DISCLAIMER: This article is preliminary research on the part of its author (Paul Strack) and does not necessarily reflect the opinions of the owner of this site. Since the source material is complex and its interpretation can be subjective, multiple conclusions are possible.
Sindarin is primarily a SVO language (subject-verb-object), with subject first, followed by the verb and then any objects of the verb. This is different from Welsh, which is mainly a VSO language. In many situations, Sindarin has a word order similar to English: the subject precedes the verb, prepositions precedes the qualified noun phrase, relative pronouns precedes the subordinate clause, etc. One major exception to this rule, however, is that Sindarin adjectives follow rather than precede their noun:
The adjectival element usually follows in Sindarin but in certain old elements the older (Quenya) order is maintained, as e.g. with mor- “black” (morgul beside gul-dûr); and in names the order could be shifted for euphonic reasons; so Fanuiðol for normal Dol-fanui (PE17/36).
Quenya preferred the (older) order in which adjectival stems preceded, while in Telerin and Sindarin the adjectival elements often were placed second, especially in later-formed names, according to the usual placing of adjectives in the ordinary speech of those languages. In names however that ended in old words referring to status, rank, profession, race or kindred and so on the adjectival element still in Sindarin, following ancient models, might be placed first (PM/346).
In S. the simple genitive was usually expressed by placing the genitival noun in adjectival position (in S. after the primary noun) (RGEO/67).
As indicated above, the placement of the adjectival element applies to (1) adjectives following nouns as in bas ilaurui “daily (ilaurui) bread (bas)”, (2) loose genitive expressions like aran Moria “king of Moria” vs. (English-style) “Moria King” and (3) compounds, where the modifying element appears after the modified element: Aeluin “Blue lake” = ael “lake” + luin “blue”.
In the case of compounds there are numerous exceptions, particular for names. Older names may follow the old order with the adjectival element second. Newer names may follow the old order when one of the elements marks things like rank (ar- “noble, royal” placed first) or a people or place (-waith “people, land” placed second), or when the putting the name in the older order simply sounds better, as with Fanuidhol above. Indeed, these “exceptions” are so numerous that the majority of attested compound names in the legendarium put the adjectival element first, though it could simply be that the published tales have a preponderance of ancient names, especially in The Silmarillion.
Object Order: It seems the normal order for objects is that the direct object follows the verb and the indirect object follows the direct object: ónen i-Estel Edain “I gave Hope [estel, direct object] to the Edain [indirect object]”. This is opposite of the default order in English, where the indirect object precedes the direct object: “I gave the Edain hope”. However, the Sindarin indirect object is generally marked with the preposition an “to, for” (in the dative), which makes the placement of the indirect object more free within the sentence (an Edain ónen estel), much the same way that using the preposition “to” or “for” makes the placement of the indirect object more free in English (“to the Edain I gave hope”).
A direct object following the verb is lenited (undergoes soft mutation) as in lasto beth lammen “listen to the word (peth) of my tongue”. There is some evidence that soft mutation is a general marker for the accusative, and is applied even when the direct object is displaced from its normal position after the verb:
- im Narvi hain echant “I, Narvi, made (echad-) them (lenited-san-plural)” (LotR/305).
- Daur a Berhael, Conin en Annûn, eglerio “Frodo (?Taur) and Sam (Perhael), princes of the west, glorify (egleria-) [them]” (LotR/953; Let/308).
The second example is less clear, since Tolkien never gave an unmutated form of Frodo’s Sindarin name, and indeed seems to have vacillated between initial t- or d- (PE17/102). There is, however, no other good explanation for the soft-mutation of Perhael “Samwise” to Berhael, since as Tolkien discussed elsewhere, none of the plausible mutations that might result from a “and” (stop mutation or spirant mutation) would produce b- from p- (PE17/41). If lenition indeed marks the direct object, then its placement in the sentence might also be somewhat free, though probably less free than the more unambiguously marked dative.
Finally, there are some examples of independent pronoun objects appearing before the verb rather than after. The pronouns le appears before the verb in both A Elbereth Gilthoniel and Lúthien’s Song:
- le linnathon “to thee (le) I will sing (linna-)” (LotR/238; PE17/20-21).
- le nallon “to thee (le) I cry (nalla-)” (LotR/729; RGEO/64).
- le linnon “*to thee (le) I sing (linna-)” (LB/354).
A prefixed pronoun also appeared in one of Tolkien’s explanations for mae govannen “well met” (LotR/209), which at one point he said was mae g(ī)’ovannen “well [art] thou (ci) met (govan-)” (PE17/16-17). Here the pronoun ci causes mutation of the verbal participle (govannen → ’ovannen), and is itself mutated by the preceding adverb (ci → gi) and then the vowel is lost between the pronoun and the first vowel of the (mutated) verb. The demonstrative pronoun hain appears before the verb in hain echant, given above. Finally, a preceding pronoun ed “it” might appear in the untranslated phrase hí in Ellath îr ed epholar “*now the Elves alone ?remember (ephola-) it (ed)” (VT50/5), but we can’t be certain because the phrase is untranslated and the “pronoun” ed appears nowhere else.
Based on these examples, it seems that an object pronoun can also appear immediately before a verb, just as it can in Quenya, and it causes mutation of the following verb. In the case of le, the preceding pronoun even functions as a dative rather than a direct object; seen the entry on direct objects for further discussion.
Summary: To summarize the above:
- The basic Sindarin word order is subject, verb, direct object, indirect object.
- Objects marked as dative (with an-) or accusative (with lenition) may appear in other places.
- Object pronouns may also appear immediately before the verb, causing mutation of the verb.
- Adjectives appear after their noun instead of before, and the adjective undergoes soft-mutation.
Conceptual Development: The practice of placing adjectives after the noun appeared all the way back in the Gnomish Grammar of the 1910s, albiet with the peculiar rule that the definite article i also sometimes followed the noun:
When a noun is qualified by both article and two or more adjectives, and especially when this again is followed by a possessive genitive, the order is usually, noun. article. adj. adj. genitive., as:
- “the beautiful white feet of Idril” talwi i’loss ar gwandra nan·Idril
- or talin i’lossi ar gwandron nan·Idril [plural].
But adjectives may and do precede the noun frequently. None the less the normal rule is that adjectives & adjectival genitives & equivalent forms succeed the qualified noun as nearly as possible.
There is no sign in Sindarin of this displacement of the article before multiple modifiers, for example: i mbas ilaurui vín “our (mín) daily (ilaurui) bread ([m]bas)” (VT44/21).
There is also an example in Gnomish where a pronoun is displaced before the verb, complicated by the fact that Gnomish uses pronominal subject prefixes rather than suffixes for verb inflection:
- G. en nin·ista mai “I (nin) am well aware of that (en)” (GL/52).
Normally, though, objects followed the verb:
- G. nôbi i·mab ’len suila ontha “*[he] took (nab-past) the slender (glen) hand (mab) of his (ontha) daughter (suil-genitive)” (GG/11).
- G. ôni cailthi mabir gleni nan·hirilion “[he] kissed [lit. ‘gave (antha-past) a kiss (cailthi) to’] the slender (glen-plural) hands (mab-plural-dative) of the ladies” (GG/11).
Note how the direct object cailthi “kiss” precedes the indirect object mabir “hands” in the second example, just as in later Sindarin, even though Gnomish has an explicit dative ending (-r). Based on cailthi “kiss” rather than **gailthi, it seems the direct object did not undergo lenition at this early conceptual stage.