We convey how a computational process should be built to "reason", successfully and consequentially, approximately its personal inferential procedures. The research proceeds in elements. First, we give some thought to the overall query of computational semantics, rejecting conventional techniques, and arguing that the declarative and procedural facets of computational symbols (what they stand for, and what habit they engender) might be analyzed independently, so that they might be coherently comparable. moment, we examine self-referential habit in computational procedures, and convey easy methods to embed a good procedural version of a computational calculus inside of that calculus (a version no longer not like meta-circular interpreter, yet hooked up to the elemental operations of the computing device in this type of approach as to supply, at any element in a computation, absolutely articulated descriptions of the kingdom of that computation, for inspection and attainable modification). when it comes to the theories that consequence from those investigations, we current a normal structure for procedurally reflective approaches, capable of shift easily among facing a given topic area, and working with their very own reasoning approaches over that area. An example of the overall resolution is labored out within the context of an applicative language. particularly, we current 3 successive dialects of LISP: 1-LISP, a distillation of present perform, for comparability reasons; 2-LISP, a dialect built when it comes to our rationalized semantics, within which the concept that of assessment is rejected in want of autonomous notions of simplification and reference, and within which the respective different types of notation, constitution, semantics, and behaviour are strictly aligned; and 3-LISP, an extension of 2-LISP endowed with reflective powers.