This paper presents a model of the human auditory system's front-end signal processing. The model is biologically plausible and provides simple explanations for a wide variety of psychoacoustic effects. It is proposed that the auditory system evolved as a threat-warning receiver, long before the development of speech. This threat-warning receiver was subsequently expropriated for use as a communications receiver. It functions primarily as an AM and FM, multi-tone demodulator. FM information is derived from the AM in a manner similar to that by which the eye derives color information. Many of the peculiar characteristics of speech signaling evolved in response to the problems encountered while attempting to use this demodulator to process information transmitted through communication channels exhibiting high-levels of both multi-path and multi-source interference. Similar problems are encountered in designing communications systems that exploit high-frequency (HF) ionospheric channels. It is not surprising then, that many analogies exist between the structure of speech signals and certain types of HF modem signals. It is proposed that these analogies are not coincidental; they reflect a common set of solutions to a common set of problems. Computer simulations have confirmed that good quality speech can be reconstructed from the model's outputs; the model may be thought of as a special form of a harmonic speech coder, designed for use in high noise/interference environments.