Since there doesn't appear to be much interest in the language, I'm not too concerned about changing things around in a substantial way. The summary is:

• Renaming: The kV series is rearranged so -ki becomes -ko, -ko becomes -ke, and -ke becomes -ki. This is to fix the weird working order of the konV series (since to start at 0 and match the kV series the vowel order seems wrong). The idea behind this is to make "not" more audible and to put the rest of the series in order from most to least specific. Additionally, the konV series may switch to have a different vowel than o. Because of these changes, I may change some kVV assignments so that those series can stay in line with the kV series thematically.

• Lenition: I'm dropping the generic nasal phoneme /n/ and the voiced stops so that the stops are voice-ambivalent (though for phonotactic purposes still unvoiced) and may be realized as flaps; their voiced counterparts can be realized as nasals, and in conjunction with the technical distinction for voice, this should make speech easier for speakers. Further, the current /x/ may move to [h] and actually just be written <h> /h/ (/ɣ/ remains unchanged as its voiced version).

• Other: <rl> is better transcribed /ɹ̠̊ɹ̠/ than as retroflex fricatives, though I'd still like a better letter than <l> to transcribe /ɹ̠/, especially since <r> looks like it could handle both, and further it allows for possibly adding /l/.

As for the current state of the fV series: A-fe describes something caused by A, so that A would be credited or at fault for whatever it is. A-fa describes something affected by A in a way that could allow one to detect A's presence (perhaps with some unreliability), but A wouldn't be credited or faulted for it; an example of this is knowledge or artistic influence. A-fi describes something which A makes more likely to happen, but it is ultimately not accountable for it, such as a tool or risk factor. As it stands, -fe is surrounded by two words that mean something more general, but each of the two are fairly similar to each other, so the arrangement doesn't make sense.

An ongoing issue is whether to reverse -kau, which I think I'd like to do because it makes sense for its spatial interpretation, except that I'm not sure how useful that would actually be. I'm also working out techniques for dealing with cases where the sense of a word may change depending on how the object it applies to is described (as opposed to just what it is), and the current kaV series seems to have some problems in that respect, so that just compounds the problem. Some generalization of space that can be made contextual in a useful way should help, since I'd like to make spatial metaphors from natural languages a well-defined part of the language that then doesn't actually rely on metaphor.

If anyone has any comments about this, as usual, feel free to leave them.