The foolishness of "natural language programming" (1978)

The article discusses the challenges of natural language programming, a concept that first emerged in the early days of automatic computing. Proponents argued that using a machine's native tongue would make it easier to use, as it would reduce the need for formal symbolism and the associated care and accuracy. However, the author argues that this is not a valid assumption.

The author points out that changing interfaces can increase the workload on both sides, a point supported by historical examples from the development of mathematics. Greek mathematics became stagnant due to its verbal and pictorial nature, while Islamic algebra failed after returning to rhetoric. Modern civilization emerged only when Western Europe broke free from medieval scholasticism, which favored verbal precision through formal symbolism.

The author argues that formal texts are an effective tool for ruling out nonsense, unlike natural languages, which are prone to errors.

The author suggests that natural language programming would require designing a machine capable of understanding nonsensical instructions, increasing the machine's burden and making it a challenging task. Furthermore, the author notes that the Western world has seen a decline in people's mastery of their native language, a phenomenon known as "The New Illiteracy."

He believes that this decline supports the argument against natural language programming, as it highlights the difficulty of using natural languages for precise instructions. Overall, the author concludes that natural language programming is not a feasible solution and that formal symbolism remains the most effective tool for programming.