Philippe Caquant <[log in to unmask]> writes:
> I had a little closer look at your site.

Ah, thanks a lot! :-)

> Whenever reading a
> documentation about a language, I like to have a
> separate chapter about semantic concepts alone,
> without any morphosyntactic, phonetic or whatever
> considerations, because I think they make the reader
> confused, but that's my opinion.

Yes, it's confusing.  But also easier to write, especially when I'm
not really 100% sure of what I want.  I hope it is at least readable.

> I noticed that you give a special meaning to "aspects"
> (but "State vs.Event" follow).

It's not complete, so a bit of confusion is still in the docu.

> "Spatiorative" doesn't sound very nice indeed (maybe
> "circumlative", or "circumjective", or something like
> that ?)

I thought about a good name but did not find a better one.
I did not like 'circumlative', but 'circumjective' sounds
ok.  I'll think about it. :-)

> maybe you could add a "notional" meaning to ?

Hmm, I could through away the essive then?  Finnish (and other
Finno-Ugric langs) is(are) supposed to have developed the
inessive,illative,elative (by adding 's') and
adessive,allative,ablative (by adding 'l') from the
essive,translative,partitive?  In the same way, my five instead of
three cases could be used like the essive/translative/partitive

Good idea, I'll think about it.

> In your example "I think, therefore I am", rather than a causative,
> I fancied something like: "thought-ABL existence-ACC I infer"

Hmm, it would be

   [I think].experience.ABL [I am].infer.PRD

Where the first clause is an ablative adjunct to the second.  ('From
that I think, I am.')  The evidence markers are not modified, so it
does not mean 'From the experience of thinking, I infer that I am',
but 'From my thinking, which I experience, I am, so I conclude.'.

> (from Thought I infer Existence) [but you have no Accusative]. "From
> thought" (notional) looks similar to me as "from the city" (spatial)
> or "from 1945" (temporal), while "thought-CAU" itches a little :

It depends on what Decartes meant. :-) Could you give me a deeper
semantical analysis of the original sentence?

> Benefactive and Malefactive might be unnecessary refinements,

Sometimes I like redundancy.  S7 is not meant to be grammatically
minimal.  It is more fun to use the two additional cases in puns than
it is to using affixes.  The latter will seem so explicit.

   'Look, I brought you same cake.'

It'd by very funny if 'you' is malefactive, no? :-) With an additional
affix, it would be much less subtle, but rather explicit.

> As to moods, as I said, I would had a "fictional" one

I reduced some of the moods and had not thought about a fictional
mood.  It is on my agenda of thoughts, though.

> consistence and laws. I also would use Conditional and
> Irrealis both in main and in subordinate clause

NONONO!  I don't like that.  I find it much more precise to mark only
the condition with conditional and transport the information about the
conditioned predication in a different mood.

> For Degree, as I said too, I would use different scales, depending
> on the situation needs.

Actually, your system of scales is too engineered for my taste.  S7 is
not supposed to be a computer language.

> About modalities, I think that it's more that just
> "personal inclination". I read (in Laurent Gosselin)
> the following scheme, which I found interesting:
> Instance of !           Ajustment direction
> Validation  ! Predicate > World   ! World > Predicate
> ============!=====================!===================
> Reality     ! Alethic / Ontic     !
> ------------!---------------------!-------------------
> Speaker's   ! Epistemic / Doxastic!
> belief      ! (subjective truth)  !
> ------------!---------------------!-------------------
> Speaker's   ! Appreciative        ! Boulic
> desire      ! (+/- desirable)     ! (volition, desire)
> ------------!---------------------!-------------------
> Institution ! Axiologic           ! Deontic
>             ! (+/- praisable/     ! (obligation)
>             !  blamable)          !

Yes, I read these (or similar) distinctions in a Lojban grammar
description, I think.  This is all not at all fully developed in S7.

> I was expecting to find something about semantic classes, as you
> refer to some in your examples (sentient, spacetime, etc), but I
> didn't ?

You didn't, because there is not yet a list of classes.  I currently
generate them on the fly when creating lexicon entries.  As soon as I
have significantly more than the maximal number of classes supported
(40), I'll combine them in a 'nice' way.  I'm affraid the chapter
about semantical classes will have to way until then. :-)

> To me it's one of the very most important concepts...

You're right!  But it's not what I start with. :-) I like morphosyntax
most.  (I like structure in general: algorithms and datastructures
is the underlying hobby for most of my hobbies abart from, maybe,
cooking. :-))

> The real question being: a language is very well, but shall we talk
> about ? It is the referent world that determines the language, IMO.

I don't understand that.  Could you explain that point in some detail?

Again, thanks for looking into the grammar and commenting!  That's all
very good input for me. :-)