answer is in the guide and is quite interesting. so, lisp development came in two phases: formal phase -- that's where original 1960 paper described lisp from

answer is in the guide and is quite interesting. so, lisp development came in two phases: formal phase -- that's where original 1960 paper described lisp from simple axioms and built on them -- and implementation phase -- formal step is of no use for us because it didn't have numbers, error handling, i/o, etc. argument here is that the formal phase might be the most important phase, but that the second phase usually takes longer and is more practical. so what if one delays the second phase for as long as possible? what could be discovered and be useful for the implementation phase? that's the raison d'etre of bel i think. i very much like it :)