--- In firstname.lastname@example.org
, David Kiltz <dkiltz@g...> wrote:
> On 04.04.2004, at 00:17, cgilson75 wrote:
> > Though we may not be certain of the status of the _-n_ in
> > _ammen_, we can at least conclude that it is somehow redundant,
> > since the translated meaning 'for us' is conveyed by the components
> > _am-_ ('for', if indeed this is parallel to _anim_ 'for myself' as David
> > suggests) and _-me-_ (1st person plural root 'we, us').
> I'm not sure. Redundancy can, I think, only be claimed if you assume
> the root to be 'sufficient'. If not, only determining the status of
> _-n_ would allow us to speak of redundancy. In _an-im_, we seem to have
> _an_ construed with a nominative or casus rectus _im_. Whereas in
> _enni_ (if < _an-ni_) we seem to have an oblique form combining with
By "redundant" I was referring to a distinctive form or formal feature
that the language could theoretically do without because there is no
context in which that distinction alone serves to convey a difference
in meaning. An example of what I am talking about is the distinction
in English between _we_ and _us_. This does convey a difference in
meaning when taken out of context, but in any syntactic construction
where either form is used the difference is also conveyed by word
order -- thus _the man saw us_ vs. _we saw the man_. In fact in
modern English the case distinctions among pronominal forms are
largely superfluous, though naturally very useful, as can be
illustrated by an example of nearly identical contexts: _mother gave
me socks_ vs. _mother gave my socks_ (to someone, away, etc.)
There is also a degree of redundancy for certain preposition + pronoun
combinations, like _to me_ and _for me_. Thus in _he bought the book
for me_ vs. _he gave the book to me_ the difference in force between
'for' and 'to' is already sufficiently conveyed by the difference in
verbs, such that either can be expressed by the so-called "indirect
object" construction: _he bought me the book_, _he gave me the book_.
The alternative **_he bought the book to me_ is ungrammatical in
English, while _he gave the book for me_ (meaning 'on my behalf')
would normally have an additional context like 'to someone else'
either expressed or implied, and this use of _for me_ cannot be
conveyed by the indirect object in English.
Given that Sindarin is a language that, within the mythology, human
beings are capable of learning to speak (see Appendix F), it seems
reasonable to hypothesize that similar cognitive strategies underlie the
interplay of redundancy and ambiguity that seems evident in its syntax,
though our insights must depend largely on a much better understanding
of the inner workings of our own native languages. Indeed such an
hypothesis is evident (sometimes explicitly but usually implicitly) in most
discussions of Elvish. Ironically, the very analytical or pattern-seeking
process, by which we can extrapolate information from a limited corpus
of data, inherently favors theories that involve fewer redundancies and
ambiguities over theories that involve more. Not that I see any way
to avoid this pitfall other than to be aware of it when attempting to
draw inferences about likelihood.
While Sindarin retains a more extensive inflexional component than
English, at a morpheme level the situation of its pronouns seems to be
similar. In so far as we can tell, the forms _im_ in _anim_ and _im
Narvi_, _nin_ in _tiro nin_, and _ni_ in _bêd enni_ are in
complementary distribution. And in this sense the final consonant of
_nin_ would be redundant, unless there is a contrasting construction
*_tiro ni_ which means something different from 'guard me!' (Which of
course there might be...)
But all this really implies is that the differences among these forms
are relicts of an earlier stage of the language where they may have
had more extensive paradigmatic parallels or occurred in contexts
where they contrasted minimally with each other. Such residual formal
distinctions could very well retain a correlation with the basic
syntactic roles they had at an earlier stage, though the fact that the
system is essentially redundant (or largely so -- the evidence is
admittedly sketchy) would render it easily susceptible to analogical
Nothing David proposes is incompatible with these observations. But
neither do the data seem incompatible with (or even unfavorable to)
the possibility that _nin_ 'me' could combine with a preposition. The
form _im_ can be used both with and without preposition, and perhaps
_ni_ also if it occurs in the sentence (a "password") _Gir..
edlothiand na ngalad melon i ni [?sevo] ni [?edran]_ (VIII:293).
> Alternatively, both _im_ and _ni_ may mean 'I' (nominative), the first
> being an emphatic form. (For _im_ cf. [LR1:402 and III:354]). _Nin_ on
> the other hand seems to be the 1st sg. accusative pronoun in Sindarin.
> _men_ is most likely: 1) Casus rectus (nominative) or 2) Casus obliquus
> (accusative), i.e. has no beneficiary connotation. In that, _nin_ seems
> to contrast Quenya where _nin_ is clearly 'for me' [R:67].
In contrast with _ammen_ ('for us' VI:413, 'of us' VII:175) part of
the context of the form _anim_ ('for myself' Appendix A I (v)) is its
co-reference with the subject ('I' in _ú-chebin_ 'I have kept
no'). In the other two occurrences of the pronoun _im_ it accompanies
the name of the subject (_im Narvi hain echant_; _le linnon im
Lúthien_ III:354). In English the subject can be emphasized this way
using the nominative, as in "I, Christopher, made them", or the
reflexive pronoun can be used (in essence meaning 'by oneself'), as
"John himself bought the book" or "I myself, Christopher, sing this
So there are two possibilities, either (1) _im_ is nominative, used
emphatically with a named subject and combining with a preposition to
indicate the object is the same as the subject, or (2) _im_ is
specifically reflexive, combinable with a preposition to indicate
reflexive object and usable with the subject for emphasis. Whichever
theory one picks there is little to suggest that _-men_ 'us' in
_ammen_ is the same case as _im_. In its sentences the subjects are
_naur_ 'fire' and _annon_ 'gate', unconnected with 'us'. So _ammen_
is neither reflexive nor nominative, judging by the contexts and the
translations. This leaves "objective'"cases.
Between the pronouns in _tiro nin_ 'look towards me, guard me' and
_guren bêd enni_ 'my heart tells me', it is difficult to say which
is semantically closer to _-men_ in _edraith ammen_ 'saving of us' and
_edro hi ammen_ 'open now for us'. We also have _ammen_ 'us' as
indirect object of _anno_ 'give!' and _díheno_ 'forgive!' and even
in the sense 'against us' in _gerir úgerth ammen_ '(who) trespass
against us', all in the Sindarin Lord's Prayer (VT44:21).
Considering the semantic range of _ammen_ and especially how close
_anno ammen_ comes to _bêd enni_ (both would use 'to' in English if
the indirect object were expanded to a prepositional phrase), it seems
quite reasonable to suggest that we could say *_anno enni_ 'give me!'
and *_bedir ammen_ '(our hearts) tell us'.
Alternatively, we may consider that _enni_ and _ammen_, though
etymologically connected by the historical derivation of _en-_ < _an-_
by affection, actually have different syntax, with the historically
later _ammen_ using a distinct case of the pronoun. If so, I would be
inclined to tie together two ideas suggested by David as
> By contrast, an abstracted _men_, while apparently exhibiting an ending
> _-n_, combines with _an-_ and thus parallels _im_ (and probably _-ni_).
> Alternatively, it may stand for the accusative. In both cases, when
> combined with _an-_, there is no redundancy as _an-_ adds the
> beneficiary notion which _men_ alone, probably doesn't convey. [...]
> I don't think *_anni_ means 'for me' but 'to me'. The meaning of _nIn_
> is crucial here. I think it may be interpreted as accusative sg. or
> (towards) me. Note, however, that Tolkien glosses "...tiro nin..."
> [loc. cit.] as '_tiro_ == 'look towards (watch over), _nin_ == 'me'.
If we assume that _nin_ 'me' is syntactically detached from its
(perhaps etymological) dative, allative, or benefactive sense in the
context of _tiro_ 'look towards, watch over, guard' and is simply the
"accusative" form of the pronoun used as the direct object of a verb;
and if we assume that the dative, benefactive, or even adversative
sense of _ammen_ is conveyed by the semantic range of the preposition
_an-/am-_ and the attached pronoun is simply marked as the object,
using this same "accusative" case-form -- then we could by implication
say *_anno annin_ 'give me!'
This is comparable to the situation in English where the same forms
(me, us, him, etc.) are used for both direct object of a verb and
object of a preposition. There is even an exceptional idiom (like
_enni_ would have to be under this theory), which uses a different
case of the pronoun. This is the possessive construction with the
preposition _of_, as in _I read that book of his_, which contrasts
with a usage like _I was speaking of him_. This is I think the only
relict in English of the older situation where different prepositions
may govern distinct cases -- a situation familiar in more highly
inflected languages, such as Latin (e.g. _ad infinitum_ vs. _de
facto_) or Quenya (e.g. _mí oromardi_ vs. _et Earello_).
-- Christopher Gilson