"Then the stem vowel of the first word is repeated at its end. "
Thank you very much! That was the rule I was searching for.
"in some of your examples of compounded words, the were not actually connecting vowels."
Yes, I assumed that too, and I wrote it.
Tamas, apparently my aims and ideas were not completely clear, I'm sorry for that, and I will repeat all of them. A bit reasonless, because Ríon understood and gave me the answers, but anyway, if one starts thinking that I'm insane
"Tom you're becoming a bit of a ahem Riddle here, and that's a bit concerning."
Then I think, I should say something.
My basic idea, which is really not complicated, is that I could use suffixes like infixes.
I wanted to create compound words. There are four possible situations:
Word one results in a vowel and word two results in a vowel.
Word one results in a vowel and word two results in a consonant.
Word one results in a consonant and word two results in a vowel
Word one results in a consonant and word two results in a consonant.
I am not interested in situation one, because I already know what happens there. I am also not interested in situation two and three, for I already know that both words just connect, without any change. Examples:
Ama + Tixe -> Amatixe
Ava + Cúma -> Avacúma
Apa + Nóna -> Apanóna
Atya + Russa -> Atyarussa
Nó + Coire -> Nócoire
Nó + Laire -> Nólaire
Caista + Nótie -> Caistanótie
Tarma + Sundar -> Tarmasundar
Tasar + ion -> Tasarion
I could tell many many more examples, but I think that the concept is obvious. I am interested in situation four: consonant + consonant. There are also four possible situations here:
4.1. The first word results in one consonant (for example Enel, Elen, Atan, etc.) and the second one starts also with one single consonant (for example Lúce, Sáma, Vande, etc.)
4.2. The first word results in one consonant and the second one starts with two (for example, Mbelekoro, Hwerme, Hwinya-, etc.)
4.3. The first word results in two consonants (for example Heceld-, Mard-, Lepetass-, etc.) and the second one starts with one consonant.
4.4. The first word results in two consonants and the second one starts with two consonants.
I do not care about situation 4.1. If one word results in a single consonant, and the next one starts with a single one, then it is ok. There are rules for that (for instance: -n + r- -> rr (Elerrína). I do not know all of them, and I could need help to find them, but that was not my question. Situation 4.2 and 4.4 are pretty seldom, so I am mainly interested in 4.3.
The problem is that it is impossible that more then two consonants are in series (although there are exceptions for semi-vowels like w and y). I'm not completely shure whether a real Quenya word can start with two consonants which are not semi-vowels, but that does not matter. At least suffixes can (-sse, -nce, -ndon, etc.)
There are two consonants in series in situation 4.1, three in 4.2 and 4.3, and four in 4.4. As far as I know, it is the most common strategy in situation 4.2, 4.3 and 4.4 to put a vowel between the two words. I just don't know, what kind of vowels are required. It is sometimes -a-, sometimes -e-, sometimes -i- and maybe -o- and -u- are possible too.
Word 1--linking vowel-Word 2---Result (compound)
Atan--------a----------mir----> Atanamir (4.1)
Er----------e----------sse----> Eresse (4.2)
Anar--------i----------nya----> Atarinya (3)
So how do we know, whether the linking vowel has to be -a-, -e- or -i-? My idea was that the linking vowel is identical to the first vowel of the second word, -e- in Eresse, because of the e of -sse. Apparently this is not true, therefore I thought that I can just use -ie- if word one will be a noun, -i- if word one will be a verb and -a- or -ya- if word one will be an adjective. Examples:
Word 1--linking vowel-Word 2---Result (compound)
Mard---------ie-------menel--> Mardiemenel (4.3)
Alfirin--------a--------sta-----> Alfirinasta (4.2)
Ric-----------i--------sse-----> Ricisse (4.2)
Ric-----------e-------sse-----> Ricesse (4.2)
Mard is a noun. -ie makes something a noun. Therefore it would be clever to give this noun with an unintentional consonant ending another ending, a vowel ending, which does not change the word's meaning. The noun of a noun is still a noun. Therefore I thought that this could be an universal method to link two words if the first one will be a noun. There might be a difference between abstract noun and noun, but I have a solution for that. I could just transform the noun into an adjective. It does not change anything, for the part of speech of a compound is defined by the second word.
Same with -a and -ya. Alfirin is an adjective. -a and -ya make something an adjective. Therefore it would be clever to give this adjective with an unintentional consonant ending another ending, a vowel ending, which does not change the word's meaning. The adjective of an adjective is still an adjective. Therefore I thought that this could be an universal method to link two words if the first one will be an adjective.
It is a bit different with verbs. As far as I know (would be interesting if I'm wrong) there is no way to make something a verb, except -ta, but that would change the meaning dramatically. So, maybe I could use the aorist ending, as Tamas suggested.
To make a verb, which results in a consonant, an aorist word, we have to add -e.
Unfortunately this -e ending changes if we will add a pronominal suffix. It becomes -i-. Therefore the aorist endings are
-e (no line on the right side, because no new word will come)
and -i- (a line on the right side, because a new word will come).
Stem---linking vowel---pronominal suffix
Therefore I was confused that you said, -e- could be the linking vowel. I asked whether it should not be -i-. I thought that maybe -e just becomes -i- if a pronominal suffix will follow, and that it would stay -e- if another suffix or word will follow. That would explain why it is
at least if the -e- really marks the aorist, as you suggested.