The words Tolkien used for “before” and “after” in Elvish were remarkably unstable. While there were some basic themes in how he conceptualized these words, the details and exact word forms underwent constant revision. In English usage, “before” and “after” have two functions: spatial (in front vs. behind) and temporal (preceding and following). Elvish words often had similar dual functions, but not always in the same arrangements as in English.
The earliest version of Elvish is represented by the Gnomish and Qenya Lexicons, published in Parma Eldalamberon #11 and #12. In the Qenya Lexicon there were three main roots for “before” and “after”:
ᴱ√NO(WO) was used for “in front = before (of place)” but “after (of time)” referring to the future. Meanwhile ᴱ√DYĒ was used for “before (of time)” referring to the past while ᴱ√POT-I was mostly used for “after (of place)”, but also seemed to refer to the immediate future: “next”, “afterwards”. Related words in the Gnomish Lexicon showed a similar distribution:
Of interest in GL was the introduction of a variant root ᴱ√BOTO serving as the basis for G. bod, which also seemed to be useable of the past: “(of time) ago”. Contrast this with Gnomish derivatives of ᴱ√POT-I which could be used of “after (of place)” but also the immediate future: “next behind or after (time or place)”.
In Early Noldorin Word-lists of the 1920s, ᴱN. bod “back” was still contrasted with ᴱQ. pote based on primitive *bot-, along with related ᴱN. bón “back” vs. ᴱQ. ponte (PE13/139). In the English-Qenya Dictionary from the same period, ᴱQ. no reappeared with a new meaning: “after (of place)” and “behind” (PE15/68, 70). The word ie or iye “ago” seems to be a conceptual continuation of ᴱ√DYĒ (PE15/68). The dictionary also had a new word la “before” (PE17/80), which was used for comparison in the contemporaneous Early Qenya Grammar (PE14/48, 81).
A later variation of la “before” appeared as ala in drafts of the Oilima Markirya poem from around 1930 (PE16/62), which initially seemed to mean “before” based an early translation “before the last night” (PE16/68) but the same line was translated “after the last night” in the version presented in the “A Secret Vice” essay (MC/214). The last version of the poem from this essay translated ala as “beyond” (MC/221), which is more in keeping with Tolkien’s use of lá “beyond” in his writings of the 1950s and 60s (PE17/65; VT44/4).
The Etymologies from 1937 didn’t have before/after words, but it does have some front/back words, namely N. nîf “front, face” based on ᴹ√NIB of the same meaning (LR/378) and Dor. dôn “back” based on ᴹ√NDAN “back” (LR/375). The second of these survived into Tolkien’s later writing of the 1950s and 60s though perhaps more in the sense “go back = retreat” (PE17/166; WJ/412; VT48/32), but ᴹ√NIB seems to have become √NEB with a different meaning: “turn towards (speaker)” (PE17/27).
In drafts of The Lord of the Rings appendices from the late 1940s, the prefix Eb- seemed to mean “before” in N. Ebloer, Ephriw as an alternate names of “June, December” [= *Before-Summer, Before-Winter], as opposed to Cad- “after” in N. Cadloer, Cathriw alternate names of “July, January” [= *After-Summer, After-Winter] (PM/136). Compare these to corresponding alternate month-names in Quenya: ᴹQ. Erlairë, Erríve [= *First-Summer, First-Winter] and ᴹQ. Nólairë, Norrívë [= *After-Summer, After-Winter] from the same draft (PM/135).
This 1940s Quenya prefix nó- “*after” was consistent with ᴱQ. nō̆ “after (of time)” from the 1910s, while the Noldorin prefix cad- “after” matched Quenya words ᴹQ. kata “after” and kato “afterwards” from the Quenya Verbal System of 1948 (PE22/124). The Noldorin prefix eb- “before” seems to contradict Primitive Elvish ĕpĕ ndĕrĕ “after the man” from Primitive Quendian Structure: Final Consonants written in 1936 (PE21/64), but this could be a difference in the spatial vs. temporal usage.
A possible late example of rear-facing pot-derivatives was ᴹQ. opto “back” from notes on the The Feanorian Alphabet from the 1940s (PE22/50 note #80).
There was a prepositional paradigm from the mid-1950s having at least two sets of words: ca, cata, cana “behind, at back of place” (VT43/30) and opo, pō, pono, poto- “in front, of place” (VT49/32 note #12). These seemed to be a continuation of the earlier KAT and POT derivations, except that PO(T) now seemed to refer to “in front” rather than the rear. Similar forms appeared in notes from the late 1960s:
The word nó also reappeared with a new meaning “before [of time]” in the so-called Merin Sentence from a letter written to a fan sometime after the publication of The Lord of the Rings: merin sa haryalyë alassë nó vanyalyë Ambarello “I hope that you have happiness before you pass from the world”. This reversal of earlier meanings is confirmed by notes from the 1960s. Notes from 1968 had nō “at back (of place), before (of time)” (VT49/32 note #12); see Section 5.0 for the dating of this note. Notes probably from the late 1960s had primitive nō “behind” (VT47/34); compare 1910s ᴱ√NO(WO) “in front; after (of time)” given above. The word noa “yesterday” in notes from the late 1960s is consistent with this paradigm (VT49/34 note #23), but another note from this period instead had noa “tomorrow” (VT49/20) indicating some ongoing vacillation.
The 1936 preposition epe “after [of place]” also reappeared with some new meanings. In the first version of the 1957 poem Löa Yucainen Tolkien used epeni as a temporal preposition “before” (CPT/1296), but in notes on comparison from 1957-59 Tolkien said epe is used as “before in all relations but time [my emphasis] where appropriate words must be used: later, after, then, earlier, before” (PE17/56). These notes on comparison also used epe much like how 1920s la “before” was used in the Early Qenya Grammar. In 1968 epe was used for “after [of time]” in epesse “after-name” (PM/339). In notes associated with the Ambidexters Sentence from August 1969 Tolkien said epe meant “after, of time” (VT49/32 note #11). In notes from October 1969 epe was used for “after of time; after of place” (PE22/167, 168); see Section 4.0 for this full paradigm.
There is another variant of epe: apa “after [of time]”. It is best known as an element in Apanónar “After-born” (S/103; WJ/386) but it also appeared in apacenye “foresight” (MR/216), another reference to the future = “*sight of what comes after”. The elements APA, PĀ appeared in notes from April 1969 glossed as both “before of time” and “after (later than) of time”, but both these references were changed (VT44/36; PE22/147). In that same April 1969 note Tolkien translated PĀ, PATA as “after, behind of place” (PE22/147).
The April 1969 paradigm mentioned above with KĀ/KATA, PĀ/PATA, and (deleted) APA/PĀ appeared in a bundle of documents referred to as Late Notes on Verb Structure (LVS) in the literature (PE22/147-168). The full paradigm is as follows (PE22/147):
Above these are written derived forms afar, khyar; katar, patar and below are some forms {pā, pai, at a prior time; kā, kai ... >> fan, afea >> } fai; afea. Almost all of these forms are new except for KĀ/KATA and (rejected) APA/PĀ and this paradigm does not seem to be long lived. There is another paradigm in the same bundle of documents from October 1969 in which older roots were restored. This final paradigm appears twice, one of which is in the upper margin of a page (PE22/167 note #117):
The other appearance is in rough notes, probably written a bit earlier (PE22/168):
The rough notes contains references to several other prepositions unrelated to “before” and “after” which are excluded here to avoid confusion. This final paradigm uses EPE for “after” of both place and time, OPO for “before” of place, and AWA for “before” of time. This use of AWA for “before” is novel, since elsewhere AWA usually means “away” (Ety/AWA; PE17/148; WJ/361; VT49/24). However AWA was already somewhat connected to the past with verbs like auta- “depart, pass away”, though that was not its main function.
There are some partially-published linguistic notes describing the Elvish conception of time. These notes were first mentioned by Patrick Wynne in the June 2007 is issue of Vinyar Tengwar, where he quoted the phrase:
The Eldar regarded all that was past as behind them, their faces being towards the future. With reference to Time therefore words with a basic sense “behind, at the back” = before; and those originally meaning “in front, ahead” = after (VT49/12).
Wynne tentatively dated this document to the mid-1950s, and used it to explain why epe might be used for both “before” of place and “after” of time. Carl Hostetter published a longer excerpt of this text in his 2021 book The Nature of Middle Earth, and he dated it to around 1968:
The Eldar regarded all that was past as behind them, their faces being towards the future. With reference to time therefore words with a basic sense “behind, at the back” = “before”; and those originally meaning “in front, ahead” = “after”.
Nonetheless in thinking of people and the generations they spoke as if the elders were leaders marching at the head of a line of followers. It thus became difficult to speak of “those behind” = “peoples of former days” being followed by later generations. But such terms as leading and following were to them pictorial analogies, only used with a definite transfer of sense: as if we were to speak of looking back into the dark mists of days before our time.
For ordinary purposes, e.g., as what happened before my time, “behind” was used, at least originally. In practice, Common Eldarin had developed four distinct prepositional or adverbial bases:
A. (1) before of time (2) behind of place
B. (1) after of time (2) in front of, ahead of place (NM/162).
Hostetter included two other texts of similar sentiments. Both Hostetter and Wynne mentioned that there were Elvish terms associated with these texts, but the only one that is currently published was mentioned by Wynne: nō “at back (of place), before (of time)” (VT49/32 note #12). Wynne also tentatively dated this to the mid-1950s, but Hostetter’s dating would put it around 1968.
These quotes include the interesting notion that Elvish used the same or similar words for “behind = after (of place)” and “before (of time)”, and also the same or similar words for “in front = before (of place)” and “after (of time)”. Tolkien then goes on to contradict this by saying Elvish had four distinct roots for these notions, but there are several examples in Tolkien’s writings where Elvish splits the spatial and temporal directions differently from English:
1910s ᴱ√POT-I was used for “after” of both place and time, while the final October 1969 paradigm used epe for “after” of both place and time, so this split arrangement was not universal.
Using the four quadrants “before (place) = in front”, “before (time)”, “after (place) = behind”, and “after (time)”, we can outline the conceptual developments of these Elvish words as follows:
| Period | Before (place) | Before (time) | After (place) | After (time) |
|---|---|---|---|---|
| 1910s | NO(WO) | DYĒ | POT-I [1] | NO(WO) |
| 1920s | la | i(y)e | bot- [1], no | |
| 1937 | NIB [2] | NDAN [3] | ||
| 1940s | N. Eb- [4] | epe [5] | Q. Nó-/N. Cad- | |
| mid-1950s [6] | opo, po(to) | ca(ta) [7] | ||
| late 1950s-1960s | epe | nō | nō | epe/apa |
| Apr. 1969 [8] | KHYĀ | PHĀ | PA(TA) | KA(TA) |
| Oct. 1969 | OPO | AWA | EPE | EPE |
For purposes of Neo-Quenya, I prefer the paradigm from the late 1950s to late 1960s, with EPE for “before (of place)” but “after (of time)”, and NŌ for “behind = after (of place)” but “before (of time)”. I think this paradigm says interesting things about Elvish psychology. I would use some additional elements as well: OPO/PŌ for “before (of place)” and APA for “after (of time)” as variants of EPE. I would also use KA(TA) for “after (of time and place)”.