Derivation
From a Root
From a verbal root, we can derive a few different words.
"-u" suffix
With the suffix "-u" we create a noun which represents the action of the verb. For example, gin "to fall" > ginu "fall"; gus "to see" > gusu "sight" (referring to the act, not one's ability to see); êr "to rise" > êru "rise".
"-se" suffix
With the suffix "-se" we create a noun which represents a thing that does the verb. For example, iml "to shine" > imlise "light". This suffix is used for "active" derivations, as oppposed to the "-as" suffix which is used for a similar, but less active meaning. Contrast the "action" of shining embodied by imlise "light" with the less active meaning embodied by balas "word".
"-as" suffix
With the suffix "-as" (underlying realization is "-ad") we create a noun which represents a thing that does or embodies the action/meaning of the verb. For example, gal "to mean, to have meaning" > galas "word, meaning"; bal "to speak" > balas "speech, utterance".
"-ê/ar" suffix
With the suffix "-ê", which surfaces as "-ar" in non-nominative forms, we create a noun which represents a person that does the action (professionally). For example, man "to cook" > manê "chef".
"-eu" suffix
With the suffix "-eu", which surfaces as "-el" in non-nominative forms, we create a noun which represents a person that does an action, but not professionally. It constrains the meaning simply to the doer of the current action. For example, man "to cook" > maneu "cook" where "cook" here refers to the doing of this particular action.