PART FOUR. QUENDI AND ELDAR. QUENDI AND ELDAR. The title Quendi and Eldar clearly belongs properly to the long essay that is printed here, though my father used it also to include two other much briefer works, obviously written at much the same time; one of these, on the origin of the Orcs, was published in Morgoth's Ring (see X.415, where a more detailed account is given). Quendi and Eldar is extant in a typescript with carbon copy that can be fairly certainly dated to the years 1959-60 (ibid.); and both copies are preceded by a manuscript page that in addition to the following preamble gives a parallel title Essekenta Eldarinwa. Enquiry into the origins of the Elvish names for Elves and their varieties clans and divisions: with Appendices on their names for the other Incarnates: Men, Dwarves, and Orcs; and on their analysis of their own language, Quenya: with a note on the 'Language of the Valar'. My father corrected the two copies carefully and in precisely the same ways (except for a few later pencilled alterations). The text printed here follows the original very closely, apart from very minor changes made for consistency or clarity, the omission of a passage of extremely complex phonology, and a reorganisation of the text in respect of the notes. As often elsewhere in his later writings, my father interrupted his main text with notes, some of them long; and these I have num- bered and collected at the end, distinguishing them from my own numbered notes by referring to them in the body of the text as Note 1, Note 2, &c., with a reference to the page on which they are found. Also, and more drastically, I have omitted one substantial section from Appendix D (see p. 396). This was done primarily for reasons of space, but the passage in question is a somewhat abstract account of the phonological theories of earlier linguistic Loremasters and the contributions of Feanor, relying rather allusively on phonological data that are taken for granted: it stands apart from the content of the work at large (and entered, I suspect, from the movement of my father's train of thought rather than as a planned element in the whole). Also for reasons of space my commentary is kept to a severe minimum. Abbreviations used are PQ (Primitive Quendian), CE (Common Eldarin), CT (Common Telerin), Q (Quenya), T (Telerin), N (Noldorin), S (Sindarin), V (Valarin). QUENDI AND ELDAR. Origin and Meanings of the Elvish words referring to Elves and their varieties. With Appendices on their names for other Incarnates. A. The principal linguistic elements concerned. 1. *KWENE. (a) PQ *kwene 'person' (m. or f.). CE *kwen (-kwen), pl. *kweni, person (m. or f.), one, (some)body-, pl. per- sons', '(some) people'. (b) PQ and CE *kwende, pl. *kwendf. This form was made from *kwene by primitive fortification of the median n ) nd. It was probably at first only used in the plural, in the sense 'people, the people as a whole', sc. embracing all the three original clans. (c) *kwendja adj. 'belonging to the *kwendi, to the people as a whole'. 2. *ELE According to Elvish legend this was a primitive exclamation, 'lo! ' 'behold! ' made by the Elves when they first saw the stars. Hence: (a) CE *el, *ele, *el-a, 'lo!' 'look!' 'see!' (b) CE *el, pl. *eli, eli, 'star'. (c) CE *elen, pl. 'elena, 'star', with 'extended base'. (d) CE *elda, an adjectival formation 'connected or con- cerned with the stars', used as a description of the *kwendi. According to legend this name, and the next, were due to the Vala Orome. They were thus probably at first only used in the plural, meaning 'star-folk'. (e) CE *elena, an adjectival form made from the extended stem *elen, of the same meaning and use as *elda. 3. *DELE. (a) A verbal base 'dele, also with suffix *del-ja, 'walk, go, proceed, travel'. (b) *edelo, an agental formation of primitive pattern: 'one who goes, traveller, migrant'. A name made at the time of the Separation for those who decided to follow Orome. (c) *awa-delo, *awa-delo, ?*wa-delo. Old compounds with the element *awa 'away' (see below). A name made in Beleriand for those who finally departed from Middle- earth. 4. *HEKE. Probably not in origin a verbal base, but an adverbial element 'aside, apart, separate'. (a) PQ *heke 'apart, not including'. (b) PQ and CE verbal derivative, transitive: 'hek-ta 'set aside, cast out, forsake'. (c) PQ *hekla 'any thing (or person) put aside from, or left out from, its normal company'. Also in personal form *heklo 'a waif or outcast'; adjectival forms 'hekla and *hekela. The element *AWA, appearing in 3(c) above, referred to move- ment away, viewed from the point of view of the thing, person, or place left. As a prefix it had probably already developed in CE the form *au-. The form *awa was originally an indepen- dent adverbial form, but appears to have been also used as a prefix (as an intensive form of *awa-, 'au-). The form *wa- was probably originally used as a verbal stem, and possibly also in composition with verbal stems. In the Eldarin languages this stem made contact in form with other elements, distinct in origin and in sense.*ABA 'refuse', 'say nay (in refusal or denial)': this is the source of the CE *abar, pl. *abari 'a refuser,' one who declined to follow Orome. *wo in forms *wo and *wo- (the latter only as a prefix): this was a dual adverb 'together', referring to the junction of two things, or groups, in a pair or whole. The plural equivalent was *jo, *jom, and as a prefix *jo, *jom. *HO in forms >ho and >ho: this was an adverb 'from, coming from', the point of view being outside the thing referred to. The principal derivatives in form (their use is discussed below) of the CE words given above were as follows: *KWEN. QUENYA. 1(a) quen, pl. queni; unstressed, as a pronoun or final element in a compound, quen. 1(b) Quendi. The sg. quende (not much used) was made in Quenya from Quendi, on the model of other nouns in -e, the majority of which formed their plurals in -i. There were also two old compounds: Kalaquendi 'Light-elves' and Mori- quendi 'Dark-elves'. 1(c) Quendya, which remained in the Vanyarin dialect, but in Noldorin became Quenya. This was only used with reference to language. TELERIN. 1(a) pen as a pronoun, and -pen in a few old compounds. 1(b) Pendi, plural only. Also in the compounds Calapendi and Moripendi. 1(c) Not found. SINDARIN. 1(a) pen, usually mutated ben, as a pronoun. Also -ben, -phen in a few old compounds. 1(b) Not found. The compounds Calben (pl. Celbin) and Morben (pl. Moerbin, Morbin) must certainly have de- scended from the same source as those mentioned above, but their final element was evidently altered to agree with the compounds of *kwen. The unaltered derivatives would have been *Calbend, *Moerbend; but though final -nd eventually became -n in Sindarin, this change had not occurred in the early records, and no cases of -bend are found. In addition, the form Morben (without affection (1) of the o) shows either an alteration to *mora- for mori-, after *kala-, or more probably substitution of S morn- from *morna, the usual S adjectival form. 1(c) Not found. *EL QUENYA. 2(a) ela! imperative exclamation, directing sight to an actually visible object. 2(b) el, pl. eli, 'star' (poetic word). 2(c) elen, pl. eleni (occasionally in verse eldi), 'star'. The normal word for a star of the actual firmament. The pl. form eleni, without syncope, is re-formed after the singular. 2(d) Elda only used as a noun, chiefly in the pl. Eldar. See also (Quenya) 3(b) below. 2(e) Elda as above. As an adjective referring to stars the form used was elenya. TELERIN. 2(a) ela! as in Quenya. 2(b) el, pl. eli. The ordinary word for 'star'. 2(c) elen, pl. elni. An archaic or poetic variant of the preceding. 2(d) Ella. An occasional variant of Ello, which was the normal form of the word. This shows contact with the products of *edelo: see further under (Telerin) 3. 2(e) Not found. The form would have been *Elna. SINDARIN. 2(a) elo! An exclamation of wonder, admiration, delight. 2(b) Not found. 2(c) el, pl. elin, class-plural elenath. An archaic word for 'star', little used except in verse, apart from the form elenath 'all the host of the stars of heaven'. 2(d) Ell-, only used in the m. and f. forms Ellon, Elleth, elf-man, elf-woman; the class-plural El(d)rim; and final -el, pl. -il, in some old compounds: see (Sindarin) 3(b). 2(e) Elen, pl. Elin, with class-plural Eledhrim, Elf, Elves. dhr is < n-r in secondary contact. On usage see further below. *DEL. QUENYA. 3(a) lelya-'go, proceed (in any direction), travel,, past tense lende. This form is due to the early change in Q of initial d > l. The change was regular in both Vanyarin and Noldorin dialects of Quenya. It occurs occasionally also in Telerin languages, though this may be due rather to d/l variation in PQ, for which there is some evidence. A notable example being de/le as pronominal elements in the 2nd person. In Q *del- seems to have become *led, by dissimilation. The past form clearly shows *led, while lelya may also be derived from *ledja, since dj became ly medially in Quenya 3(b) Eldo. An archaic variant of Elda, with which it coalesced in form and sense. Eldo cannot however be directly descended from *edelo. Its form is probably due to a change *edelo > eledo, following the change in the verb. The change of initial d > l was early and may have preceded syncope, and the loss of feeling for the etymological connexions of the word, which finally resulted in the blending of the products of 2 and 3. Unchanged *edelo would by syncope have given *edlo > *ello (which is not found). See, however, under Sindarin for reasons for supposing that there may have been a variant form *edlo (with loss of sundoma):(2) this could have produced a Quenya form *eldo, since transposition of dl in primary contact to the favoured sequence Id not infrequently occurred in the pre-record period of Quenya. 3(c) Aurel < *aw(a)delo. Oarel < *awadelo. In the Vanyarin dialect Auzel and Oazel. Oarel (Oazel) were the forms commonly used in Q. The plurals took the forms -eldi. This shows that the ending -el was associated with the noun Elda. This was facilitated by a normal development in Q morphology: a word of such a form independently as *elda, when used as the final element in a compound of early date, was shortened to *elda, pl. *elch > *eld, *eldi > historic Q -el, -eldi. In addition oar was in actual use in Q as an adverbial form derived from *AWA (see below): a fact which also accounts for the selection of oarel, oazel. TELERIN. 3(a) delia 'go, proceed'. Past tense delle. 3(b) Ello. The usual form, preferred to Ella, from which, however, it did not differ in sense. Both *edelo and *edlo regularly became ello in Telerin. 3(c) Audel, pl. Audelli. This shows the same association with -el, the shortened form in composition of ella, ello, as that seen in Q. SINDARIN. 3(a) Not found. 3(b) Edhel, pl. Edhil. The most used word in Sindarin; but only normally used in these forms. As noted above under (Sindarin) 2(d) the m. and f. forms were Ellon, Elleth; and there was also a class-plural Eldrim, Elrim (ll-r in secondary contact > ldr, later again simplified). As suggested under (Quenya) 3(b), there may have been a variant *edlo, which would regularly give ell- in Sindarin. Since this shorter form would be most likely to appear in compounds and extended forms, it would account for the limitation of Sindarin ell- to such forms as Ellon, Elleth, Eldrim. It would also account for the blending of the products of stems 2 El and 3 Del in Sindarin, as well as in Quenya. The form -el, pl. -il also occurs ' in some old compounds (especially personal names), where it may be due also to a blending of *elda and *edlo. In later compounds -edhel is used. 3(c) Odhel, pl. Odhil; beside later more usual Godhel, Godhil. Also a class-plural Odhellim, Godhellim. Odhel is from *aw(a)delo, and the exact equivalent of Q Aurel, T Audel. Godhel could be derived from *wadelo: S initial *wa- > gwo > go. But since it appears later than Odhel, and after this term had become specially applied to the Exiled Noldor, it seems most probable that it took g- from the old clan-name Golodh, pl. Goelydh, which it practically replaced. Golodh was the S equivalent of Q Noldo, both from PQ *ngolodo. *HEK. QUENYA. 4(a) heka! imperative exclamation 'be gone! stand aside!'. Normally only addressed to persons. It often appears in the forms hekat sg. and hekal pl. with reduced pronominal affixes of the 2nd person. Also bequa (? from *hek-wa) adverb and preposition 'leaving aside, not counting, exclud- ing, except'. 4(b) hehta-, past tense hehtane, 'put aside, leave out, exclude, abandon, forsake'. 4(c) hekil and hekilo m., hekile f.: 'one lost or forsaken by friends, waif, outcast, outlaw'. Also Hekel, pl. Hekeldi, re-formed to match Oarel, especially applied to the Eldar left in Beleriand. Hence Hekelmar and Hekeldamar, the name in the language of the loremasters of Aman for Beleriand. It was thought of as a long shoreland beside the Sea (cf. Eglamar under Sindarin below). TELERIN. 4(a) heca! For Q hequa the T form is heco (? < *hek + au). 4(b) hecta- 'reject, abandon'. 4(c) hecul, heculo. Also with special reference to those left in Beleriand, Hecello; Heculbar or Hecellubar, Beleriand. SINDARIN. PQ h- only survived in the dialects of Aman. It disappeared without trace in Sindarin. *hek therefore appears as *ek, identical in form with PQ *eke 'sharp point'. 4(a) ego! 'be off!' This is from *hek(e) a: a the imperative particle, being originally independent and variable in place, survived in S as o > o, though this now always followed the verb stem and had become an inflexion. 4(b) eitha-. This is in the main a derivative of PQ *ek-ta, and means 'prick with a sharp point', 'stab'; but the sense 'treat with scorn, insult' (often with reference to rejection or dismissal) may show the effect of blending with PQ *hek-ta. To say to anyone ego! was indeed the gravest eithad. 4(c) Eglan, mostly used in the plural Eglain, Egladhrim. The name that the Sindar gave to themselves ('the Forsaken') as distinguished from the Elves who left Middle-earth. Eglan is < an extended adjectival form *heklana. The older shorter form (*hekla or *hekla) survives in a few place-names, such as Eglamar (cf. Hekelmar, etc.), Eglarest. These are shown to be old from their formation, with the genitival element preceding: *ekla-mbar, *ekla-rista. *AWA. QUENYA. au- as a verbal prefix: < either *au or *awa,. as in au-kiri- 'cut off'. The point of view was in origin 'away from the speaker or the place of his thought', and this distinction is usually preserved in Q. Thus aukiri meant 'cut off, so that a portion is lost or no longer available', but hokiri (see below) meant 'cut off a required portion, so as to have it or use it'. oa, oar. Adverbs: < *awa,. the form oar shows addition of the ending -d (prehistoric -da) indicating motion to or towards a point. The form awa appears originally to have been used either of rest or motion, and oa can still be so used in Q. This adverbial oa, oar was occasionally used as a prefix in compounds of later formation. Though, as has been shown, in Oareldi, the most commonly used, the r is in fact of different origin. The verb auta- 'go away, leave (the point of the speaker's thought)' had an old 'strong' past tense anwe, only found in archaic language. The most frequently used past and perfect were vane, avanie, made from the stem *wa; together with a past participle form vanwa. This last was an old formation (which is also found in Sindarin), and was the most frequent- ly used part of the verb. It developed the meanings 'gone, lost, no longer to be had, vanished, departed, dead, past and over'. With it the forms vane and avanie were specially associated in use and meaning. In the more purely physical sense 'went away (to another place)' the regular forms (for a -ta verb of this class) oante, oantie were used. The form perfect avanie is regularly developed from *a-waniie, made in the prehistoric period from the older perfect form of this type *awawiie, with intrusion of n from the past (the forms of past and perfect became progressively more closely associated in Quenya). The accent remained on the wa, since the augment or reduplication in verbal forms was never accented even in the retraction period of Quenya (hence no form *oanie developed: contrast oante < *awa-n-te). The form vanie appearing in verse has no augment: probably a phonetic development after a preceding vowel; but such forms are not uncommon in verse. SINDARIN. The only normal derivative is the preposition o, the usual word for 'from, of'. None of the forms of the element *awa are found as a prefix in S, probably because they became like or the same as the products of *wo, *wo (see next). The form Odhel is isolated (see above, Sindarin 3(c)). As the mutations following the preposition o show, it must prehistorically have ended in -t or -d. Possibly, therefore, it comes from *aud, with d of the same origin as that seen in Q oar (see above). Some have thought that it received the addition -t (at a period when *au had already become q > o) by association with *et out, out of . The latter retains its consonant in the form ed before vowels, but loses it before consonants, though es, ef, eth are often found before s, f, th. o, however, is normally o in all positions, though od appears occasionally before vowels, especially before o-. The in- fluence of *et > ed is therefore probably only a late one, and does not account for the mutations. TELERIN. The Telerin forms are closely similar to those of Quenya in form and meaning, though the development *aua > oa does not occur, and v remains w in sound. Thus we have prefix au-, adverb au or avad; verb auta- with past participle vanua, and associated past and perfect vane and avanie; and in physical senses vante, avantie. *WO. QUENYA. This does not remain in Q as an independent word. It is however a frequent prefix in the form o- (usually reduced to o- when unstressed), used in words describing the meeting, junction, or union of two things or persons, or of two groups thought of as units. Thus: o-mentie (meeting or junction of the directions of two people) as in the familiar greeting between two people, or two companies each going on a path that crosses that of the other: Elen sila lumenna omentielvo!(3) 'A star shines upon the hour of the meeting of our ways.' (Note 1, p. 407) This prefix was normally unstressed in verbs or derivatives of verbs; or generally when the next following syllable was long. When stressed it had the form o-, as in ononi 'twins', beside the adj. onona 'twin-born', also used as a noun 'one of a pair of twins'. TELERIN use does not materially differ; but in form the su- (lost in Quenya before o) is retained: prefix vo, vo-. (Note 1, p. 407) SINDARIN. In the prefix gwa-, go- 'together, co-, com-'. The dual limitation was no longer made; and go- had the senses both of *wo and *jo. *jo, *jom- disappeared as a living prefix. gwa- occurred only in a few S dissyllables, where it was stressed, or in their recognizable derivatives: e.g. gwanun 'a pair of twins', gwanunig one of such a pair. These were mostly of ancient formation, and so retained their dual significance. gwa- is regularly developed from *wo > *wa > gwa, when stressed in prehistoric Sindarin. go- is from *wo > gwo > go, when primitively unstressed; and also from gwa- > go-, when it became again unstressed. Since PQ *wa (one of the forms of *AWA) would also have produced go-, go-, Or gwa- if primitively shortened (e.g. before two consonants), while *au would have produced o-, the same as the frequent initially mutated form of go- 'together', the prefixal forms of *AWA were lost in Sindarin. *HO. QUENYA. This was evidently an ancient adverbial element, occurring principally as a proclitic or enclitic: proclitic, as a prefix to verb stems; and enclitic, as attached to noun stems (the usual place for the simpler 'prepositional' elements in PQ). Hence Quenya ho- (usually so, even when it had become unstressed), as a verb prefix. It meant 'away, from, from among', but the point of view was outside the thing, place, or group in thought, whereas in the derivatives of *AWA the point in thought was the place or thing left. Thus Q hokiri- 'cut off', so as to have or use a required portion; whereas aukiri- meant 'cut off' and get rid of or lose a portion. hotuli- 'come away', so as to leave a place or group and join another in the thought or place of the speaker; whereas au could not be used with the stem tul- 'come'. As a noun enclitic *-ho became -o, since medial h was very early lost without trace in CE. This was the source of the most used 'genitive' inflexion of Quenya. Properly it was used partitively, or to describe the source or origin, not as a 'possessive', or adjectivally to describe qualities; but naturally this 'derivative genitive' (as English of) could be used in many circumstances that might have possessive or adjectival im- plications, though 'possession' was indicated by the adjectival suffix -va, or (especially in general descriptions) by a 'loose compound'. Thus 'Orome's horn' was roma Oromeva (if it remained in his possession); Orome roma would mean 'an Orome horn', sc. one of Orome's horns (if he had more than one); but roma Oromeo meant 'a horn coming from Orome', e.g. as a gift, in circumstances where the recipient, showing the gift with pride, might say 'this is Orome's horn'. If he said 'this was Orome's horn', he would say Oromeva. Similarly 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 expressed either by Elda-lambe or lambe Eldaiva. (Note 2, p. 407) There remained naturally many cases where either posses- sive-adjectival or partitive-derivative genitives might be used, and the tendency to prefer the latter, or to use them in place of the former, increased. Thus alkar Oromeo or alkar Oromeva could be used for 'the splendour of Orome', though the latter was proper in a description of Orome as he permanently was, and the former of his splendour as seen at the moment (proceeding from him) or at some point in a narrative. 'The Kings of the Eldar' might be either i arani Eldaron or i arani Eldaive, though the former would mean if accurately used 'those among the Eldar who were kings' and the latter 'those (kings) in a particular assembly who were Elvish'. In such expressions as 'Elwe, King of the Sindar (people), or Doriath (country)' the derivative form was usual: Elwe, Aran Sindaron, or Aran Lestanoreo. TELERIN. The Telerin use of the prefix ho- was as in Quenya. The inflexion was -o, as in Quenya, but it did not receive -n addition in the plural. It was more widely used than in pure Quenya, sc. in most cases where English would employ the inflexion -s, or of; though the possessive, especially when it concerned a single person or possessor, was expressed without inflexion: either with the possessor placed first (the older usage), or (possibly under the influence of the genitival or adjectival expressions which were placed second) follow- ing the possessed. In the latter case, the appropriate posses- sive suffix ('his, hers, its, their') was usually appended to the noun. So Olue cava; or cava Olue, usually cavaria Olue (sc. 'the house of him, Olwe'); = 'Olwe's house'. The last form was also used in Quenya with proper names, as koarya Olwe. Both languages also used the adjectival possessive suffixes in a curious way, attaching them to adjectives attributed to proper names (or names of personal functions, like 'king'): as Varda Aratarya, 'Varda the Lofty, Varda in her sublimity'. This was most usual in the vocative: as in Meletyalda, or fuller Aran Meletyalda (literally 'your mighty' or 'king your mighty'), more or less equivalents of 'Your Majesty'. Cf. Aragorn's farewell: Arwen vanimalda, namarie!(4) SINDARIN. Since initial h- disappeared in Sindarin *ho would have become u and so, clashing with the negative u, naturally did not survive. >ho as a proclitic might have given o; butg it does not occur as a verbal prefix, although it possibly contributed to the Sindarin preposition o (see under *Awa, Sindarin) which is used in either direction, from or to the point of view of the speaker. Since all final vowels dis- appeared in Sindarin, it cannot be determined whether or not this language had in the primitive period developed inflex- ional -o. Its presence in Telerin of Aman makes its former presence in Sindarin probable. The placing of the genitive noun second in normal Sindarin is also probably derived from inflexional forms. Compounds of which the first element was 'genitival' were evidently in the older period still normal, as is seen in many place- and personal names (such as Egla-mar), and was still in more limited use later, especially where the first element was or was regarded as an adjective (as Mordor 'Land of Darkness' or 'Dark Land'). But genitival sequences with the possessor or qualifier second in the later period also became fixed compounds: as Doriath, for Dor lath 'Land of the Fence'. *ABA. Though this became a verbal stem, it is probably derived from a primitive negative element, or exclamation, such as *BA 'no!' It did not, however, deny facts, but always expressed concern or will; that is, it expressed refusal to do what others might wish or urge, or prohibition of some action by others. As a verbal stem it developed the form *aba- (with connecting vowel a in the aorist); as a particle or prefix the forms *aba, *ba, and *aba. QUENYA. In Quenya the verb ava- was little used in ordinary language, and revealed that it was not in origin a 'strong' or basic verbal stem by having the 'weak' past form avane. In ordinary use it was replaced by the compound va-quet (vaquetin, vaquenten) 'to say no', sc. 'to say I will not', or 'do not', 'to refuse' or 'to forbid'. As a prefix the form used was usually ava-, the force of which can be observed in avaquetima 'not to be said, that must not be said', avanyarima 'not to be told or related', as contrasted with uquetima 'unspeakable', that is, 'impossible to say, put into words, or unpronounceable', unydrima 'impos- sible to recount', sc. because all the facts are not known, or the tale is too long. Compare also Avamanyar 'those who did not go to Aman, because they would not' (an equivalent of Avari) with Uamanyar 'those who did not in the event reach Aman' (an equivalent of Hekeldi). As a particle (the form of this stem most used in ordinary language) the Quenya form was usually va! This was an exclamation or particle expressing the will or wish of the speaker, meaning according to context 'I will not' or 'Do not! ' Note that it was not used, even in the first person, in a statement about the speaker's future action, depending on foresight, or a judgement of the force of circumstances. It could sometimes, as seen in vaquet- (above), be used as a verbal prefix. A longer form ava or ava (stressed on the last syllable), which shows combination with the imperative particle *a, was commonly used as a negative imperative 'Don't!', either used alone or with an uninflected verbal stem, as ava kare! 'Don't do it!' Both va and ava sometimes received verbal pronominal affixes of the first singular and first plural exclusive: as avan, van, vanye 'I won't', avamme, vamme 'we won't'. An old derivative of *aba- as a quasi-verbal stem was *abaro > CE *abar. This was an old agental formation, as seen also in Teler, pl. Teleri, made with the suffix -ro, added to omataina.(5) (Other forms of this suffix were -ro added to stem, with or without n-infixion; and -rdo > rd.) *abar thus meant 'recusant, one who refuses to act as advised or commanded'. It was specially applied to (or first made to describe?) the section of the Elves who refused to join in the Westward March: Q Avar, pl. Avari. TELERIN. The Telerin use was closely similar to that of Quenya. The forms were the same, except that Telerin preserved CE b distinct from v or u: hence the prefix was aba- (abapetima 'not to be said'); the particle ba; the exclamation aba. The verbal form, however, was in normal use: aban 'I refuse, I will not'. In a negative command only the uninflected aba was used: aba care 'don't do it!' SINDARIN. In Sindarin the following forms are found. baw! imperious negative: 'No, no! Don't! ' avo negative adverb with verbs, as avo garo! 'don't do it'; sometimes used as a prefix: avgaro (< *aba-kar a). This could be personalized in the form avon 'I won't', avam 'we won't': these were of course not in fact derived from avo, which contained the imperative -o < >a, but from the verb stem *aba, with inflexions assimilated to the tense stems in -a; but no other parts of the verb survived in use, except the noun avad refusal, reluctance'. Derived direct from baw! (*ha) was the verb boda- 'ban, prohibit' (*ba-ta). (With the uses of this stem, primary meaning 'refuse, be unwilling', to form negative imperatives, cf. Latin noli, nolite.) B. Meanings and use of the various terms applied to the Elves and their varieties in Quenya, Telerin, and Sindarin. Quenya. 1. quen, pl. queni, person, individual, man or woman. Chiefly used in the unstressed form quen. Mostly found in the singular: 'one, somebody'; in the pl. 'people, they'. Also combined with other elements, as in aiquen 'if anybody, whoever', ilquen 'everybody'. In a number of old compounds -quen, pl. queni was combined with noun or adjective stems to denote habitual occupations or functions, or to describe those having some notable (permanent) quality: as -man in English (but without distinction of sex) in horseman, seaman, work- man, nobleman, etc. Q roquen 'horseman, rider'; (Note 3, p. 407) kiryaquen 'shipman, sailor'; arquen 'a noble'. These words belong to everyday speech, and have no special reference to Elves. They were freely applied to other Incarnates, such as Men or Dwarves, when the Eldar became acquainted with them. 2. Quendi Elves, of any kind, including the Avari. The sg. Quende was naturally less frequently used. As has been seen, the word was made when the Elves as yet knew of no other 'people' than themselves. The sense 'the Elvish people, as a whole', or in the sg. 'an Elf and not some other similar creature', developed first in Aman, where the Elves lived among or in contact with the Valar and Maiar. During the Exile when the Noldor became re-associated with their Elvish kin, the Sindar, but met other non-Elvish people, such as Orcs, Dwarves, and Men, it became an even more useful term. But in fact it had ceased in Aman to be a word of everyday use, and remained thereafter mainly used in the special language of Lore: histories or tales of old days, or learned writings on peoples and languages. In ordinary language the Elves of Aman called themselves Eldar (or in Telerin Elloi): see below. There also existed two old compounds containing *kwendi: *kala-kwendi and *mori-kwendi, the Light-folk and the Dark- folk. These terms appear to go back to the period before the Separation, or rather to the time of the debate among the Quendi concerning the invitation of the Valar. They were evidently made by the party favourable to Orome, and referred originally to those who desired the Light of Valinor (where the ambassadors of the Elves reported that there was no darkness), and those who did not wish for a place in which there was no night. But already before the final separation *mori-kwendi may have referred to the glooms and the clouds dimming the sun and the stars during the War of the Valar and Melkor,(6) so that the term from the beginning had a tinge of scorn, implying that such folk were not averse to the shadows of Melkor upon Middle-earth. The lineal descendants of these terms survived only in the languages of Aman. The Quenya forms were Kalaquendi and Moriquendi. The Kalaquendi in Quenya applied only to the Elves who actually lived or had lived in Aman; and the Moriquendi was applied to all others, whether they had come on the March or not. The latter were regarded as greatly inferior to the Kalaquendi, who had experienced the Light of Valinor, and had also acquired far greater knowledge and powers by their association with the Valar and Maiar. In the period of Exile the Noldor modified their use of these terms, which was offensive to the Sindar. Kalaquendi went out of use, except in written Noldorin lore. Moriquendi was now applied to all other Elves, except the Noldor and Sindar, that is to Avari or to any kind of Elves that at the time of the coming of the Noldor had not long dwelt in Beleriand and were not subjects of Elwe. It was never applied, however, to any but Elvish peoples. The old distinction, when made, was represen- ted by the new terms Amanyar 'those of Aman', and Uamanyar or Umanyar 'those not of Aman', beside the longer forms Amaneldi and Umaneldi. 3. Quendya, in the Noldorin dialect Quenya. This word remained in ordinary use, but it was only used as a noun 'the Quendian language'. (Note 4, p. 407) This use of Quendya must have arisen in Aman, while Quendi still remained in general use. Historically, and in the more accurate use of the linguistic Loremasters, Quenya included the dialect of the Teleri, which though divergent (in some points from days before settlement in Aman, such as *kw > p), remained generally intelligible to the Vanyar and Noldor. But in ordinary use it was applied only to the dialects of the Vanyar and Noldor, the differences between which only appeared later, and remained, up to the period just before the Exile, of minor importance. In the use of the Exiles Quenya naturally came to mean the language of the Noldor, developed in Aman, as distinct from other tongues, whether Elvish or not. But the Noldor did not forget its connexion with the old word Quendi, and still regarded the name as implying 'Elvish', that is the chief Elvish tongue, the noblest, and the one most nearly preserving the ancient character of Elvish speech. For a note on the Elvish words for 'language', especially among the Noldorin Lore- masters, see Appendix D (p. 391). 4. Elda and Eldo. The original distinction between these forms as meaning 'one of the Star-folk, or Elves in general', and one of the 'Marchers', became obscured by the close approach of the forms. The form Eldo went out of use, and Elda remained the chief word for 'Elf' in Quenya. But it was not in accurate use held to include the Avari (when they were remembered or considered); i.e. it took on the sense of Eldo. It may, however, have been partly due to its older sense that in popular use it was the word ordinarily employed for any Elf, that is, as an equivalent of the Quende of the Loremasters. When one of the Elves of Aman spoke of the Eldalie, 'the Elven-folk', he meant vaguely all the race of Elves, though he was probably not thinking of the Avari. For, of course, the special kinship of the Amanyar with those left in Beleriand (or Hekeldamar) was remembered, especially by the Teleri. When it was necessary to distinguish these two branches of the Eldar (or properly Eldor), those who had come to Aman were called the Odzeldi N Oareldi, for which another form (less used) was Auzeldi, N Aureldi; those who had re- mained behind were the Hekeldi. These terms naturally be- longed rather to history than everyday speech, and in the period of the Exile they fell out of use, being unsuitable to the situation in Beleriand. The Exiles still claimed to be Amanyar, but in practice this term usually now meant those Elves remaining in Aman, while the Exiles called themselves Etyangoldi 'Exiled Noldor', or simply (since the great majority of their clan had come into exile) Noldor. All the subjects of Elwe they called Sindar or 'Grey-elves'. Telerin. 1. The derivatives of *KWEN were more sparingly represen- ted in the Telerin dialects, of Aman or Beleriand. This was in part due to the Common Telerin change of kw > p, (Note 5, p. 407) which caused *pen < *kwen to clash with the PQ stem *PEN 'lack, be without', and also with some of the derivatives of *PED 'slope, slant down' (e.g. *penda 'sloping'). Also the Teleri felt themselves to be a separate people, as compared with the Vanyar and Noldor, whom taken together they outnumbered. This sentiment began before the Separation, and increased on the March and in Beleriand. In consequence they did not feel strongly the need for a general word embracing all Elves, until they came in contact with other non-Elvish Incarnates. As a pronoun enclitic (e.g. in aipen, Q aiquen; ilpen, Q ilquen) *kwen survived in Telerin; but few of the compounds with pen 'man' remained in ordinary use, except arpen 'noble (man)', and the derived adjective arpenia. Pendi, the dialectal equivalent of Q Quendi, survived only as a learned word of the historians, used with reference to ancient days before the Separation; the adjective *Pendia (the equiva- lent of Quendya) had fallen out of use.. (Note 6, p. 408) The Teleri had little interest in linguistic lore, which they left to the Noldor. They did not regard their language as a 'dialect' of Quenya, but called it Lindarin or Lindalambe. Quenya they called Goldorin or Goldolambe; for they had few contacts with the Vanyar. The old compounds in Telerin form Calapendi and Moripen- di survived in historical use; but since the Teleri in Aman re- mained more conscious of their kinship with the Elves left in Beleriand, while Calapendi was used, as Kalaquendi in Quenya, to refer only to the Elves of Aman, Moripendi was not applied to the Elves of Telerin origin who had not reached Aman. 2. Ello and Ella. The history of the meanings of these words was almost identical with that of the corresponding Elda and Eldo in Quenya. In Telerin the -o form became preferred, so that generally T Ello was the equivalent of Q Elda. But Ella remained in use in quasi-adjectival function (e.g. as the first element in loose or genitival compounds): thus the equivalent of Q Eldalie was in T Ellalie. In contrast to the Elloi left in Beleriand those in Aman were in histories called Audel, pl. Audelli. Those in Beleriand were the Hecelloi of Heculbar (or Hecellubar). Sindarin. 1. Derivatives of *KWEN were limited to the sense: pronomin- al 'one, somebody, anybody', and to a few old compounds that survived. PQ *kwende, *kwendi disappeared altogether. The reasons for this were partly the linguistic changes already cited; and partly the circumstances in which the Sindar lived, until the return of the Noldor, and the coming of Men. The linguistic changes made the words unsuitable for survival; the circum- stances removed all practical need for the term. The old unity of the Elves had been broken at the Separation. The Elves of Beleriand were isolated, without contact with any other people, Elvish or of other kind; and they were all of one clan and language: Telerin (or Lindarin). Their own language was the only one that they ever heard; and they needed no word to distinguish it, nor to distinguish themselves. As a pronoun, usually enclitic, the form pen, mutated ben, survived. A few compounds survived, such as rochben 'rider' (m. or f.), orodben 'a mountaineer' or 'one living in the mountains', arphen 'a noble'. Their plurals were made by i-affection, originally carried through the word: as roechbin, oerydbin, erphin, but the normal form of the first element was often restored when the nature of the composition remained evident: as rochbin, but always erphin. These words had no special association with Elves. Associated with these compounds were the two old words Calben (Celbin) and Morben (Moerbin). On the formal relation of these to Quenya Kalaquendi and Moriquendi see p. 362. They had no reference to Elves, except by accident of circum- stance. Celbin retained what was, as has been said, probably its original meaning: all Elves other than the Avari; and it included the Sindar. It was in fact the equivalent (when one was needed) of the Quenya Eldar, Telerin Elloi. But it referred to Elves only because no other people qualified for the title. Moerbin was similarly an equivalent for Avari; but that it did not mean only 'Dark-elves' is seen by its ready application to other Incarnates, when they later became known. By the Sindar anyone dwelling outside Beleriand, or entering their realm from outside, was called a Morben. The first people of this kind to be met were the Nandor, who entered East Beleriand over the passes of the Mountains before the return of Morgoth; soon after his return came the first invasions of his Orcs from the North.(7) Somewhat later the Sindar became aware of Avari, who had crept in small and secret groups into Beleriand from the South. Later came the Men of the Three Houses, who were friendly; and later still Men of other kinds. All these were at first acquaintance called Moerbin. (Note 7, p. 408) But when the Nandor were recog- nized as kinsfolk of Lindarin origin and speech (as was still recognizable), they were received into the class of Celbin. The Men of the Three Houses were also soon removed from the class of Moerbin. (Note 8, p. 408) They were given their own name, Edain, and were seldom actually called Celbin, but they were recognized as belonging to this class, which became . practically equivalent to 'peoples in alliance in the War against Morgoth'. The Avari thus remained the chief examples of Moerbin. Any individual Avar who joined with or was admitted among the Sindar (it rarely happened) became a Calben; but the Avari in general remained secretive, hostile to the Eldar, and untrustworthy; and they dwelt in hidden places in the deeper woods, or in caves. (Note 9, p. 408) Moerbin as applied to them is usually translated 'Dark-elves', partly because Moriquendi in the Quenya of the Exiled Noldor usually referred to them. But that no special reference to Elves was intended by the Sindarin word is shown by the fact that Moerbin was at once applied to the new bands of Men (Easterlings) that appeared before the Battle of the Nirnaeth. (Note 9, p. 408) If in Sindarin an Avar, as distinct from other kinds of Morben, was intended, he was called Mornedhel. 2. Edhel, pl. Edhil. In spite of its ultimate derivation (see p. 360) this was the general word for 'Elf, Elves'. In the earlier days it naturally referred only to the Eldarin Sindar, for no other kind was ever seen; but later it was freely applied to Elves of any kind that entered Beleriand. It was however only used in these two forms. The masculine and feminine forms were Ellon m. and Elleth f. and the class-plural was Eldrim, later Elrim, when this was not replaced by the more commonly used Eledhrim (see below). The form without the m. and f. suffixes was not in use, and survived only in some old compounds, especially personal names, in the form el, pl. il, as a final element. The form Elen, pl. Elin was only used in histories or the works of the Loremasters, as a word to include all Elves (Eldar and Avari). But the class-plural Eledhrim was the usual word for 'all the Elvish race', whenever such an expression was needed. All these words and forms, whatever their etymologies (see above), were applicable to any kind of Elf. In fact Edhel was properly applied only to Eldar; Ell- may have a mixed origin; and Elen was an ancient general word. (Note 10, p. 410) 3. The Sindar had no general name for themselves as distinct from other varieties of Elf, until other kinds entered Beleriand. The descendant of the old clan name *Lindai (Q Lindar) had fallen out of normal use, being no longer needed in a situation were all the Edhil were of the same kind, and people were more aware of the growing differences in speech and other matters between those sections of the Elves that lived in widely sundered parts of a large and mostly pathless land. They were thus in ordinary speech all Edhil, but some belonged to one region and some to another: they were Falathrim from the sea-board of West Beleriand, or lathrim from Doriath (the land of the Fence, or iath), or Mithrim who had gone north from Beleriand and inhabited the regions about the great lake that afterwards bore their name. (Note 11, p. 410) The old clan-name *Lindai survived in the compound Glinnel, pl. Glinnil, a word only known in historical lore, and the equivalent of Quenya Teleri or Lindar; see the Notes on the Clan-names below. All the Sindarin subjects of King Elu- Thingol, as distinguished from the incoming Noldor, were sometimes later called the Eluwaith. Dunedhil 'West-elves' (the reference being to the West of Middle-earth) was a term made to match Dunedain 'West-men' (applied only to the Men of the Three Houses). But with the growing amalgamation, outside Doriath, of the Noldor and Sindar into one people using the Sindarin tongue as their daily speech, this soon became applied to both Noldor and Sindar. While the Noldor were still distinct, and whenever it was desired to recall their difference of origin, they were usually called Odhil (sg. Odhel). This as has been seen was originally a name for all the Elves that left Beleriand for Aman. These were also called by the Sindar Gwanwen, pl. Gwenwin (or Gwanwel, Gwenwil) 'the departed': cf. Q vanwa. This term, which could not suitably be applied to those who had come back, remained the usual Sindarin name for the Elves that remained in Aman. Odhil thus became specially the name of the Exiled Noldor. In this sense the form Godhel, pl. Godhil soon replaced the older form. It seems to have been due to the influence of the clan-name Golodh, pl. Goelydh; or rather to a deliberate blending of the two words. The old clan-name had not fallen out of memory (for the Noldor and the Sindar owing to the great friendship of Finwe and Elwe were closely associated during their sojourn in Beleriand before the Departure) and it had in consequence a genuine Sindarin form (< CE *ngolodo). But the form Golodh seems to have been phonetically unpleas- ing to the Noldor. The name was, moreover, chiefly used by those who wished to mark the difference between the Noldor and the Sindar, and to ignore the dwelling of the Noldor in Aman which might give them a claim to superiority. This was especially the case in Doriath, where King Thingol was hostile to the Noldorin chieftains, Feanor and his sons, and Fingolfin, because of their assault upon the Teleri in Aman, the people of his brother Olwe. The Noldor, therefore, when using Sindarin, never applied this name (Golodh) to themselves, and it fell out of use among those friendly to them. 4. Eglan, pl. Eglain, Egladrim. This name, 'the Forsaken', was, as has been said, given by the Sindar to themselves. But it was not in Beleriand a name for all the Elves who remained there, as were the related names, Hekeldi, Hecelloi, in Aman. It applied only to those who wished to depart, and waited long in vain for the return of Ulmo, taking up their abode on or near the coasts. There they became skilled in the building and management of ships. Cirdan was their lord. Cirdan's folk were made up both of numbers of the following of Olwe, who straying or lingering came to the shores too late, and also of many of the following of Elwe, who abandoned the search for him and did not wish to be separated for ever from their kin and friends. This folk remained in the desire of Aman for long years, and they were among the most friendly to the Exiles. They continued to call themselves the Eglain, and the regions where they dwelt Eglamar and Eglador. The latter name fell out of general use. It had originally been applied to all western Beleriand between Mount Taras and the Bay of Balar, its eastern boundary being roughly along the River Narog. Eglamar, however, remained the name of the 'Home of the Eglain': the sea-board from Cape Andras to the headland of Bar-in-Myl ('Home of the Gulls'),(8) which included the ship-havens of Cirdan at Brithonbar (9) and at the head of the firth of Eglarest. The Eglain became a people somewhat apart from the inland Elves, and at the time of the coming of the Exiles their language was in many ways different. (Note 12, p. 411) But they acknowledged the high-kingship of Thingol, and Cirdan never took the title of king.(10) *Abari. This name, evidently made by the Eldar at the time of the Separation, is found in histories in the Quenya form Avari, and the Telerin form Abari. It was still used by the historians of the Exiled Noldor, though it hardly differed from Moriquendi, which (see above) was no longer used by the Exiles to include Elves of Eldarin origin. The plural Evair was known to Sindarin loremasters, but was no longer in use. Such Avari as came into Beleriand were, as has been said, called Morben, or Mornedhel. C. The Clan-names, with notes on other names for divisions of the Eldar. In Quenya form the names of the three great Clans were Vanyar, Noldor, and Lindar. The oldest of these names was Lindar, which certainly goes back to days before the Separa- tion. The other two probably arose in the same period, if somewhat later: their original forms may thus be given in PQ as *wanja, *ngolodo, and linda /glinda. (Note 13, p. 411) According to the legend, preserved in almost identical form among both the Elves of Aman and the Sindar, the Three Clans were in the beginning derived from the three Elf-fathers: Imin, Tata, and Enel (sc. One, Two, Three), and those whom each chose to join his following. So they had at first simply the names Minyar 'Firsts', Tatyar 'Seconds', and Nelyar 'Thirds'. These numbered, out of the original 144 Elves that first awoke, 14, 56, and 74; and these proportions were approximately maintained until the Separation.(11) It is said that of the small clan of the Minyar none became Avari. The Tatyar were evenly divided. The Nelyar were most reluctant to leave their lakeside homes; but they were very cohesive, and very conscious of the separate unity of their Clan (as they continued to be), so that when it became clear that their chieftains Elwe and Olwe were resolved to depart and would have a large following, many of those among them who had at first joined the Avari went over to the Eldar rather than be separated from their kin. The Noldor indeed asserted that most of the 'Teleri' were at heart Avari, and that only the Eglain really regretted being left in Beleriand. According to the Noldorin historians the proportions, out of 144, that when the March began became Avari or Eldar were approximately so: Minyar 14: Avari 0 Eldar 14 Tatyar 56: Avari 28 Eldar 28 Nelyar 74: Avari 28 Eldar 46 > Amanyar Teleri 20; Sindar and Nandor 26 In the result the Noldor were the largest clan of Elves in Aman; while the Elves that remained in Middle-earth (the Moriquendi in the Quenya of Aman) outnumbered the Amanyar in the proportion of 82 to 62.(12) How far the descriptive Clan-names, *wanja, *ngolodo, and *linda were preserved among the Avari is not now known; but the existence of the old clans was remembered, and a special kinship between those of the same original clan, whether they had gone away or remained, was still recognized. The first Avari that the Eldar met again in Beleriand seem to have claimed to be Tatyar, who acknowledged their kinship with the Exiles, though there is no record of their using the name Noldo in any recognizable Avarin form. They were actually unfriendly to the Noldor, and jealous of their more exalted kin, whom they accused of arrogance. This ill-feeling descended in part from the bitterness of the Debate before the March of the Eldar began, and was no doubt later increased by the machinations of Morgoth; but it also throws some light upon the temperament of the Noldor in general, and of Feanor in particular. Indeed the Teleri on their side asserted that most of the Noldor in Aman itself were in heart Avari, and returned to Middle-earth when they discovered their mistake; they needed room to quarrel in. For in contrast the Lindarin elements in the western Avari were friendly to the Eldar, and willing to learn from them; and so close was the feeling of kinship between the remnants of the Sindar, the Nandor, and the Lindarin Avari, that later in Eriador and the Vale of Anduin they often became merged together. Lindar (Teleri).(13) These were, as has been seen, much the largest of the ancient clans. The name, later appearing in Quenya form as Lindar (Telerin Lindai), is already referred to in the legend of 'The Awakening of the Quendi', which says of the Nelyar that 'they sang before they could speak with words . The name *Linda is therefore clearly a derivative of the primitive stem *LIN (showing reinforcement of the medial x and adjectival -a). This stem was possibly one of the contributions of the Nelyar to Primitive Quendian, for it reflects their predilections and associations, and produces more derivatives in Lindarin tongues than in others. Its primary reference was to melodious or pleasing sound, but it also refers (especially in Lindarin) to water, the motions of which were always by the Lindar associated with vocal (Elvish) sound. The reinforcements, either medial lind- or initial glin-, glind-, were however almost solely used of musical, especially vocal, sounds produced with intent to please. It is thus to the love of the Nelyar for song, for vocal music with or without the use of articulate words, that the name Lindar originally referred; though they also loved water, and before the Separation never moved far from the lake and waterfall (14) of Cuivienen, and those that moved into the West became enamoured of the Sea. (Note 14, p. 411) In Quenya, that is, in the language of the Vanyar and Noldor, those of this clan that joined in the March were called the Teleri. This name was applied in particular to those that came at last and latest to Aman; but it was also later applied to the Sindar. The name Lindar was not forgotten, but in Noldorin lore it was chiefly used to describe the whole clan, including the Avari among them. Teleri meant 'those at the end of the line, the hindmost', and was evidently a nickname arising during the March, when the Teleri, the least eager to depart, often lagged far behind. (Note 15, p. 411) Vanyar. This name was probably given to the First Clan by the Noldor. They accepted it, but continued to call themselves most often by their old numerical name Minyar (since the whole of this clan had joined the Eldar and reached Aman). The name referred to the hair of the Minyar, which was in nearly all members of the clan yellow or deep golden. This was regarded as a beautiful feature by the Noldor (who loved gold), though they were themselves mostly dark-haired. Owing to intermarriage the golden hair of the Vanyar sometimes later appeared among the Noldor: notably in the case of Finarfin, and in his children Finrod and Galadriel, in whom it came from King Finwe s second wife, Indis of the Vanyar. Vanyar thus comes from an adjectival derivative *wanja from the stem *WAN. Its primary sense seems to have been very similar to English (modern) use of 'fair' with reference to hair and complexion; though its actual development was the reverse of the English: it meant 'pale, light-coloured, not brown or dark', and its implication of beauty was secondary. In English the meaning 'beautiful' is primary. From the same stem was derived the name given in Quenya to the Valie Vana wife of Orome. Since the Lindar had little contact with the Vanyar either on the March or later in Aman, this name was not much used by them for the First Clan. The Amanyar Teleri had the form Vaniai (no doubt taken from the Noldor), but the name appears to have been forgotten in Beleriand, where the First Clan (in lore and history only) were called Miniel, pl. Minil. Noldor. This name was probably older than Vanyar, and may have been made before the March. It was given to the Second Clan by the others. It was accepted, and was used as their regular and proper name by all the Eldarin members of the clan throughout their later history. The name meant 'the Wise', that is those who have great knowledge and understanding. The Noldor indeed early showed the greatest talents of all the Elves both for intellectual pursuits and for technical skills. The variant forms of the name: Q Noldo, T Goldo, S Golodh (Ngolodh), indicate a PQ original *ngolodo. This is a derivative of the stem *NGOL 'knowledge, wisdom, lore'. This is seen in Q Role 'long study (of any subject)', ingole 'lore', ingolmo 'loremaster'. In T gole, engole had the same senses as in Q but were used most often of the special 'lore' possessed by the Noldor. In S the word gul (equivalent of Q nole) had less laudatory associations, being used mostly of secret knowledge, especially such as possessed by artificers who made wonderful things; and the word became further darkened by its frequent use in the compound morgul 'black arts', applied to the delusory or perilous arts and knowledge derived from Morgoth. Those indeed among the Sindar who were unfriendly to the Noldor attributed their supremacy in the arts and lore to their learning from Melkor-Morgoth. This was a falsehood, coming itself ultimately from Morgoth; though it was not without any foundation (as the lies of Morgoth seldom were). But the great gifts of the Noldor did not come from the teaching of Melkor. Feanor the greatest of them all never had any dealings with Melkor in Aman, and was his greatest foe. Sindar. Less commonly the form Sindel, pl. Sindeldi, is also met in Exilic Quenya. This was the name given by the Exiled Noldor (see Note 11) to the second largest of the divisions of the Eldar. (Note 16, p. 412) It was applied to all the Elves of Telerin origin that the Noldor found in Beleriand, though it later excluded the Nandor, except those who were the direct subjects of Elwe, or had become merged with his people. The name meant 'the Grey', or 'the Grey-elves', and was derived from *THIN, PQ *thindi 'grey, pale or silvery grey', Q pinde, N dialect sinde. On the origin of this name see Note 11. The Loremasters also supposed that reference was made to the hair of the Sindar. Elwe himself had indeed long and beautiful hair of silver hue, but this does not seem to have been a common feature of the Sindar, though it was found among them occasionally, especially in the nearer or remoter kin of Elwe (as in the case of Cirdan).(15) In general the Sindar appear to have very closely resembled the Exiles, being dark-haired, strong and tall, but lithe. Indeed they could hardly be told apart except by their eyes; for the eyes of all the Elves that had dwelt in Aman impressed those of Middle-earth by their piercing brightness. For which reason the Sindar often called them Lachend, pl. Lechind 'flame-eyed'. Nandor. This name must have been made at the time, in the latter days of the March, when certain groups of the Teleri gave up the March; and it was especially applied to the large following of Lenwe, (Note 17, p. 412) who refused to cross the Hithaeglir.(16) The name was often interpreted as 'Those who go back'; but in fact none of the Nandor appear to have returned, or to have rejoined the Avari. Many remained and settled in lands that they had reached, especially beside the River Anduin; some turned aside and wandered southwards. (Note 18, p. 412) There was, however, as was later seen, a slow drift westward of the Moriquendi during the captivity of Melkor, and eventually groups of the Nandor, coming through the Gap between the Hithaeglir and Eryd Nimrais, spread widely in Eriador. Some of these finally entered Beleriand, not long before the return of Morgoth.(17) These were under the leadership of Denethor, son of Denweg (see Note 17), who became an ally of Elwe in the first battles with the creatures of Morgoth. The old name Nandor was however only remembered by the Noldorin historians in Aman; and they knew nothing of the later history of this folk, recalling only that the leader of the defection before the crossing of the dread Hithaeglir was named Lenwe (i.e. Denweg). The Sindarin loremasters remembered the Nandor as Danwaith, or by confusion with the name of their leader Denwaith. This name they at first applied to the Nandor that came into Eastern Beleriand; but this people still called themselves by the old clan-name *Lindai, which had at that time taken the form Lindi in their tongue. The country in which most of them eventually settled, as a small independent folk, they called Lindon (< *Lindana): this was the country at the western feet of the Blue Mountains (Eryd Luin), watered by the tributaries of the great River Gelion, and previously named by the Sindar Ossiriand, the Land of Seven Rivers. The Sindar quickly recognized the Lindi as kinsfolk of Lindarin origin (S Glinnil), using a tongue that in spite of great differences was still perceived to be akin to their own; and they adopted the names Lindi and Lindon, giving them the forms Lindil (sg. Lindel) or Lindedhil, and Lindon or Dor Lindon. In Exilic Quenya the forms used (derived from the Sindar or direct from the Nandor) were Lindi and Lindon (or Lindone). The Exiled Noldor also usually referred to the Eryd Luin as Eryd Lindon, since the highest parts of that range made the eastern borders of the country of Lindon. These names were however later replaced among the Sindar by the name 'Green-elves', at least as far as the inhabitants of Ossiriand were concerned; for they withdrew themselves and took as little part in the strife with Morgoth as they could. This name, S Laegel, pl. Laegil, class-plural Laegrim or Laegel(d)- rim, was given both because of the greenness of the land of Lindon, and because the Laegrim clothed themselves in green as an aid to secrecy. This term the Noldor translated into Quenya Laiquendi; but it was not much used. Appendix A. Elvish names for Men. The first Elves that Men met in the world were Avari, some of whom were friendly to them, but the most avoided them or were hostile (according to the tales of Men). What names Men and Elves gave to one another in those remote days, of which little was remembered when the Loremasters in Beleriand made the acquaintance of the After-born, there is now no record. By the Dunedain the Elves were called Nimir (the Beautiful).(18) The Eldar did not meet Men of any kind or race until the Noldor had long returned to Beleriand and were at war with Morgoth. The Sindar did not even know of their existence, until the coming of the Nandor; and these brought only rumour of a strange people (whom they had not themselves seen) wandering in the lands of the East beyond the Hithaeglir. From these uncertain tales the Sindar concluded that the 'strange people' were either some diminished race of the Avari, or else related to Orcs, creatures of Melkor, bred in mockery of the true Quendi. But the Noldor had already heard of Men in Aman. Their knowledge came in the first place from Melkor and was perverted by his malice, but before the Exile those who would listen had learned more of the truth from the Valar, and they knew that the newcomers were akin to themselves, being also Children of Iluvatar, though differing in gifts and fate. There- fore the Noldor made names for the Second Race of the Children, calling them the Atani 'the Second Folk'. Other names that they devised were Apanonar 'the After-born', and Hildor 'the Followers'. In Beleriand Atan, pl. Atani, was the name most used at first. But since for a long time the only Men known to the Noldor and Sindar were those of the Three Houses of the Elf-friends, this name became specially associated with them, so that it was seldom in ordinary speech applied to other kinds of Men that came later to Beleriand, or that were reported to be dwelling beyond the Mountains. The Elf-friends (Note 19, p. 412) were sometimes called by the Loremasters Nunatani (S Dunedain), 'Western Men', a term made to match Dunedhil, which was a name for all the Elves of Beleriand, allied in the War (see p. 378). The original reference was to the West of Middle-earth, but the name Nunatani, Dunedain was later applied solely to the Numenoreans, descendants of the Atani, who removed to the far western isle of Numenore. Apanonar 'the After-born' was a word of lore, not used in daily speech. A general term for Men of all kinds and races, as distinct from Elves, was only devised after their mortality and brief life-span became known to the Elves by experience. They were then called Firyar 'Mortals', or Firimar of similar sense (literally 'those apt to die'). (Note 20, p. 412) These words were derived from the stem *PHIRI 'exhale, expire, breathe out', which had no original connexion with death.(19) Of death, as suffered by Men, the Elves knew nothing until they came into close association with the Atani; but there were cases in which an Elf, overcome by a great sorrow or weariness, had resigned life in the body. The chief of these, the departure of Miriel wife of King Finwe, was a matter of deep concern to all the Noldor, and it was told of her that her last act, as she gave up her life in the body and went to the keeping of Mandos, was a deep sigh of weariness. These Quenya names were later adapted to the forms of Sindarin speech: Atan > Adan, pl. Edain; Firya > Feir, pl. Fir (with Firion m.sg., Firieth f.sg.), class-plural Firiath; Firima > Fireb, pl. Firib, class-plural Firebrim. These forms, which cannot for historical reasons have been inherited from CE, but are those which the words if inherited would have taken, show that they were adapted by people with considerable knowledge of both tongues and understanding of their relations to one another; that is, they were probably first made by the Noldor for use in Sindarin, when they had adopted this language for daily use in Beleriand. Fireb as compared with Firima shows the use of a different suffix, (Note 21, p. 412) since the S equivalent of Q -ima (*-ef) was not current. Apanonar was rendered by Abonnen, pl. Eboennin, using a different participial formation from the stem *ONO 'beget, give birth to'. Hildor, since the stem *KHILI 'follow' was not current in Sindarin, was rendered by Aphadon, pl. Ephedyn, class-plural Aphadrim, from S aphad- 'follow' < *ap-pata 'walk behind, on a track or path'. Appendix B. Elvish names for the Dwarves. The Sindar had long known the Dwarves, and had entered into peaceful relations with them, though of trade and exchange of skills rather than of true friendship, before the coming of the Exiles. The name (in the plural) that the Dwarves gave to themselves was Khazad, and this the Sindar rendered as they might in the terms of their own speech, giving it the form *chadod > *chadaud > Hadhod. (Note 22, p. 412) Hadhod, Hadhodrim was the name which they continued to use in actual intercourse with the Dwarves; but among themselves they referred to the Dwarves usually as the Naugrim 'the Stunted Folk'. The adjective naug 'dwarf(ed), stunted', however, was not used by itself for one of the Khazad. The word used was Nogoth, pl. Noegyth, class-plural Nogothrim (as an occasional equivalent of Naugrim). (Note 23, p. 413) They also often referred to the Dwarves as a race by the name Dornhoth 'the Thrawn Folk', because of their stubborn mood as well as bodily toughness. The Exiles heard of the Dwarves first from the Sindar, and when using the Sindarin tongue naturally adopted the already established names. But later in Eastern Beleriand the Noldor came into independent relations with the Dwarves of Eryd Lindon, and they adapted the name Khazad anew for use in Quenya, giving it the form Kasar, pl. Kasari or Kasari. (Note 24, p. 413) This was the word most commonly used in Quenya for the Dwarves, the partitive plural being Kasalli, and the race-name Kasallie. But the Sindarin names were also adapted or imitated, a Dwarf being called Nauko or Norno (the whole people Naukalie or Nornalie). Norno was the more friendly term. (Note 25, p. 413) The Petty-dwarves. See also Note 7. The Eldar did not at first recognize these as Incarnates, for they seldom caught sight of them in clear light. They only became aware of their existence indeed when they attacked the Eldar by stealth at night, or if they caught them alone in wild places. The Eldar therefore thought that they were a kind of cunning two-legged animals living in caves, and they called them Levain tad-dail, or simply Tad-dail, and they hunted them. But after the Eldar had made the acquaintance of the Naugrim, the Tad-dail were recognized as a variety of Dwarves and were left alone. There were then few of them surviving, and they were very wary, and too fearful to attack any Elf, unless their hiding-places were approached too nearly. The Sindar gave them the names Nogotheg 'Dwarf- let', or Nogoth niben 'Petty Dwarf'.(20) The great Dwarves despised the Petty-dwarves, who were (it is said) the descendants of Dwarves who had left or been driven our from the Communities, being deformed or undersized, or slothful and rebellious. But they still acknowledged their kinship and resented any injuries done to them. Indeed it was one of their grievances against the Eldar that they had hunted and slain their lesser kin, who had settled in Beleriand before the Elves came there. This grievance was set aside, when treaties were made between the Dwarves and the Sindar, in considera- tion of the plea that the Petty-dwarves had never declared themselves to the Eldar, nor presented any claims to land or habitations, but had at once attacked the newcomers in darkness and ambush. But the grievance still smouldered, as was later seen in the case of Mim, the only Petty-dwarf who played a memorable part in the Annals of Beleriand. The Noldor, for use in Quenya, translated these Sindarin names for the Petty-dwarves by Attalyar 'Bipeds', and Pikinau- kor or Pitya-naukor. The chief dwellings of the Dwarves that became known to the Sindar (though few ever visited them) were upon the east side of the Eryd Luin. They were called in the Dwarf-tongue Gabilga- thol and Tumunzahar. The greatest of all the mansions of the Dwarves, Khazad-dum, beneath the Hithaeglir far to the east, was known to the Eldar only by name and rumour derived from the western Dwarves. These names the Sindar did not attempt to adapt, but translated according to their sense, as Belegost 'Mickleburg'; Novrod, later Nogrod, meaning originally 'Hollowbold'; and Hadhodrond 'Dwarrowvault'.(21) (Note 26, p. 414) These names the Noldor naturally used in speaking or writing Sindarin, but for use in Quenya they translated the names anew as Turosto, Navarot, and Casarrondo. Appendix C. Elvish names for the Orcs. The opening paragraphs of this Appendix have been given in Morgoth's Ring p. 416 and are not repeated here. The words that now follow, 'these shapes and the terror that they inspired', refer to the 'dreadful shapes' that haunted the dwellings of the Elves in the land of their awakening. For these shapes and the terror that they inspired the element chiefly used in the ancient tongue of the Elves appears to have been *RUKU. In all the Eldarin tongues (and, it is said, in the Avarin also) there are many derivatives of this stem, having such ancient forms as: ruk-, rauk-, uruk-, urk(u), runk-, rukut/s, besides the strengthened stem gruk-, and the elaborated guruk-, nguruk. (Note 27, p. 415) Already in PQ that word must have been formed which had in CE the form *rauku or *rauko. This was applied to the larger and more terrible of the enemy shapes. But ancient were also the forms uruk, urku/o, and the adjectival urka 'horrible'. (Note 28, p. 415) In Quenya we meet the noun urko, pl. urqui, deriving as the plural form shows from *urku or *uruku. In Sindarin is found the corresponding urug; but there is in frequent use the form orch, which must be derived from *urko or the adjectival *urka. In the lore of the Blessed Realm the Q urko naturally seldom occurs, except in tales of the ancient days and the March, and then is vague in meaning, referring to anything that caused fear to the Elves, any dubious shape or shadow, or prowling creature. In Sindarin urug has a similar use. It might indeed be translated 'bogey'. But the form orch seems at once to have been applied to the Orcs, as soon as they appeared; and Orch, pl. Yrch, class-plural Orchoth remained the regular name for these creatures in Sindarin afterwards. The kinship, though not precise equivalence, of S orch to Q urko, urqui was recognized, and in Exilic Quenya urko was commonly used to translate S orch, though a form showing the influence of Sindarin, orko, pl. orkor and orqui, is also often found. These names, derived by various routes from the Elvish tongues, from Quenya, Sindarin, Nandorin, and no doubt Avarin dialects, went far and wide, and seem to have been the source of the names for the Orcs in most of the languages of the Elder Days and the early ages of which there is any record. The form in Adunaic urku, urkhu may be direct from Quenya or Sindarin; and this form underlies the words for Orc in the languages of Men of the North-West in the Second and Third Ages. The Orcs themselves adopted it, for the fact that it referred to terror and detestation delighted them. The word uruk that occurs in the Black Speech, devised (it is said) by Sauron to serve as a lingua franca for his subjects, was probably borrowed by him from the Elvish tongues of earlier times. It referred, however, specially to the trained and disciplined Orcs of the regiments of Mordor. Lesser breeds seem to have been called snaga.(22) The Dwarves claimed to have met and fought the Orcs long before the Eldar in Beleriand were aware of them. It was indeed their obvious detestation of the Orcs, and their willingness to assist in any war against them, that convinced the Eldar that the Dwarves were no creatures of Morgoth. Nonetheless the Dwarvish name for Orcs, Rukhs, pl. Rakhas, seems to show affinity to the Elvish names, and was possibly ultimately derived from Avarin. The Eldar had many other names for the Orcs, but most of these were 'kennings', descriptive terms of occasional use. One was, however, in frequent use in Sindarin: more often than Orchoth the general name for Orcs as a race that appears in the Annals was Glamhoth. Glam meant 'din, uproar, the confused yelling and bellowing of beasts', so that Glamboth in origin meant more or less 'the Yelling-horde', with reference to the horrible clamour of the Orcs in battle or when in pursuit - they could be stealthy enough at need. But Glamhoth became so firmly associated with Orcs that Glam alone could be used of any body of Orcs, and a singular form was made from it, glamog. (Compare the name of the sword Glamdring.) Note. The word used in translation of Q urko, S orch, is Orc. But that is because of the similarity of the ancient English word orc, 'evil spirit or bogey', to the Elvish words. There is possibly no connexion between them. The English word is now generally supposed to be derived from Latin Orcus. The word for Orc in the now forgotten tongue of the Druedain in the realm of Gondor is recorded as being (? in the plural) gorgun. This is possibly derived ultimately from the Elvish words. Appendix D. *Kwen, Quenya, and the Elvish (especially Noldorin) words for 'Language'. The Noldorin Loremasters state often that the meaning of Quendi was 'speakers', 'those who form words with voices' - i karir quettar omainen. Since they were in possession of traditions coming down from ancient days before the Separa- tion, this statement cannot be disregarded; though the develop- ment of sense set out above may also stand as correct. It might be objected that in fact no stem *KWEN clearly referring to speech or vocal sound is found in any known Elvish tongue. The nearest in form is the stem *KWET 'speak, utter words, say'. But in dealing with this ancient word we must go back to the beginnings of Elvish speech, before the later organisation of its basic structure, with its preference (especially in stems of verbal significance) for the pattern X-X(-), with a fixed medial consonant, as e.g. in stems already exemplified above, such as *Dele, *Heke, *Tele, *Kala, *Kiri, *Nuku, *Ruku, etc. A large number of monosyllabic stems (with only an initial consonant or consonant group) still appear in the Eldarin tongues; and many of the dissyllabic stems must have been made by elaboration of these, just as, at a later stage again, the so-called *kalat- stems were extended from the disyllabic forms: *kala > *kalat(a). If we assume, then, that the oldest form of this stem referring to vocal speech was *KWE, of which *KWENE and *KWETE were elaborations, we shall find a striking parallel in the forms of *KWA. This stem evidently referred to 'completion'. As such it survives as an element in many of the Eldarin words for 'whole, total, all, etc. But it also appears in the form *KWAN, and cannot well be separated from the verb stem *KWATA, Q quat- 'fill'. The assumption also helps to explain a curious and evidently archaic form that survives only in the languages of Aman: *ekwe, Q eque, T epe. It has no tense forms and usually receives no pronominal affixes, (Note 29, p. 415) being mostly used only 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. In this *ekwe we have plainly a last survivor of the primitive *KWE. It is again paralleled by a similar formation (though of different function) from *KWA: *akwa. This survives in Quenya only as aqua 'fully, completely, altogether, wholly'. (Note 30, p. 415) Compare the use of -kwa in the formation of adjectives from nouns, such as -ful in English, except that the sense has been less weakened, and remains closer to the original meaning of the stem: completely . (Note 31, p. 415) In Quenya the form eques, originally meaning 'said he, said someone' (see Note 29) was also used as a noun eques, with the analogical plural equessi, 'a saying, dictum, a quotation from someone's uttered words', hence also 'a saying, a current or proverbial dictum'. We may therefore accept the etymology of *kwene, *kwen that would make its original meaning 'speaking, speaker, one using vocal language'. It would indeed be natural for the Elves, requiring a word for one of their own kind as distinguished from other creatures then known, to select the use of speech as a chief characteristic. But once formed the word must have taken the meaning 'person', without specific reference to this talent of the Incarnates. Thus *nere, *ner a male person, a man was derived from *NERE referring to physical strength and valour, but it was possible to speak of a weak or cowardly ner; or indeed to speak of a dumb or silent kwen. It might therefore still be doubted that in the derivative *kwendi the notion of speaking was any longer effectively present. The statement of the Loremasters cannot, however, be dismissed; while it must be remembered that the Elves were always more deeply concerned with language than were other races. Up to the time at least of the Separation, then, *Kwendi must still have implied 'we, the speaking people'; it may indeed have primarily applied to concourses for discussion, or for listening to speeches and recitations. But when the Elves came to know of other creatures of similar forms, and other Incarnates who used vocal language, and the name *Kwendi, Quendi was used to distinguish themselves from these other kinds, the linguistic sense must have been no longer present in ordinary language. With regard to the word Quenya: an account is given above of the way in which this word became used first in Aman for Elvish speech, (Note 32, p. 416) and then for the dialects of the Eldar in Aman, and later for the language of the Vanyar and Noldor, and finally in Middle-earth for the ancient tongue of the Noldor preserved as a language of ritual and lore. This is historically correct, whatever may be the ultimate etymology of Quenya before the Eldar came to Aman. The view taken above (p. 360) is that it is derived from an adjective *kwendja formed upon the stem *kwende (of which *kwendi was the plural), meaning 'belonging to the Quendi or Elves'. Pengolodh the Loremaster of Eressea says, in his Lammas or Account of Tongues, that Quenya meant properly 'language, speech', and was the oldest word for this meaning. This is not a statement based on tradition, but an opinion of Pengolodh; and he appears to mean only that Quendya, Quenya is actually never recorded except as the name of a language, and that language was the only one known to exist when this word was first made. In any case it is clear that Quenya was always in fact particular in its reference; for when the Noldorin Loremasters came to consider linguistic matters, and required words for speech or vocal language in general, as a mode of expression or communication, and for different aspects of speech, they made no use of the element *kwen, quen or its derivatives. The usual word, in non-technical use, for 'language' was 'lambe, Q and T lambe, S lam. This was undoubtedly related to the word for the physical tongue: *lamba, Q and T lamba, S lam. It meant 'tongue-movement, (way of) using the tongue'. (Note 33, p. 416) This use of a word indicating the tongue and its movements for articulate language no doubt arose, even in a period when all known speakers spoke substantially the same language, from elementary observation of the important part played by the tongue in articulate speaking, and from noticing the peculiarities of individuals, and the soon-developing minor differences in the language of groups and clans. Lambe thus meant primarily 'a way of talking', within a common generally intelligible system, and was nearer to our 'dialect' than to 'language'; but later when the Eldar became aware of other tongues, not intelligible without study, lambe naturally became applied to the separate languages of any people or region. The Loremasters, therefore, did not use lambe as a term for language or speech in general. Their terms were derived from the stem *TEN 'indicate, signify', from which was formed the already well-known word *tenwe > Q tengwe 'indication, sign, token'. From this they made the word tengwesta 'a system or code of signs'. Every 'language' was one such system. A lambe was a tengwesta built of sounds (hloni). For the sense Language, as a whole, the peculiar art of the Incarnates of which each tengwesta was a particular product, they used the abstract formation tengwestie. Now *TEN had no special reference to sound. Ultimately it meant 'to point at', and so to indicate a thing, or convey a thought, by some gesture, or by any sign that would be understood. This was appreciated by the Loremasters, who wished for a word free from any limitations with regard to the kind of signs or tengwi used. They could thus include under tengwesta any group of signs, including visible gestures, used and recognized by a community. They knew of such systems of gesture. The Eldar possessed a fairly elaborate system, (Note 34, p. 416) containing a large number of conventional gesture-signs, some of which were as 'arbitrary' as those of phonetic systems. That is, they had no more obvious connexion with self-explanatory gestures (such as pointing in a desired direction) than had the majority of vocal elements or combinations with 'echoic' or imitative words (such ps *mama, Q mama sheep, or *k(a)wak, Q quako 'crow'). The Dwarves indeed, as later became known, had a far more elaborate and organized system. They possessed in fact a secondary tengwesta of gestures, concurrent with their spoken language, which they began to learn almost as soon as they began learning to speak. It should be said rather that they possessed a number of such gesture-codes; for unlike their spoken language, which remained astonishingly uniform and unchanged both in time and in locality, their gesture-codes varied greatly from community to community. And they were differently employed. Not for communication at a distance, for the Dwarves were short-sighted, but for secrecy and the exclusion of strangers. The component sign-elements of any such code were often so slight and so swift that they could hardly be detected, still less interpreted by uninitiated onlookers. As the Eldar eventually discovered in their dealings with the Naugrim, they could speak with their voices but at the same time by 'gesture' convey to their own folk modifications of what was being said. Or they could stand silent considering some proposition, and yet confer among themselves meanwhile. This 'gesture-language', or as they called it iglishmek, the Dwarves were no more eager to teach than their own tongue. But they understood and respected the disinterested desire for knowledge, and some of the later Noldorin loremasters were allowed to learn enough of both their lambe (aglab) and their iglishmek to understand their systems. Though a lambe was thus theoretically simply a tengwesta that happened to employ phonetic signs, hloniti tengwi, the early loremasters held that it was the superior form, capable of producing a system incalculably more subtle, precise and extensive than any hwerme or gesture-code. When unqualified, therefore, tengwesta meant a spoken language. But in technical use it meant more than lambe. The study' of a language included not only lambe, the way of speaking (that is what we should call its phonetics and phonology), but also its morphology, gram- mar, and vocabulary. The section omitted from Appendix D (see p. 359) begins here. The remainder of the text, which now follows, was all included in this Appendix. Before he turned to other matters Feanor completed his alphabetic system, and here also he introduced a change in terms that was afterwards followed. He called the written representation of a spoken tengwe (according to his defini- tion)(23) a tengwa. A 'letter' or any individual significant mark had previously been called a sarat, from *SAR 'score, incise' > 'write'.(24) The Feanorian letters were always called tengwar in Quenya, though sarati remained the name for the Rumilian letters. Since, however, in the mode of spelling commonly used the full signs were consonantal, in ordinary non-technical use tengwar became equivalent to 'consonants', and the vowel-signs were called omatehtar. When the Feanorian letters were brought to Beleriand and applied (first by the Noldor) to Sindarin, tengwa was rendered by its recognized Sindarin equivalent tew, pl. tiw. The letters of the native S alphabet were called certh, pl. cirth. The word in Exilic Quenya certa, pl. certar was an accommodated loan from Sindarin; there was no such word in older Quenya. The Sindarin certh is probably from *kirte 'cutting', a verbal derivative of a type not used in Quenya, the form of which would in any case have been *kirte, if inherited. Though Feanor after the days of his first youth took no more active part in linguistic lore and enquiry, he is credited by tradition with the foundation of a school of Lambengolmor or 'Loremasters of Tongues' to carry on this work. This continued in existence among the Noldor, even through the rigours and disasters of the Flight from Aman and the Wars in Beleriand, and it survived indeed to return to Eressea. Of the School the most eminent member after the founder was, or still is, Pengolodh,(25) an Elf of mixed Sindarin and Noldorin ancestry, born in Nevrast, who lived in Gondolin from its foundation. He wrote both in Sindarin and in Quenya. He was one of the survivors of the destruction of Gondolin, from which he rescued a few ancient writings, and some of his own copies, compilations, and commentaries. It is due to this, and to his prodigious memory, that much of the knowledge of the Elder Days was preserved. All that has here been said concerning the Elvish names and their origins, and concerning the views of the older loremasters, is derived directly or indirectly from Pengolodh. For before the overthrow of Morgoth and the ruin of Beleriand, he collected much material among the survivors of the wars at Sirion's Mouth concerning languages and gesture-systems with which, owing to the isolation of Gondolin, he had not before had any direct acquaintance. Pengolodh is said to have remained in Middle-earth until far on into the Second Age for the further- ance of his enquiries, and for a while to have dwelt among the Dwarves of Casarrondo (Khazad-dum). But when the shadow of Sauron fell upon Eriador, he left Middle-earth, the last of the Lambengolmor, and sailed to Eressea, where maybe he still abides. Note on the 'Language of the Valar'. Little is said in Noldorin lore, such as has been preserved, concerning the 'language of the Valar and Maiar'; though it has been supposed above that the application of Quenya to the speech of the Elves in Aman was due to the contrast between the tongue of the Valar and the tongue of the Elves, which they had before supposed to be the only language in the world. Consider- ing the interest of the Noldor in all matters concerning speech this is strange. Pengolodh indeed comments upon it and offers explanations. What he says in the beginning of his Lammas is here summarized; for his comment contains all that is now known of the matter. 'Even if we had no knowledge of it,' he says, 'we could not reasonably doubt that the Valar had a lambe of their own. We know that all members of their order were incarnated by their own desire, and that most of them chose to take forms like those of the Children of Eru, as they name us. In such forms they would take on all the characters of the Incarnates that were due to the co-operation of hroa with indwelling fea, for otherwise the assumption of these forms would have been needless, and they arrayed themselves in this manner long before they had any cause to appear before us visibly. Since, then, the making of a lambe is the chief character of an Incarnate, the Valar, having arrayed them in this manner, would inevitably during their long sojourn in Arda have made a lambe for themselves. 'But without argument we know that they did so; for there are references to the Lambe Valarinwa in old lore and histories, though these are few and scattered. Most of these references appear to be derived, by tradition of mouth, from "the Sayings of Rumil" (I Equessi Rumilo), the ancient sage of Tirion, concerning the early days of the Eldar in Aman and their first dealings with the Valar. Only part of these Equessi (26) were preserved in the memory of the Lambengolmor during the dark years of the Flight and the Exile. All that I can find or remember I have here put together.' The information that Pengolodh then gives is here set out more briefly. His preliminary points are these. Few of the Eldar ever learned to speak Valarin, even haltingly; among the people as a whole only a small number of words or names became widely known. Feanor indeed, before the growth of his dis- content, is said to have learned more of this tongue than any others before his time, and his knowledge must at any rate have far surpassed the little that is now recorded; but what he knew he kept to himself, and he refused to transmit it even to the Lambengolmor because of his quarrel with the Valar. Our knowledge (Note 35, p. 416) is therefore now limited (1) to statements of the 'ancients' that certain words in Quenya were actually derived from Valarin; (2) to the occasional citation of words and names purporting to be Valarin (neither adopted in Quenya nor adapted to it), though undoubtedly recorded with only approximate accuracy, since no signs or letters not already known in the Elvish alphabets are employed; (3) to statements that certain names (especially those of the Valar or of places in Valinor) were translations of the Valarin forms. In cases (1) and (3) the actual Valarin words are not always indicated. With regard to group (1) Pengolodh cites a 'Saying' of Rumil: 'The Eldar took few words from the Valar, for they were rich in words and ready in invention at need. But though the honour which they gave to the Valar might have caused them to take words from their speech, whether needed or not, few words of Valarin could be fitted to Elvish speech without great change or diminution. For the tongues and voices of the Valar are great and stern, and yet also swift and subtle in movement, making sounds that we find hard to counterfeit; and their words are mostly long and rapid, like the glitter of swords, like the rush of leaves in a great wind or the fall of stones in the mountains.' pengolodh comments: 'Plainly the effect of Valarin upon Elvish ears was not pleasing.' It was, he adds, as may be seen or guessed from what survives, filled with many consonants unfamiliar to the Eldar and alien to the system of their speech.(27) The examples that Pengolodh gives are as follows. (1) (a) words. Ainu 'one of the "order" of the Valar and Maiar, made before Ea'. Valarin ayanuz. It was from this ainu that in Quenya was made the adjective aina 'holy', since according to Quenya derivation ainu appeared to be a personal form of such an adjective. aman 'blessed, free from evil'. Chiefly used as the name of the land in which the Valar dwelt. V form not given; said to mean 'at peace, in accord (with Eru)'. See Manu e. apar, N asar 'fixed time, festival'. V apara 'appointed'. axan 'law, rule, commandment'. V akasan, said to mean 'He says', referring to Eru. indil 'a lily, or other large single flower'. V inidil. mahalma 'throne'. V maxallam (adapted to Quenya), pro- perly one of the seats of the Valar in the Mahanaxar or 'Doom Ring'. The element maxan, said to mean 'authority, authorita- tive decision', was also used in the form Mahan, one of the eight chiefs of the Valar, usually translated as Aratar. miruvore, miruvor 'a special wine or cordial'. V mirubhoze-; said to be the beginning of a longer word, containing the element mirub- 'wine'.(28) telluma 'dome', especially the 'Dome of Varda' over Valinor; but also applied to the domes of the mansion of Manwe and Varda upon Taniquetil. V delguma, altered by association with Q telume. See Note 15. Pengolodh also cites the colour-words, which he says may be found in ancient verse, though they are used only by the Vanyar, 'who, as Rumil reports, adopted many more words than did the Noldor': ezel, ezella 'green'. See Ezellohar. nasar 'red'; ulban 'blue'. V forms not given. tulka 'yellow'. See Tulkas. (b) names. Aule V Azulez (meaning not given). Manwe Reduction and alteration to fit Quenya, in which words of this shape, ending in -we, were frequent in personal names. V Manawenuz 'Blessed One, One {closest) in accord with Eru'. Oldest Q forms Manwen, Manwe. Tulkas V Tulukhastaz; said to contain V elements tuluk- ha(n) 'yellow', and (a)sata- 'hair of head': 'the golden-haired'. Osse, Orome On these two names, the only ones that became known to the Eldar before they reached Aman, see note below. Ulmo Like Manwe, a reduction and alteration to fit Quenya, in which the ending -mo often appeared in names or titles, sometimes with an agental significance: Ulmo was interpreted as 'the Pourer' < *UL 'pour out'. The V form is given as Ul(l)uboz, containing the element ul(l)u 'water'. Osse and Orome. Orome was the first of the Valar that any of the Eldar saw. Osse they met in Beleriand, and he remained long upon the coasts, and became well known to the Sindar (especially to the Eglain). Both these names therefore have Sindarin forms. To Osse corresponds S Yssion or Gaerys; to Orome the S Araw. The V forms are given as Os(o)sai (said to mean 'spuming, foaming'); and Aromez. The first name was evidently adopted in the form Ossai, which became naturally Q Osse. In S Ossai would become ossi > ussi > yssi to which the ending (of male names) -on was added; or else the adjective *gaira 'awful, fearful' was prefixed, producing Gaerys. The latter was more often used by the inland Teleri. *gaira is from *gay- 'astound, make aghast', which was also used in the oldest Eldarin word for the Sea: *gayar, Q ear, S gaear. Aromez evidently, as was pointed out by Feanor, contained the open a-like q (which did as a matter of later observation occur frequently in Valarin). This was treated as was the Eldarin q, so that the Sindarin development was > *arame > aromae > araum(a) > araum, arauv > araw. (In North Sindarin or Mithrim, where the diphthongization of o and the opening of intervocalic m did not occur, the form produced was Arum; cf. the North Sindarin transformation of the Exilic Noldorin name Hisilome > Hithlum.) The Quenya form with Orome for *Arome < *Arome, may show assimilation of the initial o to the following o before the retraction of the normal Q accent to the first syllable; but Pengolodh says that it was due to the association of the name with the native Q *rom, used of the sound of trumps or horns, seen in the Q name for the great horn of Orome, the Vala-roma (also in Q romba 'horn, trumpet', S rom). 'The Eldar,' he says, 'now take the name to signify "horn- blowing" or "horn-blower"; but to the Valar it had no such meaning. Now the names that we have for the Valar or the Maiar, whether adapted from the Valarin or translated, are not right names but titles, referring to some function or character of the person; for though the Valar have right names, they do not reveal them. Save only in the case of Orome. For it is said in the histories of the most ancient days of the Quendi that, when Orome appeared among them, and at length some dared to approach him, they asked him his name, and he answered: Orome. Then they asked him what that signified, and again he answered: Orome. To me only is it given; for I am Orome. Yet the titles that he bore were many and glorious; but he withheld them at that time, that the Quendi should not be afraid.' Nahar, the name of Orome's horse. 'Otherwise it was,' says Pengolodh, 'with the steed upon which the Lord Orome rode. When the Quendi asked his name, and if that bore any meaning, Orome answered: "Nahar, and he is called from the sound of his voice, when he is eager to run>.' But the V form that is recorded by Rumil was naexaerra. Ezellohar (also translated as Koron Oiolaire, Korollaire), the Green Mound upon which grew the Two Trees. V Ezelloxar. Mahanaxar, the 'Doom-ring' in which were set the thrones of the Valar whereon they sat in council (see mahalma above, p. 399). Reduced and altered from V maxananaskad. Also translated as Rithil-Anamo. (2) Valarin words and names, recorded but not adopted. (a) words urus, rusur 'fire'. ithtr 'light'. ul(l)u 'water'. sebeth 'air'. (b) names Arda: V Aparaphelun (said to mean 'appointed dwelling'). Arda Unmarred: Aparaphelun Amanaisal; Arda Marred: Aparaphelun Dusamanudan. Telperion: V Ibrtnidilpathanezel. Laurelin: V Tulukhedelgorus. Ithil 'moon': V Phanaikeluth. Said to mean 'bright mirror'. Anar Sun: V Aparaigas. Said to mean appointed heat . At the end of this short list Pengolodh cites another eques of Rumil, which might seem contrary to that already quoted above: 'Let none be surprised who endeavour to learn some- what of the tongue of the Lords of the West, as have I, if they find therein many words or parts of words that resemble our own words for the same or similar meanings. For even as they took our form for love of us, so in that form their voices would be likely to light upon similar tengwi.' Upon this Pengolodh comments: He knew not of Men or of Dwarves. But we who have dwelt among Men know that (strange though that seems to some) the Valar love them no less. And for my part I perceive a likeness no less, or indeed greater, between the Valarin and the tongues of Men, notably the language of the Dunedain and of the Children of Marach (sc. Adunaic). Also in general manner it resembles the tongues of the Kasari; though this is not to be wondered at, if the tradition that they have is true that Aule devised for them their tongue in its beginning, and therefore it changes little, whereas the iglishmek which they made for themselves is changeable.' (3) [Cf. p. 398: 'statements that certain names (especially those of the Valar or of places in Valinor) were translations of the Valarin forms'] Arda Q arda (< *garda, S gardh) meant any more or less bounded or defined place, a region. Its use as a proper name for the World was due to V Aparaphelun. Aratar 'the Supreme', was a version of the V maxanaz, pl. maxanumaz 'Authorities', also adapted as Mahan, pl. Mahani. Ea 'All Creation', meaning 'it is', or 'let it be'. Valarin not recorded. Ambar 'the Earth', meaning 'habitation'. Though the Eldar often used Arda in much the same sense, the proper meaning of Ambar was the Earth only, as the place where the Aratar had taken up their dwelling, and the Incarnate were destined to appear.(29) Eru 'the One'. Iluvatar was, however, a name made by the Eldar (when they had learned of Eru from the Valar), which they used more often than Eru, reserved for the most solemn occasions. It was made from iluve 'allness, the all', an equivalent of Ea, and atar 'father'. Varda 'the Sublime'. V form not given. Melkor 'He who arises in Might', oldest Q form *mbelekoro. V form not given. Namo 'Judge'; usually called by the Eldar Mandos, the place of his dwelling. Irmo 'Desirer'; usually called by the name of his dwelling Lorien. Este 'Repose'. (*SED: CE *esde > *ezde, Q Este, .T Ede (as names only); S idh 'rest, repose'.) Vala 'has power' (sc. over the matter of Ea), 'a Power'; pl. Valar, 'they have power, the Powers'. Since these words are from the point of Q structure verbal in origin, they were probably versions of V words of verbal meaning. Cf. axan (p. 399), Ea; and also Q eques. Atan, pl. Atani 'Men', meaning 'the Second, those coming next'. The Valar called them in full 'the Second Children of Eru', but the Quendi were 'the first Children of Eru'. From these terms the Q Minnonar 'First-born' and Apanonar 'After-born' were imitated; but Q Eruhin, pl. Eruhini 'Children of Eru', or 'Elves and Men', is a translation of the Valarin expression 'Children of Eru' (of which the actual Valarin form is not recorded, probably because the V equivalent of Eru is nowhere revealed). Besides the form -hin, -hini only used in composition after a parental name, Q has hina 'child', and hina only used in the vocative addressing a (young) child, especially in hinya (< hinanya) 'my child'. S has hen, pl. hin, mostly used as a prefix in patronymics or metronymics: as Hin Hurin 'The Children of Hurin'. These words are derivatives of stem *khin: khina (in composition khina > Q -hin), and khina. Kalakiryan 'the Cleft of Light', the pass in the Pelori not far from the north side of Taniquetil through which the Light of the Trees in Valinor flowed out to the shores of Aman. Taniquetil, the highest of the mountains of the Pelori, upon which were the mansions of Manwe and Varda. The name was properly only that of the topmost peak, meaning High-Snow- Peak. The whole mountain was most often called by the Eldar (Oron) Oiolosse, '(Mount) Everwhite' or 'Eversnow'. There were many names for this mountain in Quenya. A variant or close equivalent of Taniquetil was Arfanyaras(se). The Sindarin forms of the names were made by the Noldor, for the Sindar knew nothing of the land of Aman except by report of the Exiles: e.g. Amon-Uilos and Ras-Arphain. Pelori 'the fencing, or defensive Heights'. The mountains of Aman, ranging in a crescent from North to South, close to the western shores. On this list Pengolodh comments: 'These are all that I can find in old lore or remember to have read or heard. But the list is plainly incomplete. Many of the names once known and used, whether they be now found in the surviving histories or passed over, must have belonged to the first or the last group. Among those that are still remembered I note Avathar, the name of the dim and narrow land between the southern Pelori and the Sea in which Ungoliante housed. This is not Elvish. There are also the names Nessa, the spouse of Tulkas, and Uinen the spouse of Osse. These too are not Elvish, so far as can now be seen; and since the names Tulkas and Osse come from Valarin, the names of their spouses may also represent titles in the Valarin tongue, or such part of them as the Eldar could adapt. I say "so far as can now be seen", for there is no certainty in this matter without record. It is clear that some, or indeed many, of these adoptions and translations were made in very early days, when the language of the Eldar was otherwise than it became before the Exile. In the long years, owing to the restlessness and inventiveness of the Eldar (and of the Noldor in particular), words have been set aside and new words made; but the names of the enduring have endured, as memorials of the speech of the past. There is also this to consider. When words of Elvish tongue had been used to make the names of things and persons high and admirable, they seem to have been felt no longer suitable to apply to lesser things, and so passed from the daily speech. 'Thus we see that vala is no longer used of any power or authority less than that of the Valar themselves. One may say A vala Manwe! "may Manwe order it!"; or Valar valuvar "the will of the Valar will be done"; but we do not say this of any lesser name. In like manner Este or Ede is the name only of the spouse of Lorien, whereas the form that that word has in Sindarin (idh) means "rest", such as even a tired hound may find before a fire.' (Note 36, p. 416) The reasons that Pengolodh gives or surmises for the scanty knowledge of Valarin preserved in Noldorin lore are here summarized. Some have already been alluded to. Though Valarin had many more sounds than Eldarin, some alien to the Eldarin style and system, this only imposed any real difficulty upon the borrowing of words and their adaptation to Eldarin. To learn Valarin was probably not beyond the powers of the Eldar, if they had felt the need or desire to do so; references to the difficulty of Valarin are mainly due to the fact that for most of the Eldar learning it was an ungrateful and profitless task. For the Eldar had no need to learn the language of Valinor for the purposes of communication; and they had no desire either to abandon or to alter their own tongue, which they loved and of which they were proud. Only those among them, therefore, who had special linguistic curiosity desired to learn Valarin for its own sake. Such 'loremasters' did not always record their knowledge, and many of the records that were made have been lost. Feanor, who probably knew more of the matter than any of the younger generations born in Aman, deliberately withheld his knowledge. It was probably only in the very early days that the Eldar heard Valarin much spoken, or had opportunity for learning it, unless by special individual effort. The Teleri had little immedi- ate contact with the Valar and Maiar after their settlement on the shores. The Noldor became more and more engrossed with their own pursuits. Only the Vanyar remained in constant association with the Valar. And in any case the Valar appear quickly to have adopted Quenya. All the orders of Eru's creatures have each some special talent, which higher orders may admire. It was the special talent of the Incarnate, who lived by necessary union of hroa and fea, to make language. The Quendi, first and chief of the Incarnate, had (or so they held) the greatest talent for the making of lambe. The Valar and Maiar admired and took delight in the Eldarin lambe, as they did in many other of the skilled and delicate works of the Eldar. The Valar, therefore, learned Quenya by their own choice, for pleasure as well as for communication; and it seems clear that they preferred that the Eldar should make new words of their own style, or should translate the meanings of names into fair Eldarin forms, rather than [that] they should retain the Valarin words or adapt them to Quenya (a process that in most cases did justice to neither tongue). Soon after the coming of the Vanyar and Noldor the Valar ceased to speak in their own tongue in the presence of the Eldar, save rarely: as for instance in the great Councils, at which the Eldar were sometimes present. Indeed, it is said that often the Valar and Maiar might be heard speaking Quenya among themselves. In any case, to speak of the early days of the settlement at Tirion, it was far easier and swifter for the Valar to learn Quenya than to teach the Eldar Valarin. For in a sense no lambe was 'alien' to the Self-incarnate. Even when using bodily forms they had less need of any tengwesta than had the Incarnate; and they had made a lambe for the pleasure of exercising the powers and skills of the bodily form, and (more remotely) for the better understanding of the minds of the Incarnate when they should appear, rather than for any need that they felt among them- selves. For the Valar and Maiar could transmit and receive thought directly (by the will of both parties) according to their right nature;(30) and though the use of bodily form (albeit assumed and not imposed) in a measure made this mode of communication less swift and precise, they retained this faculty in a degree far surpassing that seen among any of the Incarnate. At this point Pengolodh does not further discuss this matter of the transmission and reception of thought, and its limitations in any order of creatures. But he cites, as an example of the speed with which by its aid a tengwesta may be learned by a higher order, the story of the Finding of the Edain. According to this the Noldorin king, Finrod, quickly learned the tongue of the folk of Beor whom he discovered in Ossiriand, for he understood in large measure what they meant while they spoke. 'Now Finrod,' he says, 'was renowned among the Eldar for this power which he had, because of the warmth of his heart and his desire to understand others; yet his power was no greater than that of the least of the Maiar.'(31) Pengolodh concludes as follows. 'In the histories the Valar are always presented as speaking Quenya in all circumstances. (Note 37, p. 417) But this cannot proceed from translation by the Eldar, few of whose historians knew Valarin. The transla- tion must have been made by the Valar or Maiar themselves. Indeed those histories or legends that deal with times before the awaking of the Quendi, or with the uttermost past, or with things that the Eldar could not have known, must have been presented from the first in Quenya by the Valar or the Maiar when they instructed the Eldar. Moreover this translation must have concerned more than the mere words of language. If we consider the First History, which is called the Ainulindale: this must have come from the Aratar themselves (for the most part indeed from Manwe, it is believed). Though it was plainly put into its present form by Eldar, and was already in that form when it was recorded by Rumil, it must nonetheless have been from the first presented to us not only in the words of Quenya, but also according to our modes of thought and our imagina- tion of the visible world, in symbols that were intelligible to us. And these things the Valar understood because they had learned our tongue.' Author's Notes to Quendi and Eldar. Note 1 (p. 367; referred to in two passages). Distinguish yomenie 'meeting, gathering' (of three or more coming from different directions). The Telerin form was: el sila lumena vomentienguo. Note 2 (p. 369). It was a later development in Quenya, after the elements -o and -va had become inflexions, applicable to all nouns, to pluralize -o by the addition of the plural sign -n, when added to a plural stem (as by natural function it could be): as lasseo 'of a leaf', lassio > lassion 'of leaves'. Similarly with -va; but this was and remained an adjective, and had the plural form -ve in plural attribution (archaic Q -vai); it could not, however, indicate plurality of source, originally, and the Q distinction Eldava Elf s and Eldaiva Elves þ was a Q innovation. Note 3 (p. 372). roquen is < *roko-kwen with Quenya syncope, *roko being an older simpler form of the stem, found in some compounds and compound names, though the normal form of the independent word 'horse' had the fortified form rokko. These compounds being old were accented as unitary words and the main stress came on the syllable preceding -quen: kirya:quen, kirya:queni. Note 4 (p. 373). That is, elliptically for Quenya lambe, as English for English language. When historians needed a general adjective 'Quen- dian, belonging to the Elves as a whole', they made the new adjective Quenderin (on the model of Eldarin, Noldorin, etc.); but this remained a learned word. Note 5 (p. 375). This change took place far back in Elvish linguistic history; possibly before the Separation. It is in any case common to the Telerin of Aman, Sindarin, and Nandorin. Note 6 (p. 375). The Noldorin Loremasters record that Pendi was used by the Teleri only of the earliest days, because they felt that it meant the lacking, the poor (*PEN), with reference to the indigence and ignorance of the primitive Elves. Note 7 (p. 377). The Dwarves were in a special position. They claimed to have known Beleriand before even the Eldar first came there; and there do appear to have been small groups dwelling furtively in the highlands west of Sirion from a very early date: they attacked and waylaid the Elves by stealth, and the Elves did not at first recognize them as Incarnates, but thought them to be some kind of cunning animal, and hunted them. By their own account they were fugitives, driven into the wilderness by their own kin further east, and later they were called the Noegyth Nibin (32) or Petty-dwarves, for they had become smaller than the norm of their kind, and filled with hate for all other creatures. When the Elves met the powerful Dwarves of Nogrod and Belegost, in the eastern side of the Mountains, they recognized them as Incarnates, for they had skill in many crafts, and learned the Elvish speech readily for purposes of traffic. At first the Elves were in doubt concerning them, believing them to be related to Orcs and creatures of Morgoth; but when they found that, though proud and unfriendly, they could be trusted to keep any treaties that they made, and did not molest those who left them in peace, they traded with them and let them come and go as they would. They no longer classed them as Moerbin, but neither did they ever reckon them as Celbin, calling them the Dornhoth ('the thrawn folk') or the Naugrim ('the stunted people'). [See further on the Petty-dwarves pp. 388 - 9.] Note 8 (p. 377). Though Morben might still be applied to them by any who remained hostile to Men (as were the people of Doriath for the most part); but this was intended to be insulting. Note 9 (p. 377; referred to in two passages). The implication that as opposed to Celbin the Moerbin were allies of Morgoth, or at least of dubious loyalty, was, however, untrue with regard to the Avari. No Elf of any kind ever sided with Morgoth of free will, though under torture or the stress of great fear, or deluded by lies, they might obey his commands: but this applied also to Celbin. The 'Dark-elves', however, often were hostile, and even treacherous, in their dealings with the Sindar and Noldor; and if they fought, as they did when themselves assailed by the Orcs, they never took any open part in the War on the side of the Celbin. They were, it seems, filled with an inherited bitterness against the Eldar, whom they regarded as deserters of their kin, and in Beleriand this feeling was increased by envy (especially of the Amanyar), and by resentment of their lordliness. The belief of the Celbin that, at the least, they were weaker in resistance to the pressures or lies of Morgoth, if this grievance was concerned, may have been justified; but the only case recorded in the histories is that of Maeglin, the son of Eol. Eol was a Mornedhel, and is said to have belonged to the Second Clan (whose representatives among the Eldar were the Noldor).(33) He dwelt in East Beleriand not far from the borders of Doriath. He had great smith-craft, especially in the making of swords, in which work he surpassed even the Noldor of Aman; and many therefore believed that he used the morgul, the black arts taught by Morgoth. The Noldor themselves had indeed learned much from Morgoth in the days of his captivity in Valinor; but it is more likely that Eol was acquainted with the Dwarves, for in many places the Avari became closer in friendship with that people than the Amanyar or the Sindar. Eol found Irith,(34) the sister of King Turgon, astray in the wild near his dwelling, and he took her to wife by force: a very wicked deed in the eyes of the Eldar. His son Maeglin was later admitted to Gondolin, and given honour as the king's sisterson; but in the end he betrayed Gondolin to Morgoth. Maeglin was indeed an Elf of evil temper and dark mind, and he had a lust and grudge of his own to satisfy; but even so he did what he did only after torment and under a cloud of fear. Some of the Nandor, who were allowed to be Celbin, were not any better. Saeros, a counsellor of King Thingol, who belonged to a small clan of Nandor living in eastern Doriath, was chiefly responsible for the driving into outlawry of Turin son of Hurin. Turin's mother was named Morwen 'dark maiden', because of her dark hair, and it was one of Saeros' worst insults to call her Morben. For that Turin smote him in the king's hall.(35) This resentment on the part of the Avari is illustrated by the history of PQ *kwendi. This word, as has been shown, did not survive in the Telerin languages of Middle-earth, and was almost forgotten even in the Telerin of Aman. But the Lore- masters of later days, when more friendly relations had been established with Avari of various kinds in Eriador and the Vale of Anduin, record that it was frequently to be found in Avarin dialects. These were numerous, and often as widely sundered from one another as they were from the Eldarin forms of Elvish speech,. but wherever the descendants of *kwendi were found, they meant not 'Elves in general', but were the names that the Avari gave to themselves. They had evidently continued to call themselves *kwendi, the People, regarding those who went away as deserters - though according to Eldarin tradition the numbers of the Eldar at the time of the Separation were in the approximate proportion of 3:2, as compared with the Avari (see p. 381). The Avarin forms cited by the Loremasters were: kindi, cuind, hwenti, windan, kinn-lai, penni. The last is interesting as showing the change kw > p. This might be independent of the Common Telerin change; but it suggests that it had already occurred among the Lindar before the Separation. The form penni is cited as coming from the 'Wood-elven' speech of the Vale of Anduin, and these Elves were among the most friendly to the fugitives from Beleriand, and held themselves akin to the remnants of the Sindar. Note 10 (p. 378). It is not surprising that the Edain, when they learned Sindarin, and to a certain extent Quenya also, found it difficult to discern whether words and names containing the element el referred to the stars or to the Elves. This is seen in the name Elendil, which became a favourite name among the Edain, but was meant to bear the sense 'Elf-friend'. Properly in Quenya it meant 'a lover or student of the stars', and was applied to those devoted to astronomical lore. 'Elf-friend' would have been more correctly represented by Quen(den)dil or Eldandil. Note 11 (p. 378). Lake Mithrim, meaning originally 'Lake of the Mithrim'. Mithrim was a name given to them by the southern-dwellers, because of the cooler climate and greyer skies, and the mists of the North. It was probably because the Noldor first came into contact with this northerly branch that they gave in Quenya the name Sindar or Sindeldi 'Grey-elves' to all the Telerin inhabi- tants of the Westlands who spoke the Sindarin language. Though this name was also later held to refer to Elwe's name Thingol (Sindikollo) 'Grey-cloak', since he was acknowledged as high-king of all the land and its peoples. It is said also that the folk of the North were clad much in grey, especially after the return of Morgoth when secrecy became needed; and the Mithrim had an art of weaving a grey cloth that made its wearers almost invisible in shadowy places or in a stony land. This art was later used even in the southern lands as the dangers of the War increased. Note 12 (p. 380). The language of Mithrim was also a marked dialect; but none of the dialects of Sindarin differed widely enough to interfere with intercourse. Their divergences were no greater than those that had arisen between the Quenya as spoken by the Vanyar, and as spoken by the Noldor at the time of the Exile. Note 13 (p. 380). For the late PQ gl- as an initial variation of l- see General Phonology.(37) Though this Clan-name has *glind- in Sindarin, the g- does not appear in Amanya Telerin, nor in Nandorin, so that in this case it may be an addition in Sindarin, which favoured and much increased initial groups of this kind. Note 14 (p. 382). For this reason the most frequently used of the 'titles' or secondary names of the Lindar was Nendili 'Water-lovers'. Note 15 (p. 382). A simple agental formation (like *abaro > *abar from *ABA) from the stem *TELE, the primary sense of which appears to have been 'close, end, come at the end': hence in Q telda 'last, final'; tele- intransitive verb 'finish, end', or 'be the last thing or person in a series or sequence of events'; telya transitive verb 'finish, wind up, conclude'; telma 'a conclusion, anything used to finish off a work or affair'. This was possibly distinct from *tel-u 'roof in, put the crown on a building', seen in Q telume 'roof, canopy'. (This was probably one of the earliest Quendian words for the heavens, the firmament, before the increase of their knowledge, and the invention of the Eldarin word Menel. Cf. Telumehtar 'warrior of the sky', an older name for Menelmakil, Orion.) The word telluma 'dome, cupola' is an alteration of telume under the influence of Valarin delgsima: see p. 399. But *telu may be simply a differentiated form of *TELE, since the roof was the final work of a building; cf. telma, which was often applied to the last item in a structure, such as a coping-stone, or a topmost pinnacle. Note 16 (p. 384). See above, p. 381. The proportion, per 144, of the Eldar remaining in Middle-earth was reckoned at 26, of which about 8 were Nandor. Note 17 (p. 384). Lenwe is the form in which his name was remembered in Noldorin histories. His name was probably *Denwego, Nan- dorin Denweg. His son was the Nandorin chieftain Denethor. These names probably meant 'lithe-and-active' and 'lithe-and- lank', from *dene- 'thin and strong, pliant, lithe', and *thara- 'tall (or long) and slender'. Note 18 (p. 384). The name Nandor was a derivative of the element *dan, *ndan- indicating the reversal of an action, so as to undo or nullify its effect, as in 'undo, go back (the same way), unsay, give back (the same gift: not another in return)'. The original word *ndando, therefore, probably only implied 'one who goes back on his word or decision'. Note 19 (p. 386). In Q Eldameldor, S Elvellyn. That is, 'Elf-lovers'. The words Quendili, Eldandili (see Note 10), though not excluding affection and personal loyalties, would have implied also deep concern with all lore relative to the Elves, which was not necessarily included in the words meldor, mellyn 'lovers, friends'. Note 20 (p. 387). That is, to die by nature, of age or weariness, and inevitably, not only (as the Elves) of some grievous hurt or sorrow. Note 21 (p. 387). S -eb is from *ikwa, CT *-ipa, probably related to the Q -inqua. Cf. S aglareb 'glorious', Q alkarinqua. Both are probably related to the element *kwa, *kwa-ta seen in Eldarin words for 'full'. Note 22 (p. 388). S ch was only an approximation; the Dwarvish kh was in fact a strong aspirate, not a spirant. Similarly at the time of the borrowing Sindarin did not possess either the sound z or long a. This does not mean that the Elves could not imitate or acquire sounds alien to their native speech. All the Elves had great skill in language, and far surpassed Men in this matter. The Noldor were the chief linguists of the Elves, but their superiority was shown not so much in the acquisition of new tongues as in their love of language, their inventiveness, and their concern with the lore of language, and the history and relations of different tongues. In adopting a word for use in their own tongue (which they loved) Elves fitted it to their own style for aesthetic reasons. Note 23 (p. 388) These words are derived from the stem *NUKU 'dwarf, stunted, not reaching full growth or achievement, failing of some mark or Standard', seen in *nukta-, Q nuhta- 'stunt, prevent from coming to completion, stop short, not allow to continue', S nuitha- of similar senses. An adjectival formation was *nauka, from which were derived S naug, Q nauka, especially applied to things that though in themselves full-grown were smaller or shorter than their kind, and were hard, twisted or ill-shapen. Nogoth is probably from some such form as *nukotto/a 'a stunted or ill-shapen thing (or person)'. Note 24 (p. 388). The Q h had become too weak to represent aspirate kh which was therefore rendered by k. Final d had become r, and this change was recognized in the adaptation. Medial z < s had become r in the Noldorin dialect of Q except when an adjacent syllable, or (as here) the same syllable, already contained an r. Note 25 (p. 388). Norno is a personalized form of the adjective norna 'stiff, tough', the Q equivalent of S dorn. Both are from the stem *DORO 'dried up, hard, unyielding'. With the frequent initial enrichment d > nd this appears in PQ *ndore 'the hard, dry land as opposed to water or bog > land in general as opposed to sea; a land (a particular region with more or less defined bounds)'. Hence S dor (-ndor > -nor, -nnor) 'land'. In Q this word became confused or blended with the distinct *nore from the stem *ONO (see p. 387), family, tribe or group having a common ancestry, the land or region in which they dwelt'. Thus Q nore was generally used for 'land' associated with a particular people, and the old *ndore survived only in name- compounds: as Valinore < *Valinore 'the people and land of the Valar', beside Valinor, Valandor. A particular land or region was in Q arda,. land as opposed to water or sea was nor (< *ndoro) as opposed to ear. The Q forms norna, Norno may also contain nd-, though S dorn does not; but this is probably one of the cases in which Q initial d became n-, not l-, by assimilation to an n occurring later in the word. Note 26 (p. 389). Novrod was the oldest form, and appears in the earlier annals, beside the variant Grodnof. These contain the CE elements *naba 'hollow', and (g)rota 'excavation, underground dwel- ling'. Novrod retains the older Eldarin (and the Dwarvish) order with the adjectival element first. At the time of its making *naba-grota had no doubt already reached its archaic S form *nov-3rot > novrod. Grodnof has the same elements in the later more usual Sindarin order. The form Nogrod which later became usual is due to the substitution of Nog-, taken as a form of Naug 'dwarf' (with the usual change of au > o), after the element Nov- had become obscure. The adjective *naba > nov, nof only remained current in the Northern dialect, where the name Novrod originated. In the other dialects nov, as a stressed independent word, proceeded to nauv > naw (with the usual loss of final v after au, u), and this word ceased to be used in current speech. Novrod in earlier annals is sometimes found glossed Bar-goll 'hollow dwelling', using the more current adjective coll < *kulda. Hadhodrond uses the adapted form Hadhod = Khazad. The element rond is not related to grod, -rod. The latter is from *groto 'dig, excavate, tunnel'. S rond, Q rondo are from *rono 'arch over, roof in'. This could be applied both to natural and to artificial structures, but its view was always from below and from the inside. (Contrast the derivatives of *tel, *telu men- tioned in Note 15.) CE *rondo meant 'a vaulted or arched roof, as seen from below (and usually not visible from outside)', or 'a (large) hall or chamber so roofed'. It was still often applied pictorially to the heavens after the Elves had obtained much greater knowledge of 'Star-lore'. Cf. the name Elrond 'Star- dome' (Elros meant 'Star-glitter'). Cf. also S othrond applied to an underground stronghold, made or enlarged by excavation, containing one or more of such great vaulted halls. othrond is < S ost+ rond. CE *osto, Q osto, S ost, is derived from *soto 'shelter, protect, defend', and was applied to any fortress or stronghold made or strengthened by art. The most famous example, after the great dwelling of Elwe at Menegroth, was Nargothrond < Narog-ost-rond ('the great underground burg and halls upon the River Narog'), which was made by Finrod, or completed and enlarged by him from the more primitive dwellings made by the Petty-dwarves. Though distinct in origin the derivatives of *groto and *rono naturally came into contact, since they were not dissimilar in shape, and a 'rondo was usually made by excavation. Thus S groth < *grotta (an intensified form of grod < *grota) 'a large excavation' might well apply to a rond. Menegroth means 'the Thousand Caves or Delvings', but it contained one great rond and many minor ones. Note 27 (p. 390). *(n)guruk is due to a combination of *(g)ruk with *NGUR 'horror', seen in S gorth, gorthob 'horror, horrible', and (redupli- cated) gorgor 'extreme horror'. Note 28 (p. 390) Some other derivatives are in Quenya: rukin 'I feel fear or horror' (constructed with 'from' of the object feared); ruhta- 'terrify'; rukima 'terrible'; rauko and arauko < *grauk-) 'a powerful, hostile, and terrible creature', especially in the compound Valarauko 'Demon of Might', applied later to the more powerful and terrible of the Maia servants of Morgoth. In Sindarin appear, for instance, raug and graug, and the com- pound Balrog (equivalents of Q rauko, etc.); groga- 'feel terror'; gruitha 'terrify'; gorog (< *guruk) 'horror'. Note 29 (p. 392). Affixes appear in equen 'said I', eques 'said he I she', used in reporting a dialogue. Note 30 (p. 392). *ekwe was probably a primitive past tense, marked as such by the 'augment' or reduplicated base-vowel, and the long stem- vowel. Past tenses of this form were usual in Sindarin 'strong' or primary verbs: as *akara 'made, did' > S agor. *akwa, however, was probably not verbal, but an extension or intensification of *kwa, used adverbially. Note 31 (p. 392). In Eldarin languages this is usually found in the forms -ikwa or -ukwa, or with nasal infixion -inkwa, -unkwa. The vowels i, u were probably derived from the terminations of nouns or other stems to which kwa was added, but the dissyllabic suffixal forms had become quite independent of this origin. The forms using u were mainly applied to things heavy, clumsy, ugly or bad. Note 32 (p. 393). Little is said in Noldorin lore concerning the language of the Valar and Maiar; but on this point a note is added at the end of this Appendix (pp. 397 ff.). Note 33 (p. 394). lamba is derived from *LABA 'move the tongue, lick', and may be referred to *lab-ma (with a suffix frequent in the names of implements): the group bm > mb in CE and possibly earlier. lambe is probably from *lab-me, denoting the action of *LABA, or the use of the *lamba. (Cf. *JULU 'drink', *julma, Q yulma, S ylf 'drinking-vessel'; *julme, Q yulme, 'drinking, carousal'.) These words have no original connexion with *LAMA which refers to sounds, especially to vocal sounds, but was applied only to those that were confused or inarticulate. It was generally used to describe the various cries of beasts. Hence the word *laman(a), *laman, Q laman, pl. lamni or lamani; S lavan, pl. levain, 'animal', usually only applied to four-footed beasts, and never to reptiles or birds. (This may be compared with *kwene 'user of articulate speech'.) The Sindarin glam < glamb/glamm (p. 391) is an elaboration of *LAM. Note 34 (p. 394). In genuine independent use mainly employed between persons out of earshot: the Elves had astonishingly acute eyesight at a distance. These 'signals' were really distinct from the gestures (especially those of the hands) made as concomitants to speech and additions to tone-changes for the conveyance of feeling, though some of the gestures in both systems were similar. The Elves made considerable use of the concomitant gestures, especially in oration or recitation. Note 35 (p. 398). By which Pengolodh meant the knowledge available in Middle- earth. The Lammas was composed in Eriador. Note 36 (p. 404). Other later Loremasters conjectured that Nessa was in fact Elvish in form (though archaic, on Pengolodh's own principle), being < *neresa, a feminine adjectival formation from *NER, meaning 'she that has manlike valour or strength'. They also would remove Taniquetil from the group of 'translations'. Arfanyarasse, they say, is the translation: 'high (i.e. noble, revered) - shining white - peak', but Taniquetil is an adapta- tion, though one that has probably greatly altered the original in the attempt to give the name some kind of Eldarin sig- nificance: ? high white point. As they say, ta- does not mean 'lofty' in Eldarin, though it may remind one of tara 'tall, high' (*TAR); nique does not refer to snow, but to cold; and Q tilde, -til is not a mountain peak, but a fine sharp point (mostly used of small and slender things). For nique cf. Q niku- 'be chill, cold (of weather)'; nique 'it is cold, it freezes'; ninque 'chill, pallid', nixe 'frost', niquis, niquesse 'frost-patterns' (the latter by association with quesse 'feather'). Most significant, they cite from an ancient legend of the Flight the tale that as the mists of Araman wrapped the distant mountains of Valinor from the sight of the Noldor, Feanor raised his hands in token of rejection and cried: 'I go. Neither in light or shadow will I look upon you again, Dahanigwishtil- gun.' So it was recorded, though the writers of the histories no longer knew what he meant. For which reason the strange word may have been ill transmitted. But even so it still bears some likeness to Taniquetil, though it can no longer be analysed. (In a few versions, say the Loremasters, it is written dahan- igwis-telgun.) They also cite Fionwe [read Eonwe?] (the herald of Manwe) as another name for which no Elvish etymology is known. Note 37 (p. 406). Usually in a formal and elevated style. Often, when there were differences, rather according to the Vanyarin manner than the Noldorin, for the Vanyar were most in their company; though the Noldorin writers have sometimes substituted their own forms. Editorial Notes. 1. 'affection': mutation (of the vowel o caused by the following i in Mori(quendi) ). 2. sundoma: see p. 319. 3. omentielvo: this was typed omentielmo, subsequently changed to omentielvo. The same change was made in the Second Edition of The Fellowship of the Ring (p. 90). 4. The Fellowship of the Ring p. 367 (at the end of the chapter Lothlorien); First Edition vanimalda, Second Edition vanimelda. 5. The term omataina or 'vocalic extension' is used of the addition to the 'base' of a final vowel identical to the sundoma (p. 319). 6. 'The glooms and the clouds dimming the sun and the stars': an explicit reference, it seems, to some form of the changed astronomical myth adumbrated in Text II of the section 'Myths Transformed' in Morgoth's Ring. In that text my father raised the question 'how can the Eldar be called the "Star-folk"?' if the Sun is 'coeval with the Earth' (X.375); and proposed a complex story (X.377 - 8) in which the darkening of the world by Melkor, who brought up vast glooms to shut out all vision of the heavens, is a chief element. See further pp. 423-4. 7. 'The first people of this kind to be met were the Nandor': this strangely contradicts the history recorded in the Annals (GA $19, p. 9; also AAm $84, X.93), according to which the Dwarves first entered Beleriand in Valian Year 1250, and the building of Menegroth was achieved before the coming of Denethor, leader of the Nandor, in 1350 (pp. 11-13). The following statement here that the first invasions of the Orcs followed Morgoth's return is an equally striking contradiction of the Annals: according to GA $27 Orcs entered Beleriand in 1330 (cf. also X.106, $85): 'Whence they came, or what they were, the Elves knew not then, deeming them to be Avari, maybe, that had become evil and savage in the wild.' 8. 'from Cape Andras to the headland of Bar-in-Myl': Cape Andras was entered on the map (p. 184, square G 2), but the headland to the south (itself an extension of the coastline as originally drawn) is there called Ras Mewrim (p. 190, $63). The name in the present text was typed Bar-in-Gwael; the translation 'Home of the Gulls' was added at the same time as the change to Bar-in-Myl (by a later pencilled change on one copy -in- > -i-). 9. Brithonbar, not Brithombar, is the form typed, and not cor- rected. 10. With this passage on the subject of the Eglain cf. p. 189, $ 57, and pp. 343-4. The concluding sentence 'But they acknowledged the high-kingship of Thingol, and Cirdan never took the title of king' differs from the Annals, where Cirdan either acknowledged Felagund of Nargothrond as overlord, or else was (as it seems) an independent Lord of the Falas 'yet ever close in friendship with Nargothrond' (GA $85, and commentary p. 117). 11. For the legend of Imin, Tata, and Enel see pp. 420 ff. 12. The story found in the Annals of Aman of the kindreds of Morwe and Nurwe, who refused the summons of the Valar and became the Avari (X.81-2, 88, 168), had been abandoned. 13. The name Lindar 'Singers' of the Teleri has appeared in the 'Glossary' to the Athrabeth Finrod ah Andreth (X.349); it was for long the name of the First Kindred, the later Vanyar. 14. On the waterfall of Cuivienen see p. 424. 15. In other late writing Cirdan is said to have been of the kin of Elwe, but I have not found any statement of the nature of the kinship. 16. Lenwe has replaced the long-standing name Dan of Denethor's father; from this text it was adopted in The Silmarillion. 17. The statement that the Nandor entered Beleriand 'not long before the return of Morgoth' is another remarkable contradiction of the Annals (cf. note 7 above). Earlier (p. 377) it is said that they came 'before the return of Morgoth', which no doubt implies the same. But in GA $31 there is a marvellous evocation of 'the long years of peace that followed after the coming of Denethor', and they were indeed long: from 1350 to 1495, 145 Valian Years, or 1389 Years of the Sun. I am at a loss to explain these profound changes in the embedded history. 18. On the Adunaic word Nimir 'Elf' see The Drowning of Anadune (Vol. IX, Index II, p. 473). 19. Firimar: the old form was Firimor (QS $83, V.245, footnote). An account of the development of meaning in the verb fire is given in connection with Firiel, the later name of Miriel, in X.250. 20. The name Nogoth niben was adopted in The Silmarillion (in the plural, Noegyth nibin: see Note 7 to the present text, p. 408); the word nogoth of the Dwarves has not occurred before (see note 32 below). For other names and name-forms of the Petty-dwarves see p. 187, $26. 21. In the revision of the QS chapter on the Dwarves the Sindarin name of Khazad-dum was Nornhabar, translated 'Dwarrowdelf' (p. 206). 'Dwarrowdelf' is found also in The Fellowship of the Ring; in the present text the Sindarin name was typed Hadhodrud and translated 'Dwarrowmine', but the change to Hadhodrond 'Dwarrowvault' was made immediately. Hadhodrond was adopted in The Silmarillion. 22. Cf. Appendix F to The Lord of the Rings, p. 409: 'The lesser kinds were called, especially by the Uruk-hai, snaga "slave".' 23. Feanor held that, in spite of the usual mode of spelling, vowels were each independent tengwi or word-building elements. 24. On one copy only a later pencilled correction changed *SAR to *SYAR. 25. At the head of the page is a pencilled note on one copy only: 'Change Pengolodh to Thingodhel'. 26. For the word equessi see p. 392. Both in that passage and in the present one the word was typed Equeri and then corrected. 27. For the old conception in the Lhammas of the 1930s, according to which the origin of all Elvish speech was in the language of the Valar (communicated to the Elves by Orome), see V.168, 192-3. 28. In The Road Goes Ever On, p. 61, the name miruvore (occurring in Namarie) is said to be of Valarin origin. 29. Cf. Note 2 on the Commentary on the Athrabeth Finrod ah Andreth (X.337), where it is said that 'Physically Arda was what we should call the Solar System', and that in Elvish traditions 'the principal part of Arda was the Earth (Imbar "the Habitation") ... so that loosely used Arda often seems to mean the Earth'. For Ambar see the references given in X.359, note 12. 30. Cf. AAm $164 (X.129): without voices in silence [the gods] may hold council one with another', and the passage cited from The Return of the King in my note on that passage (X.135). 31. Cf. the late QS chapter Of the Coming of Men into the West, p. 217: 'Felagund discovered ... that he could read in the minds of Men such thoughts as they wished to reveal in speech, so that their words were easily interpreted.' 32. Noegyth Nibin was a correction of the name typed, Nibinn..g, probably Nibinnoeg (see p. 187, $26). The notes being inter- spersed in the text, this note was written before the passage on p. 388 was reached. 33. It is curious that - as in the original text of Maeglin, where he was 'of the kin of Thingol' - in my father's very late work on the story Eol becomes again 'one of the Eldar' (p. 328), though consumed with hatred of the Noldor; whereas here he is a Mornedhel (one of the Avari), and moreover of the aboriginal Second Clan. 34. The name frith is found as a correction (made after the publi- cation of The Lord of the Rings) of the old name Isfin in QS $42 (X.177). When my father worked on the Maeglin story c.1970 he appears to have forgotten frith, for his notes at that time express dissatisfaction with the 'meaningless' name Isfin as if it had never been replaced (pp. 317 - 18). 35. Saeros' insulting of Turin by calling his mother Morwen Morben was a development in the story (see QS $39, V.321, and Un- finished Tales p. 80) that could only arise, of course, with the emergence of the words Calben and Morben. 36. Neither the interpretation of Mithrim as the name of a people (for the old etymology see V.383 - 4, stem RINGI) nor this explanation of the name Sindar have been met before. 37. 'General Phonology': my father was not here referring to any specific, completed work. APPENDIX. The legend of the Awaking of the Quendi (Cuivienyarna). It is said in Quendi and Eldar, p. 380: According to the legend, preserved in almost identical form among both the Elves of Aman and the Sindar, the Three Clans were in the beginning derived from the three Elf-fathers: Imin, Tata, and Enel (sc. One, Two, Three), and those whom each chose to join his following. So they had at first simply the names Minyar 'Firsts', Tatyar 'Seconds', and Nelyar 'Thirds'. These numbered, out of the original 144 Elves that first awoke, 14, 56, and 74; and these pro- portions were approximately maintained until the Separation. A form of this legend is found in a single typescript with carbon copy. On one copy my father wrote (and similarly but mare briefly on the other): 'Actually written (in style and simple notions) to be a surviving Elvish "fairytale" or child s tale, mingled with counting-lore . Correc- tions to either copy are taken up in the text that follows. While their first bodies were being made from the 'flesh of Arda' the Quendi slept 'in the womb of the Earth', beneath the green sward, and awoke when they were full-grown. But the First Elves (also called the Unbegotten, or the Eru-begotten) did not all wake together. Eru had so ordained that each should lie beside his or her 'destined spouse'. But three Elves awoke first of all, and they were elf-men, for elf-men are more strong in body and more eager and adventurous in strange places. These three Elf-fathers are named in the ancient tales Imin, Tata, and Enel. They awoke in that order, but with little time between each; and from them, say the Eldar, the words for one, two, and three were made: the oldest of all numerals.* Imin, Tata and Enel awoke before their spouses, and the first thing that they saw was the stars, for they woke in the early twilight before dawn. And the next thing they saw was their destined spouses lying asleep on the green sward beside them. Then they were so enamoured of their beauty that their desire for speech was immediately quickened and they began to 'think of words' to speak and sing in. And being impatient they could not wait but woke up their spouses. Thus, the Eldar say, the first thing that each elf-woman saw was her spouse, and her love for him was her first love; and her love and reverence for the wonders of Arda came later. Now after a time, when they had dwelt together a little, and had devised many words, Imin and Iminye, Tata and Tatie, Enel and Enelye walked together, and left the green dell of their waking, and they came soon to another larger dell and found there six pairs of Quendi, and the stars were again shining in the morrow-dim and the elf-men were just waking. Then Imin claimed to be the eldest and to have the right of (* [footnote to the text] The Eldarin words referred to are Min, Atta (or Tata), Nel. The reverse is probably historical. The Three had no names until they had developed language, and were given (or took) names after they had devised numerals (or at least the first twelve).) first choice; and he said: 'I choose these twelve to be my companions.' And the elf-men woke their spouses, and when the eighteen Elves had dwelt together a little and had learned many words and devised more, they walked on together, and soon in another even deeper and wider hollow they found nine pairs of Quendi, and the elf-men had just waked in the starlight. Then Tata claimed the right of second choice, and he said: 'I choose these eighteen to be my companions.' Then again the elf-men woke their spouses, and they dwelt and spoke together, and devised many new sounds and longer words; and then the thirty-six walked abroad together, until they came to a grove of birches by a stream, and there they found twelve pairs of Quendi, and the elf-men likewise were just standing up, and looking at the stars through the birch boughs. Then Enel claimed the right of third choice, and he said: 'I choose these twenty-four to be my companions.' Again the elf-men woke their spouses; and for many days the sixty Elves dwelt by the stream, and soon they began to make verse and song to the music of the water. At length they all set out together again. But Imin noticed that each time they had found more Quendi than before, and he thought to himself: 'I have only twelve companions (although I am the eldest); I will take a later choice.' Soon they came to a sweet-smelling firwood on a hill-side, and there they found eighteen pairs of Quendi, and all were still sleeping. It was still night and clouds were in the sky. But before dawn a wind came, and roused the elf-men, and they woke and were amazed at the stars; for all the clouds were blown away and the stars were bright from east to west. And for a long time the eighteen new Quendi took no heed of the others, but looked at the lights of Menel. But when at last they turned their eyes back to earth they beheld their spouses and woke them to look at the stars, crying to them elen, elen! And so the stars got their name. Now Imin said: 'I will not choose again yet'; and Tata, therefore, chose these thirty-six to be his companions; and they were tall and dark-haired and strong like fir-trees, and from them most of the Noldor later were sprung. And the ninety-six Quendi now spoke together, and the newly-waked devised many new and beautiful words, and many cunning artifices of speech; and they laughed, and danced upon the hill-side, until at last they desired to find more companions. Then they all set out again together, until they came to a lake dark in the twilight; and there was a great cliff about it upon the east-side, and a waterfall came down from the height, and the stars glittered on the foam. But the elf-men were already bathing in the waterfall, and they had waked their spouses. There were twenty-four pairs; but as yet they had no formed speech, though they sang sweetly and their voices echoed in the stone, mingling with the rush of the falls. But again Imin withheld his choice, thinking 'next time it will be a great company . Therefore Enel said, I have the choice, and I choose these forty-eight to be my companions.' And the hundred and forty-four Quendi dwelt long together by the lake, until they all became of one mind and speech, and were glad. At length Imin said: 'It is time now that we should go on and seek more companions.' But most of the others were content. So Imin and Iminye and their twelve companions set out, and they walked long by day and by twilight in the country about the lake, near which all the Quendi had awakened - for which reason it is called Cuivienen. But they never found any more companions, for the tale of the First Elves was complete. And so it was that the Quendi ever after reckoned in twelves, and that 144 was for long their highest number, so that in none of their later tongues was there any common name for a greater number. And so also it came about that the 'Companions of Imin' or the Eldest Company (of whom came the Vanyar) were nonetheless only fourteen in all, and the smallest company; and the 'Companions of Tata' (of whom came the Noldor) were fifty-six in all; but the 'Companions of Enel' although the Youngest Company were the largest; from them came the Teleri (or Lindar), and they were in the beginning seventy-four in all. Now the Quendi loved all of Arda that they had yet seen, and green things that grew and the sun of summer were their delight; but nonetheless they were ever moved most in heart by the Stars, and the hours of twilight in clear weather, at 'morrow- dim' and at 'even-dim', were the times of their greatest joy. For in those hours in the spring of the year they had first awakened to life in Arda. But the Lindar, above all the other Quendi, from their beginning were most in love with water, and sang before they could speak. It seems that my father had resolved (at least for the purpose of this 'fairy-tale') the problem of the name 'Star-folk' of the Elves (see late night under skies of unclouded stars, and the stars were their earliest memory. In Quendi and Eldar (p. 382) my father wrote of 'the lake and waterfall of Cuivienen', and this is explained in the Cuivienyarna: 'they came to a lake dark in the twilight; and there was a great cliff about it upon the east-side, and a waterfall came down from the height, and the stars glittered on the foam.' Through so many years he was returning to Gilfanon's Tale in The Book of Lost Tales (1.232): Now the places about Koivie-neni the Waters of Awakening are rugged and full of mighty rocks, and the stream that feeds that water falls therein down a deep cleft... a pale and slender thread, but the issue of the dark lake was beneath the earth into many endless caverns falling ever more deeply into the bosom of the world.