Python, Emacs and Lisp, revisited

Alex Martelli aleaxit at
Wed May 16 13:33:51 CEST 2001

"Oliver Korpilla" <korpo at> wrote in message
news:3B0195C1.A02BDA03 at
> Lisp allows side-effects (assignment and reassignment of values), and
> therefore no "full" functional language.
> Can we agree on the fact, that Lisp is that somewhat of a functional
> language?

I think it's correct to say that Lisp 'is' a multi-paradigm
language: it supports the functional-programming paradigm,
but also others (procedural/imperative, OO, &c).

The "IS" of identity is as inappropriate here as it tends
to be everywhere... when good support for other paradigms
such as O-O is added to Haskell, giving O'Haskell, does
this mean the result "isn't" a functional programming
language any more?

The "IS of identity" just doesn't work well with the
multifaceted nature of reality -- it implies taxonomies
everywhere, mutual-exclusion where no mutual-exclusion
actually obtains, incompatibility of things that actually
compate quite well (yeah I know 'compate' is not normally
considered a verb, but, given that 'compatible' "is" an
adjective, I claim 'compate' should also be OK!-).


More information about the Python-list mailing list