Although aiquen is glossed "if anybody, whoever", I personally believe ai- means simply "any" and nothing more.

*pī̆, pi seems defensible to me at this stage as well (contrasting Salo's *cí [1]). Of course, the longer Q forms cenai, cenasit, etc. are also good to bear in mind when wondering about the Sindarin conditional conjunction(s); cenasit even happens to look very much like an old Italo-Celtic form... (but was there long or short *na- / *si- at the CE stage?)

Word-history of the Welsh counterparts (ultimately always helpful):
https://archive.org/stream/welshgrammarhist00joneuoft#page/444/mode/2up

___
[1] http://www.elendilion.pl/2013/12/22/g-i-p-report-complete-sindarin-dialogs-from-the-hobbit-the-desolation-of-smaug/

+ܤܡܝ ܦܠܕܢܝܘܤ
is *ge possible phonologically?

Maybe. If we use "bo" (opo) and "vi" (imi) as guides... it's possible.

*Ae is perhaps not so shaky, since we also have aen as an uncertainty particle. And ge is in fact attested, as the form which was replaced by aen (PE8:13, see also TolkLang 40.46: http://tolklang.quettar.org/messages/Vol40/40.46).

Btw, these are of course two things: In alasaila ké nauva, ké expresses modality, in ké mo quete ulka (different source!) it's a conditional conjunction. But the two uses are consistent with each other, since markers of epistemic modality (= judging the likelihood of a situation) apparently may turn into conditionals, among other things (according to a paper I've been reading).
In fact, it happens in English as well, cf. 'This should be bad' (ep. modality) and 'Should one speak evil, [...]' (conditional).

Regarding ge, I should like to get ahold of PE8 - does anyone know if the Marquette draft quoted in it has any other differences besides ge / aen? It would be nice if that were the case; it could shed more light onto early 50s Nold/Sind. I'm not sure if it's just that line, though, as "Discussion of the Marquette King's Letter" or the like doesn't appear in the PE8 contents. Perhaps it's just an aside on another article, mentioning the ?one? variance. That would be a pity.

Do we really know that "aen" is an uncertainty particle though? Last I checked, we still don't really know what it means. Also, looking at VT50, it seems to mean that "aen" might be related to "en" and used for the passive voice. "i eithro en estar iChîn Húrin" vs "i sennui Panthael estathar aen" ... the similarities are striking. Also, the bit from the King's Letter strikes me as prescriptive, perhaps desirative, rather than uncertain.

+Fiona Jallings
It falls under the term 'deontic modality' (compelling external circumstances, social/ethical norms). But again, compare English may which is deontic in 'You may leave now' and epistemic in 'This may be true'.
In any case, the passive theory of en seems way more speculative to me: There is no translation. Why does one particle go before the verb and one after? Where does en come from and is it related to aen? Why is a particle even needed for an indicative form?
But regardless, in VT50:14 it is ruled out that aen and en are pronouns, based on existing paradigms. This still leaves aen as a modality particle which could conceivably develop into (or be related to) a conditional.

+Matt Dinse
It's from A survey of Eldarin pronouns by Bill Welden and it just says 'This [aen] replaced an earlier word ge'.
But there is also a variant of Aragorn's gotcha-sentence from Marquette: En ni túviet (PE8:9). The rest is otherwise attested, as far as I can see.
Which early parmar are you missing? I can send you the pdfs. (I'm missing 6 & 7 myself.)

Given that in some languages the ideas "when" and "if" overlap morphologically (as in German wenn), it seems pertinent to remark that the suffixed [with *–sĕ?] CE relative stem which became Q írë (related to ier ?) might result in sth. like *î(w), ae(w) in Sindarin in light of Morris-Jones pp. 102-103 (vs. Telerin/OS sound-history, obviously); cf. also W haearn < Celt. *eisarn-, *isarn- and S lae, glaer produced by I-roots.

Finnish jos "if" is also derived from a relative pronoun stem, with old lative -s (more or less = CE *-da, WJ:366).

Thanks, +Roman Rausch
I'm actually missing 1-10; I didn't know the earlier ones were available (out there on the Internet). Luckily I have the later ones, more or less (phys. copies of what I could get, and PDFs of the out-of-print until they're reprinted again).