The surprising predictability of source code has triggered a boom in tools using language models for code. Code is much more predictable than natural language, but the reasons are not well understood. We propose a dual channel view of code; code combines a formal channel for specifying execution and a natural language channel in the form of identifiers and comments that assists human comprehension.
Computers ignore the natural language channel, but developers read both and, when writing code for longterm use and maintenance, consider each channel’s audience: computer and human. As developers hold both channels in mind, we advance the \emph{dual channel hypothesis}: the two channels interact and constrain each other. If true, this hypothesis will overturn current, standard practice of considering only the formal channel, or, if both channels, each in isolation.
We describe how the constraints of this dual audience setting can lead to humans writing code in a way more predictable than natural language, highlight pioneering research that has implicitly or explicitly used parts of this theory, and drive new research, such as systematically searching for cross-channel inconsistencies. The dual channel hypothesis provides an exciting opportunity as truly multi-disciplinary research; for computer scientists it promises improvements to program analysis via a more holistic approach to code, and to psycholinguists it promises a novel environment for studying linguistic processes.