By Stephen T. Asma

No self-respecting professor of philosophy wants to discuss the soul in class. It reeks of old-time theology, or, worse, New Age quantum treacle. The soul has been a dead end in philosophy ever since the positivists unmasked its empty referential center. Scientific philosophy has shown us that there's no there there.

But make no mistake, our students are very interested in the soul. In fact, that is the main reason many of us won't raise the soul issue in our classes: The bizarre, speculative, spooky metaphysics that pours out of students, once the box has been opened, is truly chaotic and depressing. The class is a tinderbox of weird pet theories—divine vapors, God particles, reincarnation, astral projections, auras, ghosts—and mere mention of the soul is like a spark that sets off dozens of combustions. Trying to put out all these fires with calm, cool rationality is exhausting and unsuccessful.

Lately, perhaps sparked by Dan Brown's best seller The Lost Symbol, I have had to repeatedly extinguish confident student dogma assuring me that "noetic science" has "proven" the existence of the soul. Since the early 1900s, a handful of marginal experimenters have tried to weigh the soul—by arranging dying people on scales and taking their weight before and after the moment of death. Nothing even vaguely suggestive was discovered by that experimental approach, except a very high degree of wishful thinking. One humorous and underreported "finding," made by an Oregon sheep rancher and earnest amateur scientist, was the discovery that sheep actually gain a little weight as they die. It's hard to know where to start with all this.

Even if we could show that some energy was leaving our bodies at the moment of death, it can't really be a surprise, since thermodynamics tells us that energy is always being exchanged through physical systems. When I die, the slowing of my thermodynamic processes will become irreversible; my local entropy will increase. When I die, my energy will go on. But, of course, we can't get too excited by that fact, since we're talking only about heat and the chemical transformation of my decaying flesh, taken up and conserved in new organisms and physical systems. The conservation of energy doesn't give us any conservation of consciousness or any continuation of personal identity. And personal continuity is the hope for most soul proponents.

But if we could set aside all the problems with these badly controlled and executed experiments—if we could create highly precise measurements—we would still have the more challenging issue of coherence. Most people's concept of the soul includes the idea that it is incorporeal or immaterial (this is how the religious traditions have conceptualized it), so how could an incorporeal entity have any weight or mass or volume, any of the spatial properties we assign to matter? Thinking that the soul has weight seems like a category mistake—like saying the number 4 weighs 30 pounds, or the color blue smells bad. Weighing the soul, or searching for the soul in the brain, seems like a similar mistake.

Science seems entirely justified in its soul skepticism. But if such speculative metaphysics is bracketed out of science, then what is left of the soul issue? What remains of soul talk? Is it merely folk language that has been replaced by more-accurate descriptions of the human experience?

One response is for believers to rush headlong into a faith-based rejection of rationality and just hold fast to the traditional soul idea; another is to give it a New Age paint job with quantum-energy talk. Our students are very enticed by that response, partly because they see no other avenue for preserving their meaningful soul language. But lately I have been offering them a fresh alternative.

Instead of asking whether we can verify the soul's existence—find some empirical evidence for it—I suggest a Wittgensteinian approach. Following the Austrian philosopher, I ask: How do people actually talk about the soul? How is soul talk used in ordinary language? And here we find that the soul is alive and well in certain kinds of expressive language. When you look at actual soul talk, you find the following kinds of expressions: "He is my soul mate," or "She really sold her soul," or "That's good soul food," or "This nature hike is good for my soul," or "She is an old soul," or "James Brown has soul," or "The soul reincarnates," or "Her soul is in heaven now."

Those expressions share little similarity. Like Wittgenstein's famous example of diverse "games," they probably represent a family resemblance of meanings rather than a common essential definition. Notice, for example, that only the last two expressions have any metaphysical connotations.

But more important, the expressions are not really propositions about the world. They express emotional attitudes and resemble other kinds of imperative or aspirational speech, like, "You go, girl!," or "Don't do that," or "Have a nice trip," or "I got soul, and I'm superbad." When I say "You've got soul," it's not a description of some factual state of affairs; rather it is an evaluation. It expresses as much about the subject as the object referenced. We cannot expunge the subjective expressive/evaluative properties from the sentence and arrive at some testable proposition (as in science). Saying "James Brown has soul" is nothing like saying "The cat is on the mat" or "Water freezes at 32 degrees" or "The amygdala plays a large role in memory." Those are all testable propositions.

Soul talk is expressive in the same way as other nondescriptive utterances, like "oh my God" or "ouch" or "yuck" or (with head nodding to music) "Yeah, that's funky." There is no clear referent for those. They don't seem to refer to or represent anything—they seem somehow pre-representational (or presentational). Soul talk, like other emotive talk, bears little relation to the goals of scientific language, and probably can't be assessed with that language. Like other expressive forms, soul talk in ordinary folk language won't have much theoretical interest, because it is rarely, if ever, trying to explain a phenomenon. In the same way that a poem is not trying to explain a phenomenon, soul talk is equally uninterested in induction, hypothesis, prediction, and corroboration. Instead, soul talk tries to express our hopes and aspirations ("I hope I see my family again in the afterlife") or to identify inspiration ("This song really speaks to my soul"), or to express feelings deeper than friendship ("I've finally found my soul mate"), or to scare people into doing something ("Your soul will burn in hellfire"), and so on.

Moreover, the meaning of soul talk should not be searched for in the correspondence theory of truth. When I try to establish the meaning and the truth of the proposition "The cat is on the mat," I attempt to find a correspondence between my word "cat" and the actual feline animal, and my word "on" and the actual spatial relation of said cat to mat, and so on. I'm looking for a correspondence between propositions and the external world. In that way I can verify the meaning and truth of my proposition.

But the sentences "James Brown has soul" and "My soul is anchored in the Lord" rely on a very different system of meaning—they don't correspond to anything particularly. Instead they take their meaning from a coherence they have with other terms, concepts, values, connotations, and associations. "This song has soul" means: This music restores us, this music has integrity, there's something authentic and natural in its style, this music contains strong emotion, the repetition is hypnotic or ecstatic, there are elements of the African-American experience in this music and these lyrics, this song draws on gospel and R&B genres, this song is so funky you can smell it, and so on. That is the matrix of connotations that make up the context of soul talk—and the soul talk is coherent to the extent that it coheres in some way with all these other experiences and meanings. In that sense, the soul is meaningful to many of us without any scientific verification of its existence.

That is not the same as just having faith in the soul despite a lack of evidence. I'm not suggesting that familiar view. What I'm suggesting is more sly—the soul can be deeply meaningful whether it exists or not, and it can be deeply meaningful even if you disbelieve in its literal, metaphysical existence. That is not the usefulness of fictions and delusions. It's the usefulness of an expressive folk language that can't be replaced by a scientific language.

So why is soul talk still meaningful, and why can't it be replaced?

If we think about the human being, we can analyze ourselves into various parts and functions: the body, cognition, emotions, memory, perception, and so on. And we can make many impressive scientific claims about those parts and functions. Modern medicine is a testament to the genius of methodological materialism and a mechanical approach to the human being.

But in this matrix of human thoughts, feelings, and experiences, we also find forms of awareness and activity that call out for a different language. The kinds of awareness I'm thinking of might be described as aesthetic—feelings of ecstasy, feelings for the beautiful or the sublime, poignant stirrings that might be labeled transcendent—or, negatively, feelings of horror or dread. And the kinds of activities I'm trying to isolate might be creative acts (playing music, writing poetry, handcrafting furniture, serving tea while a Zen master whacks you with a stick) as well as ethical activities (acts of altruism, self-sacrifice). It's hard to see how a purely descriptive scientific language can find good traction in those domains, but an alternative language exists and has existed for a long time. Soul talk is a part of that successful expressive language.

Philosophers like Wittgenstein, Heidegger, and Kenneth Burke even went so far as to suggest that language is originally expressive, rhetorical, and dramatic, and only derivatively descriptive, scientific, and explanatory. If that is true, then soul talk is a part of that primordial language.

Wittgenstein's focus on ordinary language shows where we can preserve intelligent soul talk but avoid common category mistakes and tendencies toward reification. We can "debug" soul talk. We can detach it from its now unwarranted metaphysical history—and we see this already in ordinary language when we say, "That singer has soul" or "This nature hike is good for my soul."

But our tendency to turn this soul language into metaphysics is strong—Wittgenstein said that sometimes "language goes on holiday," and that we have to coax it back to its useful, functional meaning. Just as I don't hear a smell or taste a color, I also don't literally "live after death," or have a "soul mate." Those are perfectly good metaphorical uses of language, but they shouldn't be confused with literal descriptive uses of language. When I say, for example, "My soul will go on," I'm probably really saying, "I hope I live more." And when we've arrived at that naked expression of subjective yearning, then we've probably reached the end of our analysis. We're done understanding it.

The problem with some religious and New Age soul talk is that it exports the soul concept from the domain of subjective expression to the domain of objective fact, where it can have no empirical corroboration. That is the main category mistake.

Many atheists, like Richard Dawkins, will criticize soul believers as dimwits. And that is not my position. Everybody makes category mistakes, and everybody confuses subjective yearning and hope with objective matters of fact. Even the phrase "He is a dimwit" is just an expressive claim masquerading as a descriptive claim.

Once you take the metaphysics out of the language of soul, you begin to see how the soul is used in social contexts of ordinary language. When a minister tells parents at their son's funeral that they will see their son again, and his soul is in a better place, I cannot dismiss it or heap scorn on it. If we professors hear this language as a description of reality, then we're bound to be irritated by the issue of truant evidence and the lack of warrant. But if we hear it as emotive hope, then our objections fall away. The students in my class are right to want to hold on to this language. Metaphysics aside, the minister's language seems to suggest that there are emotions so deep and bonds so strong that not even death should end them. That is a beautiful sentiment no matter what you think of the soul.

Stephen T. Asma is a professor of philosophy, and a fellow in the School of Liberal Arts and Sciences Research Group in Mind, Science, and Culture, at Columbia College Chicago. His books include On Monsters: An Unnatural History of Our Worst Fears (Oxford University Press, 2009) and Why I Am a Buddhist (Hampton Roads Publishing Company, 2010).