Dear David,

It may be useful to refer back to John Newman's article about BHS in the Kālacakra. As he demonstrated there,
the author(s) were well aware of Buddhist departures from classical Skt. norms, and regarded this
as intentional, not due to poor knowledge of Skt. If we recall Buddhist derivations of, e.g., arhant from
ari + han, the treatment of Aja you find here as equivalent to Ajita/Ajaya does not seem so implausible,
even if without precedent in more classical usage.
 
best regards,
Matthew