Questions of Freedom
Sarah Jones Nelson
Consultant and Adviser to the Vatican
Pontifical Lateran University
Vatican City
A lecture presented at the United Nations University, Tokyo
The existence of free will or free action suggests that chance plays some indeterminate but unobserved role in the causal mechanisms of human consciousness. This explains why the reduction of moral choices to natural selection — or any evaluative concept of consciousness — lacks coherent explanations of factual responsibility for freely chosen acts. Such concepts presuppose beliefs taken on faith in some unknowably predetermined agency of causation. Examples abound from Greek antiquity to Sigmund Freud and Ludwig Wittgenstein on the nature of consciousness.
First, a very short story. Freud once received a powerful critique from the British chemist and epistemologist Michael Polanyi. Freud, he said, had reduced morality to science. Let’s examine analogous reductions of moral categories to causal explanations of consciousness.
Freud struggled most of his adult life with the persisting absence of experimental data to explain the relation between the brain and what we call the mind, or more precisely, the psyche. As a former medical student, Freud hoped to make a science of his discovery of the unconscious and the interpretation of dreams. Psychoanalysis would provide the method for gathering reliable data, but a scientific method would have to wait for the advent of cognitive science as a discipline. Partly because psychoanalysis was too newly formed a field to produce testable data and standards of verification by scientific methods, Freud reached back to Epicurus and seventeenth and eighteenth-century moral theory to support his chief thesis: the purpose of the human organism is sexual happiness, which the species sublimates under moral constraints of the external world. The pleasure principle changes into the reality principle in the creation of culture: family and marriage, art and science, war and peace, the works.
Polanyi’s problem with Freud was two-fold. On the one hand, Freud substituted objectivist or descriptive terms for candidly moral beliefs about purposive human life, a fallacy Polanyi called moral inversion. On the other hand, by inverting moral and scientific language, Freud invented a deterministic model for indeterminate, plural formations of evolving human existence. When a scientist asserts formally of one purposive life system or structure, that the goal of human life is happiness — or more currently, the optimal expression of natural selection — the language imposes causal, teleological statements of value upon statements of fact describing physical reality.
Freud’s analysis of causation in mental states involved memories stored in the brain, which organizes the triadic structure of its relation to the mind in the corresponding formation of the id, ego, and superego. He believed that each developed in healthy persons as a form of life determined by the necessary avoidance of pain and a pursuit of pleasure toward the goal of self preservation.
I am persuaded by Quentin Skinner that Freud took this idea from the materialist theory of morals and sensation of Thomas Hobbes, a Protestant Puritan whose writings were censored and burned, along with John Milton’s, in Tom Quad at Christ Church, Oxford. Freud, however, missed Hobbes’s important provision for the freedom of the will. Hobbes stated the obvious heresy that people act the way they want to and change their desires to accommodate changing situations. By contrast, Freud held that the superego as a system of agency would arbitrate the selection of acts made on the basis of transfer of mental events from the id to the ego. The superego, by necessity and nature of the system, would determine the instrumental means and value of transforming or sublimating erotic desire into the final ends of organized civic culture.
Freud’s mind-brain system reflects Aristotle’s method of reducing reality to predictably determined structures of causation. Except for Aristotle’s physics, in which two teleological processes can converge unpredictably, this method presupposes its Platonic counterpart of physical systems morally and necessarily determined for the cosmos to cohere in the purposive destiny of souls. All goodness and beauty of nature, a perfect unity of purpose, mirrors the true destiny of souls. It stands to reason that the Greeks produced no coherent concept for free will.
Factual valuations of purposive forms of life are so deeply woven into the fabric of human memory that some of us rarely notice how automatically we attribute value-choices to goal-directed life processes, and how tacitly we sense that such processes — like natural selection in the evolution of consciousness —“should” develop by the nature of a teleological mechanism like DNA to determine any life process.
We are fortunate to be free of the anxiety of double predestination and bondage of the will, which obsessed early-modern inventors of human nature. Let me digress a bit about how these beliefs drove John Calvin and Martin Luther to despair at the outset of the Protestant Reformation. One strand of their thought appears disguised in modern scientific language and runs through the history of ideas on teleology from Plato to Augustine and Luther, once an Augustinian monk who savagely challenged an unwilling Desiderius Erasmus to debate the freedom and bondage of the will. Luther argued for bondage not only because of salvation by grace alone, whatever you do or say to be saved, but because he believed that humans were born really depraved, immutably bound in original sin by natural law — as formalized by Augustine — unless God intervened to save the soul. Like all deterministic systems, ironically even aspects of Freud’s, reformed doctrine reduced agency and the freedom to choose a fundamental course of action to a causal mechanism external to the volition of the individual.
In western thought, so freighted with the Greek concept of τέλος and the destiny or ends of things, the freedom of the will has always been irreducible to cause or chance.
This was a topic of my last conversation with Isaiah Berlin on a brilliant sunny day in Oxford at Headington House. We worked our way past Hume’s distinction between facts and values, on to deterministic valuations of ends and goals, how they enter into the very texture of factual concepts of human nature. When we got around to the question of teleology, Berlin mused. “The purpose of life? The purpose of life is to live!” Like Polanyi, he was deeply skeptical of accounting for human choices “by the kind of causal explanations which are accepted, say, in physics or biology.” He believed that the solution of the problem of free will in the face of a daunting history of teleological explanations for human conduct — explanations that reduce moral choices to causal theory — would require “a new set of conceptual tools, a break with traditional terminology.”
The introduction of new categories of explanation need not be unscientific. Much of science is reductionist, but not all. When Faraday introduced fields into physics he employed a new category of explanation not reducible to Newtonian particle mechanics. The procedure was scientific because it led to testable consequences, to a degree Freud must have envied. The formal study of consciousness is ripe for integrating new critical concepts and categories of explanation for experiences irreducible to mechanistic models of the mind that make the idea of moral responsibility for freely chosen acts unintelligible.
Responsibility involves acts of meaningful reference to physical reality, freely chosen and constrained, for example, by knowledge of principles such as rights and basic human decency, to draw upon the moral philosopher Harry Frankfurt on the freedom of the will. We think of moral knowledge in many forms, one being the deliberative, agonistic reasoning you see in jury groups of courts of law, for instance, or of the kind you experience as an individual in making a just and fair decision. We may never possess the technology to map the brain or compute which parts of it can be correlated to acts of reference, deliberation, or freely chosen moral acts of an immense diversity of which altruism and selfishness are only two.
In the absence of coherently verifiable data for constructing a physical theory of volition, a situation Freud faced in the analysis of dreams, you can easily reduce the will, whether free or consciously constrained, to natural selection and teleological valuations of the purposive nature of the mind. Analysis of the computational properties of the mind as evolved through natural selection may explain a great deal of its complexity, but a wider analysis of agency should explain the fact that free will operates at a deeper level of activity than computation, just as dreams do. To resort to the widely-used language that ascribes conscious agency, even moral agency, to genes seems deterministic in the extreme because it factors not only moral responsibility out of the system of analysis, but also the right to self-determination.
The problems of agency, volition, and intentionality are central to the study of consciousness. To ascribe agency to a non-conscious entity such as a selfish gene is to indulge in perilous metaphor. To neglect agency in a theory of the mind is to relegate consciousness to an epiphenomenon of computation, an automaton watching passively as the world comes into being. The fact is, we can direct our lives the way we want to, however predisposed by nature, cultures, or history. Hobbes got it right. I believe that Hobbes is foundational to the mature work of Ludwig Wittgenstein and most modern forms of discourse on the freedom of the will and its relation to mental states. His adage that words are deeds is a helpful guide to the following outline of a plausible theory of mind that accounts for testable results of free will.
Language is one of the first chosen structures of human experience; intelligible communication of experience is one of the most fundamental activities of the mind from infancy on. Grammatical rules provide a fixed or closed structure for building the simple and complex combinatorial sequences of expression that Wittgenstein called language games. The combinatorial aspects of acquiring language have been worked out by psycholinguists, cognitive scientists, and experts specializing in how we learn to understand the meaning of statements organized by the rules of grammar. Such rules determine the structure of use, but do not necessarily determine the sense of meaning. For this reason Wittgenstein asserted in The Blue Book that sense is not the same as the sentence, any more than a word is the same as the thing to which it refers. A word only names the thing. More accurately, humans physically name things referring to the self and reality by thinking, speaking, writing, gesturing, and engaging freely in the infinity of uses for languages.
In his mature work, Wittgenstein compared language to a medieval city governed by imperial and feudal laws with unique civic codes that make the city autonomous, but related economically and culturally to other cities. The picture suggests Wittgenstein’s view of systemic forms of life with rules of use and physical constraints on formation. A language is a form of life bearing a family resemblance in various degrees to other languages. All share universal constituents like alphabets, names, syntax, and grammar.
From Aristotle to Augustine to Luther on, the most concrete constituents or of language have been called signs such as letters, numbers, words, and names. In The Blue Book Wittgenstein stated that understanding meaning gives life to signs, that “the life of the sign is the use.” Within the limits of use Wittgenstein called language games particular and strategic forms of life signified through words. The study of language games in the The Blue and Brown Books presupposes the operation of signs fixed within language, rather than the categories of signification set out by Aristotle and Augustine centuries before—which is why as a modern aware of Freud’s contribution, Wittgenstein defined language games to be “forms of language with which a child begins to make use of words.” At this original level of activity, he explained that the use of signs is much simpler and thus more foundational than ordinary use by adults. Both children and adults, however, are free to choose signs, uses, and reasons to communicate.
For Wittgenstein, all language games share certain properties creating a basic resemblance or family likeness at this “primitive” level of mental process. All games have rules, a beginning, end, and a shared agreement as to how to play the content and purpose. If at chess you checkmate the other’s king, the game is complete because you have played well by certain rules and strategies. The analogy of a language game suggests a chosen strategy begun at infancy — in sinless innocence — with the purpose of communicating by rules and customs of the birth family and a culture sharing interpretations and uses of signs of an infinite chosen variety. Wittgenstein focused his later study on simple or primitive language games not only because of their explanatory relation to complex forms of expression, but also because simple forms are best suited to philosophical problems and questions of meaning, truth, falsehood, and thus morality.
At the original level of a language game, a child uses words by using grammar to combine and create meanings. The child’s intention is not to represent logical formations of the world, but to choose meaningful utterance. Wittgenstein would oppose deductive analysis from the complex uses of words to simple uses because he believed this direction with its “craving for generality” would lead philosophers into the “complete darkness” of metaphysics, a domain of discourse he felt belonged to scientific, not philosophical, method. Philosophy, he wrote, is “purely descriptive,” and philosophers confuse themselves and others by attempting to represent complex phenomena. They should begin from description, from the elementary forms of a lived language and experience of the simplest and most foundational human “games.”
In the Philosophical Investigations Wittgenstein asserted, “Grammar tells us what kind of object anything is.” This means that grammar preserves constancy of meaning among the forms of life that language generates. Whatever may be the relation among language games, the grammatical structure described at its simplest operation of use coheres in the original sense of words and their meanings, because structure is a permanent and universal property of human languages. I would suggest the validity of understanding original sense and structure with respect to W. V. Quine’s remarkable argument on the indeterminacy of translation and the conjectural nature of deciding the exact meaning of words from one language to another across temporal, spatial, and cultural worlds of discourse.
Constancy of meaning in language presupposes that every human culture uses grammar, and that fundamental grammatical rules exists across cultures by virtue of use. The philosophical question of an infinite series of rules needed to translate one language to unspecified numbers of others suggests the counterexample of known finite structures or rules of grammar built over time into every human language. These structures remain valid, if slowly changing, formations of agreements among populations about sense and the precise meaning of words, however ambiguous or vague in context.
The assertion that there is no such thing as getting meanings right denies the fact that humans do, precisely, routinely, and freely. The mind is uncanny in its capacity for communicating the exact original or intended sense of things to be said or done. Richard Rorty once stated that pragmatists want our culture to “get rid of” getting things right according to grammatical rules too complex and numerous to make the enterprise of getting things right worthwhile. Philosophers’ self image of getting things right should be replaced by the self image of machines programmed to serve “undreamt-of functions.” Could the result be a powerful language to “program” other minds in service of unimagined functions?
Central to Wittgenstein’s analysis of language is its involvement with everyday action and the embodied agent—a far cry from the disembodied computer software analogy for mental life that would have been utterly foreign to Freud. Anthony Giddens has pointed out that Wittgenstein never separates consciousness and action, but always links them through the body as the locus of agency. This important distinction will become the locus of empirical study of free will and its relation to the brain and the indeterminate flow of words which determines their meaning in the context of situated activity. If the activity is situated, it is relational, as we experience it to be in ordinary speech acts. Grammar may constrain speech, but speech is not reducible to grammar or any related function of natural selection. For these reasons Wittgenstein’s analysis of language games — chosen freely to compete and win — should be foundational to the analysis of consciousness and any future science of free will as structured and constrained by neural events, but no more reducible to them than speech to grammar or fields to particles.