Essay 37. On the Soul

For some reason I am still interested in the words that for millennia had been as common and clear-cut terms of everyday speech as horse, bread, and fire, before they retired to theology and philosophy. If we use them, they mean something.

As if the subject of fate was not enough (Essay 36, On Fatalism), I am picking up another phantom from the same Addams family. The soul is so vague a concept, spread over so many meanings, that it seems just a figure of speech, even in a religious context.

Hard science has neither interest in the soul nor a place for it. Only in popular discourses on Artificial Intelligence (AI) and in related journalism the word is sometimes used as a shortcut to the property of being recognized by other humans as human. Traditionally, the soul was the term for what distinguished the human from the plant, animal, machine, and thing. The so-called strong AI extends the privilege to advanced machines, which could be built in the future.

The over fifty year old debate around the question whether machines can have mind and soul is still smoldering. The Mind's I: Fantasies and Reflections on Self and Soul by Douglas R. Hofstadter and Daniel C. Dennet (New York: Bantam Books, 1981) was a landmark anthology of science and fiction views on the subject. Can we distinguish between a real thing and its exact simulation, or, as Douglas Hofstadter commented on a sci-fi story, "what is the difference between a simulated song and a real song?" We can substitute soul for song in this question.

Can an artificial person be created? Could we treat it is equal? Will it have a soul?
The discussion on the ultimate possibilities of artificial intelligence in reproducing human nature is over half a century old. In the enormous literature, a few sources have the word soul in the title, others in the text. The sci-fi movies, like the film AI by Steven Spielberg, carry the banner on.

The witty shortcut sci-phi (J.D.Casnig; his remarkable site is now under renovation at http://knowgramming.com) is very appropriate for the whole area of modern philosophy of science where AI is only part of the discussion on the subjects not verifiable by experiment at the present time. Mental constructs, however, can be tested by logic. The arguments are about axioms and terms. The volcanic activity in sci-phi testifies that our understanding of such old words as life and mind changes: larger categories take shape.

History stores the record of our changing attitude toward "the savages," first hunted like animals and brought as zoo exhibits to Europe, later hunted for domestication as slaves, but then moved to freedom through the Underground Railroad and later elected to US Congress. Each time I see a movie about "primitive cultures," and especially about the first contacts with remote tribes, I cannot notice anything that would suggest any inborn divide between them and us. It was religion that first recognized the human soul in them, paradoxically, judging by the appearance and Natural Intelligence and not by culture. Similarly, our attitude toward our electronic creations of a very different appearance may change with time, facts, and evolution, as it has been changing regarding the whales and elephants that have a civilizing influence over us.

I am circumventing the discussions around Artificial Intelligence here not only because the debating sides do not give a definition of the soul. As a chemist, I pay little attention to the distinction between the Natural and the Artificial. Of course, there is no difference between two objects meeting the same criteria, as there is no difference between the natural and synthetic versions of vitamin C.

A pure individual chemical compound does not carry a tag certifying its origin. This is one of little appreciated laws of chemistry: the law of constant composition, first formulated by Joseph Louis Proust in 1794. It says that the composition of a pure compound does not depend on its origin (i.e., natural or artificial or made by a particular person at a particular place), which implies that neither do its properties. The question is: what is the pure and individual subject of our discussion? Curiously, the same question arises in logic: are we talking about the same subject, do we change it along the way?

Soul—chemistry—logic: could there be a stranger trio? An exciting choice for Essays Montaignesque; let us keep the two outsiders in mind.

I want to understand what it means to have a soul.

Meaning evolves as anything else. The words may walk on the surface of the earth but then decide to crawl into deep caves or even sink to the ocean floor. The meaning and the connotations of the word horse have changed, and so has the usage of the words honor, virtue, and nobility, which are now stored in the social memory of the advanced industrial state side by side with quill, crinoline, typewriter, and telephone switchboard.

In our world of man-made Things the humans are turning into enzymes in complex metabolic webs where the turnover of money, more important than the alternation of day, night, and seasons, brings the crop of products for sale from the social soil tilled by social machines under the artificial sun of burning mineral fuel.

While our human nature still holds well under the attack of stress, artificial chemicals, and the accumulation of genetic defects (I swear, I am not a social critic, please), we are starting to pay attention to the suspicious changes in our social biochemistry. I remember well how, before the advent of molecular biology, serious people believed that some new and unknown principle could be hidden in the phenomenon of life. Experimental science put an end to all such expectations. What about the soul? Entering this new world—which is of course just the next stage in the evolution of the old world—we might reconsider the meaning of some old words. As we may need a set of new terms to understand and describe the modernity, the old-modern words could be as good as the derivatives of the classical Greek and Latin. We would do with a prefix: meta-life, meta-mind, meta-soul.

Pushing aside religion, philosophy, social psychology, and AI, I am turning to Aristotle, who not only established criteria of the purity of thinking, still used today, but also left us a relatively short book On the Soul where he attempted to look at the subject, using only the powers of logic and suppressing belief, emotion, and fantasy.

Why might we need to read Aristotle who lived twenty-three centuries ago? Because we do not need to. It is a useless and impractical waste of time, a futile indulgence, which is exactly what separates us from the dapper and efficient machines. We cannot learn anything from him that would help us with work, wealth, business, research, love life, health, and beauty. Aristotle's writings are dry and, with the exception of logic, hopelessly obsolete. They serve only as the material for an occasional student of philosophy and history of science to write a thesis and climb the next career step.

Not only Aristotle but also the soul itself is beyond any practical use, utilitarian benefit, and instrumentality. Nevertheless, reading Aristotle does something to the soul of the reader who is aware that our view of the world grew from some pots on Aristotle's windowsill. Aristotle purifies the muddled soul and the mind, but if it too sterile, Aristotle spreads germs of new ideas in it .

Aristotle was at the initial building and furnishing of some most important compartments of our civilization: logic, science, art, and ethics. Most important, Aristotle, together with his teacher Plato, was the architect of the Western cult of unrestricted questions and answers. Aristotle is a whole planet and his boring and complicated texts look like a landscape of majestic cosmic beauty, which could be an intense pleasure to visit and, refreshed by a diversion, return to the familiar health, love, and money worries.

Of all our faculties, the soul is the least needed to earn a living. We cannot even sell our own soul to the extremely difficult to reach devil who is busy with other things and probably would not give a damn for it. Whether we have souls or not, whether they are immortal or die with us, and whether the heaven or the hell is their final destination is of no relevance for any practical matter in the modern world. And yet long before Aristotle and up to modern times, the fate of the soul (I have caught up the ghostly couple together!) has been a matter of big concern for many people, and, as Max Weber thought, even a motivation for the development of the capitalist way of production. What an irony: the capitalism of the third millennium, allegedly born from the Protestant ethics, is as much about the soul as entomology about whales.

Disinterested in the religious and ethical views of the soul, I am nevertheless fascinated by the questions: What does it mean to have a soul in our times and what does it mean to lose it? Is there any rational interpretation of the soul, one of the most
Losing our souls ancient and irrational creations of human mind?

Believers or not, we stick to the soul as a metaphor. Not too often, but one can run into a completely secular question "Are we losing our souls?" (search losing our souls with Google; only 558 hits in July, 2002: remarkably little concern!) on printed (see the book cover on the left) and electronic pages.

NOTE: the soul gained and lost
artificial intelligence as a philosophical project
Philip E. Agre

EXAMPLE: "Persons in commodified relationships are there to 'serve' or 'perform,'" [Jeremy] Rifkin writes. In this environment, what happens to empathy? What happens to the individual soul in relation to other souls? (See APPENDIX 11)

But every metaphor is a connection (transfer) between two objects. Is there anything tangible behind the soul or is it just an echo of the word? I am looking for the place of the idea of the soul in the changing system of our civilization. Is the soul a legitimate dimension of the process? If so, the soul is not of the all-or-nothing kind. Are we really losing our souls? If yes, to whom? The machines can have a mind of their own, but can they have soul? Does super-strong AI ("besouled") make sense?

In this Essay I deliberately limit myself to Aristotle as the only literary source. It seems to me that his De Anima (widely represented on the Web) provides an interpretation on his own terms. In the sci-phi forum, Aristotle has as much to say as anybody else.

NOTE. An excellent study of the problem of the soul in Aristotle's De Anima by Marian Hillar is available on the Web. In the world of information, it is the body, the hard copy, which is practically immortal. The weightless electronic information is as mortal as heavy human flesh. Fortunately, Marian Hillar's work is published in Contributors to the Philosophy of Humanism, M. Hillar and F. Prahl, eds, Humanists of Houston, Houston, 1994, pp. 51-82. His personality and works on humanism (for example, on universal ethics) deserve independent attention.

Evidently, the subject of the soul was difficult for Aristotle.

To attain any assured knowledge about the soul is one of the most difficult things in the world

The soul is so much unlike anything else that Aristotle discusses the method of study at a great length and often, short of rigorous logic, uses comparison, parallel, analogy, and metaphor. The reason for this is easy to see: the soul has no larger category to fall into. It is what remains in life if we subtract from it the observable material body. In the end, Aristotle takes the only possible secular way. He simply lists all aspects and species of phenomena comprised by the vague notion of the soul, as if defining the concept of the animal from all particular species of animals. His book is traditionally entitled in Latin De Anima, but if we remember that the soul in Greek is

, psyche (or psuche), the subject of Aristotle looks the same as that of modern psychology, only against a wider biological background. Classification and analysis is where Aristotle feels at home. Analysis, unlike synthesis, never generates chimeras: it dismembers them.

Today practically all the elements and blocks into which Aristotle decomposed "the soul" belong to established areas of knowledge: biology, physiology, psychology, social psychology and Artificial Intelligence. Having completed the analysis of the soul, Aristotle did not find any mystery. And yet reading De Anima, I had a feeling of the great mind's tension, struggle, and dissatisfaction, and it prompted me to look for something else. The problem for Aristotle was that while everything was clear about different parts of the soul, i.e., observable functions of the living organism and its mind, and the whole did not have any other function but life itself. The soul was just a sum of its parts. The abstract notion of soul was empty and shallow because it was circular:

From all this it follows that soul is an actuality or formulable essence of something that possesses a potentiality of being besouled.

For Aristotle, the soul was the set of all faculties of life, starting from the lowest and adding up. For example,

The soul of animals is characterized by two faculties, (a) the faculty of discrimination which is the work of thought and sense, and (b) the faculty of originating local movement.

The plants have the nutritive faculty, and so the animals and humans have it, too. The faculties of the soul, therefore, form a pyramid of a kind, with plants at the foundation and humans at the top. I have an impression, however, that Aristotle pondered on the possibility that even inanimate things could form the foundation of the pyramid:

Suppose that what is literally an 'organ', like an ax, were a natural body, its 'essential whatness', would have been its essence, and so its soul; if this disappeared from it, it would have ceased to be an ax, except in name.

Yes, let's suppose that for a moment: not an ax but a robot..

A possible interpretation of the Aristotelian idea of the soul can be found in his metaphoric explanation:

It follows that the soul is analogous to the hand; for as the hand is a tool of tools, so the mind is the form of forms and sense the form of sensible things.

The hand is the tool of tools because it can manipulate and use any tool, including an unfamiliar one. The mind is the form of forms, for example, because it can perceive the meaning of many verbal expressions, images, sounds, etc. The sense, such as vision, is capable of perceiving any visual image, not necessarily understanding it. Hearing perceives all sounds, etc.

It seems to me that in the above quotation Aristotle took some liberties with analogy. He says that the soul is analogous to the hand but further he takes only the parts of the soul, as if speaking about the hand he meant only its fingers. The complete analogy should be: the soul is analogous to the hand; for as the hand is a tool of tools, so the soul is .... is what?

Aristotle refuses to give a general definition of the soul other than in terms of its parts.

It is evident that the way to give the most adequate definition of soul is to seek in the case of each of its forms for the most appropriate definition.

Aristotle understood—it is only my guess—that, regarding the soul as a whole, he would end up in a vicious cycle: soul is soul, as life is life. And this is true about modern science, where there is a general and detailed understanding of what life is, but no satisfactory definition of life, and, for that matter, of energy, either.

The closest modern translation of the term soul in Aristotle is bio-life, which comprises all general functions of the body.

It is a fact of observation that plants and certain insects go on living when divided into segments; this means that each of the segments has a soul in it identical in species, though not numerically identical in the different segments, for both of the segments for a time possess the power of sensation and local movement. That this does not last is not surprising, for they no longer possess the organs necessary for self-maintenance. But, all the same, in each of the bodily parts there are present all the parts of soul, and the souls so present are homogeneous with one another and with the whole; this means that the several parts of the soul are indisseverable from one another, although the whole soul is divisible.

This remarkable paragraph becomes completely modern if we substitute life for soul and a life function for a part of the soul. Life divides and multiplies, while its functions are indivisible and we cannot have hearing without breathing. There is no place for the separate faculty of having a soul in the scientific and rational picture of a human being. Taking life apart, we find no such part as the soul per se.

Nevertheless, a consistent version of Aristotle's analogy would look as:

As the hand is a tool of tools, the soul is the X of Xs.

Aristotle's formal logic did not admit self-reference. But we can attempt it:

As the hand is a tool of tools, the soul is the soul of souls.

Aristotle did not say that, and could not, because it violates his formal logic. But he expressed the idea elsewhere in an uncharacteristically informal way.

The thinking part of the soul must therefore be, while impassible, capable of receiving the form of an object; that is, must be potentially identical in character with its object without being the object.

That could be generalized by substituting soul for object as:

The "soul proper" part of the overall soul must therefore be, while impassible, capable of receiving the form of another soul; that is, must be potentially identical in character with its object (another soul) without being the object.

This means that the human soul is something that recognizes souls of other beings as identical in character with the soul of the observer. The soul, therefore, could be just another separate human "faculty." To give a far-fetched metaphor, it reminds me the surprising ability of dogs to recognize another dog from afar or even by the sound of its steps.

This does not necessarily mean compassion. We may hate the guts of another person (the guts stands for the soul). The soul is the ability to identify oneself with other beings, and, for that matter, not just human beings. A person can identify himself or herself with other persons, fictional characters, poets, animals, gods, and even forests, atmosphere, and the finite resources of mineral fuel. The soul is the ability to substitute somebody's soul for one's own, albeit for a short moment. This is possible because all souls are interchangeable in the sense the electrons are in molecules. The response may be positive, as well as hostile. A terrorist watches with great satisfaction the terror of another soul even if he is driven by love to something.

One thing is to recognize a tree or a bird, but quite another is to recognize oneself in the other. The reason for that is that while, along Aristotle, we do not really have stones and birds inside, only their forms, we certainly have our selves inside our bodies.

The soul as a separate human faculty, in my opinion, means not identification with a group, as in social psychology, but with another soul. A soulless human being is strictly functional, like a machine. It has a purpose and a means to achieve it. Anything not related to the function is ignored or tackled as a distraction. A human being with the soul recognizes itself in another human being. I fear death and so does he or she. I suffer, and so does he or she. He is like myself. For a short moment, both souls—mine and the other's—are in joint possession and exchanged freely. When I look at my dog who looks at me, I feel for both the dog and myself, and so does the dog who expects me to take him for a walk. Can we look the same way into the eyes of a robot? If we can, than the robot has the soul, but only if the robot sees a soul in us and regards us as one of them, robots.

The human soul falls into a larger category, as life does: there is soul, as there is life, not necessarily of biological nature (which is one of the main motives of my Essays: the life of Things).

From the pragmatic point of view, this may seem quite irrational. One primary indivisible and singular term—self—is substituted for the other. We see an elephant in ourselves and ourselves in the elephant. The elephant does not see us as elephants. I really do not know about the dogs, but I suspect that my dog would see me as a kind of dog. For any practical functional purpose, it would be a fatal mistake to mix up myself and the other. Actually, acting as machines, we cannot mix up anything, as we cannot mix up letters while typing on the keyboard or keys while playing piano. A machine is not supposed to mix left and right.

I cannot find any scientific way to explain what I mean by the soul. As Aristotle did with the difficult topic of the soul, I would rely on a metaphor. The exchange and fusing of self and the other reminds me of the nature of chemical bond: covalent chemical bond is formed when electrons belonging to two different atoms become indistinguishable. When two atoms contribute one electron each to form a bond, the delocalization of their shared electrons lowers the energy of the combined atoms. Both electrons take the same "molecular orbital" and are indistinguishable. Curiously, such a joint possession leads to either a stable union (bonding orbital) or repulsion (anti-bonding orbital). The following picture is greatly vulgarized, to avoid technicalities.

I cannot escape the problem of the definition of the soul, and here is my definition:

The soul is the ability of a system to recognize the presence of the soul in another system.

This definition reminds of logical paradoxes because of its circularity. How to recognize the soul ("self") in another system? To check if the other system recognizes the presence of the soul in your system.

I believe this is what we mean by having a soul. The soul is not an organ but a relation. It is not the self, because the self is senseless without the other. The soul is a bond, an exchange of souls, as a chemical bond is an exchange of electrons. Whether it is a subject of psychology or social psychology, I cannot say. I would say that the soul manifests in any strong attraction to anything which is not part of an outside program of rational actions. What is programmed and nothing but programmed is soulless.

Naturally, one can have more or less soul. I would even measure the size of the soul as the size of the ethical neighborhood of the self (Essay 24, On Myself).

The soulless being is strictly functional, as all without exception existing creations of AI. It has no ethics. Nevertheless, the corollary is that it may be possible to make a robot with a soul because there must be a neurophysiological mechanism behind the soul, and any mechanism can be duplicated. I would reserve my guess (of a mathematical nature) about the mechanism for a separate Essay. Therefore, it would be inappropriate to call such a robot machine. Strictly speaking, we all are machines, but not all machines have souls.

To lose the soul means to become a machine. We can literally lose our souls to the machines who will appropriate them. Sci-fi or sci-phi?

All the above could be just a starting point. I am taken aback by what I have found. I display some tentacles of the idea in the APPENDIX.

So much for the soul.

APPENDIX

1. The related problems of identification, empathy, and consciousness have been discussed in two different areas: artificial intelligence (AI) and social psychology. Come to think about it, the two apparently distant areas might fuse one day.

2. The famous article by Thomas Nagel What is it like to be a bat? reverberated in responses entitled What is it like to be a Rock? by Aaron Sloman, where one can find also answers to the questions what is it like to be:

that rock over there?
a sunflower?
a bat?
a (normal) new-born human infant?
in the advanced stages of Alzheimer's disease?
autistic?
a seer?
a woman?
a robot?

Finally, one can find a discussion on What is it like to be a Human (Instead of a Bat) by Laurence BonJour.

3. It seems to me that the religious idea of the soul is nothing but the idea of a tiny personal god. Monotheism simply kneads all the pagan gods into a dough and gives everybody a cookie.

4. Dictionary Definition
Soul (Soul), n.

1. The spiritual, rational, and immortal part in man; that part of man which enables him to think, and which renders him a subject of moral government; -- sometimes, in distinction from the higher nature, or spirit, of man, the so-called animal soul, that is, the seat of life, the sensitive affections and fantasy, exclusive of the voluntary and rational powers; -- sometimes, in distinction from the mind, the moral and emotional part of man's nature, the seat of feeling, in distinction from intellect; -- sometimes, the intellect only; the understanding; the seat of knowledge, as distinguished from feeling. In a more general sense, "an animating, separable, surviving entity, the vehicle of individual personal existence." Tylor. "The eyes of our souls only then begin to see, when our bodily eyes are closing." Law.
2. The seat of real life or vitality; the source of action; the animating or essential part. "The hidden soul of harmony." Milton. "Thou sun, of this great world both eye and soul." Milton. 3. The leader; the inspirer; the moving spirit; the heart; as, the soul of an enterprise; an able general is the soul of his army. "He is the very soul of bounty!" Shak.

4. Energy; courage; spirit; fervor; affection, or any other noble manifestation of the heart or moral nature; inherent power or goodness. "That he wants algebra he must confess; But not a soul to give our arms success." Young.

5. A human being; a person; -- a familiar appellation, usually with a qualifying epithet; as, poor soul. "As cold waters to a thirsty soul, so is good news from a far country." Prov. xxv. 25. "God forbid so many simple souls Should perish by the sword!" Shak. "Now mistress Gilpin (careful soul)." Cowper.

6. A pure or disembodied spirit. "That to his only Son . . . every soul in heaven Shall bend the knee." Milton.

5. Today some people, myself including, have a feeling that we live in the mixed society of humans and machines where the former distinctions are being eroded: people become more machine-like and machines more human. The humans with machine-guns are responsible for unthinkable atrocities to each other, the machines directed by humans are saving human lives, and some humans are turned into destructive suicidal machines by other humans. I firmly believe that the relation between humans and machines is the major defining conflict of the near future. Fast evolving machines, with their short—and shrinking—life cycle, dictate the organization and function of the incomparably more conservative human society where the life cycle is artificially extended, not without the help of the machines.

6. "Losing Our Souls [by Edward Pessen] is the first book to sum up the consequences of the cold war for Americans - the shifting ideals of our approach to international affairs; the building of our nuclear arsenal; the tactics used to combat "communist subversion" throughout the world and within the United States; the transformation of the American economy in response to security demands. Carefully reviewing the evidence, and writing with the authority of a distinguished historian, Mr. Pessen charges that American cold war policy has been disastrous for many of our cherished values and institutions."