“Computer Science as Empirical Inquiry: Symbols and Search” was the tenth ACM Turing Award lecture, delivered by Allen Newell and Herbert A. Simon and published in Communications of the ACM, volume 19, number 3, in March 1976 (page 113). It is the clearest single statement of the philosophy that drove classical, symbolic AI, written by two of its founders shortly after they received the field’s highest honor.
The lecture advances two empirical hypotheses. The first is the physical symbol system hypothesis: a physical symbol system, meaning a machine that can create, manipulate, and combine symbol structures, has the necessary and sufficient means for general intelligent action. In other words, Newell and Simon claim that intelligence does not require anything beyond the ability to process symbols, and that any system intelligent in the general sense will turn out to be such a symbol system. The second is the heuristic search hypothesis: such systems solve problems by generating and progressively modifying symbol structures, searching a space of possibilities, and the solutions emerge from that search guided by heuristics that keep it from exploding. They present their own programs, the Logic Theorist and the General Problem Solver, as empirical evidence.
The lecture also frames computer science itself as an empirical science, in which programs are experiments and hypotheses about intelligence are tested by building and running systems. This stance, and especially the symbol system hypothesis, became the defining doctrine that later connectionist and embodied-AI researchers would push against.
Why a business reader should care: this is the foundational argument that thinking can be reduced to symbol manipulation and search, the assumption underneath every rule-based system and planner, and understanding it clarifies both the power and the eventual limits of the symbolic approach that dominated AI’s first decades.