A root in the Qenya Lexicon of the 1910s glossed “steal, take; get by stealth, unlawfully”, with derivatives like ᴱQ. nak (naq-) “anything stolen, a theft, a trick” and ᴱQ. naqar “thief” (QL/64). There are no signs of this root in the contemporaneous Gnomish Lexicon, but words like G. nig- “steal, creep, do or go by stealth” and G. nigla- “to thieve, pilfer” might be derived from a variant (GL/60); there is otherwise no ✱ᴱ√NIKI root appearing in the Qenya Lexicon. There are no signs of either of these roots in Tolkien’s later writing with these meaning; later √NIK means “small” (VT47/26; VT48/18).
Early Primitive Elvish
stainá
adjective. stainá
stalga
adjective. stalga
stalgond-
noun. stalgond-
stalqe
noun. stalqe
stakta-
verb. to split
stanka
adjective. forked, split, bifurcated
stor’onturá
masculine name. King of Eagles
hulu
root. strip
naqa
root. steal, take; get by stealth, unlawfully
atefe-
noun. strong dislike
goþo
root. strive; angry
kaẇa
root. stoop
oro
root. steepness, rising
raha
root. stretch forward
tegna
adjective. straight
tálı̯èndǝ
noun. stunt or jest
ṇðṇ
root. stretch, extend
aða
root. stretch, extend
fulu
root. strip
nřn
root. stretch, extend
ðoro
root. sit
A root in the Qenya Lexicon of the 1910s glossed “sit” with derivatives mean “sit” or “seat” (QL/85-86). In the Qenya Lexicon Tolkien gave the root as SORO with ÐORO in parenthesis, but Gnomish cognates like G. dorn “seat” and dortha- “to settle” (GL/30) make it clear ÐORO was the true form of the root. There were a variety of different roots for “sit” in later writings such ᴹ√KHAM or √KHAD.
kereke
root. *spit
A root appearing in the Gnomish Lexicon of the 1910s with variants ᴱ√kerek- and ᴱ√kereχ- and derivatives like G. crech “spittle”, G. crectha- “spit”, G. agrectha- “despise”, and G. agrectharol “despicable” (GL/27). The first of these derivatives reappeared in Early Noldorin of the 1920s as a derivative of ✶kǝ̀rekka along with the Qenya form ᴱQ. rekka (PE13/140). This Qenya form, and the meaning of the Gnomish words from the 1910s, resembles the derivatives of the root ᴱ√RETYE from the Qenya Lexicon of the 1910s such as ᴱQ. retye- “spit” and ᴱQ. retima “despicable” (QL/79). Thus ᴱ√KEREKE may be variant or replacement for ᴱ√RETYE.
In The Etymologies of the 1930s there is a later root ᴹ√PIW “spit” which may have supplanted ᴱ√K(E)REKE, but the derivatives of ᴹ√PIW are only verbal and I think it worth positing the existence of a Neo-Eldarin noun root ᴺ√K(E)REK “spittle” to salvage the Gnomish words, especially those having to do with contempt.
sḷtḷ
root. *sift
The unglossed root ᴱ√SḶTḶ appeared in the Qenya Lexicon of the 1910s with derivatives like ᴱQ. silt- “sift, sort out, winnow” and ᴱQ. silta “sieve” (QL/84). Gnomish words G. thlid “sieve” and G. thlid- “to sort out, sift, sieve, discriminate” from contemporaneous Gnomish Lexicon are clearly related, and the archaic past form salti (later thlinti) confirms the root was actually ᴱ√SḶTḶ (GL/73). A last hint of this root can be seen in ᴱQ. silte “sift” < ᴱ✶sḷt- with past salsie, a verb form appearing in the Early Qenya Grammar of the 1920s (PE14/58).
Neo-Eldarin: I think it is worth positing a Neo-Root ᴺ√SILIT “sift” to salvage some of these early words.
foʒo
root. hide, hoard, store up, lay up in secret
A root in the Qenya Lexicon of the 1910s glossed “hide, hoard, store up, lay up in secret” with derivatives like ᴱQ. foina “hidden, secret” and ᴱQ. foa “hoard, treasure” (QL/38). Gnomish derivatives also appear in the contemporaneous Gnomish Lexicon such as G. fuin “secret, dark; hoarded” and G. fuis “hoard”, though there the root was given as foχo (GL/36); this is not a contradiction, since in QL the ancient voiceless spirant χ was represented with H in roots. One of the Qenya derivatives of this root, ᴱQ. foalóke “✱hoard dragon” reappeared much later as Q. foalóke with a new etymology where the first element meant “breath” (PE17/181). Thus this early root was likely abandoned.
pḷtyḷ
root. *strike, flat of hand
An unglossed root in the Qenya Lexicon of the 1910s with derivatives like ᴱQ. paltya “flat of hand” and ᴱQ. pilty- “strike, esp. with flat hand” (QL/74). In later writings words for the flat of the hand were derived from √PAL: Q. palta and S. plad, probably a partial restoration of this early root (VT47/8).
qiði
root. *stir, spin
An unglossed root in the Qenya Lexicon of the 1910s given as √QIŘI [QIÐI] having derivatives like ᴱQ. qiri- “stir (make spin)”, ᴱQ. qindl “a spindle, top”, and ᴱQ. qirme “cream” (QL/77), the last probably from its production via stirring. It also had derivatives in the contemporaneous Gnomish Lexicon such as G. cwir- “to stir, stir round, make spin”, G. cwiril “spindle”, and G. cwîr “cream” (GL/29).
Neo-Eldarin: I think it worth positing a Neo-Eldarin root ᴺ√KWIR “stir, spin” to preserve these Early Qenya words, perhaps as a variant of √KWER “revolve”.
siripi
root. *stalk
The unglossed root ᴱ√SIRIP is one of the few triconsonantal roots appearing in the Qenya Lexicon of the 1910s, where it had derivatives ᴱQ. siripta “slender stem” and ᴱQ. sirpe “stem, stalk, slender tube, etc.” (QL/84). In the contemporaneous Gnomish Lexicon it had derivatives like G. sirp “straw” (GL/68) but also G. thribin “lean” and G. thrimp “stalk” from ✱s(i)rip- (GL/73) since initial sr- became thr- in Gnomish.
Neo-Eldarin: I think it is worth positing a Neo-Root ᴺ√THIRIP “stalk” to salvage some of these early words. I prefer ᴺ√THIRIP over just keeping ᴱ√SIRIP unchanged because (a) it better preserves the forms of more Gnomish words and (b) helps differentiate the Neo-Root from 1930s ᴹ√SRIP “scratch” (Ety/SRIP).
sṇtṇ
root. esteem
The root ᴱ√SṆTṆ “esteem” appeared in the Qenya Lexicon of the 1910s with derivatives ᴱQ. santa “dear, beloved” and ᴱQ. sinte “esteem; estimate, computation” (QL/85). There are no signs of this root in Tolkien’s later writing, but I think it is worth positing a Neo-Root ᴺ√SINIT for “esteem” and “estimate” words.
tapa
root. *taper, stretch out
viki
root. *sticky
yaya
root. *steel, possession
A root implied by the primitive form {ᴱ✶ı̯aksĕ >>} ᴱ✶ı̯aisa in the Gnomish Lexicon of the 1910s that was the basis for derivatives like G. gais/ᴱQ. yaisa “steel” and similar words (GL/37); in the contemporaneous Qenya Lexicon, the word was ᴱQ. Y̯akse “steel”, consistent with the deleted form in GL (QL/105). In the Early Qenya Grammar and English-Qenya Dictionary of the 1920s, the Qenya word was ᴱQ. akse “steel” (PE14/48, 81; PE15/77). I think it is worth positing a Neo-Root √GAK “steel” to salvage these early words.
retye
root. *spit
s’ta-
verb. s’ta-
aikwa
adjective. high, steep
eðe
root. *iron, steel
ninda
noun. *stream
nuhu
root. bow, bend down; stoop, sink
taka
root. fix, fasten, stick
tolo
root. *stick up
tuluku
root. *steady, firm
yolo
root. smell, stink, reek (intr.)
tiři
root. ‽stick up
vityi
root. *sticky
tiði Reconstructed
root. ‽stick up
A root given as ᴱ√TIŘI [TIÐI] “stick up” in the Qenya Lexicon of the 1910s (though Tolkien marked the gloss with “?”), with derivatives like ᴱQ. tinda “spike”, ᴱQ. tirin “tower”, and ᴱQ. tirion “a mighty tower, a city on a hill” (QL/93). In later writings, the city name Q. Tirion “Great Watch Tower” (RGEO/65) or “Watchful City” (MR/176) was almost certainly derived from the root √TIR “watch”.
fawa
root. smell
A root in the Qenya Lexicon of the 1910s glossed “smell” along with several Qenya derivatives, most unglossed (QL/37). It seems this was a neutral “smell” root based on words in the contemporaneous Gnomish Lexicon, such as G. faust “smell, odour (neither pleasant nor unpleasant)” (GL/34). There are no signs of this root in Tolkien’s later writing.
maþa
root. dusk
This root was given as ᴱ√MASA¹ “dusk” in its main entry in the Qenya Lexicon of the 1910s, but its Gnomish form math- indicates the true root was ᴱ√MAÞA (QL/59). This was clarified in a list of roots at the end of the M-section in the Qenya Lexicon (QL/63) and its representation as maþ- in the contemporaneous Gnomish Lexicon (GL/59). Its most notable use in the legendarium was in the name G. Umboth-muilin “Pools (muil-plural) of Twilight (umboth)”, where G. umboth or umbath “nightfall” was derived from a strengthened form of the root, ᴱ√mbaþ- (GL/75). However, in later writings this name was reconceived as Ilk. Umboth Muilin “Veiled (muilin) Pool (umboth)”, with the first element umboth meaning “large pool” (Ety/MBOTH, MUY). The name was ultimately replaced with S. Aelin-uial (S/114), by which point the early root ᴱ√MAÞA was long abandoned.
ru’u
root. ‽
An unglossed root in the Qenya Lexicon of the 1910s, probably actually ✱ᴱ√RUƷU, with derivatives like ᴱQ. rú “dwelling, village, hamlet”, ᴱQ. rue “rest, stillness, remaining, steadfastness”, and ᴱQ. ruin “peace” (QL/80). There were a number of likely-related words in the contemporaneous Gnomish Lexicon such as G. rô “enduring, long suffering; quiet, gentle, docile”, G. rû “dwelling, house”, and G. rûtha- “dwell, remain”, though Tolkien seems to have rejected the Gnomish forms beginning with rô- (GL/66). There are no signs of this root in Tolkien’s later writing, but I think it is worth positing a Neo-Root ᴺ√RUH “✱still” to preserve some of these early words.
sivi
root. [unglossed]
Unglossed roots in the Qenya Lexicon of the 1910s with variants ᴱ√SIVI and ᴱ√SIWI and a single unglossed derivative ᴱQ. sivilda (QL/84). It is difficult to guess what Tolkien intended for these forms to mean, though they conceivably reemerged as the later roots ᴹ√SIW “excite, egg on, urge” (Ety/SIW) or √SIB “rest, quiet” (VT44/35).
sṣtyṣ
root. ‽
The unglossed root ᴱ√SṢT͡YṢ appeared in the Qenya Lexicon of the 1910s with derivatives like ᴱQ. sastya “sore, galled” and ᴱQ. sist (sisty-) “ulcer, sore” (QL/86). The latter word was also mentioned as siste in “ulcer, boil” as a derivative of ᴱ✶sṣtē (PE12/14), and the primitive form was given as sistyi, though Tolkien did say it was indicative of ancient syllabic ṣ (PE12/3). However, in the Early Noldorin Dictionary of the 1920s it seems primitive ᴱ✶sṣt- = “✱hiss” (PE13/163); see the entry on ᴹ√SUS “hiss” for discussion.
Neo-Eldarin: For purposes of Neo-Eldarin I think it is worth positing a Neo-Root ᴺ√SISTI to salvage some of these early words.
tołᵂo
root. [unglossed]
An unglossed root in The Qenya Phonology of the 1910s illustrating a hypothetical series of ancient lateral approximants, with derived roots like ᴱ√TOLO and ᴱ√TOẆO [with ẇ = ɣʷ] (PE12/16). The former appeared in the contemporaneous Qenya Lexicon as the basis for island words (QL/94), but the latter appeared nowhere else in Early Qenya writings.
yere(ne)
root. *iron
This root appeared in two different forms in the Qenya Lexicon of the 1910s: as ᴱ√DYEÐE = ‘EÐE equivalent of ᴱQ. anga “iron” (QL/105), and as ERE(N) “iron or steel” = ᴱ√YEREN (QL/36). In the contemporaneous Gnomish Lexicon Tolkien had related forms G. ger “ore, metal” and G. geth “ore of metals (esp. raw iron)” (GL/38); the second of these was deleted, which probably means Tolkien selected ᴱ√YEREN over ᴱ√DYEÐE. In Tolkien’s later writing he largely used √ANGA for “iron”.
Neo-Eldarin: In a 2024-08-17 post in the Vinyë Lambengolmor Discord Server (VLDS), Luinyelle suggested restoring ᴺQ. yeren “pig iron” with the added meaning “✱ore”, which also implies the preservation of ᴺS. [G.] ger “ore, [unrefined] metal”. Presumably these words would be derived from a Neo-Root ᴺ√GYEREN of similar meaning.
oso
root. ‽
An unglossed root in the Qenya Lexicon of the 1910s with derivatives like ᴱQ. oksa “joint” and ᴱQ. oswe “hip” (QL/71). To salvage these words, I would revise this early root to ᴺ√OTH as a variant of ᴹ√OS “round, about” referring to a rotating joint. If you are not comfortable with this, derivatives of √LIM “link, join” might also be useable for joints, based on Q. málimë/S. molif “wrist, (lit.) hand-link” (VT47/6).
saw̯a
root. [unglossed]
An unglossed root in the Qenya Lexicon of the 1910s (QL/82), which may have reemerged as √SAWA “disgusting, foul, vile” in notes from the 1950s (PE17/172, 183).
ƕawa Speculative
root. snow
A hypothetical root explaining words in the Gnomish Lexicon such as ᴱQ. fáwe/G. fô “snow” and ᴱQ. fauta-/G. fôtha- “to snow” (GL/35). Given the existence of ᴱ√FAWA “smell”, I theorize this root may be slightly different, perhaps ?ᴱ√ǶAWA, but that’s just a guess. There are no signs of this root in Tolkien’s later writing.
-yǝ
suffix. [unglossed]
gwori
root. ‽
hama
root. ‽
nusu
root. *smell
oto
root. knock
sē
root. ‽
sōđā
noun. house
sṇtyṇ
root. twinkle
tini
root. twinkle
toẇo
root. [unglossed]
tum(b)u
root. ‽
tyulu
root. tall
tḷpḷ
root. *support, prop
uqu
root. wet
ẇaða
root. dwell
bē Reconstructed
root. ‽
pelek Reconstructed
root. axe
pṛtṛ Reconstructed
root. judge
ʒono Reconstructed
root. hard
-ya
suffix. present
gu̯ara-
verb. to dwell
keme
root. soil
lepse
?. [unglossed]
lopse
?. [unglossed]
nufu
root. *smell
nulu
root. ‽
po
root. ‽
pol-i
root. ‽
siŋi
root. ‽
The root ᴱ√SIŊI appeared unglossed in the Qenya Lexicon of the 1910s with derivatives ᴱQ. singe “salt” and (adjective) ᴱQ. singwa “salt” (QL/83). It had similar derivatives the contemporaneous Gnomish Lexicon: G. sing and singrin “salt”, noun and adjective (GL/67). I think it is worth positing a Neo-Root ᴺ√SIÑGI to salvage these early words.
snóra
noun. muscle
sulu
root. *support
sḷt-
verb. to sift
teled-
noun. [unglossed]
tḷkḷ
root. [unglossed]
vḷkḷ
root. ‽
An unglossed root in the Qenya Lexicon of the 1910s with derivatives like ᴱQ. valkane “torture” and ᴱQ. vilkin “bitter, evil”, serving as the basis for ᴱQ. Valkarauke, the Qenya name of Balrogs (QL/100). In later writings the initial element of this name was derived from √BAL “(divine) power”.
yn̄t
adjective. large
nawa Speculative
root. dwarf
A root in the Qenya Lexicon of the 1910s glossed “strip”, with derivatives in both Qenya and Gnomish such as ᴱQ. hulqa/G. hulc “naked” (QL/41; GL/49). It had a variant ᴱ√FULU which seems to have no derivatives (QL/38). There are quite a few later roots of similar meaning, so likely the root was abandoned.