The root √ANA was connected to “giving” and “movement toward” for much of Tolkien’s life. The root first appeared as ᴱ√ANA “give, send towards” in the Qenya Lexicon of the 1910s, and in this period was the basis for words like ᴱQ. anta-/G. antha- “give” and ᴱQ. ana “to(wards)” (QL/31; GL/19). The root reappeared in The Etymologies of the 1930s as ᴹ√ANA with an inverted variant ᴹ√NĀ¹ “to, towards”; the most notable derivative of this inverted form was N. na “with, by” (Ety/ANA¹, NĀ¹). In the first version of Tengwesta Qenderinwa from the 1930s (TQ1), √AN was given as one of the primary examples of the inversion of roots, where √AN(A) was originally a “biconsontal root” (²√) with a vocalic beginning, and the inverted form √NA was an abnormal vocalization of the basic root (PE18/38).
This same example of inversion was used in the second version of Tengwesta Qenderinwa from around 1950 (TQ2: PE18/88), and this invertible root appeared regularly in Tolkien’s later writings almost always with glosses like “to, towards, motion to”. There were a few instances in which Tolkien explored some alternate meanings for the root, however. In Common Eldarin: Noun Structure from the early 1950s Tolkien said that:
> It is true that in Q. na appears with sense “to”, but this at nearest means “towards, to a position near, alongside” ... The original sense of Eldarin ana was plainly “at side of, alongside, besides”, hence also “moreover, in addition, plus” (seen in use of an- as an intensive prefix), and so an or na in some languages has the sense “along with, with, accompanied by, provided with, associated with” and the like (PE21/79).
Likewise in Quenya Notes from 1957 (QN) Tolkien said:
> √ANA/NĀ, to, towards — added to, plu-. Quenya verbal prefix ana, na-, as in napan-, add. an “plu-”, intensive prefix, arcalima, preeminently bright, ancalima, very bright ... S an, dative chiefly with pronouns or persons. < ana, hence vocalic mutation, but takes form m before m, b. ✱ to, for (PE17/146-7).
In this note Tolkien went on to explore various non-allative meanings associated with the root such as the conjugation/adverb Q. an “for, then, or, but” and prepositions S. na(n) “with” vs. S. na “to, towards”, differing mainly in the mutations they cause (nasal versus soft). It seems the association of this root with both the intensive an- and words like Q. an “for; moreover” and S. na(n) “with” motivated Tolkien to find a more nuanced meaning for the root. This likewise probably motivated Tolkien to explore an alternate basis for the intensive prefix in the mid-1960s, deriving it from √AMA rather than √AN (PE17/90-94).
Neo-Eldarin: For Neo-Eldarin, I prefer to retain an- as the basis for intensives in both Quenya and Sindarin, and thus I think it is best to retain the extended meanings of √AN from the 1950s to include senses like “add to, more” and “up to the side of”, hence by extension “alongside” becoming (in S.) “with, accompanying, possessing, having the trait of” as in S. na(n) “of, with”.
Tolkien used a variety of roots as the basis for the meaning “long” throughout his life. Its best known forms are the adjectives Q. anda, S. and “long”. In the Qenya Lexicon of the 1910s Tolkien first used the root ᴱ√ṆÐṆ “stretch” as the basis for the ᴱQ. adjective ande(a) “long” (QL/31), but its Gnomish cognate took the very different form G. in(d)ra (GL/51). By The Etymologies of the 1930s, however, Tolkien had revised this root to ᴹ√ANAD with Quenya and Noldorin derivatives basically as given above: ᴹQ. anda, N. ann (Ety/ÁNAD).
There is a later mention of the root √ANAD in a 1959 note, but in that note Tolkien considered transferring the sense “long” to a new root √ƷAN as a variant of √YAN, so that he could use √ANAD < √ANA- as the basis for words meaning “gate” (PE17/40). This new use of √ANAD would be a replacement for the 1930s root ᴹ√AD “gate” (Ety/AD). In another set of 1959 etymological notes, Tolkien did indeed give primitive forms ✶ʒandā “long” vs. yanā/yandā “wide” as derivatives of √ƷAN and √YAN respectively (PE17/155).
Later still, in 1967 notes on comparison, Tolkien gave a new root √NDA as the basis for and(a) “long”, though he said “S †ann- [long] only preserved in certain compounds, owing to competition with ann (< annā) gift, and ann(on) gate, of different origin” (PE17/90). Thus it seems Tolkien had abandoned √ƷAN > ✶ʒandā > Q./S. and(a), and in notes from 1968 Tolkien glossed √ƷAN as “adorn”, and at this later stage √ƷAN was probably connected to or a variant of the 1970 root √HAN “add to, increase, enhance, honour (espec. by gift)” (VT47/26-27).
The ultimate fate of 1967 √NDA “long” is itself unclear. In the notes where it appeared, Tolkien was also considering it as the basis for the intensive prefix Q. an-. But Tolkien abandoned this idea and decide the intensive prefix was actually am- derived from √AMA “addition, increase, plus” (PE17/91). However, this change in the intensive does not necessarily invalidate the use of √(A)NDA for “long”, and that is the last word we have in the published corpus on this topic.
Neo-Eldarin: For purposes of Neo-Eldarin, it’s probably best to assume the meaning “long” came from √ANAD or √ANDA or some similar root, much like it did in the 1930s.