Hi Jayarava,

In my first response, I explained the "u" of "vantu" and "mantu" as an
anubandha or indicatory letter. After some further digging I now think
this is not right. It was an assumption based on the Sanskrit
counterparts "vatup" and "matup" for Pali "vantu" and "mantu" plus the
fact that the "u" is not seen in the declensional paradigms. For
example, in the locative singular form one would expect a
"bhagavantusmi.m" instead of "bhagavantasmi.m" with the stem
"bhagavantu". Well, I just found a rule in the Saddaniiti which states
that the "u" of "ntu" is changed to "a" before some of the
declensional vibhattis or case endings -- so bhagavantu + smi.m >
bhagavanta + smi.m > bhagavantasmi.m. It's not clear why the
traditional grammarians use "vantu" instead of "vanta" in the first
place and then having to add another rule just to change the "u" to
"a".

Best wishes,
Jim

> Further to my question... Actually Warder does deal with this under

primary derivatives starting on pg.251. The -u presumably
distinguishes the mant/vant taddhita suffixes from the kita
present-participle in -nt.

>
> Jayarava