so that one must a rely on the context to nd out which number is meant. The dual dative ending is given as -nt in the Plotz Letter, the dual dative form of cirya ship being listed as ciryant which would mean something like for a couple of ships. Of course, this ending -nt simply combines the dative ending -n with the dual ending -t. Already in Lesson One I briey mentioned that this -nt seems to be the sole example of a nal consonant cluster being allowed in LotR-style Quenya. It may be that only nouns with nominative dual forms in -t would have dative dual forms in -nt; in the case of nouns with dual forms in -u, it may well be that the simplest dative ending -n would be employed. Once again using Aldu Two Trees as our example, the dative form should perhaps not be Aldunt, but rather Aldun (or possibly Alduen; see below concerning Ar-Veruen). We lack examples either way, though. In some instances, a sentence may include an indirect object (in Quenya, a dative object) even though the sentence contains no direct object. In the 199
Quenya equivalent of the men work for the king, the English preposition for would be represented by the case ending -n added to aran (probably producing aranen, a helping vowel -e- intruding before the ending). There is at least one attested example of a sentence having a dative object, but no direct object: Tolkien in his Quenya version of the Pater Noster used a dative form of the pronoun we, us when rendering the phrase . . . those who trespass against us. (This, by the way, is an example of the dative case being used to identify a party adversely aected by the verbal action: The context must decide whether the dative form should be translated against us or for us. Grammatically speaking, both interpretations would be equally valid, but those who sin for us would not make sense in this particular context.) The verb to trespass or to sin can have no direct object, but obviously some party may be indirectly aected by the trespassing or sinning, so there can be an indirect object aptly presented as a dative object in Quenya. Dative forms may also turn up in sentences having no subject, a grammatical phenomenon virtually unheard of in modern English. Yet such sentences may be compared to phrases like it seems to me that. . . , where the formal subject it is actually nothing but a grammatical dummy with no real content: In informal speech it may even be omitted, seems to me that. . . , and the meaning is quite intact. Such English wordings are comparable to Quenya phrases like or nin caritas, literally [it] impels for me to do it, e expressing the meaning that in English might be worded something like I would like [or, feel moved] to do so (VT41:13). Notice that the sentence or e nin caritas has no subject, but it does have a dative object: nin to me, for me. Or- or ora- impel, urge is one of the Quenya impersonal verbs which invite such constructions; we will return to these verbs in a later lesson. An earlier application of the case ending -n: In Tolkiens long evolution of Quenya, -n was not always the dative ending. From the perspective of the LotR scenario, one of the few things that are wrong with the Quenya of the Etymologies (written in the mid-thirties) is that -n is here the genitive ending instead. For instance, the entry lep lists the names of various Valinorean weekdays, including Ar Manwen = Day of Manw, or Ar Ulmon e = Day of Ulmo (cf. the noun ar() day though later Tolkien changed e the word for day to aur or r, as indicated by the LotR appendices). e e This use of the ending -n is also found in a phrase written in the midforties, reproduced in SD:303: Quenta Eldalien, History of the Elves [Eldali, Elf-people]. In draft versions of the poem that was to become e Namri, Tolkien used Vardan as the genitive Vardas (see for instance a e the early version reproduced in TI:284-285). In one of the later manuscripts, Tolkien still wrote Vardan, but then he crossed it out and replaced it with Vardo. This may seem to pinpoint the moment when Tolkien changed the genitive ending from -n to -o. Actually the genitive ending -o turns up in earlier sources as well; a Qenya poem of the early thirties already has 200
ciryo (spelt kiryo) as the genitive of cirya ship (langon veakiryo the throat of the sea-ship, MC:216). As for the dative case, the Qenya of the pre-LotR composition F riels Song already uses -n (pl. -in) as the dative ending, as is evident from some examples quoted above. Later, -n for a while reverted to being the genitive ending, as reected by the Namri drafts. It a e would seem that Tolkien changed his mind back and forth over the decades but the nal resolution, as reected in LotR and as codied in the Plotz Letter, was that -o is to be the genitive ending, while -n is the dative ending. Some of the obsolete genitives from the Etymologies are still interesting as forms. In the entry ay, the noun ailin pool, lake is said to have the g.sg. (genitive singular) ailinen. Because of Tolkiens later revisions, the form ailinen must rather be understood as a dative singular in LotR-style Quenya meaning for a lake instead of genitive of a lake. The example ailinen is interesting insofar as it shows us what happens if the case ending -n (no matter what case it is assigned to!) is added to a noun ending in a consonant, like ailin. Since **ailinn is not a possible Quenya word, an e is inserted before the ending, producing ailinen. Though the ending -n had its meaning redened, the principle of inserting this helping vowel whenever it is required would still be valid. If the noun has a special stem-form the nal consonant turning into another consonant or a consonant cluster when endings are to be added such changes occur before this extra -e- as well: In the entry lin2 in the Etymologies, we learn that Laurelin (Laurelind-), the name of the Golden Tree of Valinor, has the g.sg. Laurelinden. In LotR-style Quenya, this would be the dative singular instead, but the form as such is presumably valid still. The same goes for lissen as a form of lis (liss-) honey; see the entry lis. We may then assume that a noun like n (niss-) woman would s be treated in a similar fashion: dative nissen. One of the genitive forms of the Etymologies may even throw some light upon what the later dative of dual forms in -u should look like. One Valinorean weekday listed in the entries bes and lep is Arveruen or ArVeruen, the Day of the Spouses, referring to the Vala couple Aul and Yae vanna. Here we have veruen as the genitive of the dual form veru spouses, married pair. It should be noticed that the helping vowel -e- is employed here as well (somewhat surprisingly: one might have thought that **verun would be an acceptable form). If this genitive veruen would still be a valid form after Tolkien redened the ending -n so that veruen is a dative form in LotR-style Quenya, it would indicate that nouns with nominative dual forms in -u should have dative forms in -uen. The dative of Aldu Two Trees would then be neither Aldunt nor Aldun, but rather Alduen. But I hardly have to say that we are not on solid ground here, and I will not construct any exercises based on such hypotheses.
201
13.2
The Gerund
Most of the time, nouns and verbs are distinct parts of speech. There are, of course, the verbal nouns discussed in the previous lesson, but they are unquestionably genuine nouns abstract formations denoting verbal actions considered as things. But verbs have one form, the gerund, which almost dees the dichotomy of noun vs. verb. One might say that a gerund is a verb masquerading as a noun. In Cirions Oath occurs the word enyalien, literally meaning for recalling (that is, in order to remember). The prex en- means re-, and the nal -n is the case ending discussed above, the dative marker corresponding to the English preposition for. Stripping away these extra elements, we are left with -yalie-, yali. In his notes on Cirions Oath, as reproduced e in UT:317, Tolkien makes it clear that yali is an innitive (or gerundial) e form of a verb yal-, meaning to call or to summon. Thus we can isolate -i e as a grammatical ending used to derive innitive or gerundial forms. Earlier in this course, we have discussed another kind of innitive, which is simply the stem of the verb (with - added, in the case of primary verbs). e An attested example is the sentence polin quet, I can speak (VT41:6), e with quet as the innitive form of the verb quet- speak. Since Tolkien e clearly identies -i as an innitive ending in UT:317, it may be that polin e queti would have be equally possible (more on this below). In her tutorial e Basic Quenya, written before the example polin quet became available, e Nancy Martsch uses -i as the Quenya innitival ending throughout. This e may not necessarily be wrong; Tolkien denitely imagined an old Elvish innitive in -ie. In the Etymologies, entry nar2 , the Old Noldorin word trenarie to recount is explicitly called an inf. form (Old Noldorin being the language Tolkien might later have referred to as Old Sindarin, after he revised his linguistic mythos in the early fties). However, I think that in many cases, the Quenya verb forms derived by means of the sux -i are best termed gerunds. e In English, gerunds are derived by means of the ending -ing, e.g. nding vs. the verb to nd. Now a form like nding can of course also be an abstract noun (synonymous with discovery) as well as an adjectival participle (as in the man nding the treasure). But what we are interested in here, is rather the word nding as it appears in a sentence like nding the treasure was wonderful. Here, nding in a way behaves like a noun, for it is the subject of the sentence. But we can tell that in some sense, nding is still a verb, for it has not lost one of the unique characteristics of a verb: the ability to take an object. In the phrase nding the treasure, the treasure is the object of nding. If nding had here been an abstract noun, one would have to use an of-construction to bring in the thing which is found: the nding (= discovery) of the treasure. Cf. our discussion of the Quenya object genitive (as in Nurtal Valinreva, Hiding of Valinor) e o 202
in the previous lesson. Since we are able to tell that in the sentence nding the treasure was wonderful, the word nding is not a verbal noun, we can conclude that it is actually a gerund. A gerund is a form of the verb which can function as a noun, with much the same meaning as a genuine verbal noun. However, a gerund is still capable of taking an object, and this goes for Quenya gerunds as well: Regarding the Quenya gerundial form in -i which Tolkien used in e Cirions Oath, he noted that it was governing a direct object (UT:317).
NOTE: In Quenya as in English, gerunds and abstract nouns cannot always be clearly distinguished. Just like the English ending -ing is used to derive both gerunds and verbal nouns, the Quenya ending -i may also be used to derive abstracts, e. g. tyali play (as e e noun) from the verb tyal- to play. Indeed -i is also used as a general abstract ending, e much like English -ness, cf. for instance morni darkness. e
As usual, there is an extreme scarcity of attested examples. But we must assume that in Quenya as in English, gerunds may often function as the subjects of sentences, perhaps something like this: Hiri harma caruva nr alya nding a treasure will make a man rich e e Tiri i aiwi anta i vendin alta alass watching the birds gives the e e maidens [vendin, dative] great joy In these examples we have equipped the gerunds with objects (harma and i aiwi), but a gerund could certainly function as subject without any e e further additions, for instance like this: Mati n i analta alass ily e a tiuc Naucoron, eating is the greatest joy of all fat Dwarves. e Presumably Quenya gerunds can also function as the object of a sentence, comparable to such English constructions as I love shing. The gerund functioning as object may in turn govern its own object: A sentence like I love watching the birds may perhaps be rendered into Quenya as melin tiri i aiwi (watching being the object of the phrase I love, and the e birds in turn being the object of the gerund watching). Perhaps the latter might also be expressed as I love to watch the birds = melin tir i aiwi e (?), using an innitive instead of a gerund. Gerunds and innitives may well be interchangeable in many contexts, in Quenya as in English. Indeed our terminology may be stricter than the one Tolkien himself used, if we reserve the term innitive for forms like tir to watch and insist on e calling tiri watching a gerund only: In UT:317, quoted above, Tolkien e himself refers to the forms in -i as both innitival and gerundial. As we e mentioned above, in the Old Noldorin of the Etymologies one form in -ie is explicitly identied as an innitive. The post-LotR example polin quet e I can speak demonstrates that -i at least cannot be a universal innitive e ending. Would polin queti be a possible wording, or would this sound e about as weird as I can speaking in English? And what about I want to nd a treasure? Would merin hiri harma be OK, or would the Eldar e nd this wording as awkward as I want nding a treasure in English? It 203
may be safer to use the simplest innitive, hir, in such a context. e When a verbal action is the subject or object of a sentence, one may to some extent choose between the innitives and gerunds in English: To err is human, to forgive is divine = Erring is human, forgiving is divine. Especially when a verbal action functions as subject, I think it would be safer to use the gerund (the form in -i) in Quenya. But since we have no e actual examples, it is presently impossible to say with any condence what Tolkien would have thought of as acceptable Quenya in this regard. There is, however, one important use of the gerund which luckily is attested in our tiny corpus. In English, the normal innitive (marked by to) is often used to indicate purpose: They have come to see the king. Whether this could be rendered directly into Quenya as ?ut lient cen i aran u e e none can say at present but I tend to doubt that this is a valid construction. Notice the wording used in Cirions Oath: Vanda sina termaruva Elennanro alcar enyalien. Tolkiens translation in UT:305 goes this o e oath shall stand in memory of the glory of the Land of the Star, but more literally the Quenya wording is something like this oath shall stand for recalling [the] Elenna-lands glory. Cf. Tolkiens comments on the form enyalien in UT:317, already quoted in part: yal - summon, in innitive (or gerundial) form en-yali, here in e dative for the re-calling, but governing a direct object [namely alcar glory]: thus to recall or commemorate the glory. So here we have a verb en-yal- re-call- = commemorate. Add the gerundial ending -i, and we get the gerund enyali, recalling. Since a e e gerund may be described as a verb functioning as a noun, it may also receive case endings as a noun. So Tolkien supplied the dative ending -n for to produce enyalien for recalling. The word can now function as the indirect object of the sentence, the benefactor of the verbal action: The oath termaruva shall stand, and this action promotes recalling (enyali). The e dative gerund enyalien for recalling in turn has Elennanro alcar, o e [the] Elenna-lands glory, as its object. Of course, in English one does not say this oath shall stand for recalling the Elenna-lands glory, but rather this oath shall stand (in order) to recall the glory of the land of Elenna. Nonetheless, this example seems to tell us that English innitives indicating purpose should be rendered into Quenya as gerunds with a dative ending attached. They have come (in order) to see the king would then translate into ut lient cenien i aran, literally u e they have come for seeing the king. (If we were to slavishly follow the word order Tolkien used in Cirions Oath, with the gerund at the end of the sentence, we would actually have to say ut lient i aran cenien = u e they have come the king for seeing! However, Quenya word order is in all likelihood quite exible.) The rule we have tried to make out may be 204
summarized like this: If in English you can add the words in order in front of a innitive without destroying the meaning (never mind the style!), this innitive indicates purpose and should be rendered into Quenya as a gerund inected for dative. Forming gerunds from A-stem verbs: All the gerunds so far exemplied have been formed from primary (ending-less) verbs. What happens if the ending -i is to be added to an A-stem verb? We have no direct, explicit e attestations to guide us, so I saved this problem for the end. But all the indirect evidence points to one conclusion: the nal -a should be dropped before -i is suxed. e The Etymologies, entry oro, lists the Quenya verb orta- rise, raise, but a form ortie is also cited, though this is Old Noldorin (/Old Sindarin) rather than Quenya. This word ortie, simply glossed rise, would be an archaic Elvish form that later involved into a Sindarin innitive. But it could very well correspond to a Quenya gerund orti rising, raising, since e Old Noldorin is relatively close to Quenya. This would indicate that when the ending -i is to be added to an A-stem verb, the nal -a drops out e before the ending. We have one possible attestation of a Quenya form which would conrm this conclusion: Listing various forms of the verb ora- urge, Tolkien did include ori (VT41:13), and while he did not clearly identify e this or any other of the forms, ori may well be intended as the gerund. Also e notice naini lament(ing) as a derivative of the verb naina- to lament e (compare RGEO:66 with the Etymologies, entry nay): Naini may be seen e either as a gerund or as a verbal noun. As we have already touched on, -i can also function as a general abe stract ending, somewhat like English -ness. Where -i is used to form e abstract nouns from adjectives, adjectives in -a lose this nal vowel before -i is added; morni darkness is apparently formed from morna dark. e e Another attested pair of this kind is lta open vs. lti openness. The a a e abstract ending -i is certainly closely related to the gerundial ending -i; e e basically it is the same ending we are dealing with (as noted above, the distinction between gerunds and abstract nouns often becomes blurred). If the ending -i causes a nal -a to drop out when it is added to adjectives, it e seems very likely that this also happens when it is added to A-stem verbs. So starting from verbs like orta- raise and nurta- hide, we may probably derive the gerunds orti, nurti and build sentences like orti Pelri e e e o nurtien Valinor um mra noa raising [the] Pelri to hide Valinor was e a o not a good idea. (It wasnt see MR:401, 405 for Tolkiens critical comments on this move by the Valar!) In the case of verbs in -ya, e.g. harya- to possess, the entire ending -ya would probably drop out before -i is suxed. Otherwise the gerund would e have to be **haryi, but yi is not a possible Quenya combination. Abstract e nouns formed by means of the ending -i from adjectives in -ya are seen to e surrender the latter ending, e.g. veri boldness from verya bold (see e 205
the Etymologies, entry ber). We may probably assume that veri would e also be the gerund of the related verb verya- to dare. So the gerund of a verb like harya- to possess is most likely hari (e.g. in a platitude like e hari malta uva car nr anwav alya, possessing gold will not make e e e e a man truly rich).
13.3
The pronoun we
We have been practicing various pronominal endings: -n or -ny I (the e short form must not be confused with the dative ending!), -ly you, -s e it, -nt they and -t them. It is time to introduce the endings for the e rst person plural, corresponding to the English pronoun we. There are actually several Quenya endings for we. One of them occurs in the Cormallen Praise: Andav laituvalmet, long shall we bless them. e Here we have a future-tense verb with pronominal endings for we (subject) and them (object): laituvalmet, blessshallwethem. The ending for we is seen to be -lm (-lme-). e However, in WJ:371 Tolkien discusses the Quenya exclamation v, signala ing refusal or prohibition: in eect no! in the sense of I will not or do not! Tolkien also indicated that this v could receive explicit pronominal a endings, such as -n(y) for I, producing the form vn or vny for I e a a e wont. But Tolkien also mentioned the form vamm, we wont. So here e the pronominal ending for we is suddenly not -lm, but rather -mm. e e
NOTE: Notice, by the way, how the long of v is shortened in the form vamm. a a e This is one of the examples indicating that Quenya cannot have a long vowel in front of a consonant cluster or a long consonant a phonological rule we have repeatedly alluded to earlier in this course. The fact that the vowel remains long in vny suggests that ny is a e perceived as a single consonant, palatalized n like Spanish n, and not as a cluster n + y.
The fact that there are two endings for we, -mm and -lm, is easily e e explained once we realize that this part of the Quenya pronoun table makes a distinction not found in English. Tolkien referred to the ending -mm e as the rst [person] plural exclusive (WJ:371, emphasis added). The ending -mm denotes an exclusive we, a we that excludes the person(s) e addressed. The exclamation vamm we wont represents a refusal as it e would be spoken to some other party (likely the one whose will we refuse to obey). This other party is not included in we, but stands outside the we group. Therefore, the exclusive we is proper here. On the other hand, the ending -lm apparently denotes an inclusive we: e The party being addressed is included in we. In the example andav e laituvalmet, long shall we bless them, the people who are praising Frodo and Sam are addressing one another, not the Ring-bearers. If they had said long shall we bless you instead, addressing Frodo and Sam directly, they would have had to use an exclusive we: ending -mm instead of -lm. e e
206
Frodo and Sam would not be part of this exclusive we; they would stand outside the we group addressing them. (Conversely, a group of people conspiring among themselves about refusing an order would have to say valm, not vamm, for we wont!) e e In written works, the word we would normally be exclusive (-mm), e unless the writer has reason to believe that the reader will be part of the we group referred to. In such a case, -lm for inclusive we would be e proper. Back in Lesson Eight, it was noted that while the author of this course may sometimes seem to refer to himself as we, this is not (necessarily!) because he has an ego of royal dimensions. The author tends to include the reader in this we, as if implying that the author and the readers somehow share this odyssey through the various aspects of Quenya grammar. (You can take it as a friendly gesture, or as a particularly cunning brain-wash technique which the author uses to somehow make you an accomplice whenever he draws conclusions that he should actually take the full responsibility for himself!) Anyhow, in Quenya there could have been no misunderstanding. A royal we, excluding the people addressed, could only be -mm. An e author using the word we to refer to himself and his readers, directly addressing them in his text, would have to use the inclusive we: -lm. e There is yet another ending for we, namely the dual inclusive, which is we referring to two persons only in eect you and I or the two of us. (The ending -lm would refer to an inclusive we involving three or more e persons.) Unfortunately, the ending denoting dual inclusive we must be extrapolated in a two-step operation, and we cannot be quite certain about its form (most likely either -lw or -lv). We will return to this problem e e later; in the exercises below, we will only practice the endings -lm and e -mm. e
NOTE: There has been much confusion regarding the Quenya endings for we. An earlier interpretation, set forth in Jim Allans Introduction to Elvish in the late seventies, had it that the ending -lm is exclusive rather than inclusive. The extrapolated ending e -lv was thought to be the inclusive we (and inclusive it would certainly have to be, but e it is a dual inclusive form, referring to a we of two persons only). The true ending for exclusive we, -mm, was unknown until WJ was published in 1994 (providing the exame ple vamm). The wrong interpretation haunts many post-Tolkien Quenya texts written e before the mid-nineties, and it also made its way into Nancy Martsch Basic Quenya.
13.4
An indenite pronoun
In English, words like one and you are often used with deliberately vague or general reference: One has to earn a living. . . or you have to wonder. . . Here, one is not the number 1 (Quenya min), and you e does not refer to the one who is addressed. For such meanings, Quenya
207
has the pronoun quen (WJ:361) essentially an unstressed variant of the noun qun, which simply means person. Though ultimately related to e Quend Elf, these words have no special reference to Elves. As usual, we e lack attested examples, but we must assume that quen would be used for one or someone in such sentences as quen milyan leryal one longed e e for release or quen hant i yulma someone broke the cup. Presumably e quen may also receive case endings, e.g. genitive queno ones or dative quenen (which in English would often be translated for you rather than for one: Mati yv n mra quenen, eating fruit is good for you e a e a a you here meaning people in general!) Most Quenya pronouns usually appear as endings, and it may be that Tolkien at one stage even reckoned with a pronominal ending for the indenite pronoun one. There exists an early Qenya text where this meaning seems to be associated with an ending -o: Kildo kirya ninqe, translated a white ship one saw (MC:220221). However, transforming this into LotRstyle Quenya would probably require rather more than altering the spelling to cildo cirya ninqu: While the two last words would certainly be ace ceptable, the verb form kildo does not seem to t Tolkiens later system, and the status of the ending -o at the LotR stage is highly doubtful. If we aim for LotR-style Quenya, it is certainly much safer to use the indenite pronoun quen from a post-LotR source.
208
gerunds inected for dative, e.g. hirien (in order) to nd. Quenya has several pronominal endings corresponding to English we. One of them is -lm, denoting a we that includes the party that is addressed, whereas the e ending -mm expresses an exclusive we, used when the speaker addresses e a party outside the we group that the speaker himself belongs to. The indenite pronoun one or someone is in Quenya quen. Presumably it can receive case endings, e.g. genitive queno ones.
Vocabulary
In each Vocabulary section, we have rst of all introduced a new number. The numbers 1-11 are explicitly mentioned in the Etymologies: min, atta, e neld, canta, lemp, enqu, otso, tolto, nert, cainen and minqu. e e e e e The Elvish way of counting, with base 12 instead of 10, would obviously require a word for twelve as well the last of the basic numbers. However, the Etymologies does not mention the Quenya word for twelve, and neither is it attested elsewhere. Etym only cites the primitive root-word for this number: rasat. No other forms are given, Christopher Tolkien notes. However, students of Elvish agree that a Quenya word derived from this root would most likely have the form rasta (the complete Primitive Elvish word being something like rsat, the accent mark here indicating stress rather a a than length). Some writers have used rasta in their own compositions, so it is at least a post-Tolkien Quenya word. To complete our survey of the basic numbers, I have included rasta in the vocabulary below but it must be understood that while this is denitely a plausible word, it is not explicitly given in published material. ?rasta twelve mahta- to ght anta- to give, irregular past tense n. (This past tense is listed in a very a e
old source, the Qenya Lexicon p. 31. It is entirely possible that in Tolkiens later Quenya, the past tense of anta- could just as well be regular: antan. However, the e Sindarin verb form onen I gave occurring in LotR Appendix A would correspond to Quenya nen rather than ?antanen. The Etymologies, entry ono, indicates that a the past tense of the verb onta- beget, create may be both on and ontan; e e perhaps the past tense of anta- may likewise be both n and antan. We will a e e use the attested form n here.) a e
suc- to drink anna gift alass joy e hroa body (related to a word introduced earlier, hrv esh; Tolkien meant them a e
to be descended from Primitive Elvish sraw and srw, respectively. See MR:350.) a a e
209
noa idea cala light (as in Calaquendi Light-elves, Calacirya/-cilya Light-cleft) mra good (in the sense of t, useful Quenya has other words for good in the a
moral sense)
quen indenite pronoun one, someone arwa adjective possessing, in control of, having, followed by genitive (see note)
NOTE ON ARWA POSSESSING, IN CONTROL OF: This adjective is listed in the Etymologies, entry 3AR (though it is derived from a variant root gar). It can be used to form compound adjectives; Tolkien mentioned the example aldarwa having trees = tree-grown (alda + arwa, tree-having). But apparently arwa having, possessing can also be used by itself, and then it would be followed by a genitive form. (As we have touched on above, the Quenya genitive ending was -n when Tolkien wrote the Etymologies, but we must assume that the rule as such was still valid when he changed the ending to -o later.) So we may probably have phrases like nr arwa collo, a man having/possessing a e cloak (colla cloak, genitive collo). Perhaps this may simply be translated a man with a cloak, and if we use arwa to mean with, it would mean that the words for with and without (arwa and u, respectively) are both followed by genitive! Yet arwa is said to be an adjective and not a preposition, so arwa presumably agrees in number, becoming arw (for archaic arwai) when pointing back to a plural word: Neri arw collo, men e e possessing a cloak, Naucor arw harmaron Dwarves possessing treasures, arani e arw ohtarion kings in control of warriors. e
Exercises
1. Translate into English: A. I nr n i nissen anna. e a e B. Anar anta cala Ambaren. C. Hiruvalm i harma, ar antuvalmes i rasta Naucoin. e D. Mati hrv car quen tiuca, ar umilm mer tiuc hroar, e a e e e e e an tiuc hroar umir vany. e e E. Lendemm mir i osto hirien i sail nissi, an mernemm e e e cenitat. F. Nr arwa mr noaron n saila ar antuva sr ar alass i e a e a e e e oston. G. Utultiemmet quetien rimb engwion. e H. Suci limp um mra queno hroan. e e e a 2. Translate into Quenya: 210
I. Someone gave the warrior a great sword. J. Making a house for the boys is a good idea. K. We (inclusive) ght for peace; ghting does not give the people [any] joy, for we (inclusive) have seen the light. L. Speaking the Elven-tongue (Eldalamb) is a great joy to Men e (Atani). M. The warriors of the two lands will ght for the [twin] peoples (dual), and we (exclusive) will go through a great darkness to nd light. N. The men having the good wine wanted cups to drink the wine, and the kings thralls gave the men twelve cups of gold. O. We (exclusive) want to go into the city to free all Men (Atani) and (to) give the gold of the king to the thralls. P. The walls of the city are great; we (inclusive) have made them to protect the people.
211
Lesson 14
The Allative and Ablative cases. Equ and auta: two e peculiar verbs. Possessive pronominal endings: -nya, -lya, -lma, -mma.
14.1
The dative case ending -n presented in the previous lesson may sometimes correspond to the English preposition to, as when it is appended to gerunds: enyalien = to recall (UT:317). Yet this is a rather abstract kind of to; as we have seen, the Quenya dative may also be translated as phrases involving the preposition for, or it may simply correspond to a specic English word order. However, Quenya does have a special case form implying to in the more basic sense of towards or against; the Latin term for such a case is allative. The relevant Quenya ending is -nna: In the entry Eldanna in the UT Index, Christopher Tolkien identies this ending as a sux . . . of movement towards. The word Eldanna itself is not a bad example; it may be translated Elfwards and was used by the Nmenoreans as the name of u a bay on the west coast of Nmenor, thus in the direction of the Blessed u Realm where the Eldar dwelt (UT:164). In Elendils Declaration, repeated by Aragorn at his coronation, the ending -nna carries the full force of to with the implication of motion towards: Et Erello Endorenna ut lien a u = out of the Great Sea to Middle-earth [Endor] I am come. Cf. also the sentence Sin Quent Quendingoldo Elendilenna (PM:401) apparently e meaning Thus Spoke Quendingoldo to Elendil (or perhaps This Quendingoldo Said to Elendil; the meaning of the word sin is not quite clear). As the allative forms of cirya ship and lass leaf, the Plotz Letter points e to ciryanna to a ship and lassenna to a leaf. (Of course, the stress now falls on the vowel of the second-to-last syllable because of the following consonant cluster introduced by the ending -nna: ciryAnna, lassEnna.) So if you want to say Ill go to the ship in Quenya, you dont normally use a separate word for to, but employ the ending -nna instead: Lelyuvan i ciryanna. 212
While the ending -nna may sometimes be rendered -wards in English, e.g. Elenna Starwards as a name of Nmenor (see below), the English u ending -wards cannot be freely applied to any noun like the Quenya ending can. But if the day ever comes when Columbus lend Americanna e can be translated Columbus went Americawards, people thinking this is acceptable English, the language shall have acquired a living allative case.
NOTE: Besides -nna, there are also traces of an older allative ending in Quenya. In primitive Elvish it had the form -da, later reduced to -d (WJ:366). In Quenya, this -d became -z and later -r, and we have already met it in the word mir into (this is literally mi-r in-to, cf. mi in!) Since this ending came to clash with the plural ending -r, as in Eldar, it only survived in a handful of words indicating motion to or towards a point. Attested examples include tar thither, oar away, yar to whom and mir into. Actually into can also be minna with the normal, modern allative ending -nna. Likewise, thither can be tanna as well as tar.
In the plural, the ending -nna changes to -nnar, hence lassennar to leaves and ciryannar to ships (e.g. lelyuvan i ciryannar Ill go to the ships). The nal -r here appearing seems to be the same plural element that we are familiar with from the nominative. Since Quenya may express to, towards by means of a case ending, the question naturally arises whether there is an ending for from as well. There is. As we pointed out in Lesson 11, the genitive ending -o may occasionally take on this meaning, as in one word in Namri : Oiolosso = from a e e Oioloss (Mount Everwhite). However, the idea of from is more regularly e expressed by the ablative case, which is marked by the ending -llo. According to Plotz, we can have forms like lassello from a leaf and ciryallo from a ship (again accented on the second-to-last syllable, of course). So we may build sentences like tulin i ciryallo I come from the ship. For a Tolkienmade example, cf. the phrase mtima hrestallo from the last shore in e the Markirya poem. Both the ablative and the allative cases are exemplied in the phrase telmello talmanna from hood [telm] to base [talma], e that is, from top to bottom. (In the entry tel of the Etymologies as reproduced in LR, this expression is actually cited as telmello telmanna, but this is plainly a typo, for as is evident from the entry tal, the word for base, foundation, root is not telma, but talma.) As for the plural form of the ablative, there are apparently several options. Just like the sux -nna for allative turns into -nnar in the plural, the ablative ending -llo may have a plural equivalent -llor: In the Markirya poem, Tolkien used elenillor as the plural ablative of elen star, hence from stars. However, according to the Plotz letter, the plural ablative is to have the ending in -llon instead. Here we have the same plural marker -n as in the ending -on for plural genitive (the basic genitive ending -o + the plural sign -n, WJ:407). One of Tolkiens earlier tables of Quenya case endings lists both -llor and -llon as possible plural ablative endings. So I 213
come from the ships could evidently be both tulin i ciryallor and tulin i ciryallon. I generally prefer -llon, the Plotz variant, since the Plotz letter is our best late source regarding the Quenya case system but -llor must be considered a valid alternative. Dual allative/ablative: The dual forms of the allative and ablative endings include the already-familiar dual element -t, which replaces one of the consonants of the suxes -nna and -llo to produce -nta and -lto instead. Thus the nominative ciryat two ships, a couple of ships corresponds to an allative form ciryanta to(wards) a couple of ships and an ablative form ciryalto from a couple of ships. These are the examples Tolkien used in the Plotz letter, but again it is uncertain whether the same endings would be suxed to a noun that forms its nominative dual in -u rather than -t. Still using Aldu as our standard example, should to the Two Trees be Aldunta or simply Aldunna? Similarly, should from the Two Trees be Aldulto or simply Aldullo? I tend to think of Aldunna, Aldullo as the likeliest forms, but lacking attested examples we cannot be sure. Additional shades of meaning of the Allative and Ablative cases: While the primary implication of these cases is to(wards) and from, they may have other shades of meaning as well. The idea of actual, physical motion towards or from something is not always present. Notice the use of the ablative in a phrase found in J. R. R. Tolkien Artist and Illustrator : Itarild Ondolindello, Itarild from One e dolind, or using the better-known Sindarin forms: Idril from Gondolin. In e English, this is best rendered Idril of Gondolin, identifying Idril as a person living in Gondolin; the Quenya wording may not necessarily imply that Idril had actually left Gondolin. Possibly, the ablative can also be used in other ways that carry no implication of motion. It may be noted that regarding the Quenya verb ruc- to feel fear or horror, Tolkien wrote that it is constructed with from of the object feared (WJ:415). He did not provide any further information or examples, but from is regularly expressed by the ablative case in Quenya. So given that the Quenya word for monster is ulundo, perhaps I fear the monster would translate something like rucin i ulundollo. (Insofar as the words from and of express related meanings, this may be compared to such an English wording as Im afraid of the monster.) As for the allative, it does not always mean to(wards), but may also imply on, upon: The meanings are related insofar as an object that rests upon something also presses towards it, though there is no actual motion. This use of the allative may typically occur in connection with the verb caita- lie, as in this sentence from the prose Namri : Morni caita i a e e falmalinnar, darkness lies on the foaming waves (falma foaming wave, here not only with the plural allative ending -nnar but also the partitive plural marker -li, in this context possibly implying a great number of waves: In his interlinear translation in RGEO:67, Tolkien analyzed falma214
li-nnar as foaming waves-many-upon). Further examples of allative forms implying on, upon are found in the Markirya poem; we have atalanti e mindoninnar (or, mindonnar) upon fallen towers and axor ilcalannar on bones gleaming. However, the student should also notice that while the allative and ablative cases may not always imply physical motion to or from something, their basic meanings of to, towards and from may also be strengthened. Instead of just indicating motion towards something, the allative may also suggest motion into it: Attested examples include ari lantier cilyanna e seas fell into a chasm (LR:56) and mannar Valion into the hands of the Vali [Valar] (F riels Song). The ablative may likewise indicate motion out of something rather than merely from it: The word sindanriello o occurring in Namri Tolkien translated out of a grey country (though in a e the interlinear analysis in RGEO:67, he broke it down as sinda-nrie-llo, o grey-country-from). These additional uses of the allative and ablative cases may lead to some ambiguities: Is lenden i coanna to be interpreted I went to the house or I went into the house? Where confusion may arise, it is probably best to use the independent word mir (or minna) if into is the desired meaning: Lenden mir/minna i coa. As for out of as opposed to merely from, Elendils Declaration demonstrates that the word et out can be placed in front of an ablative form to clarify the meaning: Et Erello . . . ut lien, a u out of [or, out from] the Great Sea I am come. Some would even analyze et out as a preposition governing the ablative case (like u without governs the genitive case). Adding the allative and ablative endings to nouns ending in a consonant: Suxes like -nna and -llo and their dual/plural variants can never be added directly to a noun ending in a consonant without creating impossible consonant clusters. For instance, the allative to Elendil cannot be **Elendilnna, for Quenya phonology does not permit the group lnn. As is evident from the actual form Elendilenna occurring in PM:401, the language may work around this problem by inserting a connecting vowel e before the case ending. The ablative and allative forms occurring in Elendils Declaration in LotR may be examples of the same: et Erello out of the a Great Sea (Er: Quenya name of the Ocean), Endorenna to Middlea earth (Endor: Quenya for Mid-land = Middle-earth). However, the word Er is also cited in the form Er (SD:305), and Endor is shortened a a e from an older form Endr, so we cannot be absolutely certain that the o e es occurring before the case endings in the forms Erello, Endorenna a are not simply part of the nouns proper. On the other hand, the example Elendilenna almost certainly includes a connecting vowel e, for there is no reason to assume that the name Elendil ever ended in -. So the main e strategy for avoiding unwanted consonant clusters before case endings is probably to insert an -e- before the ending. 215
It should be noted, though, that in the case of a plural noun requiring a connecting vowel, it seems that -i- rather than -e- is preferred. We have already mentioned that in the Markirya poem, Tolkien used elenillor as the plural ablative form of elen star. If eect, the pl. ablative ending -llor has been added to the nominative plural eleni. One version of the Markirya poem also had mindoninnar as the pl. allative upon towers (before Tolkien decided to go for a contracted form instead; see below). Here, the pl. allative ending has been added to the nominative plural mindoni towers.
NOTE: Notice, though, that nouns in - with nominative plurals in -i (e. g. lass leaf, e e pl. lassi) do not change their nal - to -i before -nnar or -llon/-llor is suxed: Plotz e indicates that the pl. allative and ablative forms of lass are lassennar and lassellon, e respectively not **lassinnar, **lassillon. In this respect, the allative and ablative cases dier from the genitive case: A noun that forms its nominative plural in -i always receives this ending before the genitive plural ending -on is added the genitive plural of lass e being lassion, not **lasson. e
If one does not insert any connecting vowel, another way of getting rid of an unwanted consonant cluster is to simply omit the nal consonant of the noun that is to receive a case ending. Especially where the nal consonant of the noun is identical to the rst consonant of the case ending, these two consonants may simply merge. As indicated above, Tolkien rst used mindoninnar as the plural allative of mindon tower. But then he decided to drop the connecting vowel intruding before -nnar and introduced a contracted form instead: Mindonnar, which simply represents mindon-nnar. As we see, the nal -n of mindon merges with the rst n of the ending -nnar. A more-well known example is Elenna (for Elen-nna) as a name of Nmenor: After following the Star of Erendil across the ocean to their new u a land, the Edain called that land Elenna, which is Starwards (Akallabth; e cf. UT:317: Elennanr = the land named Starwards). In a similar fasho e ion, perhaps the ablative of Menel heaven could or even should be Menello (for Menel-llo) rather than Menelello.
NOTE: We may wonder how certain nouns with special stem-forms would be treated. In the case of talan, talam- oor, the allative to a oor or (up)on a oor might probably be expressed as talamenna with a connecting vowel inserted (the ablative should almost certainly be talamello), but perhaps we could also start from talan and use talanna (for talan-nna) as the allative form? And what about a noun like toron, tornbrother? Should to a brother be tornenna with a connecting vowel e inserted between the stem-form and the case ending, or may we simply say toronna for toron-nna? At this stage, we cannot know what Tolkien would have accepted as correct Quenya. I would not reject any of these alternatives as wrong.
It seems that the nal -n of the four directions Formen, Hyarmen, Rmen, N men North, South, East, West quite regularly drops out o u before the case endings for allative, ablative and locative (the locative case will be discussed in the next lesson). One haven in the east of Nmenor was u 216
called Rmenna, literally Eastwards (see its entry in the UT index, and o cf. LR:47) clearly because ships sailed eastwards from it. Of course, Rmeno nna > Rmenna as such is just another example of a nal consonant of o a noun merging with the rst consonant of the case ending because they happen to be identical. However, Namri provides Rmello from the a e o East as the ablative of Rmen East, and here there can be no doubt o that the nal -n has been omitted to avoid the impossible form **Rmenllo. o It may be that Rmenello with a connecting vowel inserted would also be o a valid form, but as I said, contracted forms seem to be normal when the words for four basic directions are to be inected for allative or ablative.
14.2
The verb equ: We have earlier introduced the Quenya word for say or e speak: quet- (aorist quet, present tense quta, past tense quent). Yet e e e this verb is not always used; there is an alternative word that may be used to introduce quotations. In WJ:392, Tolkien refers to . . . a curious and evidently archaic form that survives only in the languages of Aman: [primitive] *ekw, Q[uenya] eque, T[elerin] e epe. It has no tense forms . . . being mostly used before either a proper name (sg. or pl.) or a full independent pronoun, in the senses say / says or said. A quotation then follows, either direct, or less usually indirect after a that-conjunction [e.g., Galadriel said that she wants to go to Middle-earth] So as far as inection is concerned, this equ may well be the simplest verb e in the entire language. It has no tense forms, so equ may be interpreted e either as a past tense said or as present tense say(s), depending on the context (perhaps it could even cover the future tense shall say!) It is used mainly where the subject is a full independent pronoun (to be discussed later in this course) or a proper name (not a common noun). Also notice the word order indicated by Tolkien: The word equ comes before its subject. e Tolkien gave us no actual sentences containing the word equ, but based on e the information he provided, something like the following must be possible: Equ Elendil: Et Erello Endorenna ut lien. e a u Elendil says/said: Out of the Great Sea to Middle-earth I am come. Tolkien glossed equ not only as says, but also as say. Since say e must be understood as a plural verb in English, it seems that unlike normal verbs, equ does not receive the ending -r even where it has a plural subject e or more than one subject. Notice that Tolkien stated that equ is typically e used before . . . a proper name (sg. or pl.). Now proper names normally dont appear in the plural (except in sentences like there are many Johns in this town), so when Tolkien speaks of pl. proper names, he probably 217
means several proper names occurring together. So we must assume that a sentence like this would be acceptable: Equ Altariel ar Teleporno: Ut liemm Valinorello. e u e Altariel and Teleporno [Galadriel and Celeborn] say/said: We have come from Valinor. Tolkien indicated that equ rarely received suxes of any kind, usually e not even pronominal endings (WJ:392), though forms like equen said I may also occur (WJ:415). It cannot be wrong to replace equ with a form of the verb quet-, come plete with all the normal inections (Elendil quet/quent. . . Elendil e e says/said. . . , Altariel ar Teleporno quetir/quenter. . . Galadriel and Celeborn say/said. . . ) Where the subject is not a proper name or a full independent pronoun, it would seem that a form of quet- is usually to be preferred: I n quent. . . the woman said. . . Perhaps word order is also s e signicant. Tolkien seems to be saying that equ is used to introduce a fole lowing quotation; if the speaker and the act of speaking is mentioned after the quotation, it is perhaps better to use a form of quet-, e.g. Equ Elendil: Ut lien. = Elendil said: I am come. e u but: Ut lien, Elendil quent = I am come, Elendil said. u e Also where no direct or indirect quotation is included in the sentence at all, it is probably best to use a form of quet-. Cf. the attested example Sin Quent Quendingoldo Elendilenna referred to above (PM:401) e apparently meaning Thus Spoke Quendingoldo [= Pengolodh] to Elendil. Perhaps quent could have been replaced by equ here as well but probe e ably not. The verb auta-: This verb means pass or go away, leave (the point of the speakers thought) (WJ:366). Readers of the Silmarillion will have encountered it in chapter 20, as part of a battle-cry: Auta i lm! The o e night is passing! According to the rules so far set out in the course, this verb is quite irregular, though Tolkien may not have thought of it that way: In WJ:366 he refers to its various regular forms. Anyway, the past tense of auta- is not **autan as we might expect. There are actually several possible past e tense forms. One of them is anw, formed by nasal-inxion of the primitive e root-word awa; the ending -ta seen in auta- (primitive ?awat -) does not a appear at all in this past tense form. However, the form anw was only e found in archaic language, so we will concentrate on the modern forms instead. There are two sets of past and perfect forms of the verb auta-, with somewhat dierent shades of meaning. If the meaning is went away in a purely physical sense, about someone leaving one place and going to another, 218
the past tense form oant is used. According to Tolkien, this form is regular e for a -ta verb of this class (though most verbs in -ta seem to form their past tense simply by adding the ending -n). The past tense is supposed e to descend from awant, evidently a nasal-inxed form of awat, and in e a Quenya, these words regularly developed into oant and auta, respectively. e (For the shift awa > oa, cf. one word introduced in the previous lesson: hroa body, which Tolkien derived from primitive sraw.) The perfect a tense of auta- used in the same physical sense is oanti = has gone e away [to another place]. This perfect form is obviously inuenced by the past tense oant. Tolkien noticed that the form oanti shows intrusion of e e n from the past [tense form] (WJ:366): Normally, nasal-inxion does not occur in the perfect tense. The other set of past and perfect forms of the verb auta- seems no less irregular. The alternative past tense is vn, the perfect avni. The rst syla e a e lable of vn is apparently the Quenya descendant of the stem w (WJ:366, a e a apparently another manifestation of awa), whereas the ending -n must be e simply the normal past tense ending. (Again, the perfect form seems inuenced by the past tense form the n of vn sneaking into the perfect a e avni.) a e The form vn and the corresponding perfect avni have acquired a a e a e more abstract meaning than the forms oant, oanti. Vn does not e e a e mean went away (to another place), but rather disappeared, passed. The perfect avni occurs (with the plural ending -r) in Namri, in the a e a e sentence yni ve lint yuldar avnier = long years have passed like e e a swift draughts. This sentence nicely illustrates the meaning of this perfect form, for obviously the meaning is not that the yni or long years have e gone away to another place (sc. oantier!) They long years have simply passed, and now they are gone. Where the subject is more tangible than long years, the forms vn/avni would imply that the subject has a e a e disappeared, is lost, has died o etc. Indeed Tolkien indicated that the meaning of vn/avni was inua e a e enced by the related word vanwa gone, lost, vanished, past and over. It occurs twice in Namri : S vanwa n, Rmello vanwa, Vala e a o imar = now lost, lost [to those] from the East, is Valimar. In WJ:366, Tolkien calls vanwa the past participle of auta-, though it obviously has no connection with the past or passive participles we have discussed earlier in this course (constructed with the ending -na or -ina). There is some evidence for an alternative, rarer participle in -nwa. However, for all intents and purposes, it matters little whether we call vanwa a participle or merely a verbal adjective (as does Nancy Martsch in her Basic Quenya).
NOTE 1: As pointed out in Lesson Eight, it may well be that when Tolkien wrote Namri, he thought of the word avnier as the perfect form of a verb listed in the a e a Etymologies: vanya- go, depart, disappear (see the entry wan). We should still accept Tolkiens post-LotR ideas about the verb auta-; it occurs, after all, in such a primary
219
source as the Silmarillion. Interestingly, the adjectival word vanwa gone, lost, over is found already in the Qenya Lexicon of 1915 (QL:99) and was retained throughout all stages of Tolkiens development of Quenya. NOTE 2: In the Etymologies, entry gawa, a quite distinct verb auta- to invent is listed. It would seem that the later verb auta- go away did not exist at the time Tolkien wrote Etym. If we nonetheless accept both verbs as valid Quenya, we can distinguish between them in some tenses, for auta- to invent may perhaps have the simplest past tense: autan. e
14.3
So far, we have discussed ve pronominal endings that may be suxed to verbs to function as their subject: -ny (often shortened to -n) I, -ly e e you, -lm and -mm we (inclusive and exclusive), plus -nt they. e e e We have also pointed out that Quenya pronouns usually appear as endings, not so often as separate words as in English. Pronouns may also describe possession or belonging. Among the English possessive pronouns we have my (and mine), your(s), our(s), their(s). Quenya has endings for these pronouns as well, though these endings are logically added to nouns, not to verbs. For instance, the ending for my is -nya. Thus, my house is coanya, while my land would be nrenya. The accent now falls on the syllable before the pronominal o o ending: co-A-nya, n-RE-nya. All the pronominal endings begin in a consonant cluster, and in accordance with the normal rules, the stress falls on the second-to-last syllable when its vowel is followed by a group of consonants. Notice that the subject endings that we have already introduced, all end in the vowel -: -ny I, -ly you etc. The corresponding possessive e e e pronominal endings can be derived simply by changing the nal vowel to -a. Here are the attested correspondences: -ny I / -nya my e -ly you / -lya your e -lm we (incl.) / -lma our (incl.) e -mm we (excl.) / -mma our (excl.) e So besides coanya my house we can have coalya your house, whereas coalma and coamma would both translate as our house in English.
NOTE: As for the distinction between inclusive and exclusive our, it would closely correspond to the distinction between inclusive and exclusive we, explained in the previous lesson. Hence our house is expressed as coalma if the one(s) you are addressing is (are) also among the owners of the house and thus included in the word our. Conversely, coamma is the word to use for our house if you are talking to a party that is not among the owners of the house and hence is not included in the word our.
220
It seems very reasonable to assume that the ending -nt they has a e counterpart -nta their, though the latter sux is not attested in published material. One problem may seem to be that it would clash with the dual allative ending, but in context it would hardly be very dicult to tell whether (say) ciryanta is to be interpreted to a couple of ships or their ship. Presumably the endings could even be combined: ciryantanta, to their couple of ships! I wont construct any exercises involving the unattested ending -nta their, but I think it is safe enough to be recommended to writers. Combining pronominal possessive endings with endings for case and number: Combining these two kinds of endings is what truly makes the total number of forms that a Quenya noun can assume explode. We are left with hundreds of possible combinations, but since they are just that combinations the endings involved are not nearly as numerous, and the load on the students memory is not so great after all. Here follows sambelya your room (samb room, chamber + -lya e your) inected for the all the numbers and cases so far discussed in this course. If this list appears somewhat complicated and daunting at rst glance, the student will be relieved to discover that it is actually perfectly regular and in a way contains no new information at all: Just start from sambelya your room and treat it as you would any other noun in -a, adding the normal endings for number and case. One consequence of this is that the word now has a plural in -r (sambelyar your rooms), though samb room occurring by itself would be an i-plural (sambi rooms). e Nominative/Accusative: singular sambelya your room, dual sambelyat your couple of rooms, plural sambelyar your rooms.
(In the archaic form of Quenya that had a distinct accusative, we would presumably see the acc. sing. sambely and the acc. pl. sambelyai, but in this course, a we dont use distinct accusative forms.)
Genitive: singular sambelyo of your room (the genitive ending -o regularly displacing the nal -a of sambelya even though the -a is here part of another ending), dual sambelyato of your couple of rooms, plural sambelyaron of your rooms. Possessive: singular sambelyava of your room, dual ?sambelyatwa of your couple of rooms, plural sambelyaiva of your rooms. (While we here provide the same translations for the genitive and possessive cases, there are of course certain subtle shades of meaning that distinguish them.) Dative: sambelyan for your room, dual sambelyant for your couple of rooms, plural sambelyain for your rooms. Allative: sambelyanna to your room, dual sambelyanta to your couple of rooms, plural sambelyannar to your rooms. 221
Ablative: sambelyallo from your room, dual sambelyalto from your couple of rooms, plural sambelyallon (or, -llor) from your rooms. (In the case of the allative and the ablative, the accent falls on the vowel in front of the case ending [e. g. sambelyAllo], in accordance with the normal stress rules which apply for all the forms here listed.)
NOTE: Elendils Declaration includes the words sinom maruvan, ar hildinyar in e this place will I abide, and my heirs. From the example hildinyar my heirs, one might argue that plural nouns with plural forms in -i (like hildi heirs) should assume this ending before pronominal endings and secondary plural markers (like the -nya- and -r of hildinyar) are added. If so, your rooms should actually be sambilyar rather than sambelyar as we suggested above. This is possible, but the example hildinyar may have its own peculiarities; see below.
Notice that the possessive pronominal ending is added rst, and endings for number and case are added after it: From your room is therefore sambelyallo, not **sambellolya. For a Tolkien-made example, cf. the greeting Anar caluva tielyanna the Sun shall shine upon your path (UT:22, 51): The noun ti path here appears combined with the pronominal ending e lya your, and tielya your path is further expanded with the allative ending -nna upon to express upon your path. Another example of a noun equipped with both a pronominal ending and a case ending occurs in the most famous Elvish greeting of all, a star shines on the hour of our meeting: Elen s l menn omentielmo, as the verla u sion printed in the rst edition of LotR went. With this lesson we have nally presented all the grammar one needs to fully understand this sentence: Elen a star, s shines or rather is shining (the present/continuative la tense of the verb sil-), l menn or in full l menna on (the) hour (the u u noun l m hour + the allative ending -nna on), and nally the word u e that is relevant for our present discussion: omentielmo. This must be analyzed as an abstract noun (or gerund?) omenti meeting + the ending e -lma for inclusive our (WJ:367), and omentielma our meeting is then equipped with the genitive ending -o to express of our meeting: Hence omentielmo, since the ending -o displaces a nal -a. Dual inclusive our: In the revised edition of LotR (1966), Tolkien altered the word omentielmo to omentielvo, though the translation remained the same: of our meeting. Tolkien made up an internal explanation for this change, briey referred to in one of Humphrey Carpenters notes on the collection of Tolkiens letters that he edited (Letters:447, notes on letter #205): The Elvish language Quenya makes a distinction in its dual inexion, which turns on the number of persons involved; failure to understand this was, Tolkien remarked, a mistake generally made by mortals. So in this case, Tolkien made a note that 222
the Thains Book of Minas Tirith, one of the supposed sources of The Lord of the Rings, had the reading omentielvo, but that Frodos original (lost) manuscript probably had omentielmo; and that omentielvo is the correct form in the context. The point seems to be this: In the greeting a star shines on the hour of our meeting, the word our refers to a meeting of two people (Frodo and Gildor, the former speaking to the latter). Therefore, a dual our is appropriate here our meaning of you (sg.) and me. Frodos lost original manuscript instead had a plural our (ending -lma, or with the genitive ending -lmo), but this is the kind of our that is used to refer to something belonging to three or more people. It was inappropriate when only 2 persons, Frodo and Gildor, were involved in our meeting. Frodos little grammatical error was mercifully corrected in the Thains Book of Minas Tirith, one of the copies of his original manuscript. This internal explanation is of course mere rationalizing on Tolkiens part. From an external point of view, it is obvious what is going on: After publishing LotR, the Professor couldnt resist the temptation to continue tinkering with the Quenya pronoun table (a process that had already been going on for decades). Tolkien apparently decided to introduce special dual forms of (inclusive) we and our, used when only two persons are involved: the speaker and the one who is addressed. Unfortunately, this revision conicted with a piece of already-published Quenya, but even so, the Professor changed omentielmo to omentielvo and made up a little story that would explain the discrepancy from an internal point of view: Frodo just couldnt get his grammar right! Gildors remark to the other Elves, here is a scholar in the Ancient Tongue, suddenly ring rather ironic! Perhaps we are to imagine mocking smiles twitching in their lips ah, after all these centuries, mortals still cant get their dual pronouns right. . . So let us poor mortals try to gure it out: What, really, is the ending for dual (inclusive) our and the corresponding ending for dual inclusive we? The most straightforward extrapolation would go like this: Removing the genitive ending -o from omentielvo gives us omentielva, for surely this pronominal possessive sux ends in -a just like all the other attested endings of this kind. We proceed to isolate -lva as the ending for dual inclusive our, e.g. coalva our house of a house owned by two persons, one of them speaking to the other. According to the pattern observed in such pairs as -lya your vs. -ly you, we may further assume that our reconstructed e ending -lva our corresponds to an ending -lv we a dual inclusive e we meaning you (thou) and I: Lelyuvalv i ostonna we will go to e the city [, you and I]. Notice that the ending for inclusive we introduced in the previous lesson, namely -lm, would refer to a group of at least three e people.
NOTE: As far as is known, Quenya has no specic ending for dual exclusive we sc.
223
an ending meaning I and exactly one other person! It is assumed that -mm is used e for exclusive we, and -mma for exclusive our, no matter whether two or more people are involved. Cf. one of our home-made examples above; we put the words ut liemm u e Valinorello into the mouths of Galadriel and Celeborn. On the other hand, let none be surprised if it turns out that Tolkien actually did invent specic endings for dual exclusive we and our, to make the pronoun table wholly symmetric.
As I said, the most straightforward reconstruction based on the form omentielvo of our meeting would lead us to extrapolate the inclusive dual endings -lva our and -lv we. This reconstruction may well be e correct, and the student should notice these forms, for many writers have used them. Yet there are other possibilities as well. If we are to isolate the specically dual element of the word omentielvo, it has to be the v, that must somehow be related to the dual ending -u (as in Aldu Two Trees). But if u were to be transmuted into a consonant, we would rather expect it to become w, which is the corresponding semi-vowel the vowel u used as a de facto consonant. Therefore it has been suggested that the ending for dual inclusive our may not be -lva after all, but perhaps rather -lwa. According to this theory, adding the genitive ending -o produces -lvo instead of **-lwo for the simple reason that wo (unlike wa) is not a possible Quenya combination. Therefore, w (the consonant corresponding to u) here becomes v before -o. A possible phonological parallel is provided by some of the names quoted in PM:352353. One example is Telunw Last Finw, e e which is said to have a shorter form Telvo. This is apparently the initial element of the full name, telu-, + the sux -o (here functioning as a masculine ending). Notice how the -u of telu- has become a v in the name Telvo, evidently because of the following -o. Perhaps we are to imagine a development whereby a form Teluo is reduced to two syllables by pronouncing it Telwo, and this form in turn has to be altered to Telvo to get rid of the unwanted combination wo. Similarly, omentielwo could plausibly transmute into the attested form omentielvo. If the ending for dual inclusive our is actually -lwa rather than -lva, we may also assume that the ending for dual inclusive we is -lw rather e -lv. But extrapolating from a form that is itself reconstructed is of course e a risky business. Adding pronominal endings to nouns ending in a consonant: To avoid impossible consonant clusters, an extra vowel -e- may be inserted before the pronominal ending where necessary. As we remember, this extra vowel may also turn up before case endings. Combining atar father with -lya your to express your father would probably produce atarelya (since **atarlya is not a possible Quenya word). Our father would be ataremma (exclusive our) or atarelma (inclusive our) and either atarelva or atarelwa if our is a dual inclusive pronoun (the father of the two of us). It may be that if the noun is plural, one would use i as the connecting vowel, if pronominal endings and case endings behave in the same way. Cf. Tolkiens elenillor 224
for from stars. Thus, your stars might similarly be elenilyar, and our (excl.) fathers should evidently be atarimmar rather than ataremmar. (Of course, it is primarily the nal -r which functions as a plural marker here, so there can be no misunderstanding regarding the number anyway.) It may be that hildinyar my heirs from Elendils Declaration is an example of this, if the noun heir is ?hil with a stem hild-, hence pl. hildi. However, the ending -nya my seems to be somewhat special. Where a connecting vowel is required, it apparently always prefers -i-, whether the noun it is added to is singular or plural. It seems that this connecting vowel reects the vowel of the primitive root producing the Eldarin 1st person pronouns, namely ni2 (listed in the Etymologies and simply dened I). F riels Song has Anarinya, not **Anarenya, for my Sun. Similarly, my father is atarinya (LR:61) rather than ?atarenya; we cannot know whether the latter form is valid at all. The (nominative) plural my fathers would of course be atarinyar, so the singular and plural remain distinct. In the same fashion, the singular form of hildinyar my heirs would perhaps be hildinya my heir with the same connecting vowel i, since it is always preferred by the ending -nya. In the case of another ending, like -mma our, we might conceivably see a variation between hildemma (?) our heir and hildimmar our heirs; in the latter case, the -i is the normal nominative plural ending used as a connecting vowel. (The Etymologies, entry khil, lists precisely such a plural form hildi there glossed followers, close enough to heirs in meaning.) It has been suggested that the ending -nya, added to a noun in -, would e also displace this - with -i- (much like the plural ending -i displaces a nal - e e when added to a noun). However, one Tolkien example that was published in the summer of 2000 demonstrates that this is not so: VT41:11 has renya, o not **rinya, for my heart (r: heart in the sense of inner mind). o o e According to the system we are trying to sketch, even the plural my hearts would be renyar rather than ?rinyar, since r ends in - and thus o o o e e requires no connecting vowel before suxes anyway. Cf. the Plotz Letter: lassennar, not **lassinnar, as the plural allative of lass leaf though e the nominative pl. is lassi. In a similar fashion, we would probably see lassenya my leaf vs. lassenyar my leaves (hardly **lassinyar). The connecting vowel -i- only turns up where plural nouns ending in a consonant are to receive endings; singular nouns have -e- instead, except in the case of the ending -nya my which prefers -i- whether the noun it is added to is singular or plural. (If it is plural, this will be suciently indicated by the secondary endings for number and/or case that are added after the ending -nya.)
NOTE: Of course, we must assume that nouns in - that have stem-forms in -i- would e appear in the latter form when endings are to be added. So if l e (l r ri-) means song, my song would evidently be l rinya (plural l rinyar my songs). But this is actually a quite dierent matter, for here we would evidently see l before any sux, for pronoun ri-
225
In some instances, contracted forms are used instead of inserting any connecting vowel. UT:193 provides the form aranya, untranslated but apparently meaning my king (Erendis uses this word when addressing the King of Nmenor). This is apparently aran king + -nya my, the impossible u form **arannya being simplied to aranya. Possibly ?araninya would also be acceptable Quenya, but when the noun ends in the same consonant as the pronominal ending begins in, it may be permissible to let the last consonant of the noun and the rst consonant of the ending merge a phenomenon also observed where case endings are involved. (Cf. mindonnar rather than mindoninnar as the pl. allative of mindon tower; perhaps my towers would be mindonyar rather than mindoninyar.) Especially where the ending -nya my is concerned, contracted forms may turn up even where no contraction would be necessary to achieve a permissible Quenya word. The High-elven word for son is yondo, so my son might simply be yondonya, and there is little reason to doubt that this is a valid form. Yet in LR:61 Elendil addresses his son as yonya, apparently a contracted variant of yondonya. Perhaps yonya would be used for my son primarily when addressing the son concerned. If so, it would parallel another example: One Quenya word for child is h na, or hina with a short vowel the latter only being used when addressing a (young) child (WJ:403). Tolkien went on to note that this hina, used as a form of address, often appeared in the form hinya my child the latter being contracted from hinanya (still WJ:403).
226
name or an independent pronoun. The verb auta- pass, go away, leave has rather surprising past and perfect forms: oant or oanti if the verb e e refers to physically leaving one place (and going to another), but vn and a e avni if the verb refers to disappearing, being lost, or dying o. Quenya a e possessive pronouns are normally expressed as endings added to the relevant noun (the thing that is owned). These suxes include -nya my, -lya your, -lma our (inclusive), and -mma our (exclusive). Notice that these possessive endings correspond to the subject pronominal endings sufxed to verbs, the former ending in -a whereas the latter end in - (therefore e the unattested ending for their may well be -nta, corresponding to -nt e they). There is also a specic ending for dual inclusive our, but its form cannot be condently reconstructed (either -lva or -lwa; only -lvo with the genitive ending added is attested). Where required, connecting vowels may be tted in before the noun and the pronominal ending, probably by much the same rules that apply to the case endings -nna and -llo, except that the ending -nya my seems to consistently prefer the connecting vowel -i-. Once a noun has received a possessive pronominal ending, this noun may be further inected for number or case just like a regular noun in -a would be.
Vocabulary
We have now exhausted the basic numbers 112 (including the extrapolated number rasta). Higher numbers are unfortunately rather uncertain, though we have some clues. I may add some thoughts about this later, but in this and the next couple of lessons we will introduce the attested ordinal numbers showing order or position in a series, like English rst, second, third etc. minya rst (cf. the number min one and the adjectival ending -ya. The original e
name of the First Clan of the Elves was Minyar, literally Firsts, though the Noldor later called them Vanyar or Fair Ones instead [WJ:380, 382383].)
equ say(s), said (tense-less verb introducing quotations) e auta- to leave, to go/pass away (past tense oant and perfect oanti, alternae e
tively vn and avni, the latter two forms referring to disappearing or dying a e a e o as explained above). The past participle of auta- is said to be vanwa lost, gone, passed, vanished but this word may be treated almost as an independent adjective.
menta- to send ruc- to feel fear or horror; to fear (said to be constructed with from of the
object feared, presumably meaning that what would be the direct object in English appears in the ablative case in Quenya)
mindon (great) tower (cf. the Mindon Eldaliva or Great Tower of the Eldali e e
mentioned in the Silmarillion. The rst syllable of mindon is related to the number min one, since a mindon is an isolated tower, not part of a larger structure.) e
samb room, chamber (Sindarin sam, samm-; cf. the Sammath Naur or Chame
bers of Fire inside Mount Doom)
yondo son haira far, remote et out (followed by ablative to express out of)
In addition to our traditional list of twelve new glosses we will also introduce a couple of proper names, required in these exercises. In accordance with our established policy we will avoid explicit references to Tolkiens mythos in these exercises, so no proper names coined by him will appear here. Yet we can readily coin new names using his principles. The ending -(n)dil often occurs in masculine names and signies friend or lover, e. g. Erendil Sea-friend or Elendil Star-friend (but also implying Elf-friend since a the words elen and Elda are ultimately related and were even confused by the Edain: WJ:410). So we can venture, say, Calandil Friend of Light. As for feminine names, one observed pattern is that an adjective in -a can be turned into a fem. name by changing the ending to - (not to be confused with the plural form of the adjective). For instance, one of e the queens of Nmenor was called Ancalim, transparently formed from the superlative u e form ancalima brightest, exceedingly bright. (Similarly, masculine names can be made by changing the ending -a to -o or -on, cf. Sauron vs. the adjective saura foul, putrid one suddenly understands why the Dark Lord didnt permit his servants to use the name the Elves had given him!) Starting from a suitable adjective like nessima youthful, we can derive a plausible womans name Nessim Youthful One. However, the meaning of e the names Calandil and Nessim is of no importance for the exercises. e
Exercises
1. Translate into English: A. Lelyuvalm i mindonello i coanna. e B. Ily Eldar avnier Ambarello. e a C. I Naucor ut lier i orontillon; elendient i coannar ar u e s car limpelma. u D. I umi ohtari mapuvar i malta lielmava mentien harmal e mar haira nrenna. o E. I n oant coanyallo ar lend i s s e e renna. 228
F. I minya cirya tuluva N mello. u G. Quen ruc i rvillon, an amtient i aran liemmo, ar e a a e uvant auta nremmallo. e o H. Equ Nessim Calandilenna: Yondonya avni sambee e a e nyallo! 2. Translate into Quenya (and notice that inclusive our is meant to be a plural
pronoun throughout, since the exact form of the ending for dual inclusive our cannot be reconstructed with certainty):
I. Calandil said to Nessim: Your son has gone out of the house, e for all the boys went to the hill. J. From heaven [Menel: the sky] the sun is giving light to our (inclusive) world, and the darkness has passed. K. Calandil said to the evil king: You have sent your warriors to the tower to nd my sons. My thrall will protect the boys, and they will not be lost! L. The man having [arwa] the ships wanted to leave, and all the ships went away west[wards]. M. We (exclusive) went to a two-room apartment [dual of samb!], e and the man from the hills gave your son a great sword, saying [qutala]: The sword comes from a remote land, out of the e outermost West. (Outermost: use the superlative of haira.) N. All trees died and disappeared from our (incl.) land, and Calandil and Nessim said: We (excl.) will send our (excl.) thralls to nd e a land with [or, having = arwa] many trees. O. The maiden said to the animal: I fear [/Im afraid of] your big horns (dual). P. I went to our (incl.) room to gather my things, for I wanted to give my brother my rst book; the book lay [/was lying] on the oor.
Additional exercises
involving Quenya nouns combining pronominal endings with suxes for number and/or case
(Students may check the keys to the exercises above before proceeding to these exercises.) There exercises above include several examples of nouns with both pronominal endings and case endings, e.g. lielmava of our people. A skilled Quenyaist would be able to extract the meaning of such forms at a glance, indeed perceiving a single word like ostolmallon as something 229
like a single meaning, from our cities, without having to consciously break this down into osto-lma-llo-n city-our-from-plural. Of course, a skilled Quenyaist would also be able to readily produce such words, combining the relevant endings without hesitation. 3. Translate the following list of single Quenya words into English phrases.
NOTE: In the keys to this exercise, the following simplied equivalents are used: genitives and possessive-adjectival forms are all turned into of-constructions, dative forms are represented as prepositional phrases in for, whereas allative and ablative forms are represented as phrases involving the prepositions to and from, respectively. The same system is used in the English-to-Quenya exercises below, with specication of whether of is to be rendered as a genitive (gen.) or a possessive-adjectival (poss.) form. In these exercises, there are also a few examples of the sux for dual inclusive our, though conveniently always with the genitive ending attached, since we know with certainty (from the attested example omentielvo) what form it then assumes.
a) d) g) j) m) p) s) v) y)
b) e) h) k) n) q) t) w) z)
c) f) i) l) o) r) u) x)
4. Translate into single Quenya words (of = genitive or possessive as further specied, for = dative, to = allative, from = ablative): a) To your hills b) For our (excl.) peace c) Your two-volume book (use a dual form of parma) d) To your tower / To your towers (translate the sg. and pl. separately) e) Of (poss.) our (excl.) queen f) Of (poss.) my sisters g) From my sister h) Of (gen.) our (incl.) gifts i) Of (gen.) our (incl.) gold j) For our (incl.) joy k) Of (gen.) your wine 230
l) From your world m) Of (gen.) my sun n) For my king o) Of (poss.) our (excl.) son p) Of (gen.) our (excl.) cups q) For your pair of birds (use a dual form of aiw) e r) To our (excl.) double walls (use a dual form of ramba) s) From our (incl.) double walls (same) t) From your lands u) Of (gen.) my [two twin] sisters (use a dual form of seler, sell-) v) Of (gen.) our (excl.) treasures w) To our (incl.) horses x) Of (gen.) our (dual incl.) house (= of the house of the two of us) y) For my brother z) To our (excl.) trees
231
Lesson 15
The ending -rya and more about possessive pronominal endings. The Locative case. Relative sentences. Third Person obscurities.
15.1
In the previous lesson we introduced a series of possessive pronominal endings that can be added to nouns: -nya my, -lya your, -lma our (incl.) and -mma our (excl.); furthermore, there is an ending for dual inclusive our, either -lva or -lwa (only attested in the form -lvo with the genitive ending -o added). Ignoring a strange variant of the ending -lya your (-lda, to be discussed later), only one of the attested pronominal endings remains to be mentioned: -rya. It occurs twice in Namri. The rst time it a e is followed by the genitive ending, regularly producing the form -ryo: The relevant word if maryo, translated of her voice, the genitive form of o o marya her voice. The word ma voice is attested by itself elsewhere o (Etym., entry om; VT39:16). The second time -rya occurs in Namri, it also has another ending fola e lowing it, in this case the dual marker -t: the word mryat is translated a her hands, referring to a natural pair of hands (the word m hand is a also attested by itself). Anyhow, the ending -rya is seen to mean her, and from the Namari examples it is clear that it can be used and combined e with other endings just like any of the other pronominal suxes we have already discussed and practiced (samberya her room, samberyan for her room, samberyanna to her room, samberyallo from her room, samberyo and samberyava of her room. . . and so on with plural and dual forms: samberyar her rooms, samberyat her couple of rooms/her two-room apartment, etc. etc.) For thirty years, from The Fellowship of the Ring (with Namri in it) a e appeared in 1954 until Christopher Tolkien published The War of the Jew-
232
els in 1994, her was the sole known meaning of the ending -rya. In the meantime, we had one more example of -rya = her in the Markirya poem, which was published in The Monsters and the Critics in 1983 (though in Markirya, her does not refer to a person, but to a ship). But when WJ appeared in 1994, it became evident that the sux -rya actually covers not only her, but also his: Coarya is shown to be the Quenya for the house of him or his house (WJ:369, there spelt karya). Of course, the o form coarya as such could just as well mean her house, and conversely the Namri forms mryat, maryo could in another context mean his a e a o hands and of his voice: We have to conclude that Quenya simply does not make a distinction between his and her. Indeed it is entirely possible that -rya covers its as well (see below) so that there is one single ending for the entire third person singular in the table of possessive pronouns. The English translation would depend on the context, of course. There is more to learn from the two examples of -rya in Namri. Notice a e the dual form mryat her (pair of) hands. As described in Lesson Three, a Quenya developed a system whereby -t is the normal dual ending, ordinarily replaced by -u only where euphonic concerns demand this, as when the word that is to receive a dual ending already includes t or d (Letters:427, footnote). But in Lesson Three we also argued from the example peu (pair of) lips that body-parts occurring in pairs occur in fossilized dual forms, always taking the ending -u reecting the older system in which only -u denoted a natural or logical pair. Nonetheless, the student may also remember a parenthetical warning to the eect that the other ending -t may however be used if certain other endings intrude before the dual ending itself; we will return to this in a later lesson. It is time to have a closer look on this. It has often been assumed that removing the ending -rya her from mryat her hands would simply leave us with mt (a pair of) hands. a a Yet since the dual form of p lip is attested as peu, we might reasonably e assume that the dual form of m hand is likewise mau pair of hands, a though the latter form remains unattested. If the noun that normally has a dual form in -u is to receive a possessive pronominal sux, it seems that the dual ending -u is suppressed and duality is instead expressed by means of the ending -t, suxed after the pronominal sux as in mryat. Though the a dual (pair of) lips is peu, we can assume that her (two) lips would be constructed by starting from the singular form p lip and adding -rya for e his/her and then -t for dual number, so that as a parallel to mryat we a would see pryat. (It then follows that the genitive is pryato, the dative e e pryant, the allative pryanta, the ablative pryalto, etc.) Aldu may be e e e the normal dual pair of trees, but her pair of trees would perhaps be constructed from the singular alda with the appropriate suxes, producing aldaryat. Even so, we may suspect that the dual ending -u could function as a connecting vowel where one is needed just as the plural ending -i 233
is known to function in certain instances. The word for foot is tl with a stem tal-, so perhaps the dual (pair of) feet is talu. Adding a possessive pronominal ending to tl, tal- would however require a connecting vowel a before we can even think about adding -t as a dual marker at the end of the word. Should her pair of feet perhaps be something like taluryat with double dual markers, -u- and -t, just like there would apparently be double plural markers (-i- and -r) in a plural word like (say) talimmar our feet? If so, this would be an exception to the apparent rule that the dual marker -u is not used before a possessive pronominal sux. As usual, we lack attested examples, but since Elendils Declaration indicates that my heirs is hildinyar, it would not be wildly implausible to assume that a corresponding dual form would be something very much like hildunyat. (Or maybe the rule that -nya my prefers -i- as its connecting vowel would prevail, producing hildinyat, but we might still see -u- as a connecting vowel before other pronominal endings, e.g. hilduryat her pair of heirs.) Another thing to be learnt from the Namri examples mryat her a e a hands and maryo of her voice has to do with whether ry here counts o as a consonant cluster (r + y) or as a single consonant: palatalized r. What we learn is however somewhat paradoxical. We touched on these problems already in Lesson One, but a new inquiry may be in place here, since the combinations in -y (like ry, ly, ny, ty) occur in several of the possessive pronominal endings. Tolkien repeatedly indicated that maryo is accented o on the a in the second-to-last syllable (in one of his Namri transcripts a e in RGEO, he indicated all major and minor stresses in this song, and we also have two actual recordings where he is heard to use this accentuation). For maryo to be accented in such a way, ry must count as a consonant o cluster, not as a single consonant. If ry were a single consonant, the normal rules dictate that the stress would not land on the vowel before it, but on the third syllable from the end. Yet we have repeatedly referred to another observed rule of Quenya phonology: there cannot be a long vowel in front of a consonant cluster. Thus the long vowel of m hand is logically shortened in the plural allative form a mannar into . . . hands, attested in F riels Song. **Mnnar would not a be a possible Quenya word. So if ry is also a consonant cluster as we thought we had just established, why is not shortened in the form mryat? Why a a dont we see ?maryat as a parallel to mannar? Frankly, I cant think of any obvious explanation. Apparently we must simply accept that ry as well as ly, ny, ty count as consonant clusters for the purpose of stress, but a preceding long vowel does not have to be shortened. Thus we would have mrya his/her hand, mnya my a a hand and mlya your hand with the preceding long vowel intact. Bea fore the other attested pronominal suxes, it would have to be shortened, since these endings unquestionably introduce a following consonant cluster: malma and mamma = our hand (inclusive and exclusive). **Mlma, a 234
**mmma would hardly be possible Quenya words. Such variations would a closely parallel a couple of attested forms we have referred to earlier, though they involve subject endings (-mm for we and -ny for I) rather than e e the pronominal possessive endings added to nouns: The exclamation v a signifying refusal has its long vowel shortened before the cluster mm in vamm we wont, but the long vowel is seem to persist in vny I e a e wont (WJ:371). So we can tell that while mm is unquestionably a cluster, ny may well count as a single consonant palatalized n like Spanish n. There are only a handful of nouns that can be aected by these variations in vowel-length, words of a single syllable that end in a long vowel: Besides m hand, only c bow, p lip, r day (24 hours) and l time, a u e e u occasion spring to mind if we dont bring in Tolkiens early Qenya material as well. Of course, the long vowels of these words would also be shortened before case endings introducing a following consonant cluster, as indicated by the plural allative mannar into hands in F riels Song. But into your hands would evidently be mlyannar, or mlyanta as a a a dual form since ly, ny, ry, ty do not count as consonant clusters for this purpose. On the other hand, there is also some evidence suggesting that these combinations should be taken as clusters. In a Namri manuscript reproduced a e in RGEO:76, Tolkien split the word maryo into its constituent syllables o and seemingly indicated that -ar- and -yo are separate syllables as if ry is a genuine consonant cluster after all, not just palatalized r. (Sure enough, r would probably be palatalized before y, but if y is also to be sounded as a distinct consonant, we would still have a cluster.) Likewise, Tolkien split the words fanyar clouds and ily all into fan/yar, il/y. If ry, ny, e e and ly, and by implication ty as well, really are to be taken as consonant clusters when they occur in the middle of words, that would explain the observed stress patterns. But then we are left with the problem of why long vowels are not shortened before these combinations. Luckily, these seeming inconsistencies cause no problems to people trying to write in Quenya, since we can simply imitate the system or systems that Tolkien used. Even so, I havent bored the student with the paragraphs above only as an academic exercise, for there remains the problem of how ly, ny, ry, ty occurring in the middle of words should really be pronounced: Are we dealing with single, unitary palatalized consonants, long palatalized consonants, or single consonants followed by a distinct y? It seems that we cant reach any denite answer based on what has been published so far. When Tolkien in RGEO:76 syllabied fanyar as fan/yar, it seems to demonstrate that he at least cant have the pronunciation **fa-ar in mind, though palatals n like ny and ty must always be pronounced as single, unitary consonants when they occur initially (as Quenya cannot have consonant clusters at the beginning of words: SD:416417). The choice apparently stands between fa-ar (with a long or double palatal n ) and fan-yar or rather fa-yar (a nn n 235
distinct y being sounded). In either case, a word like atarinya my father (that is, atarina or atariya) would then logically be accented on the i n n according to the normal rules. Why this combination ny, as well as ly, ry, ty, apparently lack the power to make a preceding long vowel become short remains a mystery. If they are pronounced with a distinct y, as I tend to think, these combinations may not be counted as regular consonant clusters because y is a semi-vowel rather than a quite proper consonant. Possessive pronominal endings used with innitives: In Lesson Ten we described how innitive forms of verbs have an extended form in -ta which is used when the innitive is to receive a sux denoting an object pronoun: thus car (cari-) to do, but caritas to do it or doing it. To such an e extended innitive it is also possible to add a pronominal ending denoting the subject of the verbal action. Our attested example is caritalya(s), which Tolkien translated your doing (it) (VT41:17). You is here the subject of the verbal action (that is, the doing), and this is expressed as a possessive pronominal ending -lya your. A second pronominal ending, denoting the object, may then be added at the end of the word: caritalyas, your doing it, tiritanyat my watching them. Such a phrase can probably be used as a noun, functioning, for instance, as the subject or object of a sentence. Perhaps I want you to watch them would be expressed something like merin tiritalyat, literally I want your watching them. The object of the innitive could certainly be an independent word as well, e.g. merin tiritalya i seldor, I want you to watch the boys (I want your watching the boys). By their meaning, such innitives would come very close to gerunds. Indeed we must assume that regular gerunds (in -i) may also receive possese sive pronominal endings, e.g. tulierya his coming (tuli coming). It is, e however, uncertain whether a second pronominal ending denoting the object may then be suxed (?carieryas his doing it).
15.2
In connection with the forms mir, minna into we have referred to the Quenya preposition mi in, which is sometimes combined with the denite article to produce the form (mi + i =) m in the. It occurs in Namri, a e in the phrase m oromardi, translated in the high halls (so in RGEO:66, at least the text in LotR has mi with a short vowel, though this should be a simple in with no article incorporated, and indeed the translation provided in LotR goes simply in lofty halls). Yet Quenya often dispenses with prepositions, using special case forms instead, as when to, towards is normally expressed by the allative ending -nna, whereas from is usually expressed by means of the ablative ending -llo though Quenya does have separate prepositions that could express
236
the same meanings. It should be no surprise, then, that Quenya instead of using a preposition like mi often prefers a specic case form in order to express the meaning of in (or on, upon). The relevant case is called the locative, marked by the ending -ss (probably inspired by the Finnish e ending -ssa, -ss of similar meaning). For instance, in a house can be exa pressed as coass, in the house could be i coass, in my house would e e be coanyass, etc. (Of course, the stress moves to the vowel immediately e preceding the case ending, since the ending begins in a consonant cluster.) The locative can refer to location in time as well as space: In an early version of the greeting a star shines on the hour of our meeting, Tolkien had the noun l m hour appearing in the locative case (l mess, RS:324). u e u e
NOTE 1: Students should notice that the ending -ss is not always a locative marker, e meaning in or on wherever it occurs. Sometimes -ss functions as an abstract ending. e We have already introduced the noun alass joy, merriment. Entuless is attested as e e the name of a ship, said to mean Return (UT:171; entul- would be the verb to re-come = to return). Caimass could be the locative form of caima bed, but caimass is e e also used as a noun lying in bed = sickness, and this is even the basis of the adjective caimassa bedridden, sick (Etym., entry kay). Sometimes -ss as a noun ending is not e e abstract, but it is seen to maintain the connotations of locality that it also has when used as a locative ending: The noun aicass mountain peak is derived from the adjective e aica sharp, so the term aicass basically refers to some kind of sharp place. The e ending -ss also turns up in the names of a couple of the months of the Elvish calendar, e listed in LotR Appendix D: V ress and Ltess, roughly corresponding to April and e o e May. The meaning of the word V ress is uncertain, but Ltess certainly connects with e o e lt ower and would seem to mean essentially In Flower, a tting description of the o e month of May. Whether the locative ending -ss could or should be attached to a noun e already ending in -ss is uncertain. Ltessess does seem like a rather cumbersome way e o e of expressing in May, and aicassess for on a mountain peak is not much better. e Instead of adding the locative ending to nouns of such a shape, it may be better to use the preposition mi in: Mi Ltess, mi aicass. o e e NOTE 2: As we remember, the allative case in -nna does not always denote motion towards something, but may also express the idea of on, upon. In some contexts, it would perhaps be permissible to use either the locative or the allative, resulting in pretty much the same meaning (caitan caimanyass = I lie in my bed / caitan caimanyanna e I lie on my bed). Yet Tolkien sometimes translated a Quenya locative form using the English preposition upon. Cf. for instance ciryass upon a ship (MC:216, there spelt e kiryasse); cf. also mahalmassen below.
In the plural, the simple locative sux -ss is expanded with the same e plural element -n that is also seen in the plural forms of the endings for genitive (-on) and ablative (-llon). Thus, plural locative forms end in -ssen. The plural locative of mahalma throne occurs in Cirions Oath, where the Valar are referred to as i hrar mahalmassen mi N men, those a u who sit upon [the] thrones in the West. The dual locative ending is formed by substituting the dual element t 237
for the rst of the ss of the ending -ss. The resulting ending -ts is not e e attested in any actual Quenya composition by Tolkien, but he listed it in the Plotz letter, so presumably we can have forms like sambets in a two-room e apartment or ciryanyats on my [two sister] ships. (These words may e be seen as the simplest dual forms sambet, ciryanyat with the locative ending -ss attached, though it is simplied to -s to avoid the impossible e e combination **-tss.) e Of course, endings like -ss, -ssen, -ts can never be added directly to a e e noun ending in a consonant without producing impossible consonant clusters. Lacking attested examples, we can only assume that connecting vowels would be tted in by much the same rules as the ones that are seen to apply in the allative and ablative cases: -e- is used as a connecting vowel in e the singular, whereas plural forms have -i-. Hence presumably eleness in a star, elenissen in stars. The dual in a couple of stars might prefer the connecting vowel -e- (?elenets). Contracted forms may also turn e up, e.g. eless for elen-ss. The directions Formen, Hyarmen, N men, e e u Rmen = North, South, West, East would almost certainly surrender their o nal -n in the locative, just as they are seen to do in the allative and ablative cases. Hence probably Formess in the North, etc. F e riels Song has N messier for they are in the west. This strange form seems to include u the ending -i is, pl. -ier (they) are which Tolkien probably dropped e later. Even so, an underlying locative form N mess in the West must in u e any case be presupposed here. Since the noun N men West also appears u in the shorter form N m, we cannot be certain that a nal -n has dropped u e out here, but this locative form may be noted all the same. Would nouns with dual forms in -u also form their dual locatives in ts, or does this ending occur only where we are dealing with nouns that e have nominative dual forms in -t? We may well wonder what the locative form of Aldu Two Trees would be. Alduss with the simplest ending e ss, because duality is already suciently expressed by -u? Aldats formed e e from the uninected form alda? Alduts with double dual markers, -u and e -t-? Personally I lean toward Alduss, but I would like to see a Tolkiene made example. The locative ending(s) can of course be combined with possessive pronominal endings just like the other case endings we have discussed. The Markirya poem has ringa s maryass for in her cold bosom (ringa cold, s ma u e u bosom; the reference is to the bosom of a ship).
15.3
Relative sentences
In LotR, there is one single example of a Quenya locative. The ending -ssen for plural locative occurs in Namri, in the phrase Vardo tellumar. . . , a e yassen tintilar i eleni = Vardas domes. . . , wherein [or, in which] the
238
stars twinkle. . . The word ya which, here appearing with the locative ending -ssen to imply in which, is a relative pronoun. It can be used to build relative sentences, that is, sentences embedded in other sentences as a kind of descriptive phrases. Two sentences like the treasure is great and you found it can be combined as the treasure which you found is great. Notice that the pronoun it of the sentence you found it is replaced by which. This relative pronoun is capable of referring back to the words the treasure, and which I found now becomes a descriptive phrase providing extra information about the treasure. The probable Quenya equivalents of these examples: I harma n alta the treasure is great a + hirnelyes you found it e a = i harma ya hirnely n alta the treasure which you found is great As a relative pronoun, English may also use that (the treasure that you found. . . ) In German, the denite articles der, das, die (all = English the, for various genders and numbers) are also used as relative pronouns. The Quenya article i may likewise take on this function. This is evident from Cirions Oath, the last words of which exemplify i used rst as article, then as rele e ative pronoun: . . . i Eru i or ily mahalmar a tennoio, the One who is above all thrones forever. Since the One (Eru, God) is a person and not a thing, the relative pronoun must here be translated who rather than which. If there is any distinction in meaning between i and ya used as relative pronouns, this may indeed be it: i refers back to a person (English who), while ya refers back to a thing or a situation (English which). Notice, by the way, that these glosses have nothing to do with the questionwords who and which: The word i cannot be used for who in a question, like who are you? The Quenya word for who in this sense is quite dierent (man).
NOTE: This who/which distinction is not the only possible interpretation of our scarce examples. By another suggestion, i is used when the relative pronoun is the subject of the relative sentence, while ya is used when it is the object: Elda i tir Nauco an e Elf who watches a Dwarf, but Elda ya tir Nauco an Elf whom a Dwarf watches e (English uses whom as the object form of the relative pronoun who). We need more examples before we can pick the right interpretation with condence.
The word i is in Quenya the indeclinable article the (Etymologies, entry i). That is, i = the cannot be declined; it cannot receive any case endings. We must assume that this is still true when i functions as a relative pronoun who instead. However, ya is perfectly able to receive case endings, as indicated by the example yassen in which from Namri. The locative a e ending is plural because the relative pronoun refers back to a plural word, 239
tellumar domes; in the case of a single telluma or dome, the relative pronoun referring back to it would likewise be singular: yass. Likewise with e other nouns: coa yass a house in which. . . , but plural coar yassen. . . e houses in which. . . Besides the form yassen in Namri, we have one more example of ya a e occurring with a case ending. An early Elvish poem by Tolkien includes the words tanya wende . . . yar i vilya anta miqilis, translated that maiden . . . to whom the air gives kisses (MC:215, 216). This is not quite LotR-style Quenya, so I dont regularize the spelling, but the form yar to whom is interesting. The nal -r here suxed to ya seems to be the old allative ending, as in mir into; hence yar = whom-to, to whom. The examples yassen in which and yar to whom suggest that if you need a relative pronoun to receive case endings, such endings are always attached to ya-. Even if ya means which and i means who as we theorized above, a form like yassen may well cover both in which and in whom (since we cant have a distinct form **issen for the latter meaning if i is indeclinable). We must assume that ya can receive all the various endings for number and case, being inected like a noun in -a, as in these examples: Dative: i nr yan nen annanya the man to whom I gave my e a gift, plural i neri yain. . . the men towhom. . . (The attested form yar to whom, occurring in a context involving the verb give, may evidently also take on dative-like functions but yar is properly an archaic allative, and generally I think yan, pl. yain, is to be preferred.) Genitive: i n yo yondo cennen the woman whose [= whos] son I s saw (we must assume that ya + the genitive ending -o would produce yo, a nal -a being displaced as usual), plural i nissi yaron. . . the women whose [= whos]. . . (for a form like yaron, cf. aldaron as the plural genitive of alda tree) Possessive: i aran yava malta mapuvan the king whose [= whos] gold I will seize, plural i arani yaiva. . . the kings whose [= whos]. . . Allative: i coa yanna lenden the house to which I went / the house whither I went / the house that I went to, plural i coar yannar. . . the houses to which. . . Ablative: i coa yallo tullen the house from which I came / the house whence I came / the house that I came from, plural i coar yallon [alternatively, yallor]. . . the houses from which. . . Locative: i coa yass marin the house in which I live / the house e that I live in / the house where I live, plural i coar yassen. . . the houses in which. . . In the nominative singular, the simple form ya is of course used: i parma ya etcien, the book which I have written. It may be that this would e 240
become yar (with the plural ending -r) when referring back to a plural word: i parmar yar. . . the books which. . . (Distinguish the attested relative pronoun yar to whom, MC:215, 216; this form includes the old allative ending -r instead.) Where i is used as a relative pronoun, it receives no plural ending, since i is indeclinable: Eldar i lindar Elves who sing. We have listed no dual forms, but they would presumably be quite regular: nominative yat (e.g. i peu yat. . . the [pair of] lips that. . . ), dative yant (e.g. i veru yant. . . the [married] couple to/for whom. . . ), genitive yato, possessive yatwa (?), allative yanta, ablative yalto, locative yats (e.g. i sambet yanta/yalto/yats. . . the two-room apartment e e to/from/in which. . . ) It should be noted that in some grammatical contexts, a case ending that could have been added to ya may be omitted and understood. For instance, given that the word for night is lm (lmi-), we could presumably have o e o a sentence like lmiss yass cennenyes in [the] night in which I saw o e e it = (in/on) the night when I saw it, but it is also permissible to let ya occur by itself: Lmiss ya cennenyes, a construction paralleling English o e in [the] night that I saw it (very frequently, English would drop the initial in, but in Quenya the locative ending should probably be included). Notice that the article may be dropped before the rst noun (lmiss in o e our example); it is perhaps suciently determined by the following relative phrase. Tolkien employed such a construction in his Quenya translation of the Hail Mary (he paraphrased in the hour of our death as in [the] hour that we shall die). Usually, a relative pronoun refers back to a noun so that the following relative sentence provides information about that noun, as in all the examples above. Notice, however, the example i carir quettar those who form words, quoted as a description of the Elves (WJ:391). I carir quettar by itself is a relative sentence, and we could certainly connect it with a noun and let the relative sentence refer back to it, e.g. Eldar i carir quettar Elves who form words. However, it appears that i can be put in front of a verb to express the one who (if the verb is singular) or those who, the ones who (if the verb is plural, marked by the ending -r). Cirions Oath provides another example: i hrar mahalmassen mi N men those who sit a u upon thrones in the West. We can probably feel free to build sentences like these: I linda n n [the one] who is singing is a woman e a s I hirner i malta nar aly [the ones] who found the gold are rich e Hiruvan i suncer limpenya I will nd [the ones] who drank my wine (singular . . . i sunc limpenya, [the one] who drank my wine) e If ya can also be used in such constructions, and we are right to assume that i signies who while ya means which, there may be distinctions in meaning like ecnien i t la I have seen [the one] who is coming vs. e u 241
ecnien ya t la I have seen what is coming (literally, I have seen [that] e u which is coming). The sentence what I want is wine would perhaps translate something like ya merin n limp (i.e., [that] which I want is wine). a e Word order: Some languages employ a special word order in relative sentences. German insists on placing the verb last, so that we have constructions like the man who there stands (der Mann der dort steht) for the man who stands there. For a while I wondered if Quenya employed a similar system; the verb a is, exists appears near the end of the relative sentence e concluding Cirions Oath: i or ily mahalmar a tennoio, literally who e e over all thrones is forever. However, as we see, the verb is not absolutely nal; an entirely German word order would require who over all thrones forever is. In Namri, the verb actually follows immediately after the relative proa e noun in the relative sentence yassen tintilar i eleni wherein the stars twinkle, literally in which twinkle the stars. We might think that this is just a poetic word order, but Tolkien did not change it in the prose Namri in RGEO:6667. Does it make any dierence that this is a relative a e pronoun with a case ending attached? Would it be wrong to say yassen i eleni tintilar, with the subject of the verb preceding rather than following the verb? We cannot tell. Especially in the case of yasse(n), yanna(r), yallo(n) in/to/from which, I would imitate our attested example and let the verb immediately follow the relative pronoun: I osto yass mar i e e nr the city in which the man dwells, i tol yanna c e rar i ciryar the island whither the ships are sailing, i nri yallon tulir i ohtari the o lands whence the warriors come. Otherwise, I will not try to make out any hard-and-fast rules for what word order Quenya relative sentences should have.
15.4
Above we introduced the possessive pronominal ending -rya, covering his and her. So what is the corresponding subject ending, meaning he and she? Since the ending -lya your is known to correspond to an ending -ly e you, many researchers, starting from -rya her, have extrapolated an unattested sux -ry as the subject ending = she. If, as indicated by e Namri, the Quenya for you will nd is hiruvaly, she will nd would a e e then be hiruvary. Nancy Martsch uses this extrapolated ending -ry she e e throughout her Basic Quenya and it may well be correct. Now that it is known that -rya covers his as well as her, we would have to assume that -ry may similarly signify he as well as she. e The subject endings of the Third Person Singular the endings for he, she, and it however belong to one of the more obscure parts of the
242
Quenya pronoun table. In material closely related to F riels Song, one ending for he is seen to be -ro. It occurs in the form antavro he will give, a attested in the question e man antavro? what will he give indeed? a (LR:63). Antva as the simple future tense will give occurs on the same a page (and in the full text of F riels Song as printed in LR:72). This may not be quite LotR-style Quenya; as we argued in Lesson Seven, the future tense of anta- should perhaps be antuva rather than antva according to a the system Tolkien decided upon later. Even so, the form antavro nicely a illustrates one apparent property of the ending -ro: For some reason, the vowel immediately preceding this ending is lengthened, antva becoming a antavro when -ro is added (and the original long vowel of antva is a a shortened to avoid the form **antvro: It may be that Quenya cannot a a have a long vowel in the syllable immediately preceding the vowel receiving the main accent except when this syllable is also the rst syllable of the word). Should we update antavro to something like antuvro in LotRa a style Quenya? This ending -ro also turns up in a Qenya poem reproduced in MC:220, there added to a couple of verb forms including the past-tense ending -n, e and again the vowel preceding -ro is lengthened so that it receives the accent. One of them is laustanro, which would seem to be a verb lausta- make e a windy noise (cf. MC:216) + the past tense ending -n + the ending e ro he (and it?) The whole phrase goes s ru laustanro, translated u e the wind rushed (literally perhaps [the] wind, he/it [-ro] rushed). Since this is Qenya rather than LotR-style Quenya, we shouldnt put too much emphasis on the details, but Tolkien does seem to be using an ending ro, that may mean he (but also it?), and that has the strange power of making the preceding vowel long. It has been suggested that the vowel rather remains long in this position because Tolkien imagined it to have been long in Primitive Elvish. If so, the vowel -i- seen in the aorist of primary verbs (e.g. tulin I come) should not be lengthened, since this vowel was never long (?tuliro rather than ?tul for he comes). ro Where would this ending -ro for he come from, and what is the ending for she? The entry s- in the Etymologies throws some light on what Tolkien imagined. Various Elvish words for he, she, it are there discussed. One primitive word for he is cited as s or so, cf. -so inection of verbs o apparently meaning that the primitive Elvish language might express he by means of an ending -so added to verbs. This -so could be the origin of the Quenya ending -ro, for in Quenya, -s- occurring between vowels was normally voiced to -z-, which later became -r- (the sound z merging with original r). In the Etymologies, Tolkien went on to cite one primitive word for she as s or si, cf. -se inexion of verbs. If -so produces -ro as a Quenya ending for he, we would have to assume that -se similarly yields -r (earlier -z) as an ending for she. This -r is possibly directly attested e e e in the Qenya phrase kirya kallire, translated the ship shone (MC:220, e 243
221) literally [the] ship, she shone? Turning the form kallire into LotRe style Quenya would probably take more than just altering the spelling to callir, but it may be noted that the ending -r, like -ro, seems to prefer e e e the company of a long vowel in the preceding syllable. Many writers have used the endings -ro = he and -r = she, so e students of Quenya should certainly memorize them but as far or short as we know, they are only attested in material predating the writing of LotR. In 1994, there nally turned up a tiny scrap of evidence regarding Tolkiens post-LotR ideas about the pronominal ending for he, she. In the essay Quendi and Eldar, in the discussion of the tense-less verb equ said, says, e Tolkien noted that while this form normally does not receive endings of any kind, it may occur with certain pronominal endings. He cited two examples of this: equen, translated said I, and also eques, translated said he / she (WJ:414) or said he, said someone (WJ:392). So here we have an ending -s that covers both he and she (or even someone). In the postLotR period, Tolkien demonstrably used the ending -rya for both his and her, so it is not surprising that he might have decided that Quenya used one ending for both he and she as well (cf. also the Finnish genderneutral pronoun hn.) Actually this ending -s must also cover it, for it a can hardly be kept apart from the ending -s that we have already met in object position as in tiruvantes they will keep it (Cirions Oath) or caritalyas your doing it (VT41:17). So eques could probably mean it said just as well as (s)he said. Conversely, -s may probably refer to people in object position as well: Perhaps tiruvantes might also mean they will keep [or, watch] him/her. A form like tulis would have to be translated either he comes, she comes, or it comes depending on the context. The existence of such an ending does not necessarily contradict the references Tolkien made to primitive -so inexion and -se inexion of verbs in the Etymologies: Normally, the nal short -o and -e of primitive Elvish have been lost in Quenya, so primitive forms like tuli-so he comes and tuli-se she comes might well merge as tulis (s)he comes. Where this would leave the longer, genderspecic endings -ro and -r found in early material is unclear. Tolkien may e have meant them to descend from variant endings with long vowels (-s and o -s ), nal - and - becoming -o and - in Quenya. Perhaps the gendere o e e specic endings would be used where the short, general 3rd person ending -s he, she, it is not specic enough? But there is every reason to believe that Tolkien repeatedly changed his mind about the details; we cant even rule out the possibility that the long endings -ro he and -r she were e dropped altogether. Anyhow, if -s is to be the ending for (s)he, where does this leave the unattested ending -ry that some students have (plausibly) extrapolated e from the possessive ending -rya his, her? The ending -ry may still be e valid. Perhaps the ending for (s)he alternates between -s and -ry just e 244
like the ending for I may appear as either -n or -ny. (While the endings e -s and -ry may seem less similar than -n vs. -ny, it should be understood e e that -ry would come from earlier -sye: Following a vowel, the combination e sy turns into zy and then ry. Cf. the Etymologies, entry sus; from this root, Tolkien derived the Quenya word surya spirant consonant, which must be understood to come from susy in the primitive language.) The longer a ending -ry would be used primarily when a second pronominal ending e denoting the object is to be added, e.g., tiriryet (s)he watches them whereas (s)he watches by itself could be either tiris or tiriry, but more e commonly the former. But writers who want to avoid the unattested ending -ry may opt for the gender-specic endings -ro and -r instead, to bring in e e a connecting vowel: tirirot he watches them, tiriret she watches them. In the exercises below, we will however avoid all speculative endings and constructions and concentrate on the only known facts we have at our disposal regarding the 3rd person singular of the pronoun table: In Quenya as Tolkien had come to see this language in the post-LotR period, the ending -s may be used for he, she, it, whereas -rya covers his and her. (We may plausibly assume that -rya covers its as well: Notice that in the phrase ringa s maryass in her cold bosom cited above, the reference is u e actually to a ship, so its bosom would seem to be an equally appropriate translation.) The long endings -ro and -r are not used in the exercises or e the keys, since their status in LotR-style Quenya is slightly uncertain (not that I necessarily discourage writers from using them).
245
in houses. Quenya relative sentences may be formed using the relative pronoun ya which, that. Ya may also receive endings for case and number, cf. the plural locative yassen in which or wherein occurring in Namri a e (plural because it refers back to a plural word). The article i the may also e e be used as a relative pronoun, cf. i Eru i or ily mahalmar a tennoio, the One who is above all thrones in Cirions Oath, but i apparently cannot receive endings for case or number. In front of a verb, i can be used by itself to express the one(s) who do(es) whatever the verb expresses, e.g., i carir quettar the ones/those who form words. The pronominal endings for he and she are somewhat uncertain. Early material contains verbs with the endings -ro he and -r she (often combined with lengthening of e the vowel of the preceding syllable). In post-LotR material, we have one attestation of -s as an ending covering both he and she, and since the same ending is attested with the meaning it elsewhere (as object), we may assume that -s is a general ending covering the entire 3rd person singular, as subject or object. One educated guess has it that this -s alternates with a longer form -ry (plausibly extrapolated from the possessive ending -rya e his/her), but only the attested ending -s is used in the exercises below.
Vocabulary
tatya second (The original name of the Second Clan of the Elves was Tatyar,
literally Seconds, Second Ones, though the Eldarin branch of that clan would later be called Noldor instead [WJ:380381]. A variant form of tatya is atya [attested, compounded, in VT41:10], which connects more clearly with the basic number atta two. As will be explained in Lesson 17, second was later expressed as atta, but students should know the archaic form tatya as well, and we will use e this form here.)
aur day (the actual daylight period, not a full 24-hour cycle) veru (married) e couple, man and wife, pair of spouses (an old dual form apparently lacking
any singular; there are only the gender-specic words verno husband and vess e wife from the same root)
m hand a p lip, nominative dual peu (so according to VT39:9, reproducing a post-LotR e
source. Earlier, in the entry peg of the Etymologies, the word p had been glossed e mouth instead which would be pure plagiarism of the Hebrew word for mouth! But Tolkien apparently thought better of it: in LotR Appendix E, the Quenya word for mouth is given as anto instead, which word we introduced in Lesson Eleven.)
246
mall road, street (nominative pl. maller, LR:47, 56; SD:310 as we theorized in e
Lesson Two, nouns in -l may regularly have plural forms in -ler rather than -li.) e
hr e winter v apa preposition after (only attested compounded; cf. Apannar the Afterborn o
as an Elvish name of Men, the Elves themselves being the Firstborn see the Silmarillion, near the beginning of chapter 12. We will here assume that apa may also function as an independent preposition.)
Exercises
1. Translate into English (the pronominal ending -s may have various English equivalents): A. Tuluvas i tatya auress. e B. I hr vess rimb aiwi autar marien Hyarmess; apa i hr e e e e v autant Hyarmello ar tulir nrelmanna. e o C. Hiritarya malta i orontissen n alass lieryan, an hiria e e taryas carn lierya alya. e D. Tatya hr vess ya marnes i coass hirnes harma nu i talan. e e E. Quetis lambelma, an maris nrelmass. o e F. Eques: Cennen macil i ohtaro hyarya mass. e G. I nr i hirn i harma nurtuva i engwi yar ih e e ries samberyats. e H. I amboss cenis i veru yat it e ries coaryallo, ar yant nes a annarya. 247
2. Translate into Quenya: I. She saw a couple in the street. J. I found the woman who lives in the house between the rivers, and I watched her lips (dual) and her hands (dual); in her left hand I saw a book. K. I saw his cup in his hands (dual), the cup from which he poured wine into his mouth. L. The ones who dwell in the towers to which the man is going are warriors. M. His drinking the wine was not a good idea, for what he did after his drinking it was not wise. N. After we (excl.) went away [pa.t. of auta-] from our (excl.) land in the South, we have seen many Dwarves on the roads. O. The towers on the hills are great; the one who owns [harya = possesses] the greatest tower, from which one [quen] can see the Elven-land [Eldanr], is the richest man in the city. o e P. A people whose king is wise will dwell in peace in a good land which they will love deeply.
248
Lesson 16
The Instrumental case. Verbs with an unaccented vowel + -ta. The imperative. The nai formula.
If we accept the information provided in the Plotz letter as Tolkiens denite version of Quenya case system, we have now discussed all the Quenya noun cases except two. One of them is somewhat obscure; Tolkien supplied no further information about it, not even telling us what this case is called. The relevant ending is -s, plural -is. The Plotz lay-out suggests that this Mystery Case is simply a shorter, alternative version of the locative: The word exemplifying this case is listed in a parenthesis below the locative form of the same word. So instead of coass in a house, plural coassen in houses, e one may perhaps use the shorter forms coas, pl. coais. However, since we cannot be entirely condent regarding the function of this case, I will not construct any exercises involving it. On the other hand, the function of the last Quenya case we will discuss in this course is relatively well understood. We are talking about:
16.1
The rule for how the instrumental case is constructed can (for pedagogical purposes!) be stated very simply: Just add -en to the dative form! So where the dative has the ending -n, corresponding to plural -in and dual -nt, the instrumental has the endings -nen, plural -inen, dual -nten. Before discussing the function of this case, we will ll in some more details about the instrumental forms as such. We must assume that the basic instrumental ending -nen can be added directly to nouns ending in -n and -r without creating impossible clusters, so that we could have elennen as the instrumental form of elen star, or Anarnen as the instrumental of Anar Sun. (These instrumental forms would of course be accented on the second-to-last syllable because of the consonants cluster -nn- or -rn- now following the vowel of this syllable.) Nouns ending in -s with stems in -r- (for older -z-) would probably also show -rn- in the instrumental, e.g. olornen as the instrumental form of 249
olos, olor- dream. Nouns in -n with stems in -m- must be assumed to have instrumental forms in -mnen, e.g. talamnen as the instrumental form of talan, talam- oor. But from this point, we cant be sure. Since the group ln regularly becomes ld in Quenya, it could be that the instrumental form of (say) estel trust, hope should be ?estelden for older estelnen. Otherwise, as in the case of nouns in -t, it becomes even more dicult. What is the instrumental form of a noun like nat thing? Since **natnen is not a possible Quenya form, would it turn into ?nanten with metathesis tn > nt, or would a connecting vowel (possibly -e-) materialize to produce a form like natenen? In the case of nouns with special stem-forms ending in consonant clusters, a connecting vowel must be inserted before the ending -nen; the instrumental form of n (niss-) woman may be something like s nissenen. Some otherwise long-lost nal vowel may also be preserved before caseendings, as when ambar fate, doom is seen to have the instrumental form ambartanen (the relevant example is further discussed below). The stem of ambar may be given as ambart(a)-: Presumably the word did end in -rta way back in primitive Elvish, but except when shielded by grammatical endings, the nal vowel and (later) the -t had been lost. If the plural ending -inen is added to a noun ending in one of the three vowels -a, -o, or -u, the initial -i- of the ending merges with the last vowel of the noun to form a diphthong. Constituting the new second-to-last syllable, it naturally attracts the stress. Thus WJ:391 has mainen as the plural o instrumental form of ma voice, the form mainen being accented on o o the diphthong -ai-. Nouns ending in - may originally have behaved in a e similar fashion, so that lass leaf once had the plural instrumental form e lasseinen, accented on the diphthong ei but in Quenya, older ei eventually turned into a long and the Plotz letter points to lass , nen as the current form. Of course, this long still attracts the stress, like any long vowel occurring in the second-to-last syllable of a word. It is possible that nouns ending in -i, like tri queen, would also show in their plural instrumental a forms, tri+inen manifesting as tr a a nen since two short is would merge into one long This plural form tr . a nen, accented on the in the second-to last syllable, would then contrast with the singular trinen, accented on a tr -. Nouns in - with stem-forms in -i may behave in a similar fashion. a e The singular instrumental form of the noun l e, l song is attested r riin Namri as l a e rinen (this would be simply l +nen); perhaps the plural ri form would be l nen (for l r ri+inen). For the last time in the course proper I must bore the student with the question of dual forms: Some dual instrumentals have the ending -nten as indicated by Plotz, but the dual element is obviously the t, intruding into the simplest instrumental ending -nen. So is the ending -nten peculiar to nouns with nominative dual forms in -t, so that nouns with nominative dual forms in -u would rather add the simplest ending -nen after this -u? I tend 250
to think so; the instrumental form of Aldu Two Trees would then be Aldunen rather than ?Aldunten (or ?Aldanten or whatever). As the name suggests, the function of the instrumental case is to identify the instrument (in a wide sense) by which some action is achieved. The best example available is probably the phrase i carir quettar mainen o those who make words with voices (WJ:391). This description of the Elves, involving the plural instrumental form of ma voice, identies their voices o as the instrument or means by which they make words. Lacking an instrumental case, English often uses the preposition with instead, as in Tolkiens translation of mainen: with voices. However, it should be understood o that the Quenya instrumental endings correspond to English with only where this preposition means using or by means of (i carir quettar o mainen could also be translated those who make words using voices). It is highly unlikely that the endings marking the instrumental case can be used for English with in the sense of together with (and please allow me to dwell on this point for a moment, for some writers have actually misapplied the Quenya instrumental case in such a way!) A sentence like I saw them with an Elf can hardly be translated as **cennenyet Eldanen, for to the extent this makes any sense at all, it implies that the Elf is the instrument by which I saw them! On the other hand, in a sentence like I saw them with my binoculars, it would be quite all right to use the instrumental case for the English preposition with. (Unfortunately, I cant reconstruct the actual Quenya wording, for Tolkien doesnt seem to mention any Elvish word for binoculars anywhere: Perhaps the far-sighted Elves just didnt need such artices!) The Quenya instrumental endings may also be rendered into English by means of other prepositions than with. The two instrumental forms occurring in Namri Tolkien translated as phrases involving the preposition a e in; yet it is clear from the context that the instrumental does not really intrude on the area otherwise covered by the locative. The rst instrumental form occurs at the end of the rst line of the song: Ai! lauri lantar lassi e s rinen, ah! like gold fall the leaves in the wind. Despite Tolkiens transu lation, the context indicates that the wind (s r, s ri-) is here thought u e u of as the instrument which makes the leaves fall : In the wind actually implies by means of the wind, or simply because of the wind. This example shows that the Quenya instrumental case may indicate simply the reason why something happens (the instrumental ending marking the noun denoting what makes it happen). The second example of the instrumental case in Namri is similar, involving the noun l e, l song: Quoting a e r rifrom the prose version in RGEO, reference is made to Vardo . . . tellumar, yassen tintilar i eleni maryo l o rinen, that is, Vardas . . . domes, in which the stars twinkle by the song of her voice (maryo l o rinen = her voices song-by). So the song of Vardas voice is what makes the stars twinkle, and the word for song is accordingly marked with the instrumental 251
ending -nen. Another instrumental ending translated in by Tolkien is found in F riels Song, one line of which says that the Valar gave everyone the gifts of Ilvatar u lestanen = in measure. Here the instrumental noun tells us something about how the verbal action was accomplished. The Markirya poem includes the plural instrumental form (ending -inen) of the word rma wing, the sails of a ship being poetically referred to as a its wings: The ship is described as wilwarin wilwa . . . rmainen elvi, a e meaning something like uttering like a buttery . . . on starlike wings (or, with starlike wings, by means of starlike wings). We could imagine a less poetical example using the same plural instrumental form, e.g. aiwi vilir rmainen, birds y with (or, using) wings. Talking about a single bird a we could use a dual instrumental form: aiw vil rmanten, a bird ies e e a with [a pair of] wings. One (actually the only) example of an instrumental form occurring in the Silmarillion is particularly interesting. Near the end of chapter 21, Of Trin Turambar, N u niel refers to her brother as Turambar turun ambartanen, master of doom by doom mastered. UT:138 indicates that the more proper reading is Turambar tur n ambartanen. This sentence is u peculiar for several reasons. The word for doom (= fate) is here ambar with stem ambart(a)-, as in the name Turambar master of doom and the instrumental form ambartanen by doom. Other sources point to umbar as the Quenya word for fate, doom (it is even mentioned in LotR Appendix E as the name of a Tengwa letter). Ambar elsewhere occurs with the meaning world, as in Elendils Declaration in LotR (where reference is made to the Ambar-metta or end of the world), but ambar doom only partially coincides with this noun, since the stem-form ambart(a)- is distinct. Conceivably the proper Quenya word for doom was umbar, but the variant form ambar appeared in Exilic Quenya because of inuence from the corresponding Sindarin word (ammarth or amarth). We must have faith: perhaps Tolkien explains the seeming discrepancies in some still unpublished note. Another peculiar feature of N niels cry is the word turun or more properly tur n, translated mastered. The translation would seem to indicate u that this is a passive participle, and the complete form must be tur na, u the nal -a here dropping out because the next word (ambartanen) begins in the same vowel. This form tur n[a] mastered must be related to the u verb tur- govern, control, wield that we introduced in Lesson Seven. However, according to the rules for the formation of passive participles set out in Lesson Ten, the participle of tur- ought to be turna (cf. carna made as the attested passive participle of car- make), or less likely t rina u (cf. rcina broken as the attested participle of rac- break). The form a tur n[a] is quite perplexing. It could belong to some peculiar phase in u Tolkiens evolution of Quenya, an experiment later abandoned. Since we are 252
dealing with posthumously published material here, we can never be certain that all the linguistic samples represent the Professors denitive decisions on what Quenya grammar was really like. We must hope that future publications will throw more light upon the strange form tur n[a], but if we accept it as some kind of passive particiu ple, we can make out one important grammatical rule from N niels cry: Following a passive participle, the agent who brought about the condition described can be introduced as a noun in the instrumental case. In our attested example, Trin Turambar was mastered, and since N u niel wanted to add information about what it was that mastered her brother, she used the instrumental form ambartanen = by doom. A less gloomy example could involve, say, tcina written, the passive participle of the verb e tec- to write: We could build a phrase like i parma tcina i Eldanen, e the book written by the Elf. Following a participle, the instrumental form could surely also assume its more basic function of denoting an instrument, so that we could have a phrase like tcina quessenen written with a e feather [pen] (quess = feather). e We must assume that the instrumental endings can be added to the relative pronoun ya- to express by which, with which: Singular i cirya yanen lenden amba i s e the ship by which I went up the river (amba r = up), plural i ciryar yainen. . . the ships by which. . . , dual i ciryat yanten. . . the couple of ships by which. . . Though our attested examples involve other cases, there is no reason to doubt that also instrumental endings can be combined with possessive pronominal endings producing forms like mnyanen with my hand, a using my hand (m-nya-nen hand-my-with). a Combined with a gerund (ending in -i), the instrumental case may pere haps express the idea of by doing so and so, e.g. tirinen by watche ing (for instance, in a sentence like I found out by watching). When the instrumental ending is added to nouns in -i, the vowel in front of the e ending would likely be lengthened, thus receiving the accent (a quite awkward stress-pattern being avoided): Hence I go for tirinen rather than e ?tirienen, which would have to be accented on the second i. We have no attested example involving the instrumental case, but cf. Tolkiens tyaliva e as the possessive form of tyali play. The instrumental would likely be e tyalinen, then. Such vocalic lengthening apparently to avoid cumbere some stress-patterns is also observed in other parts of speech, as we will discover in the next thrilling section:
16.2
We have earlier discussed what must be the main categories of Quenya verbs. There are some minor sub-groups of verbs that may have their own peculiar
253
features, but our knowledge is very limited since (the litany of Tolkienan linguistics:) we have so few examples. All the same, a few observations about some of these sub-categories may be made, and we will deal with one of them here. In some of my examples and exercises I have combined the verb carmake, do with an adjective, e.g. exercise C in the previous lesson: Hiritaryas carn lierya alya, his nding it made his people rich. I should e point out that we have no Tolkien-made example of an adjective being combined with car- in such a way, and it may be that I am here imposing an English idiom on Quenya. Now this may not be such a disaster: If we are ever to develop a usable form of Quenya, it would almost inevitably become somewhat coloured by modern usage (and if the Eldar return from Valinor to protest against their language being mistreated, that would not be a bad thing, either). Even so, it may be noted that Quenya vocabulary includes what may be termed causative verbs derived from adjectives; perhaps fully idiomatic Quenya would rather use such formations. These verbs express, in a single word, the idea of making an object have the properties described by the corresponding adjective. The student should already be familiar with the ending -ta, which occurs in many Quenya verbs (e.g. pusta- to stop). Often it is just a verbal ending with no particular implications, but occasionally it may take on a causative meaning; compare the primary verb tul- to come and the derived verb tulta- to summon (= to cause to come). Added to adjectives, it seems that this ending may similarly be used to derive causative verbs. We have only a handful of examples, but the adjective air holy apparently corresponds to a verb airitae to hallow that is, to make holy. (The nal - of air holy appears e e e e as -i- in airita- because the - of air descends from -i in the primitive language, and it changed to - only when nal. Cf. the similar variation in e the aorist: sil shines, but with a plural subject silir shine, because if e you add any ending the nal vowel is no longer nal at all.) The one form of the verb airita- that is actually attested is the past tense. It reportedly appears as airitn in an unpublished Tolkien manuscript a e stored at the Bodleian: According to a footnote in Vinyar Tengwar #32, November 1993, p. 7, the manuscript page in question dates to c. 1966 and gives much information about Quenya verbs. It will be published in an upcoming issue of Vinyar Tengwar. Eight years and nine Vinyar Tengwar s later, we are unfortunately still waiting to see this apparently highly interesting document but at least VT#32 cited the past tense airitn. a e It obviously includes the well-known past tense ending -n, but it should e be noted that the vowel of the ending -ta- is here lengthened when the past tense ending is added. In this way, the now long syllable -t- attracts the a stress. **Airitan with no lengthening would have a rather awkward stresse pattern (accented on -rit-), and it is perhaps for this reason the lengthening occurs. This may also imply that if some further ending is added after -n so e 254
that the stress would not threaten to land on -rit- after all, the lengthening of -ta- may not occur: Perhaps, say, we hallowed is airitanelm rather e than ?airitnelm, since the stress must here fall on -ne- and -ta- receives a e no stress at all. Some think Quenya cannot have a long vowel in a wholly unaccented syllable unless this syllable is also the rst one of the word. Whatever the case may be, we can apparently infer this rule: As long as the past tense form of such a verb (that is, a verb with an unaccented vowel in front of the verbal ending -ta) is not to receive any further endings that may shift the stress, the ending -ta is lengthened to -t- when the past a tense ending -n is added after it: Thus airitn as the pa.t. of airita-. e a e Of course, not all endings that may be suxed to -n have the power to e shift the stress, and then the lengthening of -t- must remain to prevent a the accent from going someplace it shouldnt: Airitner hallowed (with a a plural subject), airitnes (s)he hallowed, airitnen I hallowed. But a a quite possibly, it should be airitaneny with no lengthening of -t- if you e a use the longer form of the ending for I so that the stress moves to -ne-, and -ta- becomes a wholly unaccented syllable. In the Etymologies, Tolkien listed at least one more verb that seems to belong to this class. The entry nik-w- provides a verb ninquit- whiten, a i.e. make white, derived from the adjective white: ninqu (stem nine qui-; the primitive form is given as ninkwi ). By writing ninquit-, Tolkien a obviously suggested that the nal vowel is often long, and we may safely assume that the past tense is ninquitn. a e
NOTE: In the entry nik-w-, Tolkien also listed a verb ninquita- shine white that would probably inect otherwise: perhaps the past tense would rather be ?ninquint with e nasal inxion (allow me to state explicitly that this is speculation!) In the aorist, the two verbs must probably coincide as ninquita, the context determining whether this is to be interpreted whitens or shines white.
We may be able to tell one more thing about this class of verbs: how the passive (or past) participle is formed. The evidence is widely scattered, though. In The Houses of Healing, Chapter 8 of Book Five in The Return of the King, Tolkien has Aragorn saying that in the high tongue of old I am Elessar, the Elfstone, and Envinyatar, the Renewer. The Quenya title Envinyatar = Renewer is interesting. As for the nal -r seen here, this ending may be added to (A-stem) Quenya verbs with much the same meaning as the English agent ending -er, so Envinyatar Renewer points to an underlying verb envinyata- to renew. The prex en- means re-, and vinya is the Quenya adjective new, so apparently we are looking at another verb derived from an adjective by means of the ending -ta. Interestingly, what may be seen as the passive participle of this verb envinyata- to renew is attested in MR:405, in the phrase Arda Envinyanta. This Tolkien translated Arda Healed (the reference is to a future world healed from the consequences of the evil of Morgoth). Compar255
ing it with Aragorns title Envinyatar = Renewer, we can tell that Arda Envinyanta more literally means Arda Renewed. It should be noted how the passive participle is formed: by nasal-inxion intruding before the t of the ending -ta of the verb envinyata-. The resulting form envinyanta differs from the passive participles of normal verbs in -ta, which seem to have participles in -taina. (Compare hastaina marred from the same text that provides the example Arda Envinyanta Arda Healed: Arda Hastaina or Arda Marred was the world as it actually appeared, marred by Morgoth. See MR:405, cf. 408, note 14. It is important to notice that these divergent types of participles occur in the same source text, allowing us to know with certainty that the dierent formations do belong to the same version of Quenya: Otherwise, it would be tempting to dismiss some of the formations as representing merely a certain stage in Tolkiens evolution of the language ideas he later abandoned.) If envinyata- to renew has the passive participle envinyanta, we may plausibly assume that the pass. part. of airita- to hallow is similarly formed by means of nasal-inxion: airinta hallowed (rather than ?airitaina, though perhaps this form would also be acceptable). And if airita- has the past tense form airitn with lengthening of -ta- to -t-, we a e a can probably assume that envinyata- to renew becomes envinyatn in a e the past tense. Similarly, if ninquit- is the verb to whiten, with the past a tense ninquitn, the participle whitened may well be ninquinta. (The a e forms envinyanta, airinta, ninquinta would of course agree in number like adjectives in -a, changing this nal vowel to - in the plural.) e We have mentioned pretty much all the very few known verbs that may tentatively be assigned to this sub-class. There is no direct evidence for how they would behave in other forms than the past tense and the passive participle. (As for the active participle in -la, we would almost certainly see the same lengthening of the ending -ta as we observe before the past tense ending -n: hence airitla hallowing, envinyatla renewing. Again, the e a a motivation for lengthening the vowel of -ta would be to achieve euphonic stress-patterns.) It is of course dicult to know to what extent we should feel free to derive new Quenya verbs ourselves by adding -ta to adjectives (remembering that adjectives in - change this vowel to -i- before endings, as in airita- to e hallow from air holy). To return to the sentence we started with, hire itaryas carn lierya alya his nding it made his people rich, perhaps e this might better be expressed as hiritaryas alyatn lierya? We then a e assume that the adjective alya- can be used as the basis for a verb alyatamake rich or enrich, with past tense alyatn (and passive participle a e alyanta). In this as in other matters, people who want to write in Quenya face a dicult choice: Should we try to make the language work using solely the words Tolkien himself provided, introducing unattested idioms or long circumlocutions where necessary to work around gaps in the Tolkien-made 256
vocabulary? Or should we feel free to derive new words from Tolkienian elements by applying the Professors principles as far as we understand them, something that may be perceived as diluting Tolkiens actual linguistic output with fake elements (however cleverly constructed)? Some post-Tolkien creativity must unquestionably be allowed if we are ever to develop Quenya into anything like a useable language, but there are no easy answers here.
16.3
The imperative
The imperative is a form of the verb used to express commands or requests. In English, imperatives are often preceded by the word please to make them more polite, but it should be understood that an imperative form as such is not necessarily to be taken as a blunt order. In Tolkiens Quenya rendering of the Lords Prayer, several imperatives occur, and such a prayer as deliver us from evil is of course just that a prayer, not an attempt to order God around. According to Tolkien, the primitive Elvish language had an imperative particle that could be used in conjunction with a verbal stem to indicate that it was to be taken as an imperative. The particle had the form a, and it was originally independent and variable in place (WJ:365). Sometimes it was placed after the stem, and in such cases it came out as an ending -a in Quenya. WJ:364 mentions an imperative exclamation heca! meaning be gone! or stand aside! and on the next page, this is suggested to come from the primitive phrase hek(e) a. There is also the primitive exclamation el-, lo!, look!, see!, which is supposed to be the very rst thing the a Elves ever said as they awoke at Cuivinen and rst saw the stars (WJ:360). e In Quenya, this word came out as ela! It was an imperative exclamation directing sight to an actually visible object (WJ:362). If we were to be guided by examples like heca and ela, we would have to conclude that in the case of primary verbs at least, imperatives may be formed by adding -a to the verbal stem. For instance, tir- to watch would have the imperative tira! watch!, representing primitive tir- or tir(i) a. a The corresponding Sindarin form tiro! is actually attested. (Notice that the imperative tira watch! would be distinct from the present/continuative form t is watching, since in the latter form, the stem-vowel is lengthra ened.) This may be one way of constructing Quenya imperatives, but it is also possible that exclamations like heca and ela are to be taken as fossilized forms descending from earlier stages of Elvish. As for the typical modern way of forming imperatives, there is some evidence that a descendant of the original particle a was still treated as an independent word: it was placed in front of the verbal stem instead of being suxed as an ending. In the LotR itself, an example is provided by the Cormallen Praise, the crowd hailing Frodo and Sam with the words a laita
257
te. . . Cormacolindor, a laita trienna! Bless them. . . The Ring beara ers, bless (or praise) them to the height (translated in Letters:308). Notice how the verbal stem laita- bless, praise is here preceded by the imperative particle a to form an imperative phrase a laita! bless! or praise! The particle a also appears in the long form , directly from primitive a, as in a the exclamation vala Manw! may Manw order it! (WJ:404). Here, a e e the verb vala- rule, govern (the origin of the noun Valar and in later usage therefore referring to divine power only) is combined with the imperative particle : The literal meaning of vala Manw! is transparently a a e something like do rule Manw!, if we make an eort to translate as a e a separate word. Incidentally, this example demonstrates that the subject of the imperative (the one who is to carry out the order or request) may be explicitly mentioned after the imperative phrase proper. Is there any reason why the imperative particle appears in the short form a in a laita, but in the long form in vala? It has been suggested that a a is shortened to a whenever it occurs in front of a long syllable (like lai-, a because of the diphthong ai), but we cannot be sure. Perhaps vs. a is just a an example of random variation: Presumably being unaccented, the particle could well tend to become shortened if speakers dont enunciate with care (the ecstatic crowds at Cormallen, hailing the hobbits who had saved the world, hardly did!) I would normally prefer the long form , avoiding cona fusion with a as a particle of address, like English o (as in Treebeards greeting to Celeborn and Galadriel: a vanimar = o beautiful ones, Letters:308). For instance, since the verb to go is lelya-, the imperative go! would be lelya! a The imperative particle can also be combined with the negation v a a to form the word va, used in negative commands: Ava car! Dont do a e [it]! (WJ:371). This example also gives away how primary verbs behave in imperative phrases: they appear with the ending -, just like they do when e they are used as innitives (and ending-less aorist forms). So from a primary verb like tir- watch, we can probably form a command like tir! (do) a e watch! negative va tir! dont watch! a e
16.4
If one does not want to issue a command (however polite), but is merely expressing a wish that something will be done or will happen, Quenya has a special wishing formula. Near the end of Namri, we nd these lines: Nai hiruvaly Valimar! a e e Nai ely hiruva! In LotR, this is translated maybe thou wilt nd Valimar! e Maybe even thou wilt nd it! The word nai is here rendered into English as maybe, but elsewhere, Tolkien indicated that this Quenya word does not merely imply that something is possible. He noted that nai expresses
258
rather a wish than a hope, and would be more closely rendered may it be that (thou wilt nd), than by maybe. (RGEO:68) We may wonder why he used the misleading translation maybe in the rst place; possibly there are some conceptual developments involved here (i.e., Tolkien changed his mind about the precise meaning of a Quenya text he had already published!) e Anyway, his nal decision on the meaning of the phrase nai hiruvaly Valimar was that it is to be interpreted be it that thou wilt nd Valimar or may thou nd Valimar. Nai ely hiruva likewise means be it that e even thou wilt nd [it]. (The word ely even thou here occurring is an e emphatic, independent pronoun corresponding to the ending -ly thou, e you, whereas Valimar here stands as an alternative to Valinor: Galadriel singing Namri thus expresses a wish that Frodo will eventually nd or a e come to the Blessed Realm and as we remember, both he and Galadriel herself went over the Sea in the end.) We have one more attestation of the nai wishing-formula. It occurs in Cirions Oath, Cirion expressing a wish that the Valar will guard the oath: Nai tiruvantes, be it that they will guard [/watch over] it. Tolkien noted that this is the equivalent of may they guard it (UT:305, 317). As for the basic meaning of the word nai itself, Tolkien implied that it is quite literally be (it) that: He derived Quenya nai from earlier n-i a (RGEO:68). The n part would seem to be the element meaning be!, a undoubtedly closely related to the Quenya copula n is, itself a form of a the verb to be. The nal i must be the element corresponding to the that of be (it) that, and this i is certainly meant to be related to the Quenya article i the. Whatever the precise origin or basic meaning of nai may be, it is a useful word that can apparently be put in front of any sentence including a futuretense verb, turning a simple statement about the future into a wish about what the future may bring: Elda tuluva coalmanna an Elf will come to our house > Nai Elda tuluva coalmanna! be it that an Elf will come to our house! = (I) wish that an Elf will come to our house! or may an Elf come to our house! Hiruvan i malta I will nd the gold > Nai hiruvan i malta! be it that I will nd gold! = wish that I will nd the gold! Caruvantes they will do it > nai caruvantes! be it that they will do it! = wish that they will do it! In our attested examples, nai is combined with the future tense, but since we have only three examples, it certainly cannot be ruled out that nai may be used in conjunction with other tenses as well. (One may even say we have only two examples, Cirions Oath + Namri, since the two examples a e of the nai formula near the end of Namri are very similar.) Perhaps nai a e 259
can also describe the speakers hope that a certain wish is already being fullled, or has been fullled in the past the speaker still not knowing whether the wish came true or not. If so, we could have constructions like nai t ras be it that (s)he is watching = I hope (s)he is watching (with the present or continuative tense of tir- watch), nai hirnentes! be it that they found it = I hope they found it (with the past tense of hirnd), or nai ut lies be it that (s)he has come = I hope (s)hes come u (with the perfect tense of tul- come). However, in the exercises below, nai is only combined with the future tense as in our attested examples.
260
Vocabulary
nelya third (The original name of the Third Clan of the Elves was Nelyar, literally
Thirds, Third Ones, though the Eldarin branch of that clan would later be called Lindar or Teleri instead [WJ:380, 382].)
imperative particle (variant a, but we will use here) a a va dont! (i. e., the imperative particle combined with a negation. Also in the form a
av, the sole attested two-syllable Quenya word that is denitely known to be a accented on the last syllable [WJ:371] but we will use va here.) a
rac- to break envinyata- to renew airita- to hallow harna- to wound (and the passive participle is evidently also harna, dened as
wounded in the entry SKAR in the Etymologies. The adjectival or participial formation harna wounded is the primary derivative from the original root; eventually harna- also came to be used as a verbal stem to wound. Of course, if this word were to go like a regular A-stem verb, the passive participle should then be ?harnaina. But the ending -ina is just a longer variant of the ending -na which is present from the beginning, and suxing it twice to the same word should hardly be necessary!)
namba hammer eht spear e yna holy place, sanctuary a nilm friendship e Rmen the East (the initial r- is ultimately related to the or- of the verb ortao o
rise, since the Sun rises in the East.)
Exercises
1. Translate into English: A. Ut lies Rmello ninqu rocconen. u o e B. I nr harnan i r ehtenen, ar eques: Ava mat yone e a e donya! C. Quentemm i Eldanna: Nilmelma n envinyanta ane a nalyanen! D. I nelya auress quentes i vendenna: A car ya merily! e e e E. Quen um pol hir harma nurtaina Naucoinen, an Nauco e e e mel n rav i malta ya haryas. e u e F. I nr n harna rassenten i lamno; nai uvas r! e a e 261
G. Lindas alassenen. e H. A lelya i ostonna ar quet i taura trinna: Nai varyua e a valy nrelma i umi ohtarillon! e o e 2. Translate into Quenya (consistently using the independent imperative particle rather than the ending -a, which possibly only occurs in fossilized forms): I. He said to the Dwarf: Break the cup with a hammer! J. By ship I went away [past tense of auta-] to a remote land in the East. K. The city is protected by great walls, and warriors who ght with spears cannot break the walls. L. Calandil said to his wounded son: Dont die! M. May your queen nd the sanctuary hallowed by the Elves! N. The king and the queen went to my house and renewed our (excl.) friendship with great gifts. O. She seizes the boy with her hands (dual), and she says: Dont go to the river! P. The woman who lives (/dwells) in the third house in the street said to the Elf: Watch the men who are coming from the sanctuary that you see on the hill, the ones who go to the east.
262
Lesson 17
The demonstratives: Sina, tan(y)a, enta, yana. Inecting the Last Declinable Word. U-stem nouns. Ordinals in -a. e
17.1
Quenya demonstratives
Demonstratives are such words as English this or that, with the corresponding plural forms these and those. Thus, they have a stronger meaning than the mere article the (though in the languages of the world, many denitive articles descend from older demonstratives that were overused so that their meaning faded). The demonstratives may be used together with nouns, producing phrases like this house or that man. In LotR-style Quenya we have only one demonstrative attested in an actual text: Cirions Oath commences with the words vanda sina, translated this oath. The Quenya word order is actually oath this, sina being the word for this: The root si- has to do with present position in time or space (cf. such words as s now or sinom in this place = here, the latter e from Elendils Declaration in LotR). A word for that appears as tanya in an early Qenya text published in MC:215, which has tanya wende for that maiden. Here, the wordorder is English-style with the demonstrative rst and the noun it qualies following it the opposite of the word order seen in Cirions Oath. Perhaps the word order is free, so that vanda sina could just as well be sina vanda and conversely, tanya wende could also be wende tanya? Be that as it may, we cannot be quite certain that the word tanya is still valid in LotRstyle Quenya. The Etymologies lists ta as the Elvish demonstrative stem that, and the actual Quenya word for that is given as tana. Since this form seems like a perfect counterpart to sina this, we will here use tana rather than tanya as the word for that (though it is also possible that the Qenya form tanya survived into the later stages of Tolkiens conception). So given that vanda sina is this oath, we must assume that that oath would be vanda tana. Maybe we should update the Qenya phrase tanya wende that maiden to LotR-style Quenya vend tana (or wend tana e e with the older form or archaic spelling of the word for maiden). Then we 263
also implement the word-order seen in Cirions Oath, with the demonstrative following rather than preceding the noun it connects with: In the entry ta in the Etym, Tolkien actually described tana an anaphoric word for that, meaning that it refers back to something already mentioned. However, sina this and tan(y)a that are not the only Quenya demonstratives known. Though not actually observed in any Quenya texts, other demonstratives are mentioned in Tolkiens notes. Another word for that is enta, mentioned in the entry en in the Etymologies and there described as an adjective meaning that yonder. The root en itself is said to be an element or prex = over there, yonder. Still letting the demonstrative follow the noun it connects with, we may perhaps construct a phrase like coa enta, expressing that house in the sense of yonder house, that house over there. It may be that Tolkien meant Quenya to distinguish three degrees of nearness or remoteness, as do certain languages of our own world. English typically only distinguishes two degrees, this and that: To simplify matters rather drastically, we may say that this refers to something near the speaker, whereas that refers to something away from the speaker. But in some languages, the position of the listener is also considered. There are two words for that, one referring to something away from the speaker but near the person addressed (that thing over by you) and another word referring to something that is not close to either the speaker or the listener (that thing we see over there). Could it be that in Quenya, tana as a word for that refers to something close to the person addressed, whereas enta refers to something that is remote from both the speaker and the person (s)he addresses? There is presently little or no evidence to back up such a theory, but we can at least be certain that the word enta clearly connotes the idea of over there, that yonder, of something separated from the speaker by physical distance. It may be noted that one Sindarin word for there, namely ennas (SD:129 cf. 128), is understood to represent an older locative form that could correspond to a Quenya word entass = in e yonder [place]. (Perhaps tana is simply a more general word for that, merely focusing on the special identity of someone or something: that one as opposed to any other.) Yet another word for that is yana, mentioned in the entry ya in Etym: After the gloss that, Tolkien added a parenthetical specication: (the former). Perhaps aran yana would mean that king with the implication that we are talking about a former king, now dead or at least no longer ruling. There may be interesting contrasts between yana and enta as words for that: In the Etymologies, Tolkien noted that the root ya signies there, over there; of time, ago. He added that en, the root producing enta, of time points to the future. So that day may translate as aur enta if e we are talking about some future day, not yet come, whereas aur yana is e that day with reference to some day in the past. (A neutral wording, 264
with no special implications, may be aur tana.) e As for plural demonstratives, like English these and those, we have no attested Quenya forms. Yet the words sina this and tana, yana that do look like adjectives by their form (-na being an adjectival or participial ending), and enta that yonder Tolkien explicitly identied as an adjective (Etym, entry en). So in all likelihood, we can inect all of these words as adjectives, and then we can derive their plural forms simply by changing the nal -a to -: e vanda sina this oath / vandar sin these oaths e n tana that woman / nissi tan those women s e coa enta that house [over there] / coar ent those houses e aur yana that day [in the past] / auri yan those days e e As in the case of normal adjectives with the ending -a, the plural forms in - would represent archaic forms in -ai (vandar sinai etc.) Indirect eve idence conrms that demonstratives could receive the plural ending -i in older Elvish: In LotR, in the inscription on the Moria Gate, occurs the Sindarin phrase i thiw hin, translated these runes. Tolkien would have meant this to represent something like in tewi sini at an older stage and in n a a Quenya, an old plural demonstrative sini these would rst become sinai a and then sin. e It is not clear whether the demonstratives discussed above could occur by themselves, independently, and not only in conjunction with nouns. Can we use sina for this in a sentence like this is a good house? (And if we needed a plural form these, should we inect sina as a noun when it occurs by itself, so that the plural would now be sinar rather than sin?) e In PM:401, we have the sentence sin quent Quendingoldo. Tolkien proe vided no translation, but it must mean either this Quendingoldo said or thus spoke Quendingoldo. The latter interpretation has it that sin is an adverb thus, but if sin means this, it would be what we may call a demonstrative pronoun corresponding to sina, the latter however being an adjective only occurring in conjunction with a noun. By this interpretation, it would be sin, rather than sina, we should use in sentences like this is a good house or I have seen this. (And should the independent word for these be something like sini, then?) As for the other demonstratives, we have ta as an independent form of that, corresponding to the adjective tana (see Etym, entry ta). Of other such independent forms, little or nothing is known, and in the exercises below, we will concentrate on the adjectival demonstratives sina, tana, enta, yana used in conjunction with nouns.
265
17.2
Now that we have presented all the Quenya cases, we may also point out that the various case endings are not always attached to the noun they logically belong to. Where that noun is part of a longer phrase, like when the noun is followed by an attributive adjective describing it, the case ending may be added to the last word of the phrase. Cirions Oath provides the classical example. It includes a reference to Elendil Voronda, Elendil the Faithful, voronda being a Quenya adjective meaning steadfast, faithful. Wrote Tolkien: Adjectives used as a title or frequently used attribute of a name are placed after the name. (UT:317; as we have pointed out earlier, Quenya here diers from English by not inserting a denite article between the name and the adjective hence not Elendil i Voronda, at least not necessarily). In Cirions Oath, the name-and-title phrase Elendil Voronda is to appear in the genitive case: The Oath includes the words Elendil Vorondo voronw, Elendil the Faithfuls faith or (as it is translated in UT:305, e with an English-style word order) the faith of Elendil the Faithful. Notice that the genitive ending -o, which we underlined, is added to the adjective voronda (regularly displacing a nal -a) rather than to the noun Elendil. In a way, the adjective following the noun is treated as an extension of the noun proper, and so the case ending is added at the end of the whole phrase. Tolkien commented on the construction Elendil Vorondo: As is usual in Quenya in the case of two declinable names in apposition only the last is declined (UT:317). Voronda faithful here stands in apposition to Elendil as an additional name or title, and only the latter name is declined (inected for case). This principle would work with all the various cases. The allative of Elendil when the name occurs alone is attested as Elendilenna to Elendil (PM:401), but to Elendil the Faithful would apparently be Elendil Vorondanna, the last word of the phrase receiving the case ending. Where a proper name followed by some epithet (like Voronda in this case) is concerned, the system of adding any case endings to the last word of the phrase may be more or less universal. Yet common nouns, not just proper names, may also be qualied by adjectives following rather than preceding the noun. Cf. for instance a phrase like mall tra road straight = a e e straight road (LR:47). If we were to add the locative ending to express on a straight road, to what word should it be attached? Should we apply the last declinable word rule again (mall trass) or attach the locative e e e ending to the noun (malless tra)? e e It seems that both constructions would be permissible. The Markirya poem provides a string of examples of noun-phrases where the noun proper is followed by an adjective (in most cases a participle). Three consecutive examples involve the noun isilm moonlight combined with various particie 266
ples (ilcala gleaming, p cala waning, lantala falling), and all three noun phrases are inected for the locative case by attaching the locative ending to the last word of the phrase: e isilm ilcalass = in gleaming moonlight e isilm p e calass = in waning moonlight e isilm lantalass = in falling moonlight e e (Tolkiens more poetic translation in MC:215 goes in the moon gleaming, in the moon waning, in the moon falling.) Another phrase, again involving the participle ilcala gleaming but here combined with the allative case, is particularly interesting: axor ilcalannar = upon gleaming bones Notice that the noun axo bone is here plural. The plural allative upon bones occurring by itself would of course be axonnar. But here, where the plural allative ending -nnar is attached to the last word of the phrase instead, the noun axo itself receives only the simplest plural ending -r. Normally, axor would be taken as a nominative plural, but actually the -r merely marks the word as a plural form in the simplest possible way: The actual case marker follows later in the phrase. Words with nominative plurals in -i would of course receive this plural marker instead, e.g. vendi lindalaiva = of singing maidens (home-made example involving the possessive case, but the principle would be the same for all the cases: dative vendi lindalain, allative vendi lindalannar, etc.) We must assume that dual nouns would also appear in their simplest (normally nominative) form at the beginning of the phrase: The noun would merely assume the dual ending -u or -t, and the full dual case ending would follow later in the phrase. To construct a Tolkienesque example: Aldu caltalanta = upon [the] shining couple of trees. However, it is apparently not a hard-and-fast rule that you must attach a case ending to the last word of the entire phrase rather than to the noun proper. Markirya contains examples of phrases where an attributive adjective follows the noun it describes, and yet the case ending is added to the noun, not the adjective. The rst example involves a plural instrumental form (ending -inen), whereas the second example involves the locative case (the ending -ss being added to a noun that is inected for the somewhat e obscure partitive plural marked by the ending -li): rmainen elvi = on [/with] starlike wings a e ondoliss morn = on dark rocks e e Of course, the adjectives elva starlike and morna dark are here e plural (elvi, morn) to agree with the plural nouns they describe. It could e e be that in both instances, the case ending is not added to the adjective because the adjectival plural inection and the case inection would somehow 267
collide. (In the phrase axor ilcalannar upon gleaming bones there is no collision even though bones is plural, since participles in -la apparently do not agree in number.) It is less than clear how an ending like -inen could be added to a form like elvi anyway: ?elviinen seems like an unlikely and e e awkward form, prone to collapse into the quite obscure word **elv nen. Perhaps that is why Tolkien preferred to add the case ending to the noun rma instead, even though this noun is not the last word of the phrase. a Yet the system of inecting the last declinable word does seem to be a common phenomenon in the language. It seems that sometimes, only the last item on a list receives case endings that actually apply to all the nouns that are listed. Namna Finw M e riello is translated the Statute of Finw e and M riel (MR:258). Not only is the conjunction ar and that would have separated the two names omitted, but the genitive ending -o of is added to the last name (M riel, M riell-) only. The full construction would presumably have been Namna Finwo ar M e riello, but it was apparently permissible to strip the phrase down to basics to provide the Statute with a more concise title. Though we have no attested examples, the demonstratives listed above would seem to be good candidates for receiving case endings, if the word order observed in the phrase vanda sina oath this is normal. For instance, if we were to add the instrumental ending to express by this oath, it would perhaps be best to say vanda sinanen. However, vandanen sina would probably also be permissible and in the plural (nominative presumably vandar sin these oaths), consistently adding the case ending to the noun e would be the safest course: By these oaths would then be vandainen sin e rather than ?vandar sininen or sin e nen or whatever.
17.3
U-stem nouns
Apparently in the latter part of the Common Eldarin stage of Tolkiens simulated evolution of his Elvish languages, two parallel changes occurred, aecting what had earlier been short nal -i and short nal -u: they now turned into -e and -o, respectively. However, since this change only occurred where these vowels were nal, they remained -i - and -u- whenever some ending or other element followed. We have already alluded to this phenomenon earlier in this course; in particular, the student will remember it from the variation observed in the aorist of primary verbs: sil shines, but pl. silir e shine (because original -i did not change to -e when there was a following ending, like the plural marker -r in this example). Similar variation may be observed in nouns and adjectives: We have already mentioned the noun lm night, which has the stem-form lmi- (SD:415) because it descends o e o from earlier dmi - (see the entry domo in Etym). We must assume that o (say) the locative form at night would be lmiss. The adjective carn o e e
268
red descends from primitive karani (see Etym, entry karan) and therefore has the stem-form carni-, for instance in a compound like Carnistir Red-face (PM:353). The behaviour of these i -stems of course nds its parallel in the Ustems, words that end in -o when this vowel is absolutely nal, but preserve an original -u where some element follows this vowel. Such words seem to be predominantly (perhaps exclusively) nouns. One example of a U-stem noun is ango snake: Its stem-form angu- is directly observed in the compound o e angulc (simply glossed dragon, but actually combining the word for snake with the word normally translated dragon, lc: see the entry o e lok in Etym). In the Etymologies, Tolkien derived ango snake from older angu (or angwa, which would become angw and then angu), so the nal -o of this word does indeed represent an older -u. Whenever the noun ango is to receive endings for case or pronoun, it would apparently assume the form angu-, e.g. dative angun for a snake, ablative angullo from a snake or with a pronominal ending e.g. angulya your snake. The genitive would presumably be anguo of a snake. (As we have demonstrated earlier, normal nouns ending in -o do not have distinct genitive singular forms; the genitive ending -o simply merges with the nal vowel.) Where U-stem nouns end in either -go or -co, they assume a peculiar form in the nominative plural. Normally, nouns ending in -o would of course have nominative plural forms in -or. However, where -go and -co represent older -gu and -ku, it seems that adding the primitive plural ending - made the preceding u become w, so that the plurals came to end in -gw or kw Probably w merged with the g or k preceding it: The combinations . gw, kw are evidently best taken as unitary sounds, labialized versions of g and k (that is, g or k pronounced with poised lips look up Lesson One again). In Quenya, these labialized sounds persisted, though by convention, kw is spelt qu. Bottom line is, when we are told that ango snake has the stem angu-, we can also deduce that the plural form is neither **angor nor **angur, but angwi! The Etymologies conrms this; the plural form angwi is explicitly mentioned in the entry angwa/angu. An example of a -qui plural is provided by the word urco bogey, which has the plural urqui (= urcwi ). Regarding this word, Tolkien noted that as the plural form shows, urco must be derived from either urku or uruku in the primitive language (WJ:390). Thus, urco is denitely a U-stem noun, its nal -o representing older -u, and we would still see urcu- in compounds and before most inectional endings.
NOTE: The word urco bogey is akin to Sindarin orch, Orc. In WJ:390, Tolkien notes that in the lore of the Blessed Realm, the word urco naturally seldom occurs, except in tales of the ancient days and the March [of the Eldar from Cuivinen], and then e [it] is vague in meaning, referring to anything that caused fear to the Elves, any dubious shape or shadow, or prowling creature. . . It might indeed be translated bogey. Later, when the Noldor returned to Middle-earth, the word urco pl. urqui was primarily used
269
with reference to Orcs, since the kinship (though not precise equivalence) of this Quenya term to Sindarin orch was recognized. In Exilic Quenya, a Sindarin-inuenced form also appeared: Orco, the plural of which could be either orcor or orqui. The plural form orcor occurs elsewhere as well (MR:74), but if one prefers orqui, one should probably let orco Orc function as a U-stem in all respects. For instance, if one were to coin a compound e e Orc-language, it should be orculamb rather than orcolamb. In the Etymologies, far predating the source reproduced in WJ:390, Tolkien also gives the relevant word (glossed goblin!) as orco pl. orqui: stem orok. In Etym, there is no hint that this word was borrowed into Quenya from another language; orco is referred to a primitive form orku. Tolkiens precise ideas about the history of the Quenya word for Orc were apparently subject to change, but the basic idea that nouns in -co derived from primitive forms in -ku should have plurals in -qui rather than -cor is seen to persist. In accordance with our policy of avoiding specic references to Tolkiens mythos in the exercises, we will not refer to Orcs here, but we can use the word urco in its sense of bogey (it will occur in the exercises appended to Lesson Eighteen).
We will try to survey the words involved (excluding the earliest Qenya material). Ango snake, pl. angwi, seems to be our sole entirely certain example of a -gwi plural. In the Etymologies, there was also lango throat, pl. langwi (see the entry lank). The form langwi is for some reason marked with an asterisk, which would normally indicate that this form is unattested, but possibly it has another meaning here. Anyway, Tolkien decided to change the word for throat, turning it into lanco instead. It is entirely possible that this is also a U-stem, so that its plural should be lanqui rather than lancor, though we have no explicit information to this eect. One certain U-stem is the word for arm, ranco (primitive form explicitly given as ranku). Just as we would expect, the plural form is ranqui; see the entry rak in Etym. A word meaning arm would presumably often appear in its dual form to signify a natural pair of arms. We may wonder whether the dual form of ranco would be rancu (with the dual ending -u, quite unrelated to the original nal -u that later became -o) or rancut (i.e., the U-stem noun ranco, rancu- with the dual ending -t). As we have argued from the attested example peu pair of lips, nouns denoting body-parts occurring in pairs may consistently have fossilized dual forms in -u, since it was this ending that originally denoted a natural or logical pair. Once a pronominal ending is added, we may at least safely sux -t to indicate a dual form. Indeed, without this ending there would be no distinction between ranculya your arm and ranculyat your (pair of) arms, no matter what the dual of ranco may be when the word occurs by itself: Before endings, ranco must become rancu- anyway. Another U-stem is rusco fox; it our source, Tolkien mentioned both the stem-form ruscu- and the plural rusqui (VT41:10). Not all U-stems end in -co or -go, of course. One example is the word curo a skillful device (VT41:10, last word of gloss uncertain due to Tolkiens dicult handwriting). Tolkien cited the stem-form curu-, and it apparently 270
also occurs in Sarumans Quenya name: Curumo (UT:401). This name seems to combine the element curu- with the masculine ending -mo that often appeared in names or titles (WJ:400). We may wonder what the nominative plural of curo, curu- would be. Could it be curwi, paralleling angwi as the plural of ango, angu- snake? Anyhow, the special nominative plurals ending in -wi (spelt -ui when part of -qui) would also be reected in the genitive plural and the dative plural: If the nominative plural of rusco fox is rusqui (= ruscwi ) the corresponding dative and genitive forms can hardly be anything else than rusquin (= ruscwin) and rusquion (= ruscwion), respectively. One would think that we would also see rusquiva (= ruscwiva) as the plural possessive, and rusquinen (= ruscwinen) as the plural instrumental. There is one form that can be cited against the two latter assumptions: the related adjective ruscuit foxy, mentioned in the same source that gives us rusco, e ruscu- pl. rusqui (VT41:10). In the word ruscuit, which includes the ade jectival ending -it, there is no development cui > cwi = qui; we dont see e **rusquit. The ending -it may by its shape resemble the case endings e e -iva and -inen for plural possessive and plural instrumental. So if we have ruscuit, perhaps we would as phonologically parallel forms also see e ruscuiva and ruscuinen rather than rusquiva, rusquinen? We cannot know. I will not construct any exercises involving the plural form of the possessive and instrumental cases. In the other cases, where the plural case endings do not include the vowel -i, all one has to remember is to change the nal -o of a U-stem noun to -u before adding whatever ending is relevant. Using ango, angu- snake as our example, we would for instance have the plural allative angunnar to snakes (not **angwinna or **angwinnar or whatever; cf. the singular angunna to a snake). Likewise we would have the pl. ablative angullon or angullor from snakes (sg. angullo from a snake), pl. locative angussen in snakes (sg. anguss in a snake). As the corresponding dual forms, e we would presumably see angunta, angulto, anguts = to/from/in a e pair of snakes. Pronominal endings would also be added to the stem-form angu-, and any further endings for number or case would then be added after the pronominal ending as described in earlier lessons: angulya your snake, plural angulyar (hardly **angwilyar!) your snakes, dual angulyat your pair of snakes, dative angulyan for your snake, plural dative angulyain (hardly **angwilyain!) for your snakes, etc. etc.
NOTE: Nonetheless, the nominative plurals in -wi (-gwi, -qui) must be seen as the most striking feature of U-stem nouns. In at least one instance, this plural formation apparently spread to another noun by analogy: According to the Etymologies, entry t elek, the noun telco leg has the plural telqui, but this plural is said to be analogical. Presumably, Tolkiens idea is that telco is not a true u-stem noun (it does not come from Primitive Elvish teleku or telku, but rather descends from something like telek, telk ). Therefore, its o o plural should have been telcor, and the actual form telqui is merely due to inuence
271
from such pairs as ranco pl. ranqui or urco pl. urqui. However, telco seems to be exceptional in this respect. I dont think we should replace (say) Naucor as the plural form of Nauco Dwarf with **Nauqui.
17.4
Ordinals
We have already introduced three ordinal numbers, minya rst, (t)atya second and nelya third. All three include the frequent adjectival ending -ya (occurring in the word Quenya Elvish itself). However, it turns out that most ordinals end in -a, displacing the nal vowel of the corresponde ing cardinal number. Thus we have the following correspondences between cardinals and ordinals: canta four vs. canta fourth e lemp ve vs. lempa fth e e enqu six vs. enqua sixth e e otso seven vs. otsa seventh e tolto eight vs. tolta eighth (also tolda, presupposing toldo as a variant e e
word for 8)
nert nine vs. nerta ninth e e This table is based on an account of Eldarin numerals written by Tolkien in the late sixties, published in VT42:2427 (also see the editorial notes on pp. 3031). Tolkien indicated that the word for fth had earlier been either lemenya or lepenya (with the same ending as in minya etc.), but this irregular form was later replaced by lempa by analogy with the simple e cardinal lemp ve. Tolkiens notes present varying views as to when this e substitution occurred (whether already in pre-Exilic times, or later), but it is at least clear that in Frodos day, lempa would be the word to use when e you need to express fth. Even the words for second and third could have the ending -a ine stead of -ya. The ordinal (t)atya second was early replaced by atta, e which would be a regular formation compared to the cardinal atta two. Similarly, nelya as the word for three could also be replaced by nelda, e more clearly reecting the cardinal neld three (but in this case, nothing e is said about nelda wholly replacing nelya). e VT42:25 also lists a word for tenth, quaina, but this presupposes e another word for ten than the form cainen mentioned in the Etymologies. A root kay- having to do with the number ten seems to have haunted Tolkiens imagination for at least thirty years, so I hesitate to throw it over board just because a divergent form turns up in one late manuscript but this is not the place to discuss what forms we should accept as valid or canonical. The ordinal corresponding to the cardinal cainen could be either cainenya or caina (but hardly ?cainena). e e
272
Over the next three lessons, we will work our way through the attested ordinals, starting with the word for fourth (canta). e
273
Vocabulary
In addition to learning these new words, the student should notice that the noun ranco arm (introduced in Lesson Three) is a U-stem: rancu-. canta fourth e tana demonstrative that enta demonstrative that [yonder], [the one] over there (of time referring
to some future entity)
yana demonstrative that = the former (of time referring to some past entity) sina demonstrative this ango (angu-) snake sangwa poison lm (lmi-) night o e o polda adjective strong, burly (of physical strength only; the verb pol- can is
probably related)
halla adjective tall forya adjective right Formen (the) North (cf. Formenos, the Northern Fortress constructed by
Fanor in the Blessed Realm; the nal element -os is reduced from osto fortress; e city.) This concludes our listing of the four directions N men, Hyarmen, Rmen, Foru o men = West, South, East, North (this being their proper Middle-earth order). Just as Hyarmen South is related to the adjective hyarya left, so Formen North is related to the adjective forya right, since the reference-point is that of a person facing West (looking towards Valinor).
Exercises
1. Translate into English: A. Engw sina n i macil hirna Calandil Hallanen. e a B. Ily lamni avnier nr sinallo. e a o e C. Ango harnan forya rancurya, ar eques: Nai ily angwi e e ruvar! D. Lm yanass hirnent Nauco tana ambo entass. o e e e e E. I hall ciryar oantier Formenna; ciryar tan uvar tul i e e e nrennar Hyarmeno. o F. I canta auress tri yana rn anguo sangwanen. e e a e 274
G. I pold ranqui i nerion Formello polir mapa i ehti ohtari e mahtalallon. H. Hr e yanass marnent i canta coass mall tano. v e e e e e 2. Translate into Quenya: I. Watch that Dwarf, and dont watch this Elf! J. A land without snakes is a good land, for many Men [Atani] have died by (instrumental) snake-poison. K. During (locative) the fourth night I saw a terrifying warrior on that road, and I raised my arms (dual). L. Wish that [= nai] the strong son of Calandil the Tall will come to this land, for he will protect these cities in which we (inclusive) dwell! M. That tower (or, yonder tower) is the fourth tower made by Elves in this land. N. Those books are gone [vanw the pl. of vanwa]; they have dise appeared from your room. O. On that day you shall see your son. P. On that day they came from that [/yonder] mountain and went to this house.
275
Lesson 18
Independent pronouns. Impersonal verbs. U-stem verbs. The various uses of l. a
18.1
Independent pronouns
(inevitably entailing a discussion of certain Second Person obscurities)
All the pronouns so far discussed have been endings. However, Quenya also has pronouns that appear as independent words. Some of them are emphatic; the pronoun appears as a separate word to put special emphasis on it. These emphatic pronouns we will discuss in the next lesson. Here we will concentrate on the simplest independent pronominal elements. We have already cited Quenya sentences including the dative pronoun nin for me. The dative ending -n is here appended to an independent word for I, ni, attested by itself in the Arctic sentence mentioned in The Father Christmas Letters. (Though this posthumously published work of Tolkiens has nothing to do with the Arda mythos, the Arctic sentence is transparently a form of Quenya.) The relevant part of the sentence goes ni vla tye, I see you. The verb see is here apparently vel- rather than e cen- (perhaps vel- is see in the sense of meet?), but more remarkable is the fact that for the subject I, the independent pronoun ni is used instead of the ending -n or -ny. There seems to be no obvious reason for e this deviation from the normal system. It has been suggested that since the intended audience for The Father Christmas Letters was Tolkiens young children, he may have simplied the language to make it easier for them to gure out which word means what. However, since the latter part of the Arctic sentence employs a quite complex grammatical construction which is certainly not the literal counterpart of the English translation provided, we should hardly think of the language as simplied. For I as subject, the ending -n(y) added to the verb is normally to be preferred, but the e independent word ni may be a valid alternative. It may be noted that in one of Tolkiens draft versions for Elendils Declaration, the word that ended
276
up as maruvan I will remain/dwell appears as nimaruva, Tolkien using ni- I as a prex: SD:56. (It may be, however, that the idea of subject prexes was dropped; no post-LotR evidence of such prexes has ever been published. If I were to use the independent pronoun ni instead of the ending -n, I would let it stand as a separate word: Ni maruva.) Besides ni, we have a handful of other independent pronouns attested. One such pronoun is ta, meaning it or that (see Etym, entry ta the demonstrative tana that is of course related). One relatively early source suggests that it can receive case endings. The ten-word Koivienni sene tence published in Vinyar Tengwar #27 is not LotR-style Quenya in its entirety, but the short phrase Orome tanna lende (translated Orome came thither) may well have remained a valid wording after Qenya evolved into Quenya as we know it from later sources. The word tanna thither seems to be ta that, it with the regular allative ending -nna attached, hence to that [place] = thither. In Namri, one independent pronoun occurs in the phrase imb met = a e e between us. This is a dual pronoun, referring to Galadriel and Varda, so met appropriately receives the dual ending -t (also known from nouns) to indicate that two persons are concerned. Removing the dual ending leaves us with me, probably covering both we (subject form) and us (object form). In our example, this is an exclusive we/us, corresponding to the ending -mm, which is obviously closely related. The party addressed is not e included (Galadriel was singing to Frodo about herself and Varda). Me is also attested in its dative form, with the dative ending -n attached: men = for us, to us. The ending -ly you corresponds to an independent pronoun le, which e was apparently present already in early forms of Elvish (WJ:363). In Sindarin it had been lost, but it is precisely this circumstance which allows us to say with certainty that it survived in Quenya: In his notes on the Sindarin hymn A Elbereth Gilthoniel, Tolkien stated that the reverential 2nd person pronoun le occurring in this Grey-elven text had been borrowed from Quenya (RGEO:73). At Cormallen, the crowds hailed Frodo and Sam with the words a laita te, laita te, translated in Letters:308 as bless them, bless them. Thus we have te as an independent object pronoun them. (For this meaning, the Cormallen Praise also provides us with the already-discussed ending -t, as in laituvalmet = we shall bless them. Presumably the pronoun te and the ending -t are related.) Whether this te can also be used as a subject form (they) is unfortunately unclear. This te is possibly related to the word ta that, it discussed above: It may well be that ta early received the plural ending -i, the resulting form tai being as it were the plural form of that hence meaning something like those [ones] or indeed them. By this theory, the attested form te is simply the unstressed variant of tai (cf. adjectives in -a having plural forms 277
in -, simplied from older -ai). A direct attestation of a form tai eventually e turned up in VT42:34, and the context allows the interpretation that it may literally mean those; see below. Interestingly, the dative form for them, to them is apparently attested as tien. This could very well represent older taien, which would be tai those + the connecting vowel -e- + the dative ending -n. In this position, the diphthong ai is reduced to e, and as taien consequently morphs into teen = ten, this rather unstable form becomes e tien by exactly the same mechanism that also turns (laureai >) laure into e lauri (the plural form of the adjective laura golden). We may assume e e that the allative to(wards) them would likewise be tienna, whereas the ablative from them would be tiello. These forms would coincide with the corresponding case forms of the noun ti path, but in context, one should e normally be able to gure out what the intended meaning is. Another attested object pronoun is tye, translated thee or you. We have already quoted the phrase ni vla tye I see you from the Arce tic sentence. Other attestations come from a source that is more denitely Quenya or at least Qenya: In LR:61, Herendil addresses his father Elendil a e with the words atarinya tye-meln, my father, I love thee, and Elendil answers, a yonya iny tye-mla, and I too, my son, I love thee. There e e are a few strange things here (like -n rather than -ny or -n being used e e as the pronominal ending I in the rst sentence), but it is at least clear that tye is the object pronoun thee, and this is probably a valid form in LotR-style Quenya as well. At this point it should be noted that Quenya has (at least) two sets of pronouns in the second person. The object pronoun tye is not compatible with the ending -l(y) or the corresponding independent pronoun le, though e all of these may be translated you in English. We must distinguish between the L forms, represented by the ending -l(y) and the independent e pronoun le, and the T forms, represented by the object pronoun tye and also by the verb ending -t exemplied in WJ:364 (more about the latter in the next lesson; it is not to be confused with -t = them as in laituvalmet = we will bless them). All of these pronouns and endings have to do with the idea of you, thou, thee, but Tolkien seems to have been changing his mind back and forth as to what the basic distinction between the T-forms and the L-forms really consists of. Back in Lesson Eight, we quoted a passage that was originally meant to go into the LotR appendices, but which was not in the event included there: Tolkien stated that all these languages . . . had, or originally had, no distinction between the singular and plural of the second person pronouns; but they had a marked distinction between the familiar forms and the courteous (PM:4243). Within this scheme, the L forms would represent a polite and courteous you, whereas the T forms signal a familiar/intimate you used to address close friends and family members. This would agree well with the evidence: In Namri, Galadriel naturally uses L forms when politely a e 278
addressing a relative stranger like Frodo, and in Sindarin, the Quenya borrowing le is used as a reverential singular thee (as in the hymn A Elbereth Gilthoniel, where Varda is the party addressed). On the other hand, Herendil would obviously use a T form (tye) when addressing his own father. When Tolkien translated tye in the latter example as thee rather than you, he probably meant it to be an intimate rather than an overly solemn form (though confusingly, he might also use thou/thee to represent a formal or polite you; indeed this is how he rendered the L forms of both Namri a e and A Elbereth Gilthoniel ). What does not agree so well with this reconstruction is the fact that in WJ:364, Tolkien seems to imply that the L forms represent a plural you, whereas the T forms stand for a singular you sharply contrasting with his earlier statement to the eect that Elvish (just like English) fails to distinguish between sg. and pl. you. Moreover, L forms are unquestionably used in a singular sense in Namri, since Tolkien translated them using the a e distinctly singular English pronoun thou. I think the only solution that comes close to incorporating all the material would be to assume that the T forms properly denote singular you whereas the L forms properly denote plural you but the latter forms are also used as a polite singular you (so in Namri ). Bottom line is, one should not use the object form a e tye for you, thee if one otherwise uses L forms like the ending -ly or e the pronoun le: We are apparently dealing with two dierent kinds of you here, and the T forms are hardly interchangeable with the L forms. Based on the object pronoun tye you = thee (not subject thou), some writers have ventured to extrapolate a First Person object form nye me (cf. ni I). Apparently the form nye actually appears in Tolkiens papers, so we will adopt this nye = me here. It should be noted, however, that any case endings are added to the simplest form of the pronoun, that is, what functions as the subject form when it occurs by itself in this case ni I. Case endings are not added the object form nye me: The dative form to me is not **nyen, despite the English translation. As we know, the actual form is nin (ni-n = I-for). For you/for thee should likewise not be **tyen, for then we would be adding case endings to the object form again. Unfortunately, it is not clear what the subject form corresponding to tye thee really is, so the long-suering student must forgive yet another batch of Second Person Obscurities: Mechanical extrapolation based on the attested ni/nye pair would of course land us on ?ti as the subject form thou. However, the story is almost certainly more complicated than this. The Sindarin pronominal ending for you is said to be -g or -ch, indicating that these endings appeared as -k -, -kk - in earlier Elvish. In Quenya, a nal -k would turn into -t (cf. for instance lic- as the stem-form of a noun meaning small bird, closely reecting the root philik; but when this noun appears without any endings, its Quenya form turns into lit). If the abovementioned ending -t thou likewise comes from an original -k, we must 279
also assume that the object pronoun tye represents earlier kye (initial kyregularly turns into ty- in Quenya, cf. for instance the entry kyel in Etym, from which root Tolkien derived the verb tyel- end, cease). It is, then, this kye we must start from when trying to extrapolate the corresponding subject form. Its Quenya form would likely be ci (ki) or perhaps rather ce (ke): In the pronouns, the vowel i may seem to be peculiar to the 1st person (ni I), whereas e is more frequent (le you, me we etc.) Thus, the dative form for you, for thee may be something like ?cen, and likewise in the other cases, e.g. ablative cello from thee. If this is correct, what we have called the T forms must rather be termed the C/T forms, since the original k may be preserved in some Quenya forms as well (spelt c). But of course, we have now crossed over into the realm of Speculative Extrapolation. In the exercises below, only the attested object form tye appears. To summarize, we have ni I (object form nye me), le you (plural or polite singular; the object form is likely also le), tye object form thee, you (intimate; the subject form that could also receive case endings is unfortunately uncertain, perhaps ?ce or ?ci), me we (exclusive; probably this can also be used as the object form us), te object form them (the subject form they is uncertain, but perhaps identical; in any case, it may be that this pronoun appears as tai when stressed, and before at least some case endings it assumes the form tie-). This does not add up to a complete pronoun table; I hope to discuss what little can be inferred about the gaps in an appendix to this course. As for the functions of these pronouns, the examples cited above will already have provided the student with vital clues. These words (except the distinct object forms) can receive case endings; the dative form nin for me, to me is particularly well attested. Presumably we can also have allative ninna to(wards) me, allative nillo from me, locative niss in me and e perhaps even instrumental ninen by me. It should be noted that pronouns normally receive singular case endings, even if the pronoun is plural by its meaning; thus me we takes the singular dative ending -n, not plural -in even though the word me is plural insofar as we must refer to several people. The dual ending -t can however be added to independent pronouns, a e as indicated by the example met [the two of] us in Namri. Then any case endings would presumably also be dual: dative ment, allative menta, ablative melto, instrumental menten. (Another plausible dual form could be ?let = you two.) Another function of the independent pronouns would be to appear following prepositions, as in the example imb met between us [two] in Nae mari. In English, prepositions are followed by the object form (accusative e case), hence for instance as me rather than as I. If this applies to Quenya as well, the equivalent would be ve nye, but we cannot be certain; perhaps the Eldar would actually say ve ni = as I. The attested example imb e met between us [two] is of no help in this matter, since me (with or 280
without the dual ending -t) likely covers both the subject form we and the object form us. At least we cant go wrong as long as we are dealing with me and le (and te?), since these pronouns dont seem to have distinct subject/object forms. The function of the object forms (the attested words nye me, tye you/thee, te them + the probable non-distinct forms me us and le you just discussed) would obviously be to appear as the object of a sentence. After all, pronominal objects cant always be expressed as one of the two attested object endings -t them or -s it (though the full list of object endings is likely somewhat longer). These object endings may be added to extended innitives in -ta (caritas to do it) or to verbs that also has a subject ending (ut vienyes I have found it), but this grammatical u environment is not always present. The independent object pronouns may for instance be used in imperative phrases, as in the attested example a laita te bless them already cited. Presumably such pronouns can also be used following gerunds inected for dative (e.g. ut lien cenien tye I have u come [in order] to see you). We may also have to use independent object pronouns where the verb has no subject pronominal ending to which an object pronominal ending can be added because the subject is expressed as a separate word. So while we will bless them can be packed into one word as laituvalmet, a sentence like the people will bless them must perhaps be i li laituva te with a separate word for them. (We cant know whether e it is permissible to say ?i li laituvat with the ending -t added to the verb e even though it has no subject ending; using a separate word for them is therefore safer as well as clearer.) The preferred word order is somewhat uncertain. Quenya may seem to prefer placing independent pronouns in front of the verb. Sometimes Tolkien even prexed the object pronoun to the verb by means of a hyphen, as in the example tye-meln I love thee cited above. (Compare French je taime, a e literally I you love with the object preceding the verb instead of following it though French, as well as Quenya, normally has the object following the verb.) So perhaps sentences like I have come to see thee or the people will bless them should rather be ut lien tye-cenien and i li te-laituva, u e respectively? Sometimes, Tolkien even placed dative pronouns in front of the verb, as in the question occurring in the middle of Namri : S man i a e yulma nin enquantuva? = Now who will rell the cup for me? (notice how the Quenya word-order diers from that of the English translation). We even have one extreme example, involving the verb lumna- lie heavy, where a dative pronoun is not only prexed to the verb but the dative ending -n is assimilated to the initial l- of the verb itself: Mel-lumna is translated us-is-heavy (LR:47), sc. is heavy for us; this must be the dative form men for us, to us + the aorist form lumna lies heavy. The underlying, unattested form men-lumna apparently had to be altered because men completely glued itself to the following word and came to be 281
perceived as part of it and then there was suddenly a de facto cluster nl which Quenya phonology did not permit, so it had to become l-l instead. Yet such extra complications are apparently avoidable, for other examples indicate that independent pronouns may also follow the verb. In VT41:13 we have the sentence renya quta nin = my heart is saying to me (variant o e o renya quet nin, page 11). Presumably renya nin quta (or even e o e . . . nin-quta) would have been equally possible, but it is apparently not e necessary to employ such a word order, or indeed to prex object/dative pronouns directly to the verb. As for the subject forms discussed above, they may of course appear as the subject of a sentence, like ni I in ni vla tye I see you. Nonetheless, e here Quenya would more frequently use pronominal endings (like vlan e or vlany in this case assuming that the verb ?vel- see is valid in e e LotR-style Quenya!) For poets trying to maintain some meter it may be useful to be able to choose between independent pronouns and pronominal endings. However, the subject forms discussed above would most often be encountered, not as grammatical subjects, but with case endings attached! Even so, it is probable that pronouns like ni or le would frequently function as the subject of sentences when the copula is/are/am is left out and understood: Ni aran I [am] king, le halla you [are] tall, etc.
18.2
Impersonal verbs
Having investigated independent dative forms like nin for me, to me, we can fully understand our very few attested examples of sentences involving so-called impersonal verbs. In UT:396, Tolkien cites a verb la- to dream, o adding a brief remark to the eect that this verb is impersonal. Precisely what he meant by this was long obscure, but now we have at least one example that may be helpful in this regard. The sentences in question involves the primary verb or- urge, impel, move (elsewhere also given as an A-stem ora-). The regular aorist or e urges, impels, moves occurs in the sentence or nin caritas, translated e I would like to do so or I feel moved to do so. Literally it means [it] impels for me to do it. Notice that this sentence has no subject (though in our attempted literal translation, we had to ll in the dummy-subject it to achieve something like passable English it has no real meaning here!) Quenya by its very grammatical construction indicates that the urge perceived by the speaker impacts on him from the outside, so to speak. Feeling moved to do something is not a deliberate act carried out by a subject; this feeling rather aects the person involved, and in Quenya this is appropriately denoted by the dative case. In our attested example, a dative pronoun is involved, but we must assume that it could just as well be a regular noun: Or i Eldan lelya = [it] impels for the Elf to go = the e
282
Elf feels moved to go. The verb is the rst word of the sentence; normally the subject would come rst, but here there simply isnt any subject. As for the impersonal verb la- to dream, we must assume that the o underlying idea is the same: Dreaming is not an act done by a subject, rather it is something that happens to you; your dreams come to you quite independent of your own will, and therefore the dreamer is best presented as a person aected by his or her dreams: Hence dative for the dreamer! Tolkien gave us no examples involving la-, but the maiden dreams about o Elves could perhaps be rendered something like la i venden Eldaron o (notice that vend maiden here appears as a dative form, indicating that e the maiden is perceived as the receiver of the dreams rather than as their maker cf. the use of the dative to denote the recipient in connection with the verb to give). Such constructions, even in connection with the verb to dream, are not unheard of in the languages of our own world. As David Kiltz wrote on the Eling list (April 25, 2001): The dative has many more functions than just that of an indirect object. It can . . . denote the subjective experiencer as in . . . German mir ist kalt Im cold where you [would] use a nominative for the logical subject in English but not in many other languages. Mir ist kalt means to-me [it] is cold; the German dative form mir corresponds to Quenya nin. Given that the Quenya word for cold is ringa, it is entirely possible that mir ist kalt can be rendered directly into Quenya as nin n a ringa (or perhaps rather n ringa nin [it] is cold for me). a We dont know very many Quenya verbs that invite such constructions, though. In the entry mbaw- in the Etymologies, Tolkien mentioned that the Noldorin verb bui compel is impersonal (Noldorin being the conceptual predecessor of the Sindarin language exemplied in LotR). The Quenya verb corresponding to Noldorin bui is given as mauya-. If this can also function as an impersonal verb (though it can probably occur with an explicit subject as well), we may have a clue as to how I must or I need to would be expressed in Quenya. Perhaps I must go would (or at least could) be expressed as mauya nin lelya = [it] compels for me to go. In some cases, it may not even be necessary to complement an impersonal verb with a dative noun or pronoun. Regarding a Noldorin form of the verb corresponding to Quenya ulya- pour, namely oeil or eil, Tolkien noted that it was used for it is raining (Etym, entry ulu). Again, English by grammatical necessity lls in a dummy-subject it, but here there is obviously no real subject which actually does rain. Perhaps Quenya ulya, or continuative ulya, can likewise be used for [it] is raining: The naked e verb would be a full sentence in itself.
283
18.3
U-stem verbs
This is an obscure sub-group of verbs; having discussed U-stem nouns in the previous lesson, we may explore U-stem verbs now. Our data being very limited, this discussion must by necessity consist mainly of speculation. Verbal stems with the ending -u are not uncommon in Tolkiens early Qenya material, but as the decades went by, he may seem to have cut down their number. Of the well over 1,200 Quenya words mentioned in the Etymologies, there is only one single U-stem verb, namely palu- open wide, spread, expand, extend (and even this verb has an alternative form palyawith the much more common verbal ending -ya: see the entry pal). Around 1960, in his essay Quendi and Eldar, Tolkien mentioned the verb nicu- be chill, cold with reference to weather (WJ:417). Some years later, he also used a few U-stem verbs in the latest version of the Markirya poem: f ruslowly fade away (elaboration of the simpler verb r- die, fade), hlapuy or stream in the wind, nurru- murmur, grumble (MC:223). How are these verbs to be inected? Markirya as printed in MC:222 indicates that the active participle of hlapu- is hlpula, indicating that the a active participle is formed by adding the normal ending -la and lengthening the main vowel if possible (hlapu- becoming hlpu-). The participle a of nurru- murmur is attested as nurrula; here the vowel could not be lengthened because of the following consonant cluster (**n rrula being u an impossible Quenya word). The formation of the active participle is just about the only thing we can be quite sure about regarding this class of verbs (and therefore also the only thing I touch on in the Translate-into-Quenya exercises below). The passive participle is problematic. The normal ending -na or its longer variant -ina would presumably be applied somehow. Some have argued that we may have an attested example of the passive participle of a U-stem verb. We have earlier referred to the mysterious form tur n (obviously shortened u from tur na) in Nienors cry: A T rin Turambar tur n ambartanen, u u u [o Trin] master of doom by doom mastered (UT:138). A primary verb u tur- wield, control, govern does occur in Tolkiens material, but we would expect its passive participle to be turna (cf. carna made as the attested passive participle of car- make, do). Could the strange form tur na masu tered actually be the passive participle of a variant U-stem verb turu- to master? However, is not clear why adding the ending -na to turu- would produce tur na with a long vowel and other indirect evidence points in u another direction. As has been pointed out by some, the ending -(i)na that is used to derive passive participles also turns up in other parts of speech, and we have at least one example demonstrating what happens when it is added to a noun stem in -u: The adjective culuina orange is derived from a root kul, kulu gold. Here a diphthong ui arises when the nal -u of the stem is combined with the ending -ina. Carrying this principle over to 284
U-stem verbs, we could argue that the passive participle of palu- expand should be ?paluina expanded. Analogy with A-stem verbs would point in the same direction (cf. hastaina marred as the attested participle of hasta- to mar) but lacking attested examples, we cannot be sure. The innitive is quite problematic. It ought to be a stem with no additions. In the previous lesson we pointed out that U-stem nouns originally ended in a short -u. This original vowel is preserved unchanged whenever some ending follows, but in Quenya it had turned into -o when it was absolutely nal. Applying the same principle to U-stem verbs, the innitive of paluexpand could conceivably be ?palo. Of course, we would still see palubefore endings, for instance if this class of verbs may also have extended forms in -ta: hence ?paluta, or with an object ending ?palutas, to expand it. The aorist is little less obscure. As we remember, primary verbs assume the ending -i, preserved as such before further endings, but turning into - e when nal (sil shines, but pl. silir shine). Since the phonological shift e that makes an original nal short -i become - closely parallels the shift e that turns an original nal short -u into -o, we could plausibly argue that palu- expand ought to have the aorist ?palo expands (identical to the innitive), preserved as ?palu- before any ending (e.g. palur expand with a plural subject, palun or paluny I expand, palus he/she/it expands, e etc. etc.) However, one piece of evidence diverges from this scenario: After mentioning the U-stem verb nicu- be chill, cold, Tolkien also cited the form niqu, which he translated it is cold, it freezes (WJ:417). Is this verb e niqu to be understood as the aorist form of nicu-? Are we to understand e that just as in the case of primary verbs, the ending -i was added to the U-stem as well, and that a development nicui > nicwi ensued? After the change of nal short -i to -, this would indeed produce the attested form e (nicwe =) niqu. If so, the aorist of palu- could be ?palw, or with endings e e ?palwi-. However, we may wonder why U-stem verbs take the aorist ending -i when A-stems do not. This would not be encouraging for our nice little theory that the ending -i is applied to primary verbs merely as a kind of stop-gap to make up for the lack of any other ending (for U-stem verbs obviously do have another ending the -u itself!) Indeed it was the form niqu I was thinking of when I warned the student back in Lesson Seven, e This simplied view is not wholly unproblematic, but it works most of the time. We have now reached the point where it may not work anymore. While the aorist of palu- may plausibly be assumed to be ?palw or e with endings ?palwi-, perfectly paralleling (nicwe =) niqu as the aorist of e nicu-, we can only wonder how verbs like hlapu- or nurru- would behave if they received the ending -i already in the primitive language. They could hardly evolve into **nurrw or **hlapw, which would be quite impossible e e Quenya words. Perhaps the original diphthong ui would be preserved in all positions, and we would see ?nurrui and ?hlapui with no change of -i to 285
- even where the vowel is absolutely nal? However, I hardly have to tell e the student that we have now entered the realm of Extreme Speculation. The present tense must also be speculative, but Tolkien provided one excellent clue. It should be remembered that the present tense (e.g. s is la shining) actually represents a kind of continuous or continuative verbal stem, derived by lengthening the stem-vowel (if possible) and adding the ending -a. In the very last version of the Markirya poem, Tolkien replaced one of the participles with what would seem to be a continuative stem: As is evident from Christopher Tolkiens annotation in MC:222, his father altered nurrula mumbling, murmuring to nurrua. Here, the continuative stem in eect functions as a participle (still meaning mumbling), and the revision actually seems quite pointless, but at least Tolkien gave away that the ending -a may be added to a U-stem verb. In another context, nurrua could presumably have functioned as the present tense is murmuring. In this case, the stem-vowel could not be lengthened because of the following consonant cluster, but the present tense of a verb like palu- to expand would in all likelihood be plua is expanding. a In the past tense we can be reasonably certain that the regular past tense ending -n would be added. At least this was the case in Tolkiens earliest e Qenya: The Qenya Lexicon of 1915 lists allun as the past tense of the e verb allu- wash (QL:30). I use this system in the exercises below (but only in the Translate-from-Quenya section, so at least I wont seduce my students into constructing uncertain Quenya verb forms themselves!) The perfect tense is obscure. The augment (the prexed stem-vowel) would presumably be prexed as usual, while the vowel would if possible be lengthened in its normal position. So the perfect tenses of palu-, nurruwould presumably commence as apl-, unurr-. But what comes next is anya bodys guess. How can the ending -i that is associated with the perfect tense e be added to a U-stem verb? Would the initial -i- of the ending merge with the nal -u of the verbal stem to form a diphthong -ui-, so that we would see ?unurrui for has murmured? The perfect tense has expanded could e hardly be ?aplui, for the new diphthong ui would attract the stress and a e leave the syllable immediately before it completely unaccented. Then the long could hardly survive, for there seems to be a phonological rule proa hibiting a long vowel in a completely unstressed syllable unless this is also the rst syllable of the word and here it is not. Would we see ?apalui e with a short vowel, then? However, as we have argued earlier, the ending -i that is used in the perfect tense apparently displaces the nal -a when e added to an A-stem verb, so it is entirely possible that it would also displace the nal -u of a U-stem. From nurru-, palu- we would then simply see the perfect-tense forms unurri has murmured, apli has expanded. e a e (Likely, -i as a gerundial or innitival ending would likewise displace the e nal -u, so that we could have ?nurri for mumbling. But mumbling e as a mere verbal noun could almost certainly be nurrul, though attested e 286
examples of the abstract ending -l -ing involve A-stems instead.) e In the future tense we would presumably see the usual ending -uva. However, we can only speculate as to whether the initial -u- of the ending would simply merge with the nal -u of the stem, so that the future tense of paluwould be paluva, or whether the two us would combine to form one long u, so that we would see pal va instead. u
18.4
In Lesson Nine, we introduced the negative verb um- not do, not be (past tense um according to Etym., future tense uva according to F e riels Song). In all examples and exercises so far, we have used this verb + innitive whenever a sentence is to be negated. However, using the negative verb is not the only option available in this regard. Like English, Quenya does have a separate word for not, namely l (or la when unstressed). This word a may also be used for no. The negative verb um- and the separate negation l clearly coexist in a the language, since both were listed in the Etymologies (entries ugu/umu vs. la). There are hardly any very specic rules for when to use one or the other. If one uses the negative verb um-, it apparently takes the relevant endings for tense and pronoun, while the verb it negates presumably appears as an innitive: Umen lelya, I didnt [rst person past tense] go [inf.] . If one uses the separate negation l, the verb that is to be negated would itself a receive all relevant endings, just as if no negation were present: Lenden I went could be negated as l lenden or = Not I went = I didnt a go. (Our few examples suggest that the preferred word order is to place l a before the verb that is to be negated, though for all we know, lenden l I a went not would also be acceptable. But one should not use an alternative word-order where ambiguity can arise; see below.) This is obviously an easier way to negate a sentence than using the negative verb + innitive; one simply starts with the non-negated sentence and slips in one extra word. Indeed I introduce the word l this late in a the course partly because I didnt want to spoil my students with this easy-to-construct negation before they had the chance to get familiar with the negative verb. In many cases, using the negative verb may seem like the more elegant method of negating sentences, and sometimes the word l a not should perhaps be avoided because a similar form also has a quite dierent function (see below). Yet in some contexts it may be best to use l a instead of the negative verb. In particular, it may seem strange to construct the verb um- as a present or continuative form, corresponding to the English is . . .-ing construction. The form would be uma, but should the Elf is not watching the Dwarf be translated i Elda uma tir i Nauco? I e guess anything is possible, but think I would rather start with the positive
287
sentence i Elda t i Nauco and negate it by slipping in l in front of the ra a verb: I Elda l t i Nauco. Likewise, it may be best to use the negation a ra l in the perfect tense, especially since it is slightly uncertain what the pera fect form of um- would be: ? mi with no augment since the stem begins u e in a vowel, or perhaps ?um mi with the entire initial syllable repeated? u e In any case, I have not come is perhaps best expressed as l ut lien. a u Though (um) mien tul should be intelligible, it seems like a rather weird u e construction. Translating from Quenya to English one must sometimes take care to connect the negation l with the right verb. Notice the sentence alasaila a n l car tai mo nav mra, translated it is unwise not to do what a a e e a one judges good. (VT42:34; mo is there said to be an indenite pronoun somebody, one, apparently an alternative to quen. More obscure is the form tai, which as we argued above may be a plural form of ta that, it, hence those. If so, the sentence literally means unwise [it] is not [to] do those [things which] one judges good. Notice the element ala- un- here prexed to saila wise; it is of course related to the negation l.) For a a person used to English word order, it might be tempting to interpret the words n l as a connected phrase is not and mistranslate **it is not a a unwise to do what one judges good. However, if one bears in mind that l a connects with the following verb, in this case the innitive car to do, e the misunderstanding can be avoided: The phrases are correctly divided as alasaila n unwise [it] is + l car. . . not to do. . . (etc.) a a e As this example shows, l may be used to negate innitives as well, a and another example from VT42:34 indicates that it makes no dierence if the innitive is extended with the ending -ta to receive an object sux: a l caritas, navin, alasaila n literally not to do it, I deem, unwise is. a Tolkien himself oered the translation not doing this would be (I think) unwise. In one example, Tolkien even used l to negate an extended ina nitive in -ta that has no pronominal object ending attached: l carita i a hamil mra alasaila n, not to do what you judge good would be [lita a erally is] unwise (VT42:33). Here the extended innitive carita takes on the meaning of a gerund, the entire relative sentence i hamil mra that a [which] you judge good being its object. It seems, then, very likely that l can also be used to negate also the more regular gerund in -i. We have a e no examples, but perhaps l cari i hamil mra. . . would be an equally a e a possible wording. As for the unstressed variant of the negation l, namely its shorter form a la, our sole certain attestation of it occurs in the sentence la navin caritalyas mra, I dont judge your doing it good (VT42:33; this is a way a of expressing I do not advise you to do so). Here the main stress presumably falls on the rst syllable of the verb navin I judge, the negation la receiving no stress. One would think, however, that an important word like the negation (totally reversing the meaning of the sentence!) would normally 288
receive some stress, and in the exercises below, I have consistently used the long/accented form l. a The short form la would however have one thing to recommend it, namely that it would not clash with a quite distinct word l, which is used in a phrases having to do with comparison (though material appearing in Tyali e Tyellelliva #16 seems to indicate that this second l may also occur as e a a shorter form la). According to Bill Weldens article Negation in Quenya (VT42:3234), Tolkien was indeed somewhat troubled by this clash, and for a while he actually abandoned the negation l no, not. However, in the a last years of his life he reintroduced it, so we must live with the double function of this word. In context, it is hardly dicult to distinguish between the two ls. a According to a otherwise unpublished Tolkien manuscript cited by Welden in his article, the formula A (n) calima l B can be used for A is a a brighter than B (notice that the copula n is may be left out and undera stood). If we want a full sentence with no algebra, we may ll in A and B to produce (say) Anar n calima l Isil, [the] Sun is brighter than [the] a a Moon. However, this English translation diers from the actual Quenya wording in these respects: Calima is the simple adjective bright, not the comparative form brighter (we are still not certain what that would look like), and l does not really mean than. We are told that this l is propa a erly a preposition beyond, so the Quenya sentence actually means the Sun is bright beyond the Moon. We can certainly imagine sentences including l occurring with both its a a a a meanings: I mindon l n halla l i oron, the tower is not taller than the mountain (literally . . . tall beyond the mountain). Here the negative verb would certainly be preferable, if only for stylistic reasons: I mindon um halla l i oron. It may be possible to circumvent the ambiguities. e a We are told that instead of l beyond in phrases of comparison, one a may also use the preposition ep before erroneously glossed after in e Weldens article. Irrespective of this error, there actually is some evidence suggesting that ep or a similar form did mean after at one point of e Tolkiens ever-evolving linguistic scenario (apparently it was a variant of the apa introduced in Lesson Fifteen). Because of the uncertainties relating to ep, we will let it rest in peace here, and use l despite its ambiguity. e a
289
subject, but where someone is nonetheless aected by the verbal action, this someone can be mentioned as a dative form: Ora nin = [it] impels for me = I feel moved [to do something]. Quenya U-stem verbs, like hlapu- y [in the wind], form a particularly obscure group of verbs. The only thing that is known with full certainty about such verbs is that the active participle is formed by means of the normal ending -la, combined with lengthening of the main vowel of the verb (unless it has to remain short because of a following consonant cluster; thus the participle of nurrua grumble is simply nurrula, though the participle of hlapu- is hlpula). Early material suggests that the past tense of U-stem verbs is formed by adding the past tense ending -n, though we have no examples from the e more LotR-compatible forms of Quenya. The word l has various uses. It a can be used as a negation not (apparently placed in front of the word, usually the verb, that is to be negated). L is also a preposition beyond, a and as such it is used in phrases having to do with comparison VT42:32 citing the formula A n calima l B for A is brighter than B (literally a a A is bright beyond B notice that the adjective calima bright appears in its simplest form, with no ending corresponding to the -er of brighter).
Vocabulary
To cover all the attested ordinal numbers, we will introduce two at a time in this and the last two lessons. lempa fth e enqua sixth e urco (urcu-) bogey l 1) negation not, 2) preposition beyond, also used in comparison a ni independent pronoun I, object form nye me le independent pronoun you (plural or polite singular), probably unchanged when used as object tye independent object pronoun you, thee (intimate singular) me independent pronoun we (exclusive, cf. the ending -mm), probably e unchanged when used as object us ta independent pronoun it or that, probably unchanged when used as object; the allative tanna may be used = thither. This pronoun may have the plural form ?tai those, of which te them could be an unstressed form (alternatively, tai may be an archaic form which had evolved into te in all positions; we will only use the attested form te here). This plural pronoun assumes the form tie- before case endings (or at least before the dative ending -n) palu- to expand 290
or- (also ora-) to urge, to impel (impersonal verb; used with a dative form to
express [someone] feels moved [to do something])
Exercises
Notice that in the exercises below, as well as in the keys, we use you as the English equivalent of Quenya L-form pronouns (plural, or polite singular, you), whereas thee is used as the equivalent of the intimate singular object form tye. As for negations, we will here practice the independent negation l instead of using the negative verb. We do not here a use independent subject pronouns, only the pronominal endings discussed earlier. 1. Translate into English: A. Ut lient cenien me, l cenien tye. u e a B. Nai luva len Eldaron! o C. Neld neri lelyuvar tanna, ar i Naucor tiruvar te, an i e neri haryar harma i n alta l malta. a a D. I enqua auress orn tien tul ninna. e e e e E. I enqua aran i nro n saila l atarerya i lempa aran. e o e a a e F. Palunes coarya, crala ta i analta coa i mallo. a e G. Carnelyes tien; l carnelyes men. a H. Cennemm le i cilyass nu me, an lantanely mir ta. e e e 2. Translate into Quenya: I. My right arm is stronger than my left arm. J. Summon them to [allative] me! K. The bogies are watching me, for I fear them [Quenya: fear from them, ablative]. L. We [exclusive] will not come to see thee in the night. M. The boy will not dream about [genitive] bogies. N. The two women said: Your king did not want to give us [dual dative] the things taken [or, seized mapain, pl.] from us e [dual ablative] by his warriors. O. The man expanding his house is doing that [or, it] (in order) to make many rooms for [dative] all his things.
291
P. The sixth day will be [nauva] better than the fth day, and we [exclusive] will not feel moved to leave [auta]. [Better than = good beyond!]
292
Lesson 19
Pronouns in imperative phrases. Emphatic pronouns. Question-words: Man, mana, manen. Postpositions.
19.1
Imperative phrases may include pronominal elements. These pronouns may refer either the subject of the imperative phrase (the party that is asked or ordered to do something), or to the direct or indirect object of the requested action. Optional subject pronouns may be slipped in to make it clear whether the speaker wants one or several persons to do something. Starting from the imperative exclamation heca! be gone! or stand aside!, Tolkien noted that this word often appeared in an extended form with reduced pronominal axes of the 2nd person (WJ:364). If a single person is the addressee of an imperative, it can receive the sux -t (doubtless related to the singular object pronoun tye thee). So whereas the imperative heca! may be directed at one person or several, the extended form hecat! is explicitly marked as singular. It may be translated get thee gone! (Perhaps Fanors harsh e dismissal of Melkor when the latter came to Formenos, translated get thee gone from my gate near the end of Chapter 7 of the Silmarillion, went something like hecat andonyallo in the original Quenya?) Tolkien further noted that in the plural, the simple imperative heca! could receive the ending -l, so hecal! is an order with several addressees: Be gone, [all of] you! These examples also demonstrate that when he wrote this, Tolkien had come to see the distinction between the T-forms and the L-forms of the Second Person as being primarily a distinction between singular and plural you. Of course, matters are slightly muddied by the fact that L-forms like the ending -l or -ly also function as a polite singular you (translated thou e in Namri ). Of course, none can say whether the ending -l could be used a e in an imperative phrase to denote a polite order with a single addressee. Hecat! sg. and hecal! pl. are our only attested examples of this use of the endings -t and -l. The imperative heca! is perhaps not a very typical imperative. As outlined in Lesson Sixteen, instead of the ending -a an inde293
pendent imperative particle (or a) is normally employed, combined with a a following verbal stem. Attested examples include vala! rule! and a a laita! praise! If the endings -t, -l are to be slipped into such a phrase, they must probably be attached to the verbal stem, e.g. a laitat! praise! (to one person), a laital! praise! (to several persons). Laita- bless, praise is of course an A-stem. Primary verbs like carmake, do appear with the ending - in such commands, as is evident from e e the negative imperative va car! dont do [it]! (WJ:371). This - almost a e certainly comes from an earlier -i, and before endings we would still see -i-, just as in the aorist of these verbs. Hence probably va carit! dont do a [it]! (to one person), plural va caril! dont do [it!] (to several persons). a Or with a positive command like tul! come!, we could likewise have a e a tulit! if one person is asked to come, whereas tulil! would refer to more a than one: Come ye! An imperative phrase may also include pronominal elements referring to the direct or indirect object of the requested action. In the previous lesson we referred to various independent pronouns. Nye me, me us, le you, tye thee, ta it and te them can all appear as independent words; indeed our sole attestation of te them involves an imperative phrase where this word occurs by itself: A laita te, laita te! bless them, bless them! (from the Cormallen Praise). However, Tolkiens Quenya rendering of the Lords Prayer indicates that object pronouns may also appear suxed to the imperative particle . The principle can be illustrated by a home-made a example: Consider a simple imperative phrase like tir! watch! If we a e want to say watch them, slipping in the object pronoun te, it would be possible to let it follow the verb (as in the attested example a laita te bless them), hence tir te. However, it would apparently be equally a e permissible to let the object pronoun come before the verb, in which case it glues itself to the imperative particle . Watch them! could therefore be a t tir! a e e
NOTE: Since te them now becomes the nal syllable of a word of several syllables, the spelling conventions we have adopted require that the nal -e is here spelt with a diaeresis: -. The same would apply to the nal -e of the object pronouns nye, me, le, e tye if they were directly suxed to e. g. m tir watch us. Of course, this is just a a e e an orthographic complication that has nothing to do with the structure of the language: In many texts, Tolkien does not use the diaeresis at all.
Also dative pronouns (like nin to/for me, men to/for us, tien to/for them) can be directly suxed to the imperative particle ; at least Tolkiens a Quenya Lords Prayer contains an example of men being so suxed. A phrase like, say, sing for us! could thus be rendered men linda! a What happens if an imperative phrase contains two pronouns, denoting both the direct and the indirect object? We have no Tolkien-made example to guide us, but the imperative particle can hardly receive more than one pronominal sux, and the example a laita te indicates that a pronoun does 294
not necessarily have to be suxed to the particle. So it would certainly be permissible to let one of the pronouns, for either direct or indirect object, remain an independent word and sux only the other pronoun to the particle a e a e . Do it for me! could then be either nin car ta! or ta car nin!, a according to preference. (Perhaps one could also say nin caritas!, using a an extended verbal stem carita- and the sux -s for it.) Tolkiens translation of do not lead us into temptation in his Quenya Lords Prayer reveals that the negated form of the imperative particle (do not!) can also receive pronominal axes. By ca. 1960, in Quendi and Eldar, Tolkien used va as the negative imperative, combining the particle a a with the negation v (-va). In his earlier Lords Prayer rendering he used a a slightly dierent word for do not (incorporating the negation l instead a of v), but there is no reason to doubt that pronouns denoting the direct a or indirect object can be attached to the later form va as well. So while a watch us! could evidently be translated m tir!, the negative command a e e dont watch us! may probably be translated vam tir! a e e
19.2
Emphatic pronouns
In the previous lesson, we discussed a number of independent pronouns (also mentioned in the section above). There also exist certain other pronouns that likewise appear as separate words, though they are closely related to the corresponding pronominal endings. These words function as emphatic pronouns. Consider the nal lines of Namri : Nai hiruvaly Valimar! Nai ely a e e e hiruva! In LotR, Tolkien provided the translation maybe thou shalt nd Valimar! Maybe even thou shalt nd it! Of course, hiruvaly means thou e shalt nd thou being expressed by means of the pronominal ending -ly. But then this is repeated as nai ely hiruva, maybe even thou shalt e e nd [it]. Notice how the ending -ly is replaced by the independent word e ely, which is obviously closely related. Ely means simply thou or you, e e but with special emphasis on this pronoun. Using such an independent form is like italicizing the pronoun in English: Maybe [none other than] thou shalt nd. Tolkien used the translation even thou, adding an extra word, to bring out the emphatic quality of the pronoun. (The shorter independent word for you, thou, le, is apparently not emphatic.) Another independent emphatic pronoun is iny I or, since it is eme phatic, rather I myself or even I, I and no one else. In LR:61, Herendil tells his father Elendil that he loves him, and Elendil responds in Quenya: A yonya iny tye-mla, and I too, my son, I love thee (the initial a, e e apparently translated and, would seem to be a variant of the more usual word ar though a can also be a particle of address). Here the pronoun iny, translated I too by Tolkien, gives special emphasis to Elendils own e
295
identity: I love you [just as you love me]. Iny is more emphatic than the e shorter form ni, just like ely is more emphatic than le. e The form iny is obviously related to the pronominal ending -ny, and e e Namri clearly demonstrates that ely corresponds to the ending -ly. a e e e (Draft versions of Namri have the pronoun ell and the ending -ll ina e e e stead. Some think this has another meaning, perhaps plural you instead of singular thou, but I tend to believe that Tolkien simply revised the form of the pronoun without altering the sense.) It is evident from these examples that emphatic pronouns can be derived from the corresponding pronominal endings by adding the ending to some vowel. But what vowel? e e We have an e- in ely you, but an i- in iny I. This emphatic word for I may be exceptional in preferring i- as its initial vowel. The student may remember that the ending for my, -nya, seems to prefer -i- as its connecting vowel where one is required (as in atarinya my father, LR:61). The close association of these First Person pronominal forms with the vowel i seems to reect the stem-vowel of the most basic stem ni I, listed in the Etymologies. We may tentatively conclude that the other emphatic pronouns show the initial vowel e-, as in ely. This is supported by Tolkiens Quenya e version of the Lords Prayer, where he used emm as the emphatic proe noun corresponding to the pronominal ending -mm for exclusive we. It e occurs in his translation of the line and forgive us our trespasses as we forgive those who trespass against us. Here, the pronoun is emphatic to draw special attention to it (stressing the parallelism: we want God to forgive us just like we in turn forgive others). It feels natural to let this pronoun be emphatic in English as well, though this is indicated merely by putting extra stress on it, and the distinction is not represented in writing (unless one uses italics!) The Quenya system of using distinct emphatic pronouns is undeniably more elegant. While iny, ely (changed from ell) and emm are all the emphatic e e e e pronouns occurring in published and soon-to-be-published material, we can condently extrapolate at least some more forms. The exclusive emphatic pronoun for we, emm, would obviously correspond to the inclusive varie ant elm (just like the exclusive ending for we, -mm, corresponds to e e the inclusive ending -lm). Possibly the ending -nt for they would core e respond to an independent word ent. e Other forms are somewhat less certain. The ending for dual inclusive we is uncertain, but if it is -lw, we would expect elw as the independent e e emphatic pronoun. The form ely can function as a singular you, as is e evident from Tolkiens translation even thou in his LotR rendering of Namri. As I have suggested earlier, the L forms of the Second Person a e may seem to indicate either a polite singular you or a plural you; perhaps ely could assume the latter sense as well. The T forms of the Second e Person, exemplied by the object pronoun tye and the ending -t that may be used in the imperative, seem to denote an intimate and usually singular 296
you, but it is uncertain what the emphatic form would be. It depends on how we reconstruct the longer form of the ending -t. If it is -ty, the e emphatic pronoun would most likely be ety. But I have also given some e reasons for believing that Tolkien meant -t to represent an older -k, in which case the longer variant of the ending might well be -cc and then emphatic e pronoun would probably be ecc. e Things are also rather obscure in the Third Person. It is clear that the emphatic pronouns are derived from the long forms of the pronominal ende ings. The word iny I corresponds to the long subject ending -ny, not e its shorter (and more common) form -n; likewise, ely thou corresponds e to the long ending -ly, though this ending may also appear in the shorter e form -l. There are hardly any short emphatic pronouns **in, **el to go with the short endings. Our problem is that in the Third Person, the short ending -s is the only form attested in material Tolkien certainly intended to be LotR-compatible. We have earlier theorized that -s may correspond to a longer ending -ry (for earlier -sy, -zy); it would then have the same e e e relationship to the ending -rya his, her as the ending -ly you has to e the ending -lya your. If such an ending -ry exists, it could correspond e to an emphatic pronoun ?ery (even) he/she. But when we start extrape olating from extrapolations, the risk of losing contact with Tolkiens actual intentions obviously becomes acute. We know a few more things about the emphatic pronouns, though. They can receive case endings; in Tolkiens Quenya rendering of the Lords Prayer, the emphatic pronoun emm we once occurs with the dative ending -n e attached: Emmen. (This also conrms that pronouns take the simplest case endings, that is, the ones used with singular nouns even if the pronoun is plural by its meaning: We must refer to several persons, but the simplest dative ending -n is used instead of the plural form -in.) As pointed out in the previous lesson, the 1st person dative form nin to me, for me is particularly well attested, but if one wants to say for me myself, for me [and no one else], it would be better to start from the emphatic form iny e and derive the dative form inyen. We have already cited the nal lines of Namri, where part of one sena e tence is repeated with special emphasis on the pronoun: Nai hiruvaly e Valimar! Nai ely hiruva! From this example it appears that if a pronome inal ending is replaced by an independent pronoun, the ending is removed from the verb (not nai ely hiruvaly with both an independent pronoun e e and the corresponding ending still attached to the verb). Yet in the draft version, when the pronoun ely was still ell, Tolkien did use precisely that e e system: Nai ell hiruvall. He apparently decided that this version was e e somewhat over-complete, and writers should probably avoid this system. Yet the verb following an emphatic pronoun should perhaps receive at least the plural ending -r if the pronoun is plural. If we want to transform (say) hiruvamm we shall nd into two words to put special emphasis on e 297
the pronoun we, I guess the resulting phrase would be emm hiruvar, e not **emm hiruva. If the pronoun ely does denote both singular poe e lite you/thou as well as plural you, the following verb may then reveal in what sense it is used. In nai ely hiruva! the pronoun must be singue lar (thou) since there is no -r attached to the verb. Perhaps nai ely e hiruvar! would also be a possible sentence, but here it would be clear that a plural you is intended. If this is so, Frodo could after all be certain that Galadriel addressed Namri to him alone (thou), not to the entire Fela e lowship (you) though the pronoun ely as such is perhaps ambiguous e in this regard. As pointed out above, the emphatic pronouns can receive case endings, as can the shorter independent pronouns discussed in the previous lesson. It is, however, somewhat uncertain whether the endings for genitive and possessive should be added to such independent pronouns. Obviously, some kind of independent words for (say) mine or yours would be required to have a fully functional language. The ending -lya can be used to express your, as in parmalya your book, but how do we say the book is yours? Published examples of Quenya provide no clues. It has long been theorized that not only the subject endings that may be suxed to verbs, but also the possessive endings that may be added to nouns (like -nya my or -lya your), have corresponding emphatic forms. This has never been explicitly conrmed. However, given the fact that the ending -ly you, thou corresponds to an independent form ely e e you (emphatic), it certainly seems plausible to assume that the ending -lya your could correspond to an independent form elya yours. This word could then be used in a sentence like i parma n elya, the book is yours. a But it could also be used for special emphasis, so that while parmalya means simply your book, elya parma would mean your [and no one elses] book. If the theory holds, other emphatic possessive pronouns would be erya his, her, elma our [inclusive] and emma our [exclusive] of course corresponding to the endings -rya, -lma, -mma. The attested word elya for your, thy would be either a plural your or a polite singular your (or thy); the intimate singular your could be something like ecca or etya. If the ending for dual inclusive our is -lwa, we could have elwa as the corresponding emphatic pronoun. An even bigger if relates to the emphatic word for their, since it must be extrapolated in two steps: Assuming that the ending -nt for they corresponds to an unattested ending e -nta their, we may assume that it this ending in turn corresponds to an emphatic possessive pronoun enta. As for the independent form of the possessive pronoun my, corresponding to the ending -nya, we would probably expect the form inya (since the subject ending -ny corresponds to an e independent form iny). e These extrapolations are not entirely unproblematic. Some of these forms 298
actually occur in Tolkiens published writings, but with quite dierent meanings. Inya, for instance, is mentioned in the Etymologies but there it is not an emphatic word for my, it as an adjective female, of a quite different derivation (see the entry ini). Erya, which we have suggested as an emphatic word for his or her (corresponding to the ending -rya), would coincide in form with the adjective erya single, sole (Etym, entry ere). The student may already have noticed that enta as an emphatic word for their would clash with the demonstrative that, yonder (Etym, entry en). Since demonstratives may seem to follow the noun they connect with (cf. UT:305: vanda sina = this oath), we could perhaps distinguish coa enta yonder house from enta coa their house. As for our extrapolated forms inya and erya coinciding with Tolkien-made adjectives of quite dierent meaning, what can I say? Some, no doubt, would feel that this throws considerable doubt on this whole business of extrapolating emphatic possessive pronouns to go with the attested subject pronouns. Actually I think these extrapolations are about as plausible as any forms not directly attested can be. Given the uid state of the Quenya language in general and its pronouns in particular, dierent stages of Tolkiens ever-changing conception are quite likely to involve coinciding word-forms of dierent meaning. However, in the exercises below, I focus on the better-attested subject pronouns, like iny e and ely. e
19.3
Question-words
Many English words frequently used in questions show an initial wh-: who?, what?, where?, which?, whose?, why? etc. In Tolkiens Elvish, an initial ma- has similar connotations; he referred to the Eldarin interrogative element ma, man (PM:357). This element Tolkien seems to have borrowed from Semitic languages; cf. for instance Hebrew ma = what? We have nothing like a complete list of Elvish interrogative words, but some of them do occur in published material. Best attested is the word for who, man, which occurs in a question in the middle of Namri : S man a e i yulma nin enquantuva? Now who will rell the cup for me? Man = who occurs repeatedly in the Markirya poem, e.g. in the question man tiruva fna cirya[?] Who will heed [/watch] a white ship? (In one out a of ve occurrences, Markirya as printed in MC:221222 has men instead, but this must be a misreading of Tolkiens manuscript; there seems to be no grammatical variation that could explain the variant form.) Perhaps man can receive case endings, so that we can have (say) the genitive form mano = whose? If man is who (referring to people), what is the word for what (referring to things)? In LR:58, Tolkien apparently lets man cover what? as well: Man-i? is translated what is it? This is hardly LotR-style Quenya; e
299
the ending -i = is was almost certainly abandoned later. F e riels Song also uses man in the sense of what: Man . . . antva nin Il vatar[?] a u = What will the Father [Ilvatar, God] give me? Whether Tolkien at one u point intended man to mean both who and what, or whether he simply changed his mind back and forth regarding the precise meaning of man, cannot be determined now. Turning to a post-LotR source, we nd the question mana icoimas Eldaron[?], which is translated what is the coimas [life-bread = Sindarin lembas] of the Eldar? (PM:395; a variant reading occurs in PM:403). How are we to interpret this word mana, that would seem to correspond to what is in the translation? It could be a word ma what (which would be a direct borrowing from Hebrew!) + the copula n is, here directly suxed and shortened to -na. However, the form ma a is known to have at least two dierent meanings in Quenya (see the next lesson), so I tend to doubt that Tolkien intended it to mean what? as well. Rather mana is simply a word for what, and there is no explicit copula is in the sentence mana icoimas[?] = what [is] the life-bread? It may be that mana what can receive case-endings. Published material provides no word for why?, but by one suggestion we can work around this gap by adding the dative ending -n to mana the resulting form manan meaning literally what for? By its form, mana resembles the demonstratives sina this, tana that and yana that (yonder). Since vanda sina means this oath (UT:305, 317), perhaps vanda mana? would mean what oath? or which oath? Mana vanda? would rather mean what [is] an oath?, given Tolkiens example mana icoimas[?] = what [is] the life-bread? The same late document that provides this example also includes a word e for how?, appearing as part of the question manen lamb Quendion ahyan[?] = How did the language of Elves change? (PM:395). Intereste ingly, this word manen how? seems to include the instrumental ending -nen, added to the interrogative element ma- (PM:347). Manen could mean, literally, what by? (it may even be a contraction of mananen for all we know). It seems highly plausible that other case endings than that of the instrumental can be added to ma-. Perhaps we can have locative mass? where?, ablative mallo whence?/where from? and manna e whither/where to?, lling further gaps in our vocabulary. However, the exercises below only involve the attested forms man who?, mana what? and manen how?
19.4
Postpositions
We have introduced various prepositions, such as nu under, or over or ve as, like. Prepositions are so called because they are typically positioned before (pre) the word(s) they connect with. In Quenya and English
300
alike, one would say nu alda under a tree not **alda nu a tree under. (Of course, the word order may be quite garbled in poetry, as when the LotR version of Namri has Vardo tellumar nu luini Vardas domes a e under blue for under Vardas blue domes. Here, we are rather discussing the normal, non-poetic word order.) There are also postpositions, similar in function to the prepositions, but coming after (post) the word or words they connect with. In English, the word ago may be seen as a postposition, since it comes at the end of phrases like three years ago. Some languages would use a preposition instead, expressing this meaning something like before three years. Interestingly, a word simply glossed ago is mentioned in the Etymologies: Y. It is not a mentioned or exemplied elsewhere, and since its English gloss is all we have to go on, we may assume that y is a postposition in Quenya as well. If so, a three years ago could be translated directly as neld loar y. e a At an older stage, Elvish apparently had many postpositions. Tolkien stated that at the oldest stage, prepositional elements were normally attached apparently meaning suxed to noun stems (WJ:368). Many of the Quenya case endings Tolkien clearly meant to represent originally independent elements that had merged into the noun they followed. For instance, the ending -nna for allative is obviously related to the preposition na or ana, likewise meaning to, towards. Yet Quenya had at least a few postpositions that had not evolved into case endings, but still appeared as independent words. Whereas y discussed above is only attested as an isoa lated word mentioned in Etym, the word pella beyond appears in actual Quenya texts, and it does seem to be a postposition. Namri has And n a e u e pella for beyond the West, and this is apparently not just another example of a poetic word order, for Tolkien did not alter this phrase in his prose version of the song. (The word And n West is an alternative to u e the more usual word N men. And n may also be dened as sunset or u u e even evening.) Pella is used as a postposition in the Markirya poem as well, where it connects with a noun inected for plural ablative: Elenillor pella is translated from beyond the stars (literally from stars beyond). Whether pella could also be used as a preposition, just like its English gloss beyond, is impossible to say. Another word for beyond, the l that a is also used in comparison, does seem to be a preposition (and perhaps the nal syllable of pella is meant to be related to this l). Even so, I would use a pella as a postposition only, employing the word order observed in Namri a e and Markirya alike.
301
to the verb of the imperative phrase; if several people are asked to do something, the ending -l (basically meaning you, plural) may be used instead. Pronouns denoting the direct or indirect object of the imperative may glue themselves to the imperative particle . The (long forms of the) pronomia nal subject endings that may be suxed to verbs may also be used to derive independent emphatic pronouns by adding the relevant ending to e-. For instance, the subject endings -ly you, thou and -mm we correspond to e e the independent emphatic pronouns ely (even) thou and emm (even) e e we. The emphatic word for I, corresponding to the ending -ny, however e shows i- rather than e- as its initial vowel: iny. Quenya interrogative e words apparently show an initial ma-; attested examples are man who, mana what and manen how. Postpositions are similar in function to prepositions, but follow rather than precede the word(s) they connect with. The word pella beyond seems to function as a postposition; Namri has a e And n pella (not ?pella And n) for beyond the West. The word y u e u e a is glossed ago (Etym, entry YA), so perhaps it functions as a postposition just like its English gloss.
Vocabulary
otsa seventh e tolta eighth (In a late document Tolkien actually changed this form to tolda: e e
VT42:25, 31. Apparently he considered changing the word for 8 from tolto to toldo. We may accept toldo 8 and tolda 8th as valid variants, but in the e exercises below I prefer tolta to go with tolto, the form of the word 8 as e listed in the Etymologies.)
iny I (emphatic pronoun) e ely you, thou (emphatic pronoun) e emm we, exclusive (emphatic pronoun) e elm we, inclusive (emphatic pronoun). (This form is not directly attested, but e
since it forms the logical counterpart of emm, this extrapolation from the ending e -lm seems so plausible that I will use it in the exercises below.) e
man who? mana what? (according to one interpretation of the sentence where this word occurs) manen how? pella beyond (postposition) y ago (postposition like its English gloss?) a e when r
302
NOTE on the word e when: This is (almost certainly) not an interrogative word, r despite its English gloss. A question like when will you come? can hardly be translated ** e tuluvaly? No Quenya word for when? as a genuine question-word has been r e published, though it may be possible to work around this gap for instance, we could use a circumlocution like l mana(ss)? = (at) what time? The word e is used for when u e r in the sense that it introduces information about when something occurs; it is attested in F riels Song. The relevant phrases are not quite LotR-style Quenya, but the use of e r may be noted all the same: Yva tyel ar i-narquelion, ir ilqua yva ntina = there e e e o a u will be an end and the Fading, when all is counted; man . . . antva nin Il vatar . . . e Anarinya queluva? = what will the Father give me . . . when my Sun faileth? At r present we have no later attestation of such a word, and since it is highly useful, we may well adopt e when into our attempted Unied or Standard Quenya. Many post-Tolkien r writers have used it already. It has been suggested that the initial of e is related to the - r denite article i the, whereas the nal -r can be equated with the word r [24-hour] e e day (the word aur refers to the daylight period only). If so, e basically means the e r day and of course, what will the Father give me . . . the day my Sun faileth would still make sense. However, it may seem that the word e when existed long before the word r r day; the latter apparently emerged as Tolkien was writing the LotR Appendices. I e wouldnt hesitate to use e for when in general (not limiting its application to the day r when. . . )
Exercises
As described above, Tolkien sometimes added extra words when translating emphatic pronouns, e.g. ely = even thou (to bring out the emphatic e quality of the word). However, in the keys to the exercises below, as well as in the Translate into Quenya section, we have adopted the simpler system of italicizing emphatic pronouns (e.g. ely = you; by the way, we will use e this translation rather than thou since the L-form pronouns may seem to be ambiguous as to number, just like English you). 1. Translate into English: A. Man marn i coass cainen loar y, e iny l marn e e a r e a e tass? e B. Mana elm polir car? e e C. Ata antat nin, ar vata nurtat nillo! a D. Man ely cenn i otsa auress? e e e e E. Antuvant ily i annar inyen ar l elyen! e e a F. Mana i neri hirner i nress i oronti pella? o e G. Manen ely poluva orta i alta ondo e iny um pol e r e e e caritas? 303
H. Sellemma mar coa yanass; emm marir i tolta coass e e e e e mall sinass. A tulil ar cenil coamma! e e a 2. Translate into Quenya: I. Who has done that [ta]? J. What did you nd in the seventh room when you went there [or, thither = tanna]? K. When you have come, I want to leave [auta]. L. Give us the wine! (an order explicitly addressed to several persons). M. Give the wine to us and not [l] to the warriors! (Us is exa clusive here. To = dative, not allative. This order is explicitly addressed to emphone person only.) N. The seventh warrior has come from beyond the great mountains. O. Beyond [or, behind] the eighth door [or, gate, ando] you found a great treasure; we [exclusive] who came after you [apa le] did not nd a thing! P. How did the evil Dwarves nd them seven days ago?
304
Lesson 20
The obscure verb to be. Ma as a possible interrogative particle. Sa introducing nominal clauses.
We are now fast exhausting the reasonably certain facts about Quenya (though as I have tried to make clear, even many of the facts must be considered tentative deductions). This last regular lesson is already moving far into the twilight zone of linguistic obscurities.
20.1
The verb to be
(including some remarks on the form a) e
The various forms of the verb to be have always been a problem. The LotR version of Namri includes the copula n is. A draft version of a e a the same poem combines the copula with a plural subject, resulting in the form nar are. In July 2001, the future-tense form nauva will be nally turned up in Vinyar Tengwar 42 p. 34. We have mentioned and practiced all of these forms before. Several tense-forms, as well as the innitive, are missing. To start with the innitive, what is to be in Quenya? We have no other clue than the apparent fact the innitive of A-stem verbs is identical to the verbal stem itself, with no additions. In Etym, na is listed as the stem of [the] verb to be in Q[uenya]. So conceivably, n may also function as an innitive: I a want to be an Elf = ?Merin n Elda. But of course, Tolkien may have a imagined something entirely dierent. Another question lacking a clear answer: Can n, nauva and other tensea forms be combined with pronominal endings? We may have one example to guide us. In the very early (some would even say rst) Qenya poem Narqelion, written in 1915 or 1916, Tolkien used the form nre. Since there is a no straightforward Tolkien-made translation of this poem, we cannot be entirely certain what it means. Christopher Gilson, analyzing the entire poem in light of the almost contemporaneous Qenya Lexicon, concluded that nre a may mean it is (VT40:31). A 3rd person marker -re (or if you like, -r) e was perhaps present in Tolkiens later forms of Quenya as well, though at 305
the later stages we might expect it to mean she rather than it see Lesson 15. If the regular pronominal endings can be attached to n, we would a expect forms like the following: nn or nny I am a a e nt you [intimate singular] are, perhaps also with a longer form (nacc a e or nty???) a e nl or nly you [plural or polite singular] are a a e ns he/she/it is, conceivably with a longer form ?nry; there may a a e also be gender-specic forms nro he is, nr she is (the latter being a a e attested, sort of, in Narqelion) namm we [exclusive] are, nalm we [inclusive] are, and perhaps e e something like nalw we [dual inclusive] are e nant they are e Notice that the long of n would be shortened whenever a consonant a a cluster follows, as when the endings -mm, -lm, -lw, -nt are attached. e e e e I assume that the endings with a consonant + y (-ny and -ly, plus the e e two extrapolated endings ?-ty and ?-ry) would not have the power to e e shorten a preceding vowel. Compare the of m hand remaining long a a before the possessive pronominal ending -rya her in Namri : mryat a e a her hands, dual. (The future tense nauva will be would be able to receive all pronominal endings with no modications: nauvan or nauvany e I will be, nauvalm we will be, etc.) e Perhaps the verb n can even receive two pronominal endings, for subject a and predicate, e.g. nnyes = I am he. (Compare the Tolkien-made form a ut vienyes = I have found it, the second ending denoting the object.) u But of course, all of the above may be nonsense. The word n is and a pronominal endings like -n I are attested separately, but combining them to produce nn for I am could for all we know be just as wrong as saying a I is in English. In the case of the verb to be, Tolkien may very well have dreamed up some wonderfully irregular forms, which only his unpublished manuscripts can ever reveal. Luckily, there is a way to get around the whole problem, and that is to simply use an independent pronoun and leave out the entire copula to be: It is understood. When translating blessed art thou in his Quenya rendering of the Hail Mary, Tolkien simply wrote the word for blessed immediately followed by the pronoun ely thou. There is no copula art e in the Quenya text. So we can apparently feel free to build copula-less sentences like iny Elda I [am] an Elf, ely uma you [are] evil or thou e e e [art] evil. (If a plural you is intended, the adjective should probably be pl.: umi.) The shorter, less emphatic pronouns would presumably work e just as well: Ni Elda, le uma, etc. e Yet we cant always do without the verb to be, and another problem has to do with the past tense was. We have no certain attestations of it. Adding 306
the normal past-tense marker -n to the stem na would of course produce e something like ?nn, but this awkward form seems most improbable. In all a e likelihood, Tolkien actually envisioned an irregular form. A form n was e has long been rumored to occur in Tolkiens unpublished papers. The closest we have ever got to having this assumption conrmed would be Christopher Gilsons analysis of the same very early Qenya poem referred to above, Narqelion. In Vinyar Tengwar #40, page 1213, we nd Gilson trying to decipher what he calls an especially enigmatic phrase from the poem. It incorporates the form n, and Gilson argues (p. 13): e The preterite [= past tense] of the verb to be is not given in [the] Q[enya] L[exicon], but this tense-form is listed for many verbs, and frequently identied as such. The Qenya preterite has a variety of formations, but one of the familiar types is seen in kanda- blaze, pret. kandane. . . It is possible that forms like kanda-ne actually arose as a construction with the verb-stem plus a form of the verb to be, i.e. that a meaning like blazed derived syntactically as in English phrases like was blazing or did blaze, with an originally tenseless form of the stem kanda(to) blaze, blazing combined with the past tense expressed in the ending -ne was. . . The present tense of the verb to be is given in QL as n it is . . . and if the preterite is n or ne, there a e would be a parallel with certain other verbs where the present vs. preterite is marked solely by a change of -a to -e, as in pantaopen, unfold, spread, pret. pante, or sanga- pack tight, pret. sange. Since it so happens that Mr. Gilson has access to virtually all of Tolkiens linguistic papers, he must know perfectly well whether a form n was oce curs in the material or not. We may assume, then, that Gilson feels that he would be violating somebodys copyright if he were to say loud and clear that n is the Qenya word for was and so he has to pretend that he is e merely deducing this word from already published material. Remembering that Tolkien must surely have discussed the various forms of to be in the vast amount of linguistic manuscripts that he left behind, and combining Gilsons article with more recent statements made by his group to the effect that they would not write something they know to be wrong, we can apparently treat n = was as an as-good-as-attested word. e However, even if such a word did exist in the early forms of Qenya, it may of course have been abandoned in the more LotR-compatible forms of the language that emerged decades later. It may be noted, though, that the past tense ending -n was never abandoned and if Gilsons theory that e there is a connection between this ending and the verb was does reect Tolkiens actual ideas, the word n was may have survived into LotRe style Quenya. Anyhow, writers can hardly do without a word for was, and 307
currently we have no better alternative than n. Some writers have used it, e and based on the pair n is / nar are, people have also extrapolated a a form ner were to be used in conjunction with plural or multiple subjects. If the present tense n is can be combined with pronominal endings, a this would presumably be true of n as well (it would then be reduced to e ne- with a short -e- before the endings -mm, -lm, -lw, -nt, just like e e e e we must assume that n- is shortened to na- before these endings). But a rather than being combined with pronominal endings, it may well be that the copula would most often simply be omitted in the past tense as well, an independent pronoun being used instead of an ending: Ely uma = e e past tense you [were] evil or present tense you [are] evil, according to context. (However, I suspect that the future-tense copula nauva will be would rarely be omitted like this.) Five Quenya tenses are attested: Present (or continuative), aorist, past, future and perfect. The verb to be may not make any distinction between present and aorist (n is covering both); the past tense was may be n a e as discussed above, and the future tense is attested as nauva will be. This leaves only the perfect tense has been. Normally, the perfect is formed by prexing an augment similar to the stem-vowel, lengthening the stemvowel in its normal place and adding the ending -i, e.g. ut li has come e u e from the stem tul come. However, it is less than clear how the stem n a could be tted into this pattern. A form ?ani would be quite unstable; ae the group i would tend to become a normal diphthong ai. Yet ?anai still a e does not strike me as a particularly likely form, and I could recommend it to really desperate writers only. Presently it is simply impossible to tell how the Eldar would say has been (presumably a very frequent word, since Tolkienian Elves were ever more and more involved in the past VT41:12!) The imperative of n is also a matter of conjecture. It is far from obvious a how to say be! as in be good! I have used the form na, combining n a a (-na) with a prexed variant of the imperative particle . By its form, this a imperative na be! would have the same relationship to n as Tolkiens a a word va dont! has to the simple negation v no! Yet na remains an a a a extrapolation, and if we could have asked Tolkien what he actually had in mind, I would have been somewhat surprised if this had been his answer. So please think of na as a form which would t into Tolkiens general system, a but which is not (as far as I know!) Tolkienian as such. In addition to the N forms of the verb to be (n/nar, n, nauva) a a e few entirely dierent forms of related meaning occurs in the material. The pre-LotR text F riels Song has ye for is and yva for will be. Instead e of appearing as independent words they may also be turned into endings, -i e and -iva, attested in such forms as mri is good and hostainiva will e a e e be counted (cf. mra good, hostaina gathered, countered). However, a as I pointed out in Lesson Four, Tolkien may seem to have abandoned such 308
forms. The ending -i has so many other meanings (gerundial ending as e in enyali recalling, abstract ending as in veri boldness, feminine e e ending as in Vali female Vala) that Tolkien may have decided that it e should not be burdened with the meaning is as well. A few writers have used the ending -i is, but my advice would be to let these formations e from F riels Song rest in peace. Another verb we should consider is denitely not a word Tolkien abandoned, for it is found in writings postdating the publication of LotR with Namri, incorporating the verb n is, in it. Thus it is clearly meant a e a to coexist with n, and probably expresses a somewhat dierent shade of a meaning. We are talking about the verb a (or with a capital E, E). e a Readers of the Silmarillion will remember this word from Tolkiens creation myth. God, Eru Ilvatar, grants objective existence to the Music of the u Ainur with this word: I know the desire of your minds that what ye have seen should verily be . . . even as ye yourselves are, and yet other. Therefore I say: E! Let these things Be! (Ainulindal ). Tolkien explained that E a e a as a name of the universe is not originally a noun, but actually a verb: The Elves called the World, the Universe, E It is (footnote in Letters:284). a This world, or Universe, [the Creator] calls E, an Elvish word that means a It is or Let it Be (MR:330). So a can be either the present (or aorist) tense [it] is or the imperative e let it be! (In the latter sense it would parallel such one-word imperatives in -a as heca! be gone! or ela! look!) How does a is dier in meaning e from n? It has been suggested that one of these verbs means is referring a merely to some particular instance, while the other refers to a permanent or habitual state. In a sentence like the man is drunk, one word for is would simply indicate that the man is drunk right now, whereas the other would imply that he is a drunkard by habit. Parallels to such a system can be found in Spanish (a language Tolkien loved). Given the extreme scarcity of source material, nothing can be ruled out at this stage, but I would put my money on another theory. It should be noted that Tolkien translated a not only as is, he also used the rendering e exists (VT39:7). This suggests that a has a more absolute meaning than e n. The verb a is related to the noun engw thing (see below), a thing a e e being perceived as something that exists. It may be that n is a mere a copula used in phrases describing the state of something, introducing a noun (samb sina n caimasan this room is a bedchamber), an adjective e a (samb sina n pitya this room is small) or even a prepositional phrase e a (samb sina n ve i samb yass hirnenyet, this room is like the room e a e e where I found them). On the other hand, a refers to the solid, independent e existence or presence of some subject, and it can perhaps be used with no other additions than this subject (e.g. Eru a = God exists). Tolkien e informs us that the word ala, by its form obviously the active participle of e a, was also used as a noun being denoting a spirit whose natural state is e 309
to exist without a physical body. Balrogs, for instance, were alar (MR:165). e Basically, the word only refers to existing ones. At the beginning of a sentence, the verb a may possibly be used in the same sense as English e there is, to assert the existence or presence of something: E malta i a orontissen [there] is gold in the mountains, a nr i sambess [there] e e e is a man in the room, ar neld nissi i coass, [there] are three women e e e in the house. However, all the sentences in the previous paragraph were constructed by me. One of our very few Tolkien-made examples of a occurring in an actual e sentence forms part of Cirions Oath. E turns up as the verb of a relative a sentence: i Eru i or ily mahalmar a tennoio, the One [God] who is e e above all thrones forever. Since the literal meaning could well be that Eru exists in this sublime position, this does not contradict the interpretation set out above. It should be noted that a, rather than n, is the verb to e a use when describing the position of something (the position being specied either by a phrase including a preposition, like or ily mahalmar above e all thrones in the sentence above, or by a noun appearing in the locative case). Perhaps we could have sentences like i samb yass a i harma a e ee e or i samb yass a i nr s the room where the treasure is, is above e e e e the room where the man is now referring not so much to mere states as to existence, presence, position. This is the best the present grammarian can do with so few examples. How is a inected? E itself would seem to be the present or aorist form; e a the imperative (used by Eru in the Ainulindal ) is identical. Perhaps a can e e also function as the innitive. The future tense could be something like euva. The perfect has existed seems impossible to reconstruct with even a shadow of condence. As for the past tense, there are several possibilities. The regular form would of course be an, which form I might well use e e myself. Another possibility is suggested by material published in Vinyar Tengwar #39, pp. 67. Tolkien discusses the sound he represented as n (= the ng of king, a sound that in Quenya had been lost between vowels) and how it may be possible to deduce where it had occurred earlier: The former presence of intervocalic n, later lost in Quenya, could be detected by consideration of the relations between ta indicates and teng indicated, tengwe sign, e e and comparison with a exists beside engwe thing. Tolkiens point is e that before w, the sound n had been strengthened to ng, and this combi nation was not lost. The Quenya word tengw sign is explicitly said to e represent primitive te-w (VT39:17), and we are clearly to understand that n e engw thing likewise comes from e-w, though Tolkien did not explice n e itly mention this primitive form. Thus it can be established that the original roots are ten- and en-, respectively, though in the verbs ta indicates and e a exists the n has been lost because it occurred between vowels: These e words are obviously meant to descend from primitive tea, ea. What is n n 310
interesting is that Tolkien also mentioned the form teng indicated. Une fortunately, this English gloss is ambiguous. Is this indicated as the past tense of indicate, or indicated as the passive participle of the same verb? The Quenya form could be a kind of adjective of similar meaning, since many Quenya adjectives end in -. But if teng indicated is to be understood e e as the past tense of ta indicates, we must conclude that the past tense e of a exists should probably be eng rather than an: Perhaps teng e e e e e - + the past tense marker is meant to come from te-n i.e. the root ten n e -n, the combination nn evolving into a long or double nn which was not e lost even between vowels, but came out as ng in Quenya. If so, a primitive past tense e-n existed might likewise produce Quenya eng. n e e The verb a can presumably receive pronominal endings like any other e verb, e.g. amm we exist. e e
20.2
In the previous lesson, we introduced the words man who?, mana what? and manen how? They can be used to construct certain kinds of questions, but the commonest kind of question is not exemplied in published material: How do we construct the kind of question that may be answered with a simple yes or no? English uses various procedures to get from a simple assertion to a question about whether something is actually true. Assertions like it is so or he has come can be turned into a questions by fronting the verb: Is it so?, Has he come? However, in contemporary English, this procedure only works with a few verbs. An assertion like he wrote the book is turned into a question by adding a form of the verb to do at the beginning of the sentence and turn what used to be the nite verb into an innitive: Did he write the book? Obviously, the simplest way of turning an assertion (a declarative statement) into a question would be to simply slip in some kind of particle that merely signals, This is not an assertion that something is so and so, but a question about whether it is so and so. Many languages of our own world do employ such particles (e.g. Polish czy), and this simple and elegant way of constructing yes/no questions seems to have considerable appeal to language-constructors as well. Esperanto has the interrogative particle chu (cxu), apparently based on the Polish word, and the sentence he wrote the book li skribis la libron is turned into a question did he write the book? simply by adding chu at the beginning: Chu li skribis la libron? But what about Quenya? Still clinging to this example, how do we turn the declarative statement tences i parma (s)he wrote the book into a question? Does Quenya have an interrogative particle we can slip in? In PM:357, quoted in the previous lesson, Tolkien refers to ma or man
311
as an Eldarin interrogative element (PM:357). Man is the Quenya word for who, but may the shortest possible interrogative element ma function as an interrogative particle? Some writers have made this assumption. It may be noted that one Quenya verb for ask is maquet- (past tense maquent, PM:403). Since maquet- transparently means ma-say, it may e be suspected that in some way or another, Quenya questions often involve the element ma. It has come to my knowledge that there is a Tolkien manuscript which does refer to ma as an interrogative particle. If Tolkien meant a particle like Polish czy or Esperanto chu, we may apparently turn a declarative statement into a yes/no question simply by slipping in a ma, most likely at the beginning of the sentence (no attested examples): Tences i parma (S)he wrote the book > Ma tences i parma? Did (s)he write the book? N enta n Elda That woman is an Elf > Ma n enta n Elda? s a s a Is that woman an Elf? E malta i orontess [There] is gold in the mountain > Ma a malta a e e i orontess? Is [there] gold in the mountain? e I nr caruva coa The man is going to build a house > Ma i nr e e caruva coa? Is the man going to build a house? Conceivably ma might also be used in so-called dependent questions, with u the force of English whether: Umin ista ma ut lies, I dont know whether (s)he has come. This is speculation, however, and it must be emphasized that we have yet to see the interrogative particle ma in any actual Quenya sentence written by Tolkien. The manuscript in which he refers to ma as an interrogative particle is apparently quite early, so this particle may well belong to some variant of Qenya rather than the more LotRcompatible forms of Quenya. Even assuming that the system I presupposed when constructing the examples above does indeed correspond to Tolkiens intentions at some stage, he may very well have decided upon something else later. Indeed certain fragments of post-LotR Quenya material includes a word ma of a quite different meaning: it functions as an indenite pronoun something, a thing (VT42:34). Whether this implies that ma as an interrogative particle had been abandoned is impossible to say (PM:357 at least conrms that ma as an interrogative element survived into the post-LotR period). Whether the two mas can coexist in the same version of Quenya is a matter of taste, unless it turns out that Tolkien actually addressed this question in some manuscript (and I wouldnt hold my breath). Presently, the system sketched above is probably the best we can do when it comes to constructing yes/no questions in Quenya.
312
20.3
We have introduced several words that may be translated that: the pronoun ta and the demonstratives enta and tana (tanya). There is, however, another kind of that as well very common in any substantial text. We are talking about that as a particle introducing socalled nominal clauses. As we know, nouns can take on various functions in a sentence. Very often they appear as the subject or object of a verb, as when the noun Elda Elf functions as the object of the verb ista- know in the sentence istan Elda I know an Elf. Sometimes, however, it is useful to treat an entire sentence as a noun, so that it can take on noun-like functions in a sentence. Consider a simple sentence like you are here. If we want to treat this sentence as a noun and slip it into a longer sentence to function as (say) the object, English may signal the noun-like status of the words you are here by placing the word that in front of them: That you are here. Now this entire phrase, a so-called nominal clause, can function as the object of a verb: I know that you are here. It could also be used as the subject of a sentence, as in that you are here is good. (But in the latter case, English would often prefer to slip in a meaningless dummy-subject it at the beginning of the sentence and place the true subject at the end: It is good that you are here.) What, then, is the Quenya equivalent of that as a particle forming such nominal clauses? Our sole attestation of this important particle comes from a rather obscure source. A few years ago, a person who called himself Michael Dawson made a posting to the Tolkien Internet mailing list. He claimed to be quoting at two removes from an old Tolkien letter which could not be dated more precisely than years earlier than 1968. It was a Quenya greeting including the words merin sa haryaly alass which is supposed to mean, literally, I wish e e that you have happiness. (The initial merin actually appeared as meriu in Dawsons post; lower-case n and u are often very dicult to distinguish in Tolkiens handwriting. Meriu would be a quite meaningless form, and the translation provided settles the matter.) As we see, the word sa is here used as a particle turning the sentence haryaly alass you have happiness e e into a nominal clause, so that it can function as the object of the verb merin I want/wish. There are several questionable points here. For one thing, not everyone is convinced that the Merin sentence, as it is often called, is genuine at all. I am told that various eorts to get in touch with this Michael Dawson have so far proved futile, and it is somewhat disturbing to notice that his posting was made on April 1. On the other hand, Carl F. Hostetter (who has seen nearly all of Tolkiens linguistic manuscripts) briey commented on this sentence in VT41:18 and apparently recognized it as genuine, though 313
he has later specied that it does not occur in any manuscript he knows of. The word sa is not inherently implausible; it could mean basically it (related to the ending -s), so that merin sa haryaly alass originally e e or basically signies I wish it [, namely that] you have happiness. But of course, even if the word sa is genuine Tolkien, it is impossible to say what stage of Q(u)enya it belongs to. Assuming that the Merin sentence is actually written by Tolkien, I would say that it is probably quite early, since its vocabulary corresponds so well to that of the Etymologies (of the midthirties notice the use of the verb harya- have, possess, otherwise only attested in Etym). So given the ever-changing nature of Tolkiens conception, sa as a particle forming nominal clauses may well have been abandoned by the time Tolkien published LotR. Yet writers can hardly do without this important word, and presently sa is our sole alternative. Accepting sa as a word for this meaning, our example above I know that you are here could perhaps be rendered into Quenya something like istan sa aly sinom (cf. ista- to know, sinom in this place; here e e e e and above we theorized that a rather than n is the word used for is e a when a certain position is discussed). That you are here is good could presumably likewise be rendered sa aly sinom n mra. It is good e e e a a that you are here could correspond to something like n mra sa aly a a e e sinom (if the verb n is can be fronted). In this or any other context, e a Quenya would hardly need a dummy-subject like the it of the English sentence, so I would not expect to see ns or whatever. a In (slightly archaic) English, a that-clause may describe an intention; here is a Tolkienian example: The titles that [the Vala Orom] bore were e many and glorious; but he withheld them at that time, that the Quendi should not be afraid (WJ:401; modern idiom would slip in a so before that, but the meaning remains the same). It would be interesting to know whether a Quenya sa-clause can be used in this sense. If not, we do not really know how to express this meaning in Quenya.
314
the presence, existence or position of something (in the latter case a may e connect with a prepositional phrase, as in the attested example i or ily e mahalmar a [God,] who is above all thrones). According to certain e pieces of evidence, Quenya (or at least some variant of Qenya) had an interrogative particle ma. Presumably it can be added (at the beginning?) of declarative statements to turn them into yes/no questions. According to one (possibly dubious) source, the form sa can be used for that as a particle introducing nominal clauses (as in I know that you are here, he said that this is true, etc.)
Vocabulary
nerta ninth e quaina tenth (So according to a text on Eldarin numerals published in VT42. This e
presupposes another word for ten than the form cainen occurring in the Etymologies perhaps quain, not attested by itself. Maybe cainen would correspond e to an ordinal tenth something like cainenya, and perhaps writers should use either cainen/cainenya or quain7quaina, but to complete our list of Quenya e e ordinals 1st10th we will use the attested form quaina here.) e
ma, possible interrogative particle sa, that introducing nominal clauses (according to a source of somewhat questionable value)
nm place o e sinom in this place or simply here (apparently combining si- as in sina e
this with -nom, a shortened form of nm place, hence sinom = [in] this e o e e place)
tenna preposition until, as far as a verb is = exists (past tense either an or eng, future tense perhaps euva) e e e e nan conjunction but n has been hinted to be the past tense of n is, hence was e a ista- to know (notice irregular past tense sint instead of **istan; perhaps the e e
perfect has known should similarly be isinti) e
lerta- can in the sense of be allowed to (English often uses may in this
sense; see note below) NOTE on the Quenya verbs for can: As explained in a Tolkien manuscript published in VT41, at least three Quenya verbs can be used to express the idea of be able to. The verb pol- that we have introduced earlier primarily means to be physically able to do something (cf. the adjective polda [physically] strong, apparently related to this verb). The verb lerta- means to be allowed to do something, to be free to do something because there is no prohibition though in some contexts it may also be interchangeable
315
with pol-. The verb ista- to know (pa. t. sint) may be combined with an innitive to e express can in the sense of know how to, referring to intellectual ability. Thus istas tec would mean he can write [because he knows the letters of the alphabet]. Polis e tec would be he can write in the sense of he is physically able to write [because his e hands are not paralyzed or tied up or something]. Lertas tec would mean he can e write in the sense of he may [= is allowed to] write.
Exercises
1. Translate into English: A. Ely Nauco, l Elda. e a B. Cennen sa i nero ranco n rcina. e a C. I aran n taura, nan i tri n saila l i aran. e a e a D. Ma sintely sa nu i coa a nurtaina harma? e e E. Ma lertan lelya nm sinallo? o e F. Ava suc, an a sangwa yulmalyass! e e e G. Ma euvant sinom i quaina auress e tuluvan? e e e e r H. Istamm sa a nulda samb coa sinass, nan l ih e e e e a riemmes, ar tenna hirimmes uvamm ista mana a i sambess. e e e 2. Translate into Quenya (for convenience using L-forms rather than T-forms to translate you): I. I was rich [several possible translations]. J. The king said: You may not go to the place whence [yallo] you have come, but I know that I will go thither [tanna]. K. I can [= know how to] read, but I cannot read in the darkness. L. We (incl.) know that Elves exist. M. Did they dwell [mar-] here until he ninth year when the warriors came? N. We (excl.) know that the men could speak the Elven-tongue [Eldalamb], but not the Dwarf-tongue [Naucolamb]. e e O. The women said that you have seen the great worm [ango] that is in the mountain. P. On the tenth day the Sun was bright.
316
Appendix A
Keys
Lesson One
1. Marking the accented vowel or diphthong: A. Alcar B. Alcar e C. Alcarinqua D. Calima E. Oronti o e F. Untim G. Envinyatar H. Ulund e I. Eruil a e e J. Ercass As for Christopher Lees accentuation nai yarVAXa RASSelya e TALTuva notto-CARinnar, the words yarvaxa and taltuva are e correctly pronounced. However, rasselya should have been accented rassELya rather than RASSelya, and notto-carinnar should have been notto-carINNar rather than notto-CARinnar. Perhaps we are to assume that Saruman in this scene uses some special meter employed in magical invocations, discarding the normal stress rules? 2. K. Ohtar: C (ach-Laut) L. Hrv: D (hr originally denoting unvoiced r, later becoming nora e mal r) M. Nahta: C (ach-Laut) N. Heru: A (English-style breath-H, though in Valinorean Quenya it had been ach-Laut) 317
O. Neht: B (ich-Laut) e P. Mahalma: In early Exilic Quenya probably C (ach-Laut), but by the Third Age it had evidently become A (breath-H). Q. Hell: A (breath-H) e R. Tihtala: B (ich-Laut) S. Hlc: D (the group hl originally denoting unvoiced l, later beo e coming normal l) T. H e: A (breath-H) si
Lesson Two
1. A. Horses B. Either just king, or a king with an indenite article, depending on what English grammar demands in the context where the word occurs. C. The horse D. The horses E. Kings F. One people under one king. G. The king and the queen. H. Maidens 2. I. Tasari J. Eldar K. I arani L. Lier M. I rocco nu i tasar. N. Vend ar tri. e a O. I tri ar i vendi. a P. Anar ar Isil (probably not i Anar ar i Isil, since in Quenya the words
denoting these celestial bodies seem to count as proper names, requiring no denite article)
Lesson Three
1. A. (Two) eyes, (natural pair of) eyes. B. Two eyes (= atta hendi, referring to two eyes only casually related, like
two eyes of two dierent persons, one eye from each. The dual form hendu, on the other hand, refers to a natural pair of eyes.)
318
C. Two trees. D. Two trees (= atta aldar, referring to any two trees. Aldu, on the other
hand, refers to some kind of closely related pair of trees, like the Two Trees of Valinor in Tolkiens mythos.)
E. One man and one woman. F. The stones. G. Floors. H. Mountains. 2. I. Atta ciryar. J. Ciryat. K. Rancu (if the example peu pair of lips holds, the dual ending -u rather
than -t is always used in the case of bodyparts occurring in pairs, even where there is no d or t in the noun)
M. Andu (ending -u rather than -t because of the d occurring in this word) N. Aiwet. O. Atta aiwi. P. Neri ar nissi.
Lesson Four
1. A. A black horse. B. Bright eyes (hendu = a natural pair of eyes) C. Three dead men. D. Beautiful birds. E. A queen is a mighty woman. F. The mountains are great. G. Best interpreted a king [is] mighty, the copula being left out and understood, but it could also mean a mighty king with a somewhat unusual word-order (an attributive adjective would more often come before the noun it describes: taura aran rather than aran taura). H. The man and the woman are wise. Theoretically at least, exercises A, C, and D could also be interpreted black [is] a horse, bright [are] eyes, beautiful [are] birds, the copula being left out just as in Exercise G. But when the adjective 319
comes immediately in front of the noun it describes, it must normally be assumed that it is used attributively and not predicatively. On the other hand, when the order is noun + adjective, as in G, a copula is/are may well be left out. 2. I. I ninqu ando. e J. Alta cirya. K. I talan n carn. a e L. Min mor sar ar neld ninqui sardi. e e e M. Sail arani nar taur neri. e e N. I taura nr ar i vanya n nar umi. e s e O. Eldar nar vany. e P. Eldar nar vanya li. (Notice that here, the adjective agrees in number e
with the singular noun li a people, which it describes attributively. It does e not agree with the plural noun Elves, as in the previous exercise.)
(In exercises K, M, N, O, P, the copula n/nar may be left out and a understood.)
Lesson Five
1. A. The woman is laughing. B. The fattest Dwarf is eating. C. The queen is watching the king. D. The greatest mountain is mighty. E. The man is summoning the most beautiful maiden. F. The bird is singing. G. The Dwarves are seizing the four Elves. H. The mightiest king is wise. 2. I. I n t i analta cirya. s ra J. I an mi neri nar rini. u e K. I Elda mpa i parma. a e L. Canta neri caitar nu alda. e M. I assaila Elda cenda parma (an-saila becoming assaila by assime
ilation)
N. I aran ar i tri cendar i parma. a e O. I aiwi lindar. e P. I canta Naucor t rar aiw. e 320
Lesson Six
1. A. The man read the book. B. The Dwarves ate. C. The king summoned the queen. D. A woman sang. E. The maidens watched the Elf. F. The ve horses lay (/?were lying) under the big willow. G. The stars shone. H. The Dwarf saw a horse. As suggested in F, it may be that it is also permissible to translate Quenya past tenses using the was/were . . . -ing construction, e.g. B) the Dwarves were eating, D) a woman was singing, F) the ve horses were lying. However, Quenya may well have distinct verb forms for this meaning. Published material provides no clues in this matter. 2. I. Nauco hirn i harma. e J. I Elda quent. e K. I rocco camp. e L. I aran mell Eldar (or . . . i Eldar with the article if the phrase the e
Elves is taken as referring to some particular Elves rather than the Elvish race in general)
M. Nr tenc lemp parmar. e e e N. I tri ortan. a e O. I arani haryaner alt harmar. e P. I aran ar i tri tultaner canta Eldar ar lemp Naucor. a e
Lesson Seven
1. A. Many Dwarves possess treasures. B. The sun will rise and the birds will sing. C. Six men will watch (/guard) the gate. D. Every Man (= non-Elf human) will die. E. All Men die. F. A wise man reads many books. G. Every star shines above the world.
321
H. The Elf seizes the Dwarf. In A, B, E, F, and G, the aorist tense is used to describe various general truths that are more or less timeless. In H, the aorist is used to describe a momentary, duration-less action. 2. I. J. K. L. M. N. O. P. Ilya Elda ar ilya Atan. I Elda hiruva i Nauco. I rocco cap or i Nauco. e I aran tur rimb ohtari ar turuva ilya Ambar. e e I aran ar i tri cenduvar i parma. a I ohtar tur macil. e Ily rvi matir hrv. e a a e Enqu rvi mtar hrv. e a a a e
In K, the aorist describes a momentary, duration-less action. In L and N, the aorist (tur) describes a general characteristic or habit of an e individual: the king (always) controls many warriors, the warrior (generally, habitually) wields a sword. In O, the aorist describes a general truth about lions, contrasting with the present (continuative) tense in P (mtar = are eating), describing the ongoing activity of some a particular lions instead.
Lesson Eight
1. A. B. C. D. E. F. G. H. I. J. K. L. M. N. O. P. The man has found the treasure. The lions have eaten the esh. The king has summoned the queen. The women have read the book. The evil queen has seized the seven Dwarves. You have written seven books. I have spoken. You have seen it. I nr ut li. e u e I otso Naucor amtier. a I seldor ecnier r imb i aldar. e a e I enqu Eldar oroitier i otso Naucor. e I Nauco unurti harma. e Alaitien [or, alaitieny] i aran, an i aran elri ily mli. e e e e o Alantiel [or, alantiely], ar ecnienyes. e e Emtienyes. e 322
2.
Lesson Nine
1. A. [The] shining moon is rising over the world. B. The jumping Dwarf fell through the oor. C. I can hear a singing maiden. D. One man wielding one sword will not terrify the eight mighty warriors. E. A thrall seizing a mighty man is not wise. F. The eight lions lying under the trees (a)rose, for the lions wanted to eat the men. G. A lion cannot stop eating [/cease to eat] esh. H. The terrifying warrior stopped watching [/ceased to watch] the people, for the warrior was not wise. (Another possible interpretation:
stopped guarding instead of stopped watching.)
2.
I. I nr roitala i Nauco n ohtar. e a J. I aran mern lelya. e K. I vend um verya cen i tri. e e e a L. I llala nissi lender mir i coa. a M. I tolto lelyala Naucor polir hir rimb harmar. e e N. Umel(y) laita i Elda, umil(y) laita i Atan, ar uval(y) e e e laita i Nauco. O. Merin(y) lelya ter Ambar ar lerya ily lier. e e P. Veryala nr lend ter i ando ar mir i oron. e e The key to Exercise K (the maiden did not dare to see the queen) is the only possible translation using the vocabulary I have provided to far, but I cannot say for certain that cen- to see can also be used in the sense to meet, which is how an English-speaking person would normally interpret this word used in such a context. But then see = cen- may of course be used in its most basic sense, so that i vend e um verya cen i tri may be interpreted the maiden did not dare e e a to look at the queen.
Lesson Ten
1. A. I love them deeply. B. They sing beautifully, like (the) Elves. C. All nine gates are watched. D. They want to nd it swiftly. 323
E. You have two books, and nally you have read them. F. I have really [/truly/actually] seen an Elf. G. The hidden treasure will not be found. (Possibly, the Quenya wording
uva hirna would suggest: . . . will not have been found, referring to some future situation.)
H. They did not want to do it, for seeing it was enough [/suced]. 2. I. Elendient nuldav ter i nr. (Surely you understood that have e e o e gone was to be rendered by the perfect form of the verb lelya-, or did you start messing with lelyaina or something, desperate to bring in a separate word for gone? No need. . . ) J. I hostain Eldar merner cenitas. e K. Tcina lamb um ve qutina lamb. e e e e e L. Lemp ciryar umer farya; nert farner. e e M. Anwav pustuvan [or, pustuvany] caritas. e e N. Lintav hostanent i nert ruhtain Naucor. e e e e O. Teldav cenuvalyet ve emriel(y) cenitat. e e e P. Umint mer hlaritas. e e The word order is certainly somewhat exible; the adverbs in M, N, and O could probably also follow the verb (e.g. hostanent lintav e e for they swiftly gathered). Cf. my own key to I. But when an object or an innitive is to follow, I nd it slightly awkward to separate it from the nite verb by inserting an adverb between them. Of course, you can always have the adverb at the end of the sentence as well.
Lesson Eleven
1. A. They found the dead warriors sword. (Genitive of former possessor.) B. The stars of heaven are shining. (Genitive of location: the stars are in
heaven.)
D. They shall see the King of Men and (of) all lands. (Genitive describing the relationship between a ruler and the ruled people or territory.)
F. The queens evil brothers want to rule the peoples of the world.
(I trio umi torni: genitive of family relationship. Ambaro lier: genitive a e of location the peoples are in the world.)
324
H. The ten lions quickly ate the esh of the horse. (I rocco hrv the a e
horses esh genitive of source, the esh coming from the horse. Notice that the noun rocco horse is unchanged in the genitive singular, since it ends in -o already.)
2.
I. Menelo aiwi [or, (i) aiwi Menelo] cenuvar cainen ohtari imb i alt s (Menelo aiwi the birds of heaven genitive of locae e ri.
tion.)
J. I arano ml [or, (i) ml i arano] ulyan limp mir (i) analta o o e e i yulmaron [or, mir i yulmaron analta]. (I arano ml the kings o
thrall genitive denoting the relationship between the ruler and the ruled, or various relationships between people in general. Notice ulyan as the past e tense poured in the transitive sense. [I] analta i yulmaron or i yulmaron analta: the biggest of the cups, partitive genitive the biggest cup being one of all the cups mentioned. Cf. Tolkiens elenion ancalma brightest of [/among] stars.)
K. I Eldo toron [or, (i) toron i Eldo] hostan (i) cainen parmar e elenion.(I Eldo toron the Elf s brother: genitive of family relationship.
Notice that when the genitive ending -o is added to a noun like ending in -a, like Elda, it displaces this nal vowel. (I) cainen parmar elenion the ten books about stars: the genitive being used in the sense about, concerning. Perhaps the word order elenion cainen parmar is also possible, but it feels less natural.)
L. (I) alta s e i nro [or, i nro alta s e] ull mir cilya. ([I] r o e o e r e
alta s e i nro the great river of the land genitive of location. Notice r o e ull as the past tense poured in the transitive sense; contrast ulyan in e e Exercise J above.)
Lesson Twelve
1. A. Both phrases may be rendered the wine of the Elves. However, the genitive phrase i limp Eldaron implies the wine coming e 325
from the Elves, sc. wine somehow originating with or obtained from the Elves. On the other hand, the possessive phrase i limp e Eldaiva implies wine owned by the Elves at the time which is being considered, irrespective of the origin of the wine. B. You have (/possess) a cup of gold.(Yulma maltava cup of gold:
the possessive-adjectival case used in its compositive sense, denoting what something is made of.)
C. The horse of the Elf [/the Elfs horse] has fallen into the deep gorge. (I rocco i Eldava the Elf s horse: possessive case used of current
ownership. One could argue that Tolkienian Elves seem to be so close to their horses that to them, their steeds are more like family members than possessions, and then it would be more appropriate to use the genitive case: i rocco i Eldo or i Eldo rocco. But as I said in the Introduction, the Elves of these exercises are not necessarily Tolkienian Elves.)
E. Great walls of stone hid the houses of the ten richest men of the city. (Rambar ondova walls of stone: compositive -va. I coar i cainen
analy neriva the houses of the ten richest men: possessive case used of e current ownership. [neri] i osto [men] of the city: genitive of location, the men being in the city. Notice that the word osto is here inected for genitive, though the ending -o is invisible since this noun already ends in -o. Cf. also Exercises L and N below.)
F. The house of the kings sister [or, the kings sisters house] is red.
(In the phrase i coa i arano selerwa, the genitive i arano the kings is dependent on selerwa sisters, which possessive form in turn points back to i coa the house. The genitive refers to a family relationship, the possessive to the current ownership of the house. I coa i selerwa i arano, the house of the sister of the king, would be a clearer wording.)
G. One of the thralls has seized the sword of the king. (Min i mlion e o
one of the thralls: partitive genitive; i macil i aranwa the sword of the king: the possessive case used of current ownership. Of course, if the thrall runs away with the sword of the king, it eventually turns into i macil i arano instead, the genitive indicating former possession. If the rebellious thrall kills the king with his own sword, this action would produce the same eect immediately, the king instantly being reduced to a former possessor: I macil i aranwa enters the kings chest, i macil i arano comes out through his back.)
H. The maidens brother found all the treasures of the eleven Dwarves between the four horns of the white mountains. (I vendo toron e
the maidens brother: genitive of family relationship; i harmar i minqu e Naucoiva the treasures of the eleven Dwarves: possessive case denoting
326
current ownership. I canta rassi i ninqui orontion the four horns of the white mountains, either partitive genitive if the horns are perceived as being part of the mountains, or genitive of location if the horns are thought of as being in the mountains.)
2.
I. S limpeva uller mir i nero anto [or, mir (i) anto i nero]. ri
(S limpeva rivers of wine: compositive -va. I nero anto the mans ri mouth: partitive genitive, the mans mouth being part of him. Also notice uller, not ulyaner, as the intransitive past tense of ulya- to pour.)
J. I seldoron seler [or, (i) seler i seldoron] hostan (i) engwi e i seldoiva ar lend mir (i) coa i triva. (The genitive phrase e a
i seldoron seler the boys sister refers to a family relationship; the possessive forms seldoiva and triva have to do with current ownership of the a things and the house, respectively.)
K. One possibility: (I) muil i nissiva varyan alta harma male e tava. (In the phrase harma maltava treasure of gold, the -va case is used
in the same sense as in Exercise B above but the secrecy of women can be rendered in various ways. Using the possessive-adjectival case as suggested here, it refers to the secrecy of the women as a more or less permanent attribute of theirs. But one might also use the genitive, i nission muil or (i) e muil i nission, focusing rather on the womens secrecy at the particular e time in the past which is being related. One might even interpret it as a kind of subject genitive, the women being the ones who are secretive and thus the subjects of the secrecy.)
L. One possibility: I minqu ohtari umer pol varya (i) sr e e e e i osto [or, i osto sr], an alta morni lantan. (Rendering e e e e
the peace of the city as i sr i osto, using the genitive case, would focus e e on the peace of the city as its attribute at one specic time the peace emanating from the city, so to speak. Conceivably it could also be interpreted as a genitive of location, the peace being in the city. Certainly one might also say (i) sr i ostova, using the possessive case, but then we are rather e e talking about peace as a permanent attribute of the city, and the message of this sentence is that the peace did not prove to be quite permanent after all. But a Quenya-speaking Mayor, expressing a pious wish may the peace of the city last forever, might well say ostova.)
M. One possibility: Lelyuvant ter nr alt aldaiva ar rimb e o e e e ondoiva, an merint cen (i) osto i taura ohtarwa. (Nr e e o e
alt aldaiva ar rimb ondoiva a land of great trees and [of] many rocks: e e the possessive-adjectival case describing characteristic features of the land. (I) osto i taura ohtarwa is the most natural translation of the city of the mighty warrior if we imagine him to be still alive, somehow owning the city where he dwells. But of course we may also be talking about a long-dead warrior who has brought fame to the city where he once lived, and then it
327
would be more natural to use the genitive case, denoting a former possessor: (i) osto i taura ohtaro or i taura ohtaro osto. This wording might also be appropriate if the warrior happens to have founded the city in question, since the genitive case may denote an originator living or dead.)
N. One possibility: Ramba muilva varyan (i) nurtaina malta e e i osto [or, i osto nurtaina malta], ar umen hiritas. (Ramba
muilva a wall of secrecy: the -va case is used in its compositive sense, e the metaphorical wall being made of secrecy. Notice the lengthening of the nal vowel in muil secrecy when the ending -va is added, as seems to be e characteristic of words with ui in their second-to-last syllable; cf. the attested example huinva of gloom. If we translate the hidden gold of the city e using a genitive as suggested here (i) nurtaina malta i osto it would probably be a genitive of location: the hidden gold is in the city. But if we take the word city as referring primarily to the people of the city, we might rather use the possessive case of current ownership: i nurtaina malta i ostova.)
O. (I) nr (i) Eldaiva n nr rimb vany engw o e a o e e e va; nr o e u Eldaron n nr morniva, an i Atani i nro [or, i a o e e o e nro Atani] umir hlar (i) alya lamb (i) Eldaiva. (Possibly o e e e
Eldaiva should here receive the article i in both of its occurrences, since the reference may not be to Elves as a race, but rather to the particular Elves living in a particular country. Anyhow, these possessive forms refer to current ownership of the land [nr] and the language [lamb]. In the o e e phrases nr rimb vany engw a land of many beautiful things and o e e e va nr morniva land of darkness, the possessive-adjectival case describes o e e characteristic features of the land; cf. Exercise M above. Notice the long vowels of engw and morniva. The former represents earlier engweiva va e [engw + -iva], the diphthong ei later becoming long whereas in morniva e , e the nal - of morni darkness is lengthened because the word ends in two e e short syllables. In the phrase u Eldaron without Elves, the preposition u regularly governs the genitive case. In accordance with Tolkiens usage in one late source, one might also use the genitive in the phrase language of the Elves, hence Eldaron instead of Eldaiva, but this would contradict what Tolkien wrote elsewhere.)
328
Lesson Thirteen
1. A. The man gave the woman a gift. B. The sun gives light to the world. C. We (inclusive) will nd the treasure, and we (inclusive) will give it to the twelve Dwarves. D. Eating esh makes one fat, and we (inclusive) dont want fat bodies, for fat bodies are not beautiful. E. We (exclusive) went into the city (in order) to nd the wise women, for we (exclusive) wanted to see them. F. A man having good ideas is wise and will give [or, bring] peace and joy to the city. G. We (exclusive) have summoned them (in order) to speak of/about many things. H. Drinking wine isnt good for ones body. I. Quen n i ohtaren alta macil. (Notice the helping vowel -e- intruda e
ing between ohtar and the ending -n, to avoid the impossible form **ohtarn.)
2.
J. Cari coa i seldoin n mra noa. (Here the gerund cari making e a a e
takes both a direct and an indirect object coa and i seldoin, respectively.)
K. Mahtalm sren; mahti um anta i lien alass, an ecnie e e e e e elm i cala. e L. Queti i Eldalamb n alta alass Atanin. (Possibly alta alass e e a e e
is not a perfect translation of great joy, since the adjective alta primarily means great = big with reference to physical size but we have no word for great in a less concrete sense.)
M. (I) ohtari i atta nrion [or, i atta nrion ohtari] mahtuvar i o o lient, ar lelyuvamm ter alta morni hirien cala.(Alternativee e
ly the two lands, not just the [twin] peoples, might also be expressed as a dual form here: i nret, genitive i nreto, instead of i atta nrion.) o o o
N. I neri arw i mra limpo merner yulmar sucien i limp, e a e e ar i arano mli ner i nerin rasta yulmar maltava.(Notice o a
genitive following arwa [here pl. arw], hence limpo.) e e
O. Merimm lelya mir i osto lerien ily Atani ar antien (i) e e malta i aranwa i mlin. (Notice that while the phrase we want to o
go certainly expresses a purpose, go should here be rendered as a simple innitive lelya, not as a gerund in dative, since **merimm lelien = we e want [in order] to go would make no sense. On the other hand, the in order test reveals that the verbs lerya- to free and anta- to give should appear as gerunds in dative, sc. lerien and antien: We want to go into the city [in order] to free all Men and [in order] to give the gold of the king to the thralls.)
329
P. (I) rambar i osto [or, i osto rambar] nar alt; acrielmet e a varien i li. (Varien: dative gerund of varya- to protect.) e (In these keys we do not list all the possible variations in word order, like saying acrielmet i li varien instead of acrielmet varien i a e a li.) e
Lesson Fourteen
1. A. We (inclusive) will go from the tower to the house. (Alternative
interpretation: out of the tower, into the house.)
B. All Elves have passed [/disappeared] from the world. C. The Dwarves have come from [or, out of] the mountains; they have gone to [or, into] the houses and are drinking our (incl.) wine. D. The evil warriors will seize the gold of our (incl.) people [in order] to send our (incl.) treasures to a remote land. E. The woman went away from my house and went to the river. F. The rst ship will come from the West. G. One fears the lions [Quenya: . . . feels fear from the lions], for they have eaten the king of our (exclusive) people, and they will not go away from our (exclusive) land [or simply: . . . will not leave our land.] (Since r lion has the stem-form rv-, and **rvllon a a a
is not a possible word, the plural ablative would presumably require a connecting vowel, which is -i- in the case of plural words: Hence we used rvillon a as the pl. ablative of r. Also notice liemmo as the genitive of liemma our a people.)
H. Nessim said [or, says] to Calandil: My son has disappeared from e my room! 2. I. Equ Calandil Nessimenna: Yondolya elendi [or, oanti] e e e et i coallo, an ily i seldor lender [or, oanter = went away] e i ambonna. (The word et may be omitted, since the simple ablative i
coallo can express out of the house by itself but without et, the ablative might just as well be interpreted [away] from the house.)
J. Menello Anar anta cala Ambarelman [dative!] , ar i more ni avni. (Perhaps Menelello with a connecting vowel -e- would also e a e
be a valid ablative form of Menel. Notice that to our world should in this context be a dative rather than an allative form; cf. Exercise B in Lesson Thirteen above. But perhaps allative Ambarelmanna would also be possible, the meaning being: the Sun is giving [out] light [which is going] to our world. The dative and allative cases are closely related; the lay-out of the Plotz letter
330
suggests that the dative in -n may have originated as a shorter variant of the allative in -nna.)
K. Equ Calandil i uma aranna: Ementiel(y) ohtarilyar i e e e mindonna hirien yondonyar. Mlinya varyuva i seldor, ar o uvant vanw!(Alternative allative forms of aran, mindon might be e e
aranenna, mindonenna. Notice how the ending -nya always prefers -i- as its connecting vowel where one is required, hence my thrall = mlinya. On o the other hand, ohtarilyar your warriors shows -i-only because the word is plural, cf. also mlimmar our thralls in Exercise N below. According to o the system we have tried to make out, the singular forms would be ohtarelya your warrior, mlemma our thrall. Possibly my sons could also be o expressed as a contracted form yonyar [instead of yondonyar], but yonya my son [LR:61] may primarily be used as a form of address.)
L. I nr arwa i ciryaron mern auta, ar ily i ciryar oanter e e e N menna. u M. Lendemm sambenta, ar i nr i ambollon [or, ambollor] e e n yondolyan [dative!] alta macil, qutala: I macil tul a e e e haira nrello, (et) anhaira N mello. o u N. Ily aldar rner ar vner nrelmallo, ar equ Calandil e a o e ar Nessim: Mentuvamm mlimmar hirien nr arwa e e o o e rimb aldaron. (Notice that the verb equ does not receive the ending e e
-r even where it has multiple subjects.)
P. Lenden(y) sambelmanna hostien engwenyar, an mere nen(y) anta torninyan minya parmanya; i parma caitan e e i talamenna. (My brother: we go for torninya [here with the dative
ending -n], formed from toron, torn- brother with the connecting vowel -ithat is preferred by the ending -nya my. Perhaps toronya, for toron-nya, would also be possible [dative toronyan]. Talamenna as the allative of talan oor takes into account the stem-form talam-, but perhaps talanna for talan-nna would also be an acceptable form.)
c) For our (incl.) bodies d) Our (incl.) tongues (nominative) e) To your land f) Our (excl.) things (nominative) g) From your king h) Of my thrall i) Of my thralls j) To our (incl.) cities k) Of our (dual incl.) people [= of the people of the two of us] l) Of your son m) Our (excl.) two-room apartment (nominative) (Translating dual forms
of samb room as two-room apartment!) e
n) For my two-room apartment o) Of your two-room apartment p) To our (incl.) two-room apartment q) From your two-room apartment r) Of my people s) Of our (incl.) sons t) For your queen u) Of our (excl.) peoples v) Of my men w) Of my man x) For my boys y) From our (excl.) [two sister] ships (dual) z) Of our (dual incl.) son [= of the son of the two of us] 4. a) Ambolyannar b) Sremman e c) Parmalyat d) Mindonelyanna / mindonilyannar (Notice how -e- functions as a
connecting vowel in the singular, whereas -i- is used in the plural.)
e) Trimmava a f) Sellinyaiva g) Sellinyallo (Notice how the ending -nya my prefers -i- as its connecting
vowel even in the singular, as here following seler, sell- sister. Cf. also exercises M, N, U, Y below.)
332
h) Annalmaron i) Maltalmo j) Alasselman k) Limpelyo l) Ambarelyallo m) Anarinyo n) Aranyan (for aran-nya-n; alternatively araninyan with a connecting
vowel inserted)
o) Yondommava p) Yulmammaron q) Aiwelyant r) Rambammanta s) Rambalmalto t) Nrelyallon [alternatively, nrelyallor] o o u) Sellinyato(we assume that the ending -nya my prefers -i- as its connecting vowel in dual forms as well)
v) Harmammaron w) Roccolmannar x) Coalvo y) Torninyan (or perhaps toronyan for toron-nya-n, irrespective of tornas the normal stem-form of toron brother)
z) Aldammannar
Lesson Fifteen
1. A. (S)he [or, it] will come on the second day. B. In the winter[,] many birds go away to dwell in [the] South; after the winter they go away from [the] South [or, leave the South] and come to our (incl.) land. C. His/her nding gold in the mountains gave joy to his/her people, for his/her nding it made his/her people rich. D. In [the] second winter that (s)he lived in the house[,] (s)he found a treasure under the oor. E. (S)he speaks our tongue, for (s)he dwells (/lives) in our (incl.) land. F. (S)he says/said: I saw a sword in the warriors left hand. (Notice
the shortening of the long vowel of m before a consonant cluster: locative a mass.) e
333
G. The man who found the treasure will hide the things which he has found in his two-room apartment (. . . if we continue to translate dual
forms of samb room as two-room apartment, that is. Notice that the e second i of the Quenya sentence is the relative pronoun who, not the article the. The relative pronoun ya which here appears in the form yar, a plural ending being attached, because it refers back to the plural word things: We assume that ya is inected as a noun in -a. This yar should not be confused with the attested form yar to whom, which is not plural but has the old allative ending -r [as in mir into] attached.)
H. On the hill[,] (s)he sees the couple whom (s)he has watched from his/her house, and to whom (s)he gave his/her gift. (We assume
that the relative pronoun ya would appear with dual endings when referring to a dual word: yat, dative yant.)
2.
I. Cennes veru i malless. e J. Hirnen(y) i n i mar i coass imb i s ar tirnen(y) e s e e e ri, e pryat ar mryat; hyarya mryass cennen(y) parma. e a a e e
(Notice that the second i of this sentence functions as a relative pronoun who, not as an article. Cf. exercise G above: i nr i. . . the man who. . . ) e
K. Cennen(y) yulmarya mryats, i yulma yallo ulyanes e a e limp mir antorya (or, antoryanna, using a simple allative instead e
of the preposition mir).
M. Sucitarya i limp um mra noa, an ya carnes apa sucie e a taryas um saila.(Maybe the initial sucitarya could just as well be e
sucierya the pronominal ending being attached to the gerund of suc- to drink.)
N. Apa oantemm nremmallo (i) Hyarmess, ecniemm e o e e e rimb Naucor i mallessen. e O. I mindoni i ambossen nar alt; i harya i analta mindon, e yallo pol quen cen i Eldanr, n i analya nr i ostoss. e e o e a e e
(. . . yallo pol quen cen, literally from which can one see: I assume e e that in a phrase like pol cen can see, with a nite verb followed by an e e innitive, it is only the nite verb that is relocated to immediately follow a relative pronoun like yallo. But we lack attested examples, of course; perhaps it should be yallo pol cen quen with the subject following the entire verb e e phrase. And for all I know, yallo quen pol cen with an English word e e order may also be permissible.)
334
P. Li yo aran n saila maruva sress mra nress ya e a e e a o e meluvant n rav. (If ya can really be inected in the same way as a e u e
noun in -a, as suggested by the example yassen, the genitive whose would in the singular be yo the group -ao being simplied to -o as usual.)
Lesson Sixteen
1. A. (S)he has come from the East on a white horse. (Of course, the
instrumental form in -nen does not really mean on but implies by means of the horse being identied as the means of travel. It is not entirely obvious how rocconen is best translated in English, so the precise wording of the translation is unimportant as long as the student clearly grasps the
meaning of the instrumental form itself. One might well translate: . . . riding a white horse, though no element actually meaning riding is present.) B. The man wounded the lion with a spear, and he said: Dont eat my son! C. We (excl.) said to the Elf: Our (incl.) friendship is renewed by your gift! D. On the third day (s)he said to the maiden: Do what you want!
(. . . ya merily = [that] which you want.) e
E. One cannot nd a treasure hidden by Dwarves, for a Dwarf loves deeply the gold that he possesses. F. The man is wounded by [the] horns (dual) of the animal; wish that he will not die! (literally: be it that he will not die! Possible, but
rather less probable interpretation: . . . that it will not die! referring to the animal! If the ending -s is to cover both he, she and it, one cant always be quite certain where the sympathies of the speaker really lie. . . )
G. (S)he is singing with joy(/because of joy). H. Go to the city and say to the mighty queen: May you protect our (incl.) land from the evil warriors! (Nai varyuvaly nrelma. . . = e o
literally be it that you will protect our land. . . ) NOTE ON EHTE SPEAR: In exercise B, I used ehtenen as the instrumental form of this noun. In the Etymologies, Tolkien rst derived this word from ekte, but according to a note by the editor, a variant reading ekti was also introduced. If we accept this latter version, so that Quenya eht spear is to be derived from e primitive Elvish ekti, the Quenya word should probably have the stem-form ehti- so that the instrumental form would be ehtinen rather than ehtenen. But the plural instrumental form should probably be eht nen (as in Exercise K below) no matter which etymology we prefer, since eht nen could represent both ehti+inen (two short is merging into one long and ehte+inen (ei regularly becoming long ) ).
335
2.
I. Eques [or, quentes] i Nauconna: A rac i yulma nambae nen! J. Ciryanen oanten(y) haira nrenna Rmess. e o o e K. I osto n varyaina alt rambainen, ar ohtari i mahtar a e eht nen umir pol rac i rambar. e e L. Equ Calandil harna yondoryanna: Ava r!(Alternatively: e e
Calandil quent harna yondoryanna. . . ) e
M. Nai trilya hiruva i yna airinta (i) Eldainen! a a N. I aran ar i tri lender coanyanna ar envinyatner nilmema a ma alt annainen. e O. Mapas i seldo mryanten, ar eques [or, quetis]: Ava lelya a i s renna! P. I n i mar i nelya coass i malless quent i Eldanna: s e e e e A tir i neri i t lar i ynallo ya cenil(y) i amboss, i e u a e e lelyar Rmenna. o
Lesson Seventeen
1. A. This thing is the sword found by Calandil [the] Tall. (Calandil
Hallanen by C. [the] Tall: notice how the case ending for instrumental is added to the last word of the phrase.)
B. All animals have disappeared from this land (nr sinallo = from o e
this land; again notice how the case ending, here for ablative, is added to the last word the demonstrative sina this. However, nrello sina would o perhaps be equally possible.)
C. A snake wounded his/her right arm, and (s)he said: Wish that all snakes will [or, would] die! (Rancurya = his [or, her] arm; notice
how ranco arm appears in the form rancu- before endings, since it is a U-stem. Possibly, the dual pair of arms would also have the form rancu, but if that form were intended here, we would also see the extra dual marker -t following the ending -rya. Cf. mryat = her (pair of) hands in Namri.) a a e
D. That night, they found that Dwarf on the hill over there [literally: on yonder hill]. (Lm yanass = in that night, but English would simo e e
ply say that night in such a context. Yana may signify that of something that existed formerly or in the past, and since the reference is to a night in the past, this word is appropriate here if we have reconstructed Tolkiens intentions correctly! Cf. also hr e yanass in/during that winter in exercise H v e below. Nauco tana: that Dwarf , the word tana that simply focusing on individual identity. Ambo entass: on yonder hill, on the hill over there: e the word enta apparently means that with emphasis on spatial position.)
336
E. The tall ships have passed away north(wards); those ships will not come to the lands of (the) South. (Ciryar tan those ships: we e
surmise that tana that can have the plural form tan those, the demone strative behaving like an adjective.)
F. On the fourth day, that queen died by [/from] the poison of a snake. (Tri yana: that queen of a queen who is dead, yana being used a
of something that is past but that queen could probably also be tri tana, a merely focusing on personal identity.)
G. The strong arms of the men from (the) North can seize the spears from ghting warriors. (Ohtari mahtalallon from ghting warriors:
notice how the word ohtari warriors receives only the simplest plural ending, normally associated with the nominative case, but the plural ablative ending -llon added to the participle mahtala ghting indicates that the whole phrase is in the ablative case. The word order mahtala ohtarillon would be equally possible: Here the ablative ending is added to the noun warriors, since it is now the last word of the phrase.)
H. That winter, they lived (/dwelt) in the fourth house of that road (/street). (Hr e yanass: locative during/in that winter; cf. Exercise v e
D above. Mall tano of that road, genitive of mall tana that road e e the genitive ending -o displacing a nal -a as usual.)
2.
J. Nr u angwion n mra nr, an rimb Atani if o e a a o e e rier an gusangwanen. (U angwion without snakes: as the student hopefully
remembers, the preposition u without governs the genitive case. Angusang wanen by snake-poison: notice how the U-stem noun ango snake appears as angu- in a compound.)
K. I canta lmiss cennen(y) ruhtala ohtar mall tanass, e o e e e e ar ortanen(y) rancunyat. (Alternatively mall entass = on yone e e
der road, on that road [over there]. Rancuryat my arms is here unmistakably dual, because of the dual ending -t following the pronominal ending -nya my. Compare/contrast Exercise C above.)
L. Nai Calandil Hallo polda yondo [or, . . . i polda yondo Calandil Hallo] tuluva nr sinanna, an varyuvas ostor sin o e e yassen marilm!(As in Exercise A above, the case ending is added to e
the last word of the phrase Calandil Halla Calandil [the] Tall, and as in the attested example Elendil Vorondo of Elendil the Faithful [Elendil Voronda], the genitive ending displaces a nal -a.)
M. Mindon enta n i canta mindon carna Eldainen nr a e o e sinass. e N. Parmar tan nar vanw; avnient sambelyallo. e e a e 337
Insofar as the system of adding case endings to the last declinable word is not necessarily obligatory, the endings could probably be added to the relevant nouns as well, e.g. malless tana (or, enta) e rather than mall tanass (or, entass) for on that road in Exere e e cise K.
Lesson Eighteen
1. A. They have come to see us, not to see thee. B. May you dream about Elves! (Wish that you will dream about Elves!) C. Three men will go thither, and the Dwarves will watch them, for the men possess a treasure that is greater than [lit. great beyond] gold. D. On the sixth day they felt moved to come to me. [Lit. . . . (it) impelled for them to come to me.] E. The sixth king of the land is wiser than [wise beyond] his father the fth king. F. (S)he expanded his/her house, making it [or, that] the biggest house of the street. G. You did it for them; you did not do it for us. (Alternatively, you
made it for them; you did not make it for us: Car- covers both do and make.)
H. We saw you in the gorge under us, for you fell into it. 2. I. Forya rancunya n polda l hyarya rancunya. a a tulta te ninna! J. A K. I urqui t rar nye, an rucin(y) tiello. e L. L tuluvamm cenien tye i lmiss. a e o e M. L luva i seldon urquion. ao 338
N. I atta nissi quenter: Aranelya l mern anta ment i ena e gwi mapain melto ohtariryainen. e O. I nr plula coarya cra ta carien rimb sambi ily enge a a e e weryain. P. I enqua aur nauva mra l i lempa aur, ar l oruva e e a a e e a men auta. Other word-orders would apparently also be possible, such as prexing some independent pronouns to verbs: K) nye-t rar rather than t rar nye, L) tye-cenien rather than cenien tye, N) ment-anta rather than anta ment, O) ta-cra rather than cra ta, P) men-oruva a a rather than oruva men.
Lesson Nineteen
1. A. Who lived [or, dwelt] in the house ten years ago, when I did not live in it? B. What can we [inclusive] do? C. Give it to me, and dont hide it from me! (The ending -t in antat,
nurtat indicates that only one person is addressed.)
D. Who did you see on the seventh day? E. They will give all the gifts to me and not to you! F. What did the men nd in the land beyond the mountains? G. How will you be able to lift the great stone when I could not do it? H. Our sister lives in that (yonder) house; we live in the eighth house in this street. Come and see our house! (A tulil ar cenil = a come and see; the ending -l indicates that several people are being addressed.) 2. I. Man acri ta? a e J. Mana ely hirn i otsa sambess e lendel(y) tanna? e e e e r e e ely ut li, iny mer auta! (If ely can function as a plural K. Ir e u e e e e
pronoun and it is interpreted as a plural you here, the verb should perhaps also be plural: ely utlier.) e u
L. Amen antal i limp! e M. A antat i limp emmen ar l i ohtarin! e a N. I otsa ohtar ut li i alt orontillon (or, -llor) pella. e u e e O. I tolta ando pella ely hirn alta harma; emm i tuller e e e e apa le umer hir [or, l hirner] engw! (Alternatively ely hirner e a e e 339
if the pronoun is taken as plural; cf. Exercise K above. Notice that the who of the relative sentence we who came is translated by means of the relative pronoun i: The word man denotes a dierent kind of who, used in actual questions.)
Lesson Twenty
1. A. You [or, emphatic you] are a Dwarf, not an Elf. (Ely Nauco: the e
copula are is left out and understood. Perhaps nly Nauco would have a e been a possible alternative, if n can receive pronominal endings.) a
B. I saw that the mans arm was broken. C. The king was mighty, but the queen was wiser than the king. D. Did you know that under the house [there] is a hidden treasure? E. May I leave [literally, go from] this place? F. Dont drink, for [there] is poison in your cup! G. Will they be here on the tenth day when I come [literally, when I shall come]? H. We (excl.) know that [there] is a secret room in this house, but we have not found it, and until we nd it we shall not know what is in the room. 2. I. I was rich can be translated in several ways. The safest solution would perhaps be to leave out the copula and use an independent pronoun (ni or iny) for I: Ni alya or iny alya. e e Of course, this may just as well mean I am rich, since there is no tense-marker. If we want to include one, we must use the not entirely well-attested form n was. Assuming that it can ree ceive pronominal endings, I was rich could be rendered nn(y) e e alya. J. I aran quent [or perhaps equ]: L lertal(y) lelya i ne e a e o menna yallo ut liel(y), nan istan(y) sa lelyuvan(y) u e e e tanna. (Alternative translation of you may not go: umil(y) lerta lelya.) e K. Istan(y) cenda, nan l polin(y) cenda i morniess. (Ale a e e
ternative translation of I cannot read: umin(y) pol cenda. Notice the e e dierence between ista- can = know how to and pol- can = be physically able to: The speaker knows how to read, but is unable to read in the dark.)
L. Istalm sa Eldar ar. e e M. Ma marnent sinom tenna i nerta loa e i ohtari tuller? e e e r 340
N. Istamm sa i neri sinter quet i Eldalamb, nan l i Naue e e a colamb. e O. I nissi quenter sa ecniel(y) i alta ango i a i orontess. e e e e P. I quaina auress Anar n calima. e e e
341