Reconstructed Sindarin Pronominal System: A New Theory

Three Rings
Aaron Shaw & Florian “Lothenon” Dombach — March 2003
article out of date
Theoretical ArticleTheoretical Articles: A comprehensive knowledge of the works of J.R.R. Tolkien is needed to fully understand articles in this category, the subjects treated being studied in minute detail by their authors.


It has been my wonderful opportunity to work on this pronominal reconstruction with Florian “Lothenon” Dombach. What follows is a conglomerate of my ideas mixed with his (and the insights of Taramiluiel and Elena_s_g: special thanks to them). This work will most likely be changed from this form as new insights appear. I am, however, going to keep this chart in its unmodified condition as reference for future work and as a history of what has been done. May the tongue of the Elves lives ever on.

Initial Thoughts

  • From the attested examples of the Corpus, the derivation of new personal pronouns seems to follow a general pattern. When we look at nîn besides lín and mín or ten besides men and even nin it seems clear that there is a distinct basic shape for each case, nominative, dative and possessive to which is added a characteristic consonant.
    Demonstratives and neutral forms, on the other hand, seem to have a base form which displays singular and can completely regularly be formed into plural as well, which seems to be what Tolkien originally planned even for the personal pronouns as they can be found for his early Noldorin (which are used in the movie-dialogues).
  • From the examples of nin and nîn besides conjugation-ending –n for the 1st and men, mín besides conjugation-ending –m for the 1st it seems likely that the characteristic consonant is the same that is added for conjugating verbs, which makes a reconstruction easier.
  • A form that often seems not to follow those “rules” is the 1st This is, however, not a big problem because the conjugation of a-stem verbs is also irregular during conjugation for the 1st p. (i.e. the irregular change of final –a to –o in conjugation of a-stem verbs)
  • We have den attested in an accusative context, where, if we follow the general “rules” of Sindarin phonology we could clearly expect lenition. Therefore, it seems logical to conclude that the original form must have been ten; and now matching this with dative men I think it can be expected that for the accusative case, the normal dative pronouns are used in a lenited form.
  • I tend to believe that we have two separate forms of 3rd p. pronouns based off of gender. In my humble opinion, these are a Masculine/Feminine combination and Neuter/Neutral combination. The fact that we have hain attested in LR seems to support this theory.

The following is an attempt to explain some of the reasoning for the large chart at the bottom of this page. Sindarin words in blue are unattested, but can be easily reconstructed from the corpus. Words in red are completely unattested.

Personal pronouns

Possessive Pronouns

Most of the known pronouns are possessive so I will list them first because it will make coming reconstructions easier. I do not think there should be seen a difference between í and î, I expect the first as an older form.

possessive Singular Plural
1st p. nîn mîn
2nd p.r. lîn
3rd p. tîn

In the corpus, tîn is attested as dîn. We see this form multiple times in the King's Letter. It must be a lenited form because it is used an an adjective to describe to whom something/someone belongs. Sellath dîn “his daughters”, ionnath dîn “his sons”, bess dîn “his wife”.

Dative Pronouns

Next are the dative forms which shall be listed here with the accusative forms for the above mentioned reason:

dative Singular Plural
1st p. nin men
2nd p.
3rd p. ten
accusative Singular Plural
1st p. nin ven
2nd p.
3rd p. den

Men is only attested in ammen < an men “for us” and by its shape we can expect it to be dative (cp. den). Another hint to this might be enni < annin[?] “to me” while anim < an im, using a nominative pronoun, is translated as “for myself” (clearly reflexive and will be discussed in its own due time). The changed vowel of the 1st nin (e > i) might be seen as an almost expected irregularity, similar to that of a > o with Sindarin pronominal endings. Now, matching up with possessives, it is easy to fill in the gap of the 2nd as len. This clearly corresponds to the attested le “to thee” which is said to be of Quenya origin, so we can expect len to be its pure Sindarin counterpart. Whether this form should be used in actual Sindarin writings is hard to say. We know that le was present in the dialect of Imladris, but does this mean it would have taken the place of len in all Third Age Sindarin? Given our small corpus, I do not think it is possible to answer at this time. Future publishings will hopefully shed more light on this subject.

Concerning the use of san, sain: While we do not readily have evidence to prove otherwise, it seems logical to conclude that these must be neutral pronoun forms instead of masculine/feminine. The fact that we have e, den and dîn all attested as third person forms in the corpus makes it hard to otherwise fit these in phonologically. These may therefore be neutral forms, which could feasibly behave in a separate manner from the other pronoun forms. The only evidence against this proposition seems to come from Ae Adar nín, which translates den as “it”: this seems to indicate that there is no differentiation between masculine, feminine or neutral forms. However, I am not convinced that this was a very final draft. We see numerous inconsistencies in this prayer, making it awfully hard to give much weight to its use of pronouns. In my humble opinion, it is better to rely more heavily upon what we have been given in LR (i.e. the Moria Gate Inscription) which we know Tolkien carefully considered.

Nominative Pronouns

We only have two clear examples of nominative pronouns:

nominative Singular Plural
1st p. im
2nd p.
3rd p. e

That makes reconstructing the nominative forms rather difficult. The problem being that both the 1st is seemingly irregular; which does not lend itself to easy manipulation and stable conclusions. It may be that the irregularity of the i and the irregular m (where we seem to have n attested in every other case) are in some way related to the general irregularity of these forms (perhaps it may be directly tied into phonological evolution of the pronouns from CE where the n may have become assimilated in some way. In either case it is clearly irregular). Seeing the — until now — regular form of the 3rd, we might expect that nominative is formed by e-, plus the characteristic vowel. This we do not find, which is, with careful thought, to be expected. Instead we see simply e (which might correspond to în, also not using any characteristic consonant). This lack of a characteristic consonant seems to correlate quite well with Sindarin verb conjugations, in which we see no ending in the 3rd

nominative Singular Plural
1st p. im em
2nd p. el
3rd p. e

Now we might take a look at the conjugational/pronominal endings to fill in the other gaps. The attested endings are:

endings Singular Plural
1st p. -n -m
2nd p. -ch
3rd p. -r

The 1st p. (sg. and pl.) fit into this concept perfectly. What about the unkown 2nd forms though? While we have the ending -ch attested in the “Turin Wrapper” this phrase is, unfortunately, not translated for us (technically it isn’t published yet). I still believe that it is a viable pronominal ending though, along with what appears to be an alternate form -g. David Salo says that these forms are attested but have not yet been published. So then which one should we use? Since we really only have an example of –ch, and this example seems to be in reference to a single person (assuming that the assumed translation is correct) I think we can venture to guess that this form may have been used for the 2nd pronouns. This might suggest that –g is a plural form, but since we don’t have any material at hand to work with, the use of –g would be rather speculative at this time. Assuming that -ch is the correct form to use, we must look at what its development might have been. In Old Sindarin, -kke would, because of the loss of the final vowel and form shift kk > ch, give us the necessary form for the nominative pronoun; affixing as usual an e- before this characteristic consonant. In the dative, however, we would see a different form. When kk appears at the beginning of an Old Sindarin word, it would yield c-. This would therefore give us a dative form of cen in relation to the other attested forms men, ten. From this form it is simply a matter of extrapolation to produce the lenited possessive, accusative and also the long dative and reflexive dative forms for the 2nd

Now what do we do about a possible 2nd form? While it certainly may be that -g was intended to take this place, with -ch being the 2nd form, we see no evidence in Sindarin that such a 2nd form even exists. Until more text is published, the best thing, and really the only thing, for us to do is to assume that the 2nd and 2nd forms are identical to each other (as in English). This is however, pure speculation. Where now do the forms le, lîn fit? Given that the uses of le and lîn are attested when addressing the Valar or higher, I think it would be a safe assumption that Sindarin possesses both reverential and familiar pronoun forms.

Now, solely using the 3rd pronominal suffix –r, we can reconstruct all the missing pronouns for this form (assuming that the above named assumptions are not entirely wrong). Let us now turn to our reconstructed chart:

Person Nominative Possessive Dative Accusative Long Dative Reflexive
1st im nîn nin nin enni anim
1st em vîn men ven ammen anem
2nd p.f. ech chîn cen chen achen anech
2nd p.r. el lîn len (le) len (le) allen anel
3rd (m, f) e dîn ten den athen ane
3rd (n) as hîn san han assan anas
3rd (m, f) er rîn ren ren adhren aner
3rd (n) ais hîn sain hain assain anais

The possessive and accusative are both “pre-lenited”. Do not lenit these pronouns a second time.

Other Pronouns

Demonstrative Pronouns/Adjective General Reflexive Pronoun
sen “this”
sin “these”
în reflexive possessive

The use of în. This pronoun seems to be a general reflexive pronoun that can be used in place of the normal possessive pronouns. If this is the case, we would expect this pronoun to be used when referring to actions that affect oneself or the subject of the sentence (as opposed to some other person). This is similar to “the man drank his juice” problem. Does this mean “the man drank his own juice”? Or does it mean “the man drank the person across the room’s juice”? This reflexive pronoun seems to refer back to the subject of the sentence. Let’s look at some examples from the King's Letter:

  • Ar e aníra ennas suilannad mhellyn în
  • ar Eirien sellath dîn
  • ar Baravorn, ionnath dîn

It is important to keep in mind that we have two different people being discussed in this letter (both in the third person!). The King Elessar, and Sam (Perhael). The first sentence is used in reference to the king himself, and is believed to be reflexive. The final two are used in reference to Sam, who is essentially “the other man across the room” from our juice problem. Clearly these pronouns are not being used in reference to the King. I therefore think it is safe to conclude that în is a general reflexive possessive pronoun.

We also find the demonstrative pronoun hin in the corpus in the Moria Gate Inscription; Celebrimbor o Eregion teithant i thiw hin. Due to its adjectival use, we would expect this form to be lenited and pluralized (in accordance with plural noun tiw). We can therefore suppose the un-mutated form to be sin “these”, and the singular form to be sen “this”.

Concerning the Noldorin Pronouns

Ha, ho, he - I do not include these pronouns and their various given forms in this chart because I am not convinced that they made it into mature Sindarin. These pronouns have differing forms for masculine, feminine, and neutral genders, where we seem to have evidence in Sindarin of only two: m/f with neutral only appearing in 3rd p. (in which instances these pronouns behave unlike the m/f). Secondly, these pronouns do not fit very well with the overall phonological scheme that we have here devised (assuming that it is at least partially correct). Perhaps future publications will prove one way or the other.

Short Dative vs. Long Dative?

We now come to a somewhat serious problem. What do we do now that we have two separate dative forms? Which one does one use? The corpus seems to indicate that it is possible to do two things:

  • Use a short dative and place it before the verb as we see Tolkien do numerous times in the corpus. These forms seem to be used mostly when there is an implied direct object (i.e. “To thee I sing”).
  • Use a long dative and have it follow the verb. These forms seem to follow the direct object.

Then, it seems very feasible that we could see either form in a sentence. It is also very likely that we would see the short dative form used when forming other noun cases in Sindarin via prepositions. In these cases, the pronoun would acquire whatever mutation the preposition would cause. Only the context that you use such a word in will determine which form to use.

Reflexive Datives?

We also seem to have reflexive dative forms in Sindarin. These are used to express such ideas as “for myself”, “for themselves”, “for itself”, etc. The only attested form that we have from the corpus is anim “for myself”. This is quite clearly the nominative form im “I” plus the dative pronoun an “to, for”. If we follow this general pattern we can reconstruct what the other various reflexive forms may have been.

A last guess

We have the example of a suffixed possessive pronoun, namely guren = gûr nîn. As Ryszard Derdziński also suggests I think this might be (if Tolkien did not drop this idea anyway) a hint that the characteristic consonant can be used as a suffix (after –e) to work as possessive, too. These are very speculative forms, and should be treated as such. Nevertheless, I do believe that the general theory holds water, so it would not be disastrous to use such a form in a composition. It may be that these forms are used when a possessive is in the nominative (i.e. “my heart tells to me”). So to stay with the given example I expect:

guren “my heart”; gurech “your heart”; gurel “thy heart”; gure (gured?) “his/hers heart”; gures/guras “its heart” (n.); gurais “their heart” (n.); gurem “our heart”; gurer “their heart”.

See also on Tolkiendil

langues/english/i-lam_arth/reconstructed_sindarin_pronominal_system.txt · Dernière modification: 14/06/2011 19:09 par Elendil
Partager sur
Nous suivre sur
Tolkiendil - - Tous droits réservés © 1996-2017