You can ignore most shit I've written about grammar, especially the early stuff which has probably been revised later on [1]. Establishing roots is something I found useful later on, yeah - especially as it means the community could build more new words that way. I think the direction I was heading was to possibly have several roughly synonymous roots, which would have come from different cultures (Amarr, Udorian, Khanid, Takmahl), possibly giving rise to some roots being more official, technical, or noble; and others being utilitarian, lower-class, or crude. [2] If you have ideas for explaining or revising, or at all making sense of the grammar, I cannot welcome it enough.
Ad 1: Okay!
Ad 2: Well, first you could make a Swadesh list of words, or roots. Then you've got basic vocabulary items plus a starting point for creating those 'webs' I was talking about. As you say, this system is easy to build upon for other players indeed, plus it provides a more firm foundation for the fictional history, because it is easy to use a root, change affixes, some vowels and in this way creating a new word with a new language (Udorian, Khanid whatever).
Actually this is your project, so I didn't wish to interfere too much. But here are some concrete ideas.
As I read your work, it seems that you are going for a Semitic look, and for a more synthetic (lots of affixes in stead of loose words) structure. If you do this, you should look at
- The 'case hierarchy': the most common cases are the nominative (blank, zero) and accusative (object: 'he sees me'). Genetives, datives, locatives etc. are more secundary. This means that íf a language has a genetive, like yours, it will certainly have an accusative as well. The question now is: how many cases do you want?
- Agreement. It is common for synthetical languages to have a lot of agreement of e.g. particles and adjectives with the inflected noun. So: des heren, Middle Dutch for 'of the lord', and in Modern Dutch its 'van de heer'. Des was a case form of de.
- Of course there doesn't have to be a strict line between analytical and synthetical. On the one and you find English, on the other Latin, but somewhere in between is Italian. Italian has clear conjugated words, like in your language, that allows pronouns to be deleted. It also has a lot of agreement, e.g.: La casa piccola ('the small house') and il ragazzo piccolo ('the small boy'). Here's how Arabic does it: http://en.wikipedia.org/wiki/Arabic_nouns_and_adjectives
- If you wish to 'rebuild' your glossary for the sake of a harmonious, logical and consistent stem-based system, you should consider the sounds you want to put into the language. No language has all sounds. A consonant based language, like Arab and Hebrew (the ones you are aiming for) has a lot of consonant phonemes (28), since consonants bear the most information. This is exactly the reason why the alphabet was written with consonants only. You could simply copy the Arabic list of phonemes, if you like. Besides, not all combinations of those recognized phonemes are allowed. English does have [š], [k] and [r], but starting a word with škr- is not done. http://en.wikipedia.org/wiki/Arabic_phonology
- About those stems, if you're going to do that. I'm not sure, but it seems likely to me that vowels aren't inserted at will, randomly. There should be patterns. Not only for grammatical reasons, but also because consonantal context may trigger specific vowel changes. And of course affixes can trigger this too.
Take for instance the stem
w-r-d you proposed, meaning 'word'. A web could contain the following concepts (and phonotactis depend on your inventory of sounds):
word: worod
informing: tarafworda (taraf: ‘by’; second o is deleted because of word length; a indicates an infinitive verb)
support: tarvrawrada (taraf + verkh (‘other’); f and v assimilate)
reason: vamilorod (vamil: ‘think’, w is deleted, because wl is akward)
reasoning: vamilorda
utterance: traworod (I made up tra ‘to’, with same stem as taraf, because by and to share semantic features)
uttering: traworda
language: denirod (den: ‘all’, deletion of w)
linguistics: vishednurod (stem from vishen: ‘knowledge’ + d-n)
tongue: uflorod
rhetoric: vumlorod
I made these words up simply as an example, nothing more. You don't have to use any of it.
For tongue I made up the stem f-l, ‘do’, fela ‘doing’ and z-l as a affix, from your glossary entry ziel, ‘goal/purpose’. Doing is atelic, that is, it is continuous and does not semantically result in a specific state, product etc. With these morphemes we can make a telic verb, making: zelfela. On the other hand f-l results in moving, ufula. From there ufel, ‘muscle’, and hence uflorod.
You could also consider vowels to have some semantics themselves: a appears in active verbs; e in passive verbs; u in outward movements (vumlorod is something you do in front of others, based upon reasoning, which is an internal process); o in ditransitive verbs. And so on.
Some semantical ideas:
- There are two types of we, namely the inclusive we (that includes the hearer to whom one is talking) and the exclusive we (that excludes the hearer) (same goes for the plural form). The distinction is made clear by a suffix. Since I here a lot about Amarrians as being zelots with brotherhoods and striving for devine enlightenment or whatever, I think both we-forms fit in really nice. If one says: 'We+SUF. go to war', this gives a whole new meaning to the sentence, because the speakers know that we includes God, for example.
- For instance, English 'to live' is basically intransitive (exception: to live a great life etc.), but a semantical shift towards the transitive spectrum in a religious community would be interesting: to live something, namely God, or heavenly rules, or whatever.
I think your verb conjugations are absolutely fine. Strong and clear conjugations, however, will usually result in pronounless constructions, simply because the verb ending indicates person, number, tense etc. perfectly.
O yes, typologically speaking an SOV-language tends to have postpositions, and to have auxiliary verbs placed behind the main verb. So:
I table beautiful upon stand, meaning 'I stand upon the beautiful table'.
I don't have enough time to dwell upon this any further, due to examns and a Pre/Proto-Gallentean project.