Meaning Explained (Finally)
It’s really very strange that we can’t say what meaning is. Surely, we know what we mean! Meaning is a mental act and we know our mental acts, don’t we? Yet all attempts hitherto have foundered, often embarrassingly so. Mental images, sensations, definite descriptions, objects in the world, rules of use, behavioral dispositions, words in the language of thought, infusions from above (okay, I made that one up)—all these theories have come and gone, covered in shame. One might therefore suppose that the answer must be unobvious, hidden, a matter of far-out conjecture, because nothing on the surface does the trick (or else we go eliminative—the “myth of meaning”). We might even stipulate it as a condition of adequacy that the nature of meaning should be something far away from our common conceptions (compare consciousness). It should be something not evident to ordinary thought—a bit like the chemical structure of the genes. Meaning is clearly somewhat of a mystery, so we would expect that some ingenuity is going to be required in order to get it right, and some adjustment of theoretical expectations. We might have to reconfigure what we think a theory of meaning should look like—its methodology and structure. The key concept may not be what we have been led to expect; paradigms might shift under our feet (like tectonic plates). In other words, prepare to be shocked, even outraged (like being told the center of the universe is not the earth).
Let’s go back to basics. Take a simple proper name, say one that belongs to a family member or close acquaintance; and imagine the early days of human language. You make a sound, or perhaps a hand gesture, in hopes of securing reference to someone. This (putative) name is supposed to “stand for” a certain individual. How does it do so? Well, we might imagine a pre-linguistic background: up to now you have impersonated or mimicked the individual in question, and you are quite talented in this regard. You perform certain actions intended to resemble the individual; this will cause your audience (the term is apt) to bring that person to mind. If you utter the would-be name and get looks of incomprehension, you might trot out your little impression, thus securing uptake. Your hearers will then remember your act and associate it with the sound or gesture you produce. The reference of your utterance is the person you impersonated. It is in virtue of this association that the name means what it does in your close-knit community. If you are a good mimic, the association will stick, perhaps eliciting mild amusement. Here we see the germ of what I will boldly call the impersonation theory of names. Names mean (express) the impersonation they are linked to: not knowledge of a description but an ability to mimic. This is the psychological background to the institution of naming; one could write an article called “Naming and Mimicry”. We might loosely say that names are impersonations of their bearers—they evoke memories of such impersonations, or actual impersonations. People use them so as to avoid having to do an impersonation—they are “short for” such impersonations. These names are synonymous with impersonations. They “stand in” for impersonations. Impersonations symbolize the individual impersonated, and names partake of this symbolic power. Names rely on an underlying psychological capacity—the capacity to copy or imitate other people.[1] This capacity predated names and language generally, and indeed goes back to primitive capacities to mimic (like butterflies that mimic other butterflies because those others are poisonous to predators). Mimicry is common in the animal world and would no doubt have been present in pre-linguistic humans. It provides the foundation of meaning in simple cases. Language doesn’t picture objects; it impersonates them. Mimicry is the cradle of reference and hence meaning. It is the “cognitive architecture” on which meaning rests—what the meaning mind must be like in its deep structure. This structure is hidden, as promised, and it will take some persuasion on my part to get the theory accepted as a general theory of meaning.
I think that if it works for names of persons it will work for other words, because it provides necessary and sufficient conditions for meaning in one area of language—and meaning is uniform. But how? Here I must be brief. Consider names of places: these can be handled by invoking the notion of personification and extending impersonation to geographic formations (e.g., imitating the shape of the British Isles or the manners of its people). Names of shapes and colors are explained by the mimicking of shapes with the hands and pointing to color samples so as to provide a performance denoting the color intended (the sample mimics the color). Consider what people do when trying to communicate with others in the absence of a common language: they put on a theatrical performance attempting to mimic what they intend to communicate—for example, imitating someone eating to indicate hunger. They use the common language of mankind embodied in acts of mimicry. In this way facts can be impersonated not just objects—say, an accident witnessed. This repertoire of skills underlies the communication achieved by sounds and marks. Linguistic meaning piggybacks on this. It is the machinery of meaning, supplemented by sensorimotor skills, memory, etc.
There is a further component to be added to this theory, viz. the social nature of meaning. It may be objected that many users of language cannot impersonate the things they mean and don’t know anyone who can. Here we must invoke causal chains, experts, the division of linguistic labor, semantic deference, and linguistic history—the whole social web in which meaning occurs. Impersonation is essential at some point, but it need not be available to every speaker; we must allow for derived meaning, parasitic meaning. We can also bring in analogy, metaphor, intelligible extensions (impersonating a number with the fingers or a moral value with a facial expression). Human language is complex and made up of many things; I am only considering the basic mechanisms here—the deep roots of meaning. The great advantage of this theory is that it views meaning as a special case of something more general and antecedent, something practical and bodily, where it belongs. We can easily imagine communicative acts, in humans and animals, that involve the impersonation of dangers, like predators or precipices; and meaning can spring from these elementary beginnings (it has to spring from somewhere). The basic idea is to derive one kind of symbolism from another—meaningful words from imitative acts. Syntax no doubt stems from somewhere else, but the imbuing of meaning is a matter of imitative symbolism—meaning from mime. It is mime extended, elaborated, and attenuated. If butterflies had a language, it would be closely tied to their imitative wing coloration (we might even say that this is their language—“Keep away if you know what’s good for you!”). Nature is actually highly communicative, mainly by dint of imitation; language is one form of this general trait (plus some). Meaning is what happens when syntax meets impersonation in pragmatic acts of speech. For impersonation (imitation, mimicry) takes us from the thing doing the impersonating to the thing being impersonated—from the individual to the environment. Imitation is the root of reference. The other theories fail, in various ways, to achieve this result.
Humans are great imitators, the best of the best: it’s how we learn. The child in learning to speak imitates his or her elders, remarkably well. It is as if we are all born ready to imitate those around us, professionally. And we can imitate many things—noises, movements, facial expressions. Clearly, imitation comes into language acquisition; according to the impersonation theory, it also comes into the creation of meaning. It is odd that Wittgenstein never mentions this fact of human “natural history”, along with “walking, eating, drinking, playing” (PI 25). It is part of our “form of life”—the “imitation games” we play. Imitation clearly has much in common with speaking; sign language makes this particularly obvious. In the Tractatus meaning is held to consist in picturing, but picturing is a form of imitation. He could have preserved this insight in the Investigations by invoking natural mimicry instead of pictorial geometry; then he would have had an imitation theory of the basis of meaning. Would this be vulnerable to the semantic skeptic? It is hard to see how: the impersonation relation is not indeterminate. Language may not be all onomatopoeia (pure impersonation), but it’s not far off. Speech acts are acts of impersonation, or rely on such acts as a pre-condition. The linguistic brain is really an expert mimic. To understand a sentence is to know what imitative acts would convey its meaning, or to be suitably connected to someone who knows that. Other animals are limited in their powers of imitation (mentally or physically), so they are unsuited to be speakers except to a very limited extent. Humans, though, have an elaborate imitation module in their head, which they put to work when they speak and understand speech. We are certainly extremely good at interpreting impersonation and other forms of imitation; we can see it instantly without thinking. To be impersonation-blind is to be virtually subhuman (I doubt that other animals impersonate each other as we do). Our distinctive sense of humor is bound up with it (jokes often involve impersonation). We have a language-imitation-humor mental faculty. Meaning arises from this suite of capacities and acts back on them. Could there be a language without the possibility of jokes?[2]
[1] You might object that normal human powers of impersonation only extend to a few people, whereas we are typically masters of many names. I reply that probably names were first introduced only within families so that only a few needed to be backed by a distinctive impersonation (see below on the role of social interactions in extending the range of names).
[2] You see what I mean about paradigm shifts and surprising surmises; this is a far cry from formalistic denotation and connotation theories of what meaning ultimately consists in. Meaning is part of human nature in the round not an isolated formal system. It would be hard to develop a scientific theory of it, still less a mathematical theory. It is biologically messy. This is not ordinary language philosophy but ordinary people philosophy—a species of animal indeed. Not Frege but Darwin.
