Human freedom and the limitations of scientific determinism.

Recent developments in science have stimulated the centuries-old debate between proponents of determinism and those who defend human freedom. Determinism implies that every event, including every human action, is causally necessitated by prior events, so that no person could have acted otherwise. If so, it would seem that free will is an illusion. It we could not have avoided acting as we did, it is pointless to hold people morally responsible for their acts. Man becomes merely an agglomeration of molecules in motion or a product of genetic evolution. Even mind and consciousness, under the determinist view, will ultimately be explained by the laws of inanimate matter. Because of the prestige of science, non-scientists arc inclined to assume that what scientists say about the natural world is likely to be true. The prestige accorded to science unci scientists is well deserved. Science, like all other human endeavors, nevertheless has its limitations. Some of these have been noted by the renowned physicist Stephen Hawking: "Godel's theorem, Heisenberg's uncertainty principle, and the practical impossibility of following the evolution of even a deterministic system that becomes chaotic, form a core set of limitations to scientific knowledge that only came to be appreciated during the twentieth century." (1) More broadly, it can be said that science ceases to be fully applicable whenever the question at issue depends on phenomena that cannot be measured or empirically observed and therefore cannot be reduced to quantitative analysis. Such common terms as "meaning," "beauty" and "justice" fall into this non-quantifiable category.

Standard definitions of freedom include at least four principal meanings: 1) the state of not being coerced or constrained by necessity or circumstances beyond one's control; 2) the absence of antecedent causal determination of human decisions; 3) exemption from slavery or involuntary restraint by others; and 4) political meanings: e.g., the right to speak, vote, or otherwise participate in public affairs. In this essay, we will be concerned primarily with the second meaning. The relevant question has been succinctly put by Steven Pinker: "...how can we have both explanation, with its requirement of lawful causation, and responsibility, with its requirement of free choice?" (2)

The debate over "free will" has been a staple of Western philosophy for centuries. Aristotle assumed as self-evident that freedom is a presupposition of moral action, since one incurs responsibility only for voluntary actions. (3) This assumption has been generally accepted by moral philosophers, since moral philosophy would make little sense without it. In Kant's moral writings, for example, morality presupposes freedom of the will: "the will of a rational being can be a will of its own only under the idea of freedom. ..." (4)

The rise of modern science, however, has led some observers to question the presumption of freedom. Many (perhaps most) scientists believe that science progresses through a process of reduction, in which the world is seen as an assemblage of physical parts that can be broken apart into their elementary constituents, whose behavior is or will eventually be explained entirely by the laws of physics. Determinism assumes that the state of the world at any instant of time, including our disposition to act, follows necessarily from the state of the world at the immediately preceding instant of time by reason of the laws of physical causation. The past completely determines the present. If determinism is true, such human feelings as resentment, blame, remorse, and forgiveness are pointless and human beings are objects to be manipulated rather than subjects of personal relationships. The deterministic point of view thus renders meaningless human dignity, human rights, and much of the discourse of customary human interaction. No one can deny, of course, that there are serious constraints on human freedom. These constraints may be due to genetic, environmental, or pathological factors or to coercion by other people. What the proponents of freedom assert is that, despite all of the side-constraints, there remains a core of choice, a sphere of action in which we may freely choose one alternative over another.

I.

Stephen Hawking's core set of limitations to scientific knowledge comprise the constraints on mathematical reasoning imposed by Godel's Theorem and the unresolvable physical facts of quantum mechanics and chaos, which imply that some aspects of the world are not exclusively governed by the reductionist-determinist model.

After Bertrand Russell and A.N. "Whitehead, in their Principia Mathematics (1910-13), had purported to reduce arithmetic to symbolic logic, it was widely assumed that this formulation was, or could be made to be, internally consistent. It would also be complete, in the sense that the axioms of the system are sufficient for generating all logical truths expressible within the system. However, Kurt Godel showed in 1931 that, given any finite set of rules for doing mathematics, there will always be some mathematical statements that can neither be proved nor disproved by using these rules. (5) It would be necessary, therefore, to go outside the system to prove its completeness--that is, one would have to construct a "metamathematical" proof--but this cannot be done within the axioms of the system in question. Expressed differently, given any consistent set of arithmetical axioms, there are true arithmetical statements that are not derivable from the set. There will always be some problems that cannot be solved within the existing set of rules. It follows that our mathematical understanding cannot be reduced entirely to computational mechanisms. The system can be understood, as a whole system, only by intuition "outside the box" of computation. The effort to reduce science to strict deterministic rules is therefore doomed to failure. (6)

The puzzles posed by quantum mechanics (QM) are among the most fascinating problems in the history of science, which reinforce the problematic status of determinism. By the early twentieth century, it had been experimentally proven that atoms were not indivisible but instead were composed of smaller particles, including a positively charged nucleus surrounded by negatively charged electrons. But there were serious problems with the model. According to Maxwell's equations of electromagnetism, the orbiting electrons should exhibit certain behavior that docs not, in fact, occur. (7)

Einstein, Max Planck, Niels Bohr and others then developed a theory to account for these phenomena by arguing that the energy in a light wave is not spread uniformly over the wave but is concentrated in bundles or "quanta." But this raised a further dilemma: is light a wave or a series of particles? The two are separate concepts in classical physics. A particle (a photon in the case of light) is discrete and localized, while a wave is spread out. Astonishingly, it turned out that light manifests both wave-like and particle-like behavior. It seems to follow from the empirical results that light is both a wave and a particle (the "principle of complementarity"). Whether it is a particle or a wave depends on the experiment that is done. A particle is said to be in a "superposition state" until we detect it, whereupon the superposition collapses to a single definite state. Prior to detection, there is only a probability that the particle can be found in one place or another. This probability is described by a "wave function," which is not a physical state but a mathematical function quantifying the probability that the particle will have a given position or momentum. When the position is measured by an observer, the wave function "collapses" and the position is determined. This finding indicates that, at the quantum level, the observer and the observed are inextricably intertwined.

Further, two electrons emerging from a common source (such as a random discharge from a decaying atom), may have different characteristics. But there is no way to tell which electron has what characteristics until you measure one of them. If the measurement of electron A shows a particular characteristic, you immediately know that electron B has a different characteristic even though the two electrons may be very far away from each other. The result with respect to particle B depends entirely on the measurement performed on particle A. Moreover, there is no interaction or communication between the two because any such interaction would violate Einstein's theory of special relativity (which depends on a finite speed of light). Scientists call this phenomenon "entanglement" or "nonlocality"; Einstein called it "spooky action at a distance."

It is sometimes argued that the "uncertainty principle" derived from QM establishes an absence of determinism at the heart of nature and, therefore, supports the case for free will. Others argue that indeterminacy alone does not imply free will. What believers in free will seek to establish is not indeterminism but self-determination by individual persons. Quantum occurrences depend on random events, such as the random emission of electrons from an atom. But random or chance events are not under the control of anything, and thus not under the control of the actor. (8) Being at the mercy of chance is no more comforting than being at the mercy of necessity. Another difficulty is posed by the fact that quantum events occur at the very small sub-microscopic level, and the quantum uncertainties tend to cancel each other out at the larger, macroscopic level.

A simpler answer to the determinist hypothesis lies in the role of the conscious observer, whose free choice of the means of observation turns what is merely probable into a physical state that is definite and measurable. To this extent, matter is dependent on mind. The free action of the observer presupposes that freedom, is an intrinsic characteristic of rational beings. To be rational is to reason; and a conclusion cannot be rational unless the agent is free to decide among incompatible reasons. There is, accordingly, a strong link between volitional causation and rationality. If our conclusions are the result of physical necessity rather than the free judgment of reason, then our conclusions cannot be supported by rational argument and "are not worth taking seriously." (9)

The philosophical conclusion is that we, the observers, turn potential reality (for example, potential momentum and position) into actuality through our observation. The observer and the observed are part of a single reality--a "participatory universe," as physicist John Wheeler called it. (10) This reality includes the process of observation; it is a reality which has been created by the free activity of human consciousness. This conclusion takes us beyond the sphere of measurable, quantitative science and into the realm of metaphysics. Among other philosophical consequences, this analysis provides a possible solution to the problem of dualism: the "mental" and the "material" are interrelated aspects of an overall process. Reality is not a conscious subject (mind) observing an external thing (matter), but is a single process ("the object-as-observed-and-conceptualized"). Interaction between mind and matter is a part of the overall reality. The knower always interacts with and affects the known. John Polkinghorne calls this approach "dual-aspect monism," in which mental and material are opposite poles of a single reality. (11)

What QM leaves us with, then, is a feeling that quantum indeterminacy opens a gap in the tightly-woven network of deterministic causation, and that the human observer has a creative role to play in the construction of physical reality. We can take some comfort from the conclusion that there are events at the quantum level which have no physical cause in the classical sense. Therefore, "nature" must include non-material factors (such as quantum wave functions) or else we may have to give up the notion of causation.

Classical physics assumed that if the position, mass, and velocity of any object could be measured at a given instant, subsequent positions and velocities could be precisely determined. Quantum mechanics, as we have seen, casts doubt on this conclusion. A further difficulty with classical determinism is posed by so-called "chaotic systems." A chaotic system is a complex or dynamical system whose behavior exhibits a very high degree of sensitivity to initial conditions. Examples are storm systems, the flow of a stream, and a game of billiards. In each case, the overall impact of an uncountable number of impacts cannot possibly be predicted, except within broad limits of probability. In chaotic systems, tiny differences in input (e.g., changes in wind, temperature, and humidity) can quickly become large distinctions in output. In chaotic systems, there will always be a small uncertainty which will be amplified with successive iterations. (12)

The human brain is a chaotic system, since it operates through billions of unpredictable quantum events, including "quantum tunneling" (a quantum occurrence in which a particle suddenly appears in an unexpected location). These quantum effects are amplified chaotically through the system. Accordingly, the brain appears to share some of the indeterminism and non-locality of the quantum world. Precisely how the amplified quantum events manage to produce order and mind-directed activity remains unknown. Whatever the mechanism may be, there does not appear to be any way to test the mathematical model of a chaotic system so as to predict precise results. We can safely conclude that chaos theory imposes a limitation on any attempt to support a deterministic hypothesis.

II

Taken together, Godel's theorem, QM, and chaos theory support the conclusion that there are elements of indeterminacy in the natural world that leave room for human choice. But much of the recent controversy about determinism and freedom has occurred in the field of biology, especially in the subset of problems connected with evolution and natural selection. Biology is the science of living organisms. Non-organic objects are for the most part subject only to exterior forces, while organic entities are, in substantial part, the product of internal, autonomousforces. The French biologist Jacques Monod, in his influential book, Chance and Necessity (1971), identified three properties of living beings: teleonomy ("The transmission from generation Co generation of the invariance content characteristic of the species"); autonomous morphogenesis (a structure that develops from interactions within the object itself); and reproductive invariance (the ability of an organism to reproduce and transmit its own basic structure from one generation to the next). (13) Reproductive invariance is not absolute but is subject to random mutations that may cause a "scrambling" of the genetic text. These mutations are based on microscopic events whose accumulation may alter the organism's structure at the macroscopic level. Once incorporated in the molecular structure of the organism, the accidental mutation will be replicated according to the laws of genetic development. All organic evolution, Monod concluded, is therefore the product of chance and necessity. There does not appear to be any source of human value in this process. The implications are severe: "man knows at last that he is alone in the universe's unfeeling immensity, out of which he emerged only by chance." (14) Chance, from this perspective, is wholly purposeless--something that happens without any human intervention or divine dispensation.

Contemporary biologists have spelled out the implications of the determinist model. E. O. Wilson, in his book, Consilience, asserts that the "cutting edge" of science is reductionism: the breaking apart of nature into its physical constituents. All of nature is organized by simple, universal laws of physics, to which "all other laws and principles can eventually be reduced." (15) Mind is nothing more than the brain at work. Later, he adds that "Mind is a stream of conscious and subconscious experience. It is at root the coded representation of sensory impressions." It involves "internal mapping," "finely differentiated patterns in the brain circuits," and "linked concepts." (16) But this analysis exposes a flaw in materialistic determinism: coded representations, mapping, patterns and concepts are not physical things, and it is not clear how physical events alone could give rise to them.

It follows from the reductionist position that free will is an illusion. If you could know every process of every neuron in the nervous system, Wilson argues, you could predict every action. As a practical matter, total predictability is impossible because the inputs into the nervous system are incalculably numerous and the principles of chaotic systems (discussed above) therefore apply. Because of this practical unpredictability, "the self can go on passionately believing in free will," even though we know that conscious experience is a physical phenomenon. Even morality and religion are ultimately explainable by the interaction of genes and environment. "Ought is the product of a material process," and religions exist because "[t]here is a hereditary selective advantage to membership in a powerful group united by devout belief and purpose." (17) For those who value moral responsibility, there is very little comfort in "consilience."

Richard Dawkins, one of the world's most influential biologists, has popularized the "selfish gene" theory of evolution. Dawkins asserts that the body of an animal is merely a vehicle used by genes in order to preserve themselves unaltered. "They [genes] are in you and in me; they created us, body and mind; and their preservation is the ultimate rationale for our existence....we are their survival machines." (18) Dawkins concedes that the human survival machine has the property of "apparent purposiveness" and that it has evolved "the property we call 'consciousness'." Human consciousness, he admits, is "the most profound mystery facing modern biology." (19) But biology itself cannot solve this mystery; a solution would involve an "ontological leap" to a wholly different realm of discourse. (20) Dawkins simply evades the philosophical issue and concludes that consciousness is the culmination of an evolutionary trend.

Professor Daniel Dennett, in his book, Freedom Evolves, calls freedom of choice "a frankly mysterious doctrine, positing something unparalleled by anything we discover in the causal processes of chemical reactions,...or such biological processes as metabolism, growth, immune reactions, and photosynthesis." (21) For most of us, life itself and especially consciousness are indeed mysteries. But most of us believe, if we think about it, that consciousness and the ideas that result from it are not material things. The materialist, on the other hand, assumes that we are creatures of the natural order whose mental activity and decisions are entirely dependent on the physical operations of our brains. These are metaphysical positions that divide thoughtful people and will continue to do so. But we are able to choose (freely) between them. This places the determinist in the paradoxical position of having freely chosen determinism.

One of the most perplexing problems in the field of evolutionary science is the relative weight that should be given to culture, which for present purposes can be defined as those behavioral attributes of humans that are transmissible by social practices rather than by genetic evolution. Darwin in the Descent of Man (1871) argued that any animal endowed with social instincts by natural selection "would inevitably acquire a moral sense or conscience, as soon as its intellectual powers had become as well, or nearly as well developed, as in man." After the power of language had been acquired, "the common opinion how each member ought to act for the public good, would naturally become in a paramount degree the guide to action." (22) "The more efficient causes of progress," Darwin concludes, "seem to consist of a good education during youth whilst the brain is impressible, and of a high standard of excellence, inculcated by the ablest and best men, embodied in the laws, customs and traditions of the nation, and enforced by public opinion." (23) This is a purely cultural description and represents a sharp departure from explanations of progress or regression based on biological selection. Reconciling natural selection and cultural change remains a perennial problem for evolutionary biology. As E. O. Wilson concedes, the reductionists' confidence in the ultimate success of the materialist model is, at least for the present, an act of "faith." (24)

Could the choices we make, although undetermined, be simply random acts, like the emission of an electron from an atom of radium, and therefore not a product of our free will? Possibly so, if the chain of causation could be traced directly back to the random event. But the previous discussion of quantum phenomena shows that the effort to do so (at least for that category of phenomena) can result only in probabilities, not certainty. Moreover, and more directly relevant to human experience, the choice could not be "random" because our process of weighing alternatives is based on principles that are rationally related to our choices. Such a decision "will not be a random brute fact; it will be explained as an instance of the conception and weights chosen." (25)

Of all the decisions we make, moral choices are among the most fundamental. Intuitively, we tend to resist the conclusion that the scientific model of cause-and-effect applies to moral discourse. In moral reasoning, "the giving of intelligible reasons displaces the ultimately brute-factish quest for causal regularities." (26) Evolution has endowed men with many disparate faculties and characteristics; yet biological science cannot tell us which of these faculties ought to prevail in any given circumstances. The moral realm, which is the realm of freedom and responsibility, is not driven by natural causation but by the "mutual interaction of self-conscious beings through publicly assessible [sic] actions and norms." (27) Biological evolutionists contend that norms of conduct have a genetic basis arising out of the evolved human traits of sociality and cooperation, which have survival value. But cooperation as such has no ethical value. Criminals and terrorists cooperate. If rules of cooperation are merely genetic, why is there an obligation to obey them, or any others? The answer lies in recognition of the worth and dignity of every person, which implies a mutual respect for others similarly endowed. This respect for other persons differs from the instinctive sociality of animals in that self-conscious agents demand mutual recognition of basic human rights and duties. In short, we cannot accept the determinist evolutionary model of socio-biology as the totality of human experience without sacrificing moral responsibility.

Such human traits as altruism and self-sacrifice pose a particular problem for evolutionary biologists, since it is hard to see why "selfish genes" would benefit when the host organism denies itself the advantages of reproductive success or acts so as to favor other organisms. Why, for example, should humans ever follow moral principles of sexual restraint that result in leaving fewer descendants? Some biologists argue that altruism for the benefit of kin may be biologically based because the altruist is closely related to his kin genetically, but this seems implausible in the case of pure altruism, such as feeding people on the other side of the world. It is much more likely that altruistic behavior is transmitted by cultural rather than genetic means--for example, through moral education and imaginative sympathy.

If morality means anything, it must rest to some extent on general principles and not solely on what is in my (or my kinfolks') genetic interest. These moral principles cannot be deduced from the findings of sociobiology. Since free moral choice is a vital aspect of human behavior, this gap in the scientific model provokes a degree of frustration among some biologists. As Steven Pinker observes, morality, consciousness and free will are "deep enigmas" and "elusive quantities [sic] whose origins still remain wrapped in substantial mystery...." (28) This honest admission tells us that we have reached the limit of science and have crossed over into philosophy--specifically, into what Ernst Cassirer called "the philosophy of symbolic forms." (29) A creature constituted solely by the physical flux of nature could not stand outside of, and bring order and meaning to, that flux. Mind cannot be identical with the matter it describes. To be conscious of natural forces, we must be to some extent independent of those forces through a reasoning faculty that transcends the matter it seeks to understand. Reason is the creative power that brings intelligible order to the raw material of the experienced world. The practice of science itself is evidence of conscious choice.

The scientist does his work by applying mathematical models to make sense of phenomena. In addition to mathematics, scientists cannot do without abstract, conceptual terms such as category, pattern, code, representation, and other symbolic expressions used to denote the formal aspects of physical reality. The intelligibility of the natural world, and abstract relationships and forms such as consistency, correlation and order, are assumed as necessary postulates of science. The theory and the terminology, as well as the experiments, are freely chosen by the scientist in order to produce an intelligible model of reality. Indeed, the major premise of philosophy as well as science is that the world is intelligible.

How and when consciousness appeared remains a mystery, but it is the basis of language, mathematics and other symbolic forms, without which science (and other conceptual activities) would be impossible. It is this realm of ideas, hypotheses, abstractions and choices that gives us the freedom to transcend the rigid constraints of physical causation and biological selection, and only if we have this kind of freedom can we be held morally responsible for our acts. For biological determinists, the "self" is reduced to a complex array of electro-chemical events; we are neurologically programmed with impulses to do certain things and avoid others. Such an entity could not be "free" in any meaningful sense of the term. An entity that is nothing more than a physical system comprising randomly evolved components would not be capable of self-consciousness, and certainly could not explore or explain a physical system. (30)

The extraordinary complexity of species, especially Homo sapiens, and the apparently progressive evolution of man in the direction of social organization and intelligence, leads us to suspect that the evolutionary process may have been designed. If the evolution of intelligence was probable from the start, the best explanation for such a process may be that the result was intended. Most biologists reject the design hypothesis. But their inevitable recourse to such terms as "pattern," "process," "code" and "direction" implies the existence of some plan or purpose. (31)

Since it is self-evident to most people that they have free will, why is it important to pursue the topic of determinism vs. freedom? The answer is that the belief in universal causation, seemingly justified by the scientific world-view, casts doubt not only on human freedom but also on individual responsibility and ultimately on the dignity and worth of the person. The debate is a real one, as evidenced by the outpouring of serious books and articles on the subject. (32) If our acts are predetermined by material causes over which we have no control, the implications for law and morality would be severe. The only basis for punishing criminal conduct would be to deter others--meaning that we would be treating people as objects, not as responsible subjects. Science has proved what we knew all along, that there are a great many materially necessary conditions to any human act. But these conditions are not sufficient for a free act; there must be at least one step involving free agency. Without this step, we would have to discard the entire moral history of the human race.

The world we live in is quirky, mysterious, and dangerous. Although we live in a hostile environment, we are sometimes able, with conscious effort, to transcend it. In doing so, we become aware that we are part of an order of being which transcends the bare struggle for existence. Consciousness is a fundamental component of reality, like mass and electromagnetism. It is in some respects even more fundamental, since we could not understand physical forces without it. Consciousness formulates the mathematical systems that make physical reality intelligible. It may be a mystery to scientists and a scandal to materialists, but it is a fact of human life. Consciousness, in turn, brings choice and responsibility. These are the characteristics that separate us from robots. They are worth defending.

[AUTHOR'S MOTE: I am grateful for the helpful comments of Professor Nicholas Capaldi, Dr. William Martin, and Wayne Rogers; and for the invaluable help of my assistant, Valli Roberts. They, of course, bear no responsibility for the analysis or conclusions herein.] (1.) Stephen Hawking, The Universe in a Nutshell (New York: Bantam 2001), 139, (2.) Stephen Pinker, The Blank State (New York: Viking Penguin, 2002), 180. (3.) Aristotle, Ethics, III, 1109b. (4.) Immanuel Kant, Critique of Practical Reason and Other Writings in Moral Philosophy (Chicago; University of Chicago Press. 1949), 103. (5.) See Ernest Nagel and James R. Newman, "Godel's Proof," in Newman, ed., The World of Mathematics (New York: Simon & Schuster, 1956), 3, 1668. (6.) Roger Penrose, Shadows of the Mind: A Search for the Missing Science of Consciousness (Oxford and New York: Oxford University Press, 1994), 192. (7.) Roland Omnes, Quantum Philosophy (Princeton, NJ: Princeton University Press, 1999), 137-38. (8.) Robert Kane, The Oxford Handbook of Free Will (Oxford and New York: Oxford University Press, 2002), 23. (9.) Karl R. Popper and John C. Eccles, The Self and Its Brain (London: Routledge, 1984), 75. (10.) Quoted in Victor J. Stenger, Timeless Reality (Amherst, NY: Prometheus Books, 2000), 180. (11.) John Polkinghorne, "'['he Metaphysics of Divine Action," in Robert John Russell et al., Chaos and Complexity: Perspectives on Divine Action (Vatican City: Vatican Observatory Press, 1997), 154-55. (12.) James Gleick, Chaos: Making a New Science (New York: Viking Penguin, 1987). (13.) Monod, Chance and Necessity (New York, 1971), 13-14. (14.) ibid., 180. (15.) E. O. Wilson, Consilience: The Unity of Knowledge (New York: Vintage, 1999), 60. (16.) Ibid., 119-21. (17.) Ibid., 274, 281-82. See also Pinker, 271, concluding chat our moral sentiments are strategies for survival, due to "neurobiological and evolutionary design.'' Pinker argues that what we call mind or soul is the information reprocessing activity of the brain, an organ governed by the laws of biology (229). (18.) Dawkins, The Selfish Gene (Oxford and New York: Oxford University Press, 1989), 20. (19.) Ibid., 50, 59. (20.) Francis Fukuyama, Our Posthuman Future (New York: Farrar, Strauss and Giroux, 2002), 166 ff. (21.) Daniel C. Dennett, Freedom Evolves (New York; Viking Penguin, 2003), 100. (22.) Darwin, The Descent of Man (Amherst, NY: Prometheus Books, 1998), Chap. IV, 101. (23.) Ibid., Chap, V, 148. (24.) Consilience, 147, 171, 285. (25.) Robert Nozick, Philosophical Explanations (Cambridge, MA: Harvard University Press, 1981), 300-01. (26.) Anthony O'Hear, Beyond Evolution (Oxford and New York: Oxford University Press, 1997), 10. (27.) Ibid., 129. (28.) Pinker, 240, 426. (29.) Cassirer, The Philosophy of Symbolic Forms, (New Haven, CT: Yale University Press, 1965). (30.) I am indebted for this insight to Professor Nicholas Capaldi. (31.) The case for design is presented in Michael J. Behe, Darwin's Black Box (New York: Simon & Schuster, 1996), Chapter 9. (32.) Oxford has published a 630-page volume of essays on the subject. See above n. 8.

J. F. JOHNSTON, Jr., is a retired partner in a Washington, D.C. law firm and author of The Limits of Government.