Lisp-Stat used to be a big deal in the late-80s/early-90s, and had a shot at becoming the dominant open-source statistics package inspired by Bell Labs S. It eventually lost mindshare to R, though, and was pretty solidly eclipsed by the late 1990s. It'd be interesting if the R developer responds by trying to build something like Lisp-Stat again. ;-)

if you use Clojures primitives you pretty much end up with laziness by default. for, map and most of the other list processing functions all return lazy lists. So while Clojure does technically require you to "turn on" laziness in most cases you'll find it's turned on already.

in R, arguments are passed around lazily, akin to scala's =>. otherwise, evaluation is not lazy (as in Haskell) unless done so explicitly (as in force/delay for scheme).

i think essentially no one using R professionally would accept Haskell or Clojure as a substitute. having spoken to a few: lisp-like syntax is "unreadable", and Haskell is too much effort.