Theoretical Analysis of the Sindarin Pronominal System

Three Rings
Aaron Shaw — August 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article out of date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.


This article is a derivative of my previous work with Florian “Lothenon” Dombach in which I aided in a “reconstruction” of the Sindarin Pronominal System. What follows is not a reconstruction per se but rather a collection of thoughts organized into a semi-coherent structure. I do not attempt to reconstruct any form because any such reconstruction would be biased one way or another in a field that had too little support for any conclusion. Thus I will present my thoughts and analysis in hope that someone may find it of use.

Initial Thoughts

  • While we have very few attested personal pronouns in either Sindarin (Noldorin) or even Quenya, I have attempted to analyze the various similarities between the currently attested forms. Such a synchronic viewpoint is unfortunately far less useful than that of the diachronic - nevertheless, it is in all practicality the most reasonable means of analysis at this point in time. Needless to say, the diachronic will be used as well.
  • Unfortunately, we have absolutely no way of knowing whether the currently attested Sindarin pronouns are part of a continuous system (indeed they are most likely not). We may indeed also see variants of pronouns that are not in their “pure” forms, having been influenced by the speech of men, various sylvan dialects, and/or Quenya.
  • Assuming that we can trust the translations of the corpus to not be entirely misleading, it seems that each form of pronoun seems to possess a unique phonological “shape”. Each form appears to be associated with a unique use in Sindarin syntax (we have for instance differing forms of nominative and accusative pronouns etc).

Introductory Theory

To begin, I would like to discuss some various theories that will have consequences on our analysis of each individual pronominal form later in this article. With such an opening we will then be able to more effectively analyze the way these forms may interact one with another. It is commonly assumed that possessive pronominal forms should have an acute accent. This is often initiated by the forms we see in the corpus such as nín, lín and vín (mín). Ironically, we have just as many attested forms using the circumflex as we do the acute accent. What form then should be natural in a non-contextual situation? Or is there a minute difference that we have not yet picked up on?

It has been noted that tengwar do not distinguish between the circumflex and the acute accent. However, this author deems such a difference in roman orthography to be important as it mayhelp reflect grammatical features.

Some have suggested that this may be an example of Tolkien shifting his ideas about accent around. While this certainly may be the case, I find that there is another theory that is equally applicable (though not necessarily more “correct”). Let's take a minute to look at these forms in context:

  • Pater Noster:
no aer i eneth lín
tolo i arnad lín
caro den i innas lin
bo Ceven sui vi Menel.
Anno ammen sír i mbas ilaurui vín
ar díheno ammen i úgerth vin
sui mín i gohenam di ai gerir úgerth ammen.
  • Voronwe's Cry (untranslated):
Alae! Ered en Echoriath, ered e.mbar nín!
  • Lúthien's Song (untranslated):
si loth a galadh lasto dîn!
  • King's Letter:
edregol e aníra tírad i Cherdir Perhael (i sennui Panthael estathar aen) Condir i Drann, ar Meril bess dîn; ar Elanor, Meril, Glorfinniel, ar Eirien sellath dîn; ar Iorhael, Gelir, Cordof, ar Baravorn, ionnath dîn.
Ar e aníra- ennas suilannad mhellyn în phain:

Unfortunately, both Voronwe's cry and Lúthien's song are not translated for us. Therefore we have two sets of data that are reasonable to base an analysis off of - the Kings Letter and Pater Noster. What I find intriguing is that these “pronouns” seem to be enclitic. Indeed, we could very likely substitute in another constituent for an entire phrase of N+Adj. The fact that we see what look like lenited forms seems to at least give the theory some validity. If then pronouns are proclitic, why would we expect them to behave radically different from compounds? We see a fairly systematic trend of shortening vowels in compounds. Therefore I would suggest that the “natural” non-contextual form should probably be the circumflex. This, of course, poses two problems with the corpus:

  1. We see the circumflex in the King's Letter when we might expect the pronouns there to be enclitic as well.
  2. We have the interesting forms in Pater Noster that are not accented whatsoever (and still appear to be possessives).

These problems are perhaps not so problematic after all. If we are to believe that possessive pronouns are enclitic then we are in essence saying that the pronoun accent is dictated by their syntactic role. Interestingly, both lin and vin appear when they are non-final (i.e. between clauses!). For one reason or another these pronouns seem to be affected by their position in the sentence - loosing entirely their accent from (1) the noun they describe and (2) their phrasal position.

What then about dîn and în? These may, perhaps, be seen as archaic forms (perhaps Doriathrin? notice the use by Lúthien). This of course does not entirely explain a lack of cliticism but it is something to ponder. The fact that we see these forms in use in Gondor suggests that they may be natural forms for western Sindarin in general and may not be influenced excessively by outside forces.

Another possibility is that these forms might be used to make an exophoric vs. endophoric distinction, or a differentiation between present/nonpresent in which case it we would not “necessarily” have see a modification thereof. If this were so, it might explain why we seem to have attestations of both the Noldorin system and these forms as the Noldorin pronouns may have been merely neutral. Or perhaps this is nothing more than Tolkien's modification of his ideas.

  Lúthien & Huan (© Catherine Karina Chmiel)

Another interesting feature to note about the Pater Noster is the use of what appears to be a possessive pronoun in almost a reflexive sense:

>sui mín i gohenam di ai gerir úgerth ammen

Here we see the form mín in what appears to be an emphatic position – strengthening the nature of the verbal conjugation (i.e. redundency). Would this mean that “possessive” forms could be used as emphatic reflexives!? An interesting topic to ponder at least…

Regarding the use of medial i in the first person: I was recently thinking about this and it hit me that this may be a determinant between inclusive/exclusive pronouns. Perhaps the first person shows up this way as it is inherently exclusive while other forms attested in the corpus might be inclusive… lets take a look at men, den, nin:

  • edro hi ammen! - ‘open now for us’
  • A tiro nin, Fanuilos! - ‘o look towards me, Everwhite!’
  • Naur an edraith ammen. - #‘Fire [be] for saving to us’
  • caro den i innas lín - #‘make it the will your’

In the first Gandalf is calling upon the gate of the dwarves to open for him and the fellowship. If this were inclusive what might that mean? Could this be Gandalf and the door and then including the fellowship? Surely Gandalf did not wish for the door to open for the enemies that were tracking them… not very conclusive so far. The second contains nin which appears to be dative and which is singular (and therefore inherently exclusive). Not much more to say though we don't have a solid answer yet… The third falls back into the same category as the first. If this were inclusive, who would that entail? Hard to say exactly. The fourth and final could, by this theory, be inclusive. Though this doesn't prove much either. So far nothing concretely confirms or denies the theory; but its basis does look potentially weak. It is, nevertheless, something to ponder. This of course, may also be due to the adition of -ni to some base form (which might cause i-umlaut which would yield intermediate i) Or maybe it is a singular marker?

# = altered from the given definitions to fit the text better in my opinion

Corpus Tolkieni

The natural place to continue from is the corpus as attested in LR, the HoMe series and various other semi-attested fragments. The major problem is now, where do we begin? Our first order of business must be an analysis of what types of pronoun forms we have in Sindarin - we shall therefore begin by looking at each form we have not already discussed individually.


By the standards of Tolkienian linguistics, this form is pretty well attested. We have three separate instances where we find this “pronoun” - the Moria Gate Inscription1), Lúthien's song2) and quite possibly in Gilraen's Linnod3). The good thing is that two of these attestations come directly from LR itself - a good plus in an attempt to find consistency. Lúthien's Song, however, is not translated, though it probably has the same meaning as the other attestations. What then is it's meaning? Lets look at the corpus:

  • Im Narvi hain echant - ‘I narvi made them’
  • le linnon im Tinúviel - untranslated
  • ú-chebin estel anim - ‘I have kept no hope for myself’

The Moria gate inscription appears to be fairly clear in its meaning; seemingly giving us a nominative form. Whether this pronoun is emphatic or not is hard to say. The inscription itself seems to indicate as much due to its use of both a subject and what appears to be a subject pronoun.

Luthien's Song, while untranslated by Tolkien could mean ‘To thee I sing [am singing?], I, Tinuviel’. If this is an accurate translation, then the idea of the pronoun being emphatic seems to be enforced - giving us a second attestation where both subject and a subject pronoun are used together. The apparent fact that Sindarin is a pro-drop language also supports this idea. Many pro-drop languages have similar techniques to impart emphasis upon the subject (Italian for instance).

Our third attestation is quite interesting. Here we find the apparently nominative/emphatic pronoun used in conjunction with the preposition an “to, for”4). Tolkien however, gives us a reflexive translation. We might therefore translate the final prepositional phrase as “to, for I” which while ungrammatical in English, may certainly be a way to attain reflexives in Sindarin.

What would be interesting to know is whether Sindarin makes any distinction between emphatic pronouns and nominative forms. In the ‘Etymologies’ under the stem NI2 we are told: ‘NI2 = I’. What is interesting is that we see the traces of this derivation in all of our other attested first person pronouns, yet not necessarily in im. Perhaps im once was formed from NI with some sort of emphatic suffix? If so it might explain the shift from m > n though this could easily be any other type of suffix as well.

There is another option that we have not here covered. Im could be a reflexive form meaning something like “myself” instead of a nominative. If so it may or may not be related to the stem NI at all. There is, however, no immediate support for this idea.


Nin also appears to be a first person pronoun. Unfortunately, it its far less attested than im. Our only attestation luckily comes from Sam's cry to Elbereth which Tolkien translates for us in L:278 and RGEO:72:

  • A tiro nin, Fanuilos! - ‘O look towards me, Everwhite!’

If we go with Tolkien's English translation of this phrase we would probably conclude that this is a “Dative Pronoun” (though the chances that this is actually an objective form are just as likely, if not more so). This form, having no accent, might be conceived as enclitic. I am, however, not convinced that “imperative Lenition” really occurs. It is quite easily demonstrated that the syntactic relationships differ between those of a pronoun/article and that of the verb and noun. Tolkien tells us that the Sindarin imperative developed from a primitive particle â that had then become part of the verbal conjugation (WJ:365 - roughly around the Old Sindarin time period). For imperatives to create lenition we would have to have some deal of cliticism between the postverbal particle and the following noun. This, however, does not appear to be the case: >Verb stem + â + noun > [VP [NP]]

The particle therefore would cliticize with the verb - becoming part of the verbal conjugation, while the noun resides within its own noun phrase and would theoretically be unaffected by the modifications of the verb (though we would probably see Direct Object (DO) lenition in this case - but that doesn't change the point). Therefore it seems likely that this form truly should appear without accent - giving us some reason to believe that a separate “Dative” form of pronouns existed at some point.


Enni appears to be a form that I would prefer to term a “long dative”. This “long dative” seems to consist of an + ni. In VT41:11 we are given the following translation:

  • Guren bêd enni - ‘My heart tells [to] me’

This seems to fundamentally differ in phonological “shape” and meaning from anim, which is translated with a reflexive meaning. This could further suggest that im truly is a reflexive form, though there certainly is no conclusive proof in any direction.


Le is fortunately one of the most attested pronouns in Sindarin. We see this form numerous times - all in poetry. We have three separate attestations:

  • Fanuilos le linnathon – ‘to thee, everwhite, I will sing’
  • Le nallon sí di-nguruthos! – ‘to thee I cry beneath the shadow of death’
  • Le linnon im Tinúviel – not translated

Le seems to be quite regularly transcribed as an “object pronoun” though it doesn’t match up with nin. Perhaps this is due to Quenya influence. Perhaps the syntactical use of this pronoun is what gives it a dative feel. It’s hard to know which without more examples of objective forms.


Hain is an interesting mystery because it quite clearly suggests the involvement of the Noldorin pronouns found in the ‘Etymologies’ (possibly being the plural form of han “it” in a neutral sense). This clearly conflicts with the forms dîn and în. Again, this may be explained by various meaning differences as stated above:

  • Im Narvi hain echant

Hain in this case is clearly plural in reference to the “signs” or tengwar upon the gate. This would correspond quite nicely with the Noldorin pronouns. The only problem with this is that if we truly believe in “Direct Object” lenition we would expect chain from the forms Tolkien gives us in the ‘Etymologies’. Of course, there is also the possibility that h resists lenition as has been speculated about hollen but we don't have anything conclusive against the form sain either. There is also the possibility that due to hain's position in the sentence it would not undergo lenition though this would seem to be contradicted by Daur a Berhael … Eglerio.


Den is what appears to be an accusative pronoun. We really don't have anyway to prove one way or another, though the evidence seems to be strong for accusative due to its translation. We might therefore expect the form ten to be the unmutated, base, accusative form.


In the End we really don't know anything for certain. It is my hope that this article may have been of some use to the public. I am sure it won't satisfy the avid Sindarin poet as I do not provide “suggested” forms but I do hope that it may spark some debate and interesting research.

See also

On Tolkiendil

On the net

1) LR:II/4
2) LB:354
3) LR:App.A
4) LR:II/4; UT:39; SD:129-3
langues/english/i-lam_arth/sindarin_pronominal_system.txt · Dernière modification: 15/06/2011 06:16 par Elendil
Partager sur
Nous suivre sur
Tolkiendil - - Tous droits réservés © 1996-2017