I just found this topic. Its been sitting here for the past four years. I've been studying phonosemantics (the connection between sound and meaning) in natural human languages for the past 40 years. For starters, sounds are often indexes (reliable signals) of various physiological or physicomechanical states. For example, if you hear a stone crunched between your teeth this is a reliable sign that something at least somewhat brittle has broken. Similarly, if you hear a floppy noise like a rug being laid out on a floor then you know that a relatively coherent surface has lost trapped air from beneath it as it settles down onto the surface. And a pop as of a balloon marks an explosion. Many languages have onomatopoeic words which try to take advantage of such sound-meaning linkages, but using a somewhat arbitrary code. This is why imitations such as of dogs barking vary so much between different languages- but not equally in every possible dimension of possible phonetic contrasts. In examining hundreds of languages, using relatively comprehensive dictionaries and also becoming at least somewhat familiar with grammar and word formation processes, over the past couple of decades, I discovered an interesting pattern. For languages that have CVC-type root structure (which can be more or less developed, with consonant clusters, or extra syllables, or reduced ones, etc.), it is usually the case that the first consonantal position has a semantic opposition with the last one. So for example if C1- is a labial stop, it will be far more likely than not that there will be some association, in state or action verb roots, that some sort of loss of containment, confinement, possession, and similar notions involving mass, energy, responsibility, etc. But if -C2 is a labial we find GAIN, or RETENTION, of same is being depicted. On the other hand if C1- is a velar stop, then GAIN is the main sense, but final -C2 as velar stop we find LOSS. The labial C1- and velar -C2 don't have identical senses on other scores- the labial initials are usually 'about' loss of relatively low value materials- stuff you DON'T really want or respect, and the loss is all-at-once, and generally unwelcome by others as well. Think intestinal digestive wastes, gas, etc. The C1- velars refer to relatively HIGH VALUE materials which have to be fought over since others (animate or inanimate) will attempt to retain them. So say you're digging hard rock to extract gemstones or precious metals. There is also a textural association- the C1-labials associate with material looseness, while the C1- velars associate with tightness and hardness. Animal signals may have the beginnings of similar mappings, but they haven't yet been demonstrated. The Motivation Structure Theory by Eugene Morton (he was an animal behavioral ethologist at the National Zoo for many years) took vocal signals form many different vertebrate species and renormalized them for time and frequency range, so they could be compared more easily. What he found was that there was a frequency and amplitude-based code for social interactions. One can deepen a voice to make oneself seem larger to intimidate rivals. Not all species can do this- it evolved. Frogs for example have honest voice pitch so that a rival knows exactly what to expect when competing for mates or territory from another conspecific. A voice can also be varied for how tonally smooth or noisily rough it is. The latter associates strongly with agitation of the speaker, and the former with calmness and self-assuredness. Loudness goes with emphasis, and quietness with understatement. These types of vocal variation are found across mammals and birds, and even to a lesser extent to some reptiles. While these vocal signals are generally limited to social statuses and interactions, some species also have warning signals which can differentiate between different types of predator threat. And apes have food calls which alert conspecifics to the availability of abundant food discovered by the caller. Often the caller will try to prevent him/herself from sending the call. They can't stop it from coming up out of their throats, so they put their hands over their mouth to keep it inside. Zoo experiments on captive apes have shown that the acoustic structures of food calls differentiate foods of different qualities such as size, shape, color, and nutritional quality (such as ripeness). My own research seems to indicate that this type of iconicity laid the basis for the rise of language, given that there is a strong correlation between the articulatory positions of phonemes in the oral cavity, and the masticational/deglutitional (chewing and swallowing) function of the mouth at those positions in mammals. The lips gather materials inwards and keep them contained. The tongue moves food bits about from one process to another, helps to separate more from less processed materials, and consolidates uniformly processed stuffs into a food bolus for swallowing, among other functions. The tooth types are differentiated for function as well. Incisors are nippers and pinchers. Canines pierce. Bicuspids (at least in primates) slice, and molars crush. These actions associate with the material properties of the food bits being processed between them. And the location of the taste buds on the tongue are also differentiated. Bitter taste is in the rear, adjacent to the molars, which crush hard materials which are often bitter. Sweet is in the front, where ripe fruit melts in your mouth and may only need to be pressed between lips and tongue surface. And so on.
If you try again with paragraphs it is possible somebody might read this.