“Computer Science as Empirical Inquiry: Symbols and Search” is the lecture Allen Newell and Herbert Simon delivered on receiving the 1975 ACM Turing Award, published in Communications of the ACM in March 1976 (volume 19, number 3). A full-text scan is available online. It is the clearest single statement of the philosophy that drove the first decades of artificial intelligence.
Its central claim is the physical symbol system hypothesis: that “a physical symbol system has the necessary and sufficient means for general intelligent action.” A physical symbol system is any system - a computer, in principle a brain - that builds and transforms structures made of symbols. Necessary means nothing can be generally intelligent without being such a system; sufficient means that being one is enough, given the right organization. Newell and Simon offered this not as a theorem but as an empirical hypothesis, a claim about the world to be tested by building systems and seeing what they can do.
The second half of the lecture argues that intelligence, in practice, comes down to search: an intelligent system represents a problem as a space of possible states and searches through it, using heuristics to avoid examining the astronomically many possibilities exhaustively. They pointed to their own programs - the Logic Theorist and the General Problem Solver - as evidence that symbol manipulation plus heuristic search could reproduce genuine problem-solving behavior.
The paper defined a research program. For roughly thirty years, “symbols and search” was what artificial intelligence largely meant, and the physical symbol system hypothesis was the bet underneath it. Later critics - Dreyfus from philosophy, the connectionists from engineering, the embodied-AI movement from robotics - each attacked a different part of that bet, which is why this lecture remains the reference point for what they were arguing against.