The illusion of free will is itself an illusion

While debates about free will remain prevalent in theology, philosophy, and the popular imagination, the concept of free will does not do any meaningful work in modern science. Even philosophically-inclined neuroscientists who write about free will do not evoke this concept in their technical work about the brain. Similarly, we talk about “nature versus nurture” not “nature versus nurture versus free will.” According to writer, philosopher, and neuroscientist Sam Harris, free will cannot be made conceptually coherent. In his little book “Free Will” he writes that “either our wills are determined by prior causes and we are not responsible for them, or they are the product of chance and we are not responsible for them.” Sam Harris is not the first person to debunk the idea of free will but what makes his treatment of the subject stand out from most hard determinists (or hard incompatibilists) is his no-nonsense treatment of “compatibilism” and his smart take on the view that free will is an “illusion.” He also has a talent for using effective metaphors to make his cases as evidenced by sentences such as, “you are not controlling the storm, and you are not lost in it. You are the storm.

Harris is not a “compatibilist” and follows philosophers such as Immanuel Kant (“wretched subterfuge” and “word jugglery”) and William James (“quagmire of evasion”) in identifying this position as a (subtle) attempt to change the subject. About the vast compatibilist literature he writes that “more than in any other area of philosophy, the result resembles theology.” Compatibilists like Daniel Dennet have spent considerable time in twisting the meaning of free will and putting it in an evolutionary context but as some of his critics have noted, the “free will” that is compatible with determinism does not capture the kind of free agency and moral responsibility that philosophers feel is worth talking about (for example, see Paul Russell’s article “Pessimists, Pollyannas, and the New Compatibilism“). “Compatibilism amounts to nothing more than an assertion of the following creed: A puppet is free as long as he loves his strings,” writes Harris.

Harris follows philosophers such as Derk Pereboom in noting that neither determinism nor indeterminism can give rise to free will or moral responsibility. This also includes more recent attempts to find “free will” in quantum mechanics. “Chance occurrences are by definition ones for which I can claim no responsibility…how would neurological ambushes of this kind make me free?

While Harris still recognizes free will as an illusion, there are some passages in his book that reveal that he does not seem to agree that disciplined introspection is a credible source for a belief in free will. “If you pay attention  to your inner life, you will see that the emergence of choices, efforts, and intentions is a fundamentally mysterious process…I do not choose to choose what I chose…there is a regress here that always ends in darkness.” This is a distinctly refreshing perspective because most literature is plagued by the belief that regardless of whether free will exists (or can exist) it is nevertheless an illusion, or worse, a necessary illusion. This “illusion of the illusion of free will” remains a mainstay of most discussions of the topic, despite its shaky foundation in introspection or logical analysis. In a rather Buddhist perspective on the matter, Harris concludes his book by observing that

“our sense of our own freedom results from our not paying close attention to what it is like to be us. The moment we pay attention, it is possible to see that free will is nowhere to be found, and our experience is perfectly compatible with this truth. Thoughts and intentions simply arise in the mind. What else could they do? The truth about us is stranger than many suppose: The illusion of free will is itself an illusion.”

So what then gives rise to the belief in free will and the desire to prove its existence? According to Harris, a belief in free will is closely associated with the concept of “sin” and retributive punishment. One might also add that “compatibilist” philosophy arises from the recognition that most normative ethical theorizing requires some kind of compatibilism. It is not a coincidence that the most exotic treatments of free will can be found in theological, ethical, and ideological writings. Obviously, Harris denies that a belief in free will is necessary for morality and justice. “Certain criminals must be incarcerated to prevent them from harming other people. The moral justification for this is entirely straightforward: everyone else will be better off this way.” The fact that no criminal has free will does not mean that all crime should be treated the same. The reason why we are interested in, for example, whether the cause of a crime can be attributed to a brain tumor or a psychopathic personality type is because it is important to know what kind of person we are dealing with and under which conditions we should expect such crimes most likely to occur. There is no need for a complete overhaul of our criminal system but in a society in which there would be less emphasis on free will there would be more room for intelligent treatment of crime instead of hatred and retribution.

There is a brief chapter in the book where Harris discusses free will in the context of politics. He identifies modern conservatism as embodying an unrealistic belief in free will, as evidenced by the tendency to hold people responsible for their own choices and to glorify “individualism” and the “self-made man.” It is certainly the case that the concept of free will has clouded the mind of many political thinkers. For example, two writers that are closely associated with radical capitalism, Ayn Rand and Murray Rothbard, have offered rather obscure defenses of free will. Ultimately, however, most dominant ideologies can be restated without a belief in free will. A denial of free will in conjunction with postulating values such as”egalitarianism,” “impartiality,” and “universalism” can give rise to modern liberalism but a denial of free will is also compatible with an elitist, aggressive, anti-democratic pursuit of human enhancement through state coercion.

Anarcho-capitalism does not require a belief in free will either as evidenced by recent attempts to derive it from Hobbesian contractarianism (Jan Narveson) or economic efficiency arguments (David Friedman). Incoherent discussions of free will in moral and political theory are easy targets for ridicule, and often an indicator of belief in other mysterious concepts such as “natural rights.” In fact, anarcho-capitalism cannot only be restated without any appeals to “free will” or “natural rights” but it does not even require the postulation that “freedom” is valuable (or needs to be be maximized) as has been shown in the recent writings of Anthony de Jasay.

Voting, cheering, and exploitation

In his little book ‘Game Theory: A Very Short Introduction‘ Ken Binmore writes:

Real people seldom think rational thoughts about whether to vote or not. Even if they did, they might feel that going to the polling booth is a pleasure rather than a pain. But…the pundits who denounce the large minority of people who fail to vote in presidential elections as irrational are talking through their hats. If we want more people to vote, we need to move to a more decentralized system in which every vote really does count enough to outweigh the lack of enthusiasm for voting which so many people obviously feel. If we can’t persuade such folk that they like to vote and we don’t want to change our political system, we will just have to put up with their staying at home on election night. Simply repeating the slogan that ‘every vote counts’ isn’t ever going to work, because it isn’t true.

Later in the book Binmore returns to this topic when he discusses the “myth of the wasted vote”:

If a wasted vote is one that doesn’t affect the outcome of an election, then the only time that your vote can count is when only one vote separates the winner and the runner-up. If they are separated by two or more votes, then a change in your vote would make no difference at all to who is elected. However, an election for a seat in a national assembly is almost never settled by a margin of only one vote….Naive folk imagine that to accept this argument is to precipitate the downfall of democracy. We are therefore told that you are wrong to count only the effect of your vote alone – you should instead count the total number of votes cast by all those people who think and feel as you think and feel, and hence will vote as you vote…This argument is faulty for the same reason that the twins fallacy fails in the Prisoner’s Dilemma . There may be large numbers of people who think and feel like you, but their decisions on whether to go out and vote won’t change if you stay home and watch the television.

Faced with the criticism that game theorists who openly disseminate such observations lack “public spirit” he responds by drawing an analogy between voting and cheering at a football game.

“No single voice can make an appreciable difference to how much noise is being made when a crowd of people is cheering. But nobody cheers at a football game because they want to increase the general noise level. They shout words of wisdom  and advice at their team even when they are at home in front of a television set. The same goes for voting. You are kidding yourself if you vote because your vote has a significant chance of being pivotal. But it makes perfectly good sense to vote for the same reason that football fans yell advice at their teams.”

Whether this analogy is accurate or not, it is doubtful that such explanations of voting can salvage the idea that participation in an election is a meaningful public activity, let alone a civic duty. A recent Reason article, Your Vote Doesn’t Count,’ is a good survey of this topic and the desperate attempts to rehabilitate the case for voting. With the exception of economist Gordon Tullock, few scholars are known for publicly admitting to the futility of voting, let alone admitting to not voting themselves.

One explanation why people vote is that many do not explicitly recognize that they are no longer deciding an issue in a small hunter gatherer tribe. The fact that the scale of our decision making has changed substantially throughout the history of mankind is increasingly being discussed though. For example,  one presentation at the 2012 Ancestral Health Symposium reads as follows:

Richard Nikoley, B.S. – Paleo Epistemology and Sociology

Primitive peoples evolved to account for the values and actions of a relatively small tribe of family and close acquaintances comprising of 30-60 members whereby, every individual had a critical role and opportunity to influence the behavior and actions of the group or tribe as a whole. This is far removed from the unhealthy social trends in modern society where individuals are fooled into believing that they have real power at the voting booth and other activism when in reality, their influence is insignificant and pales in comparison to the social power a primitive hunter-gatherer would have wielded.

A more sophisticated argument about conditions under which which it would be rational to vote was recently expressed by the social philosopher Anthony de Jasay in a 2011 interview:

It is in fact widely held that because millions vote, no voter can rationally expect to influence the result. Millions nevertheless keep on voting, which looks a bit strange. Many parapsychological stories have been written to explain why they do so. I am not sure that we need them. In a well-oiled democracy, the perfect election result yields a wafer-thin majority because that outcome maximizes the size of the losing coalition ready to be exploited and minimizes the size of the winning coalition whose members share the spoils. This idea, of course, is the well-known median voter theorem. When the majority is literally wafer thin, the displacement of a single vote turns the majority into a minority, and vice versa. Thus, the perfectly oiled democratic mechanism produces outcomes with a majority of one vote; a single vote is decisive; and, hence, the voter is quite rational to cast it. In a less perfectly oiled democracy, where the majority is thicker than a wafer, the probability of a single vote’s being decisive is less than unity (the median voter theorem does not quite hold), but it need not be negligible. Because voting is not very costly, to affirm that it is irrational to vote is much too strong a claim.

Jasay’s explanation why people vote, or under which conditions it would be rational to vote, deserves closer scrutiny because it aims to do more than coming up with a “feel-good” story about voting. What Jasay is saying here is that in elections that are purely distributive in nature, it can be rational to vote. Real-world elections, however, do not take place in such “well-oiled” democracies and virtually all large elections are decided by majorities much thicker than a wafer.

The problem with Jasay’s argument about the rationality of voting is not just that it has little relevance for actual existing democracies but it also raises questions about whether such a cynical form of democracy could be viable at all. Although most people recognize the redistributive component in politics, it is doubtful that a democracy in which politicians operate without any illusion about serving the general interest could persist. Just like it is doubtful whether voting would survive widespread recognition that it is just another form of cheering (or signalling), it is also doubtful that a democracy that would function in the way that Jasay describes it would be able to secure stable compliance, especially from its victims.

Public choice scholars often praise themselves as doing politics without romance by stripping the political process of all its lofty rhetoric and just analyzing it in terms of interest. But if people would actually recognize public institutions solely as vehicles to form coalitions to exploit others, the resulting governments would have little resemblance to the Western governments scholars and philosophers currently analyze. In other words, it may be a mistake to assume a distinctly different view of human nature and social interaction but keep political institutions unchanged.

While economists sometimes recognize the futility of voting in technical works, Bryan Caplan has been one of the few scholars who has developed this fact (and its implications for public policy) into a general theory about the microfoundations of political failure. In two excellent blog entries for EconLog he further reflects on the illusion of choice in American elections and how politics discourages self-correction (as opposed to markets).

The Better Angels of Our Nature

The Summer 2012 issue of the Independent Review features my review essay (PDF) of Steven Pinker’s The Better Angels of Our Nature: Why Violence Has Declined. There can be little doubt that this work constitutes one of the most ambitious and credible contributions to social science to date. Although the review essay was written from a classical liberal perspective, I think that one of the main criticisms of Pinker’s project can be sustained without any kind of “ideological” perspective. In fact, one of the concerns I have about his project is that insufficient attention has been given to providing a neutral definition of violence. Why is this important?

If we would go back in time and challenge some of the violence that was routine in those days inevitable objections would be that these acts of cruelty should not be condemned because they simply involved the execution of God’s will, proper punishment, served “the common good,” etc. One of the themes of Pinker’s book is that we have become less tolerant of these kinds of justification for violence and acts of extreme cruelty. Naturally, this raises the question of whether there are still many acts of violence, cruelty, and punishment that are being rationalized with poor reasoning. In my review I suggest that most of what we consider the normal operation of government, such as collecting taxes and “regulation,” is sustained through violence and threats of violence.

One might object that this perspective reflects a minority position on violence that does not conform to common use of the term violence. I do not believe that this response would be credible because the common opinion is not that government operates without threats of violence (and punishment if one fails to obey) but that in this case the use of force is legitimate and socially sanctioned. In that case, however, Pinker’s project would not be about the decline of violence but the decline in violence not approved by governments. Pinker does not go that far because he does not exclude warfare by democratic governments from his review of violence, but there is something rather arbitrary about what matters to him.

For example, Pinker writes that “early states were more like protection rackets, in which powerful Mafiosi extorted resources from the locals and offered them safety from hostile neighbors and from each other” but does not give good reason why we should view contemporary states much differently. In fact, one can even argue (as individualist anarchists like Lysander Spooner have done) that modern democratic states do not only extort protection money but in turn use this against the victim in the form of “regulation.”

I suspect that what makes Pinker exempt force associated with normal government operations is that the actual use of violence is rather rare. But that is not necessarily because most people prefer paying taxes or complying with regulations but because individual resistance is not rational. As Anthony de Jasay writes in his essay Self-Contradictory Contractarianism (collected in his book Against Politics: On Government, Anarchy, and Order).

If the cost of rebellion is high, if the expected (“risk-adjusted”) value of its success is not very much higher, and if the very possibility of collective action against the sovereign is problematical (at least in normal peacetime conditions), then two plausible conjectures suggest themselves. The equilibrium strategy of the sovereign will be to use its discretionary power to satisfy its preferences, perhaps by exploiting all its subjects in the service of some holistic end, perhaps by exploiting some of them to benefit others. The equilibrium strategy of the subjects will be, not to resist, but to obey, adjust, and profit from the opportunities for parasitic conduct that coalition forming with the sovereign at the expense of the rest of society may offer.

A potential rejoinder to this argument is that the operation of government is necessary to prevent even more violence. Leaving the usual problems with utilitarian arguments like this to the side, such a perspective can at best confer legitimacy to a very minimal form of government and would not exempt most other operations of government. If social order and peaceful commerce can arise without government, there is no reason at all to exempt any operations of government from a critical perspective. Pinker does recognize the existence of anarchist perspectives but his treatment of this topic does not indicate a thorough familiarity with the literature on conflict resolution without the state. This is problematic because reason and commerce (two of Pinker’s drivers of the decline in violence) may be sufficient for a peaceful society. In fact, the advantage of commerce versus government (or ‘democracy’) is that commerce itself is a peaceful activity.

One might further object that there is a difference between war and collecting taxes on the one hand and regulating on the other. In a real showdown between individuals and government officials, however, the priority of government is to prevail using as much force as necessary. As mentioned above, that does generally not require a lot of force because most individuals recognize the futile nature of individual resistance. In fact, it may be the increase of intelligence and individualism that Pinker also discusses in his book that makes more people less inclined to mount heroic but ineffective forms of resistance.

This does not mean that Pinker’s claims are completely arbitrary and dependent on whether one includes normal government operations in his definition of violence. For example, it is indisputable that the nature of violence and the cruelty of punishment has seen substantial changes since the middle ages. Also, in spite of the increase of public force associated with the growth of modern governments, the tolerance of people for violence is still declining. In fact, many public debates concern forms of harm that can hardly be construed as violence (discrimination, ‘hate speech’, insensitivity, poverty, etc.). This phenomenon itself raises a rather interesting question. How can the widespread tolerance of government force co-exist with increasing sensitivities about acts of human behavior that do not even involve physical harm (or threats thereof)?

There are a lot of other interesting topics in Pinker’s book such as his treatment of the sociobiology of violence, morality, and ideology. On the topic of morality he writes:

The world has far too much morality. If you added up all the homicides committed in pursuit of self-help justice, the casualties of religious and revolutionary wars, the people executed for victimless crimes and misdemeanors, and the targets of ideological genocides, they would surely outnumber the fatalities from amoral predation and conquest.

The Better Angels of Our Nature is not a treatise on (meta)ethics but Pinker’s evolutionary perspective leaves little room for grandiose moral theories and is more in line with classical liberal views in which morality is an emergent phenomenon that allows for peaceful human interaction, which is evidenced by his observation that “modern morality is “a consequence of the interchangeability of perspectives and the opportunity the world provides for positive-sum games” and that “assumptions of self-interest and sociality combine with reason to lay out a morality in which non-violence is the goal.”

He also observes that “to kill by the millions, you need an ideology.” At the same time he notes that “intelligence is expected to correlate with classical liberalism because classical liberalism is itself a consequence of the interchangeability of perspectives that is inherent to reason itself.” He does not discuss the potential tension between his (wholesale) rejection of ideology and his identification with classical liberalism. Perhaps Pinker believes, as does the author of this review, that classical liberalism, conceived in a non-dogmatic fashion, is not so much an ideology but a perspective that starts from the recognition that individuals have different interests and that reason can provide guidance to coordinate these interests to mutual advantage.

Jacques Monod’s Ethics of Knowledge

Nobel Prize winner Jacques Monod concludes his seminal essay on the natural philosophy of modern biology, Chance and Necessity (1970), with a chapter of reflections on evolution, the place of man in nature, culture, ideas, and the nature of morality. He writes:

During entire aeons a man’s lot was identical with that of the group, of the tribe he belonged to and outside of which he could not survive. The tribe, for its part, was able to survive and defend itself only through its cohesion…This evolution most not only have facilitated acceptance of tribal law, but created the need for mythical  explanation which gave it foundation and sovereignty. We are the descendants of such man. From them we have probably inherited our need for an explanation, the profound disquiet which goads us to search out the meaning of existence. The same disquiet that has created all the myths, all the religions, all the philosophies, and science itself.

He then goes on to explain how religions, philosophical systems, and ideologies (such as Marxism) that see nature or history unfolding according to a higher plan can be traced back to this innate disposition to look for Meaning. And while science, and the associated postulate of objectivity, has gradually replaced those myths and beliefs, most of our contemporary thinking about values still reflects this kind of animism:

No society before ours was ever rent by contradictions so agonizing. In both primitive and classical cultures the animist tradition saw knowledge and values stemming from the same source. For the first time in history a civilization is trying to shape itself while clinging desperately to the animist tradition to justify its values, and at the same time abandoning it as the source of knowledge, of truth. For their moral bases the “liberal” societies of the West still teach – or pay lip-service to- a disgusting farrago of of Judeo-Christian religiosity, scientistic progressism, belief in the “natural” rights of man, and utilitarian pragmatism…All the traditional systems have placed ethics and values beyond man’s reach. Values did not belong to him; he belonged to them.

Obviously, this perspective on the futile attempts to ground values in something beyond man (beyond practical reason one might say) raises the question of “who shall decide what is good and evil.” Monod clearly struggles with this question because he does not want to admit that “objective truth and the theory of values constitute eternally separate, mutually impenetrable domains.” His answer, however, may strike contemporary readers as something of a cop-out when he tries that argue that the pursuit of science itself implies an ethical postulate:

True knowledge is ignorant of values, but it cannot be grounded elsewhere than upon a value judgment, or rather upon an axiomatic value. It is obvious that the positing of the principle of objectivity as the condition of true knowledge constitutes an ethical choice and not a judgment arrived at from knowledge, since, according to the postulate’s own terms, there cannot have been any “true” knowledge prior to this arbitral choice. In order to establish the norm for knowledge the objectivity principle defines a value: that value is objective knowledge itself. Thus, assenting to the principle of objectivity one announces one’s adherence to the basic statement of an ethical system, one asserts the ethic of knowledge. Hence it is from the ethical choice of a primary value that knowledge starts.

This attempt to derive (or distill) universal normative claims from an activity or pursuit itself is not unique in ethics. Some have tried to derive morals and rights from the nature of human agency (Alan Gewirth), the activity of argumentation (Hans-Herman Hoppe) and so forth (one might argue that there are even traces of such an approach in Jasay’s argument for the presumption of liberty). Either such attempts produce trivial conclusions or are stretched beyond credibility to make them do a lot more work than they are capable of, such as deriving specific socio-economic norms concerning welfare rights or absolute property rights. At the end of the day, these writers fail to recognize the fact that morality is an emergent property of social interaction in nature (that is to say, morality is conventional) and attempts to “justify” moral rules is as futile as trying to “justify” the laws of physics (although one might argue that certain “strategic” advantages can accrue to those who are successful in persuading others of such moral “truths”).

Monod’s ‘ethics of knowledge’ is simply “justified” by pragmatic advantages (a similar thing might be said about accepting the principle of causality – as has been proposed by the philosopher of science Hans Reichenbach). Such a pragmatic explanation for the pursuit of knowledge (and the emergence of values) places morality in the realm of individual practical reason and evolution, where serious philosophers, economists, and biologist have been making efforts to understand it.

In his introduction to the 1997 Penquin edition of Chance and Necessity, the evolutionary biologist and geneticist John Maynard Smith, briefly alludes to Monod’s rather clumsy (and dated) attempt to link his ethics of knowledge to scientific socialism in the final pages of the book, which only shows how vacuous the ethics of knowledge is for deciding moral and socio-economic questions.

A more specific concern for Monod is the end of natural selection and degeneration in man:

To the extent that selection is still operative in our midst, it does not favor the “survival of the fittest” – that is to say, in more modern terms, the genetic survival of the “fittest” through a more numerous progeny. Intelligence, ambition, courage, and imagination, are still factors in modern societies, to be sure, but of personal, not genetic success, the only kind that matters for evolution. No, the situation is the reverse: statistics, as everybody knows, show a negative correlation between the intelligence quotient (or cultural level) and the average number of children per couple…A dangerous situation, this, which could gradually edge the highest genetic potential toward concentration within an elite, a shrinking elite in relative numbers.

This is not all. Until not so very long ago, even in relatively “advanced” societies, the weeding out of the physically and also mentally least fit was automatic and ruthless. Most of them did not reach the age of puberty. Today many of these genetic cripples live long enough to reproduce. Thanks to the progress of scientific knowledge and the social ethic, the mechanisms which used to protect the species from degeneration (the inevitable result when natural selection is suspended) now functions hardly at all, save where the defect is uncommonly grave.

And since Monod seems to categorically rule out gene therapy in germ cells (“the genome’s microscopic proportions today and probably forever rule out manipulation of this sort”), his only hope resides in “deliberate and severe selection.”

Notwithstanding Monod’s unduly pessimistic perspective on human genetic engineering  and the missed opportunity to recognize the evolutionary and conventional nature of morality, Chance and Necessity remains a classic, uncompromising, exposition of modern evolutionary biology and the scientific view of the world that has made this knowledge possible.