Some books remain in my queue longer than others. Anarchy, State, and Utopia by Robert Nozick has been there since shortly after it was published in the mid 1970’s. Nozick, a libertarian and Harvard professor of philosophy, died this past year at age 63. His death, and the retrospectives that followed, convinced me to finally bite the bullet and read his most famous book.
He wrote the book in response to John Rawls’ A Theory of Justice which had been published two years earlier. Though Nozick praises that book as being the most important work of political philisophy since Mill, Nozick’s own book is devoted to tearing down Rawls’ theory of justice. In my view, he largely succeeded.
The book has three sections:
- The first section defines the concept of a minimal state and shows how such a state could arise from a Lockean ‘state of nature’.
- The second section examines the problem of distributive justice - that is the relationship of justice to holdings, and issues of justice. as they pertain to acquiring, retaining, and disposing of holdings.
- The final section seeks to show how a minimal state can be a desirable state.
The Minimal State
Nozick takes John Locke’s 2nd Treatise of Government as his starting point. He starts with the notion of a state of nature in which there is no state whatsoever, though there are family units and perhaps larger groupings. He argues that one possible direction of evolution from a state of nature is that ‘protective associations’ would form. Individuals would see that they could not adequately defend themselves without forming mutual-protection associations. In the course of time these associations would grow and as they grow it would become increasingly difficult for each member to effectively carry on his own life as well as be on constant call to defend all other members of the association. And so the association would evolve into a business - members would pay a fee, and the association would hire specialists, policemen, to carry out the functions of the association.
Through competition, the number of associations would dwindle as one association becomes dominant within a territory. There would be less incentive to join the smaller associations because the larger, more dominant association can use its larger size to be more effective in performing its role.
This condition, of one dominant protective association with a few smaller protective associations and perhaps a significant number of ‘independents’ - people who choose not to join any association, Nozick terms the sub-minimal state . He argues that eventually the dominant protective association would gain a monopoly, and would enforce that monopoly with force. This situation he terms the minimal state and argues that it is actually a state in the normally accepted sense.
Through a set of carefully constructed arguments, Nozick claims that the sole purpose of the minimal state is to ensure that each individual’s rights are preserved up to, but not exceeding, that point where the exercise of a right would conflict with another individual’s exercise of his rights. He works out a fairly comprehensive theory of rights, prohibitions, and compensation.
Nozick’s theory of justice stands on two main pillars:
- Justice can only be determined by history. A set of holdings is just if the holdings were acquired in a just manner, and subsequently transferred or disposed of in a just manner. It is not possible to take a snapshot of societyand determine from that whether there is a just distribution of holdings.
- Broad claims about societal justice should not be admitted unless the principles underlying those claims apply to individual cases. He is (rightly, in my view) suspicious of claims made about society as a whole which cannot be sustained if applied to a smaller entity.
He argues that taking money from one person (in the form of taxes, for example) in order to benefit another person is theft and is unjust. He cites Kant’s dictum that “we should never treat a person only as a means, but always also as an end.”
With the same reasoning he states that taxation of labor is identical to forced labor. If a person has to work an additional 5 hours per week to be able to pay the tax, then those 5 hours are forced labor, done not for his own ends but for someone else’s.
The central part of the second section of the book deals explicitly with Rawls’ Theory of Justice . Nozick begins with great praise of Rawls and his book, preparatory to ripping Rawls’ arguments to shreds.
Rawls poses a hypothetical situation. A group of people with no knowledge of their relative capabilities are put into a room and asked to decide what kind of society they will live in. None of them have any initial advantage, and in the rules of this game, they cannot even have an inkling of what advantages they might gain in the future. From this situation Rawls argues that the participants would come up with a scheme in which everyone gets ‘enough’ and nobody gets ‘too much’. The participants would agree, for example, that those who become better off would be more highly taxed in order to benefit those who do not fare so well.
Nozick attacks this position from every angle that it can be attacked, and leaves really nothing standing.
First - why would we think that such a hypothetical situation, even if possible, would lead to a just society? A group of people who have no idea of their relative capabilities, and therefore no knowledge even of their personal histories, do not seem good candidates for deciding how society ought to be organized.
Second - why the great stress on outcomes? If I have many possessions and you do not, how can that situation, by itself, be just or unjust? Nozick argues that if I acquired my possessions justly (for example, perhaps I invented something useful, patented it, and sold it profitably) then how is it just that my holdings should be transferred (against my will) to you?
Third - if we apply this kind of justice to society, why should we not apply it to other situations. For example, in a classroom, why don’t we transfer high marks from the A students to the F students, so the latter will not be left so far behind?
In the remainder of the second section Nozick addresses related issues. One claim that is made by end state theorists is that medical care is something that should be universally available. Nozick asks why that should be so. Should barbering be universally available? If not, why should we say that doctors should be forced to provide care to everyone, whether they can pay or not, but barbers should not be so forced? How about carpentry services - why shouldn’t we insist that everyone should receive whatever carpentry work they need?
Nozick’s arguments seem to be unassailable. Of course the same can be said of radical skepticism. As Hume said of philosophical skeptics, “their arguments are unassailable, yet have no force of persuasion.” Nozick’s arguments have great force of persuasion, yet I remain unconvinced.
Nozick dismisses the idea that new rules of justice arise when applied to society as a whole as distinct from smaller groups or individuals. I am inclined to agree. But I think it is also evident that we cannot predict the outcome of simple formulas (such as Nozick’s theory of distributive justice) applied to complex systems. You’ve undoubtedly seen the very simple recurrence equation used to generate the Mandelbrot set. It would be impossible to conceive that a formula so simple can result in a set of such infinite complexity.
Nozick would reply that the end result doesn’t matter - if we arrive at that result by a sequence of just acts, then we should accept the result.
What would be different under the minimal state?
Let’s consider what would be different under the mininal state as compared, say, to present day America.
- There would be no public universities or public schools. Parents would be expected to pay for the education of their own children, and would be permitted to do so in any way that they see fit, or to not do so.
- There would be no public roads or highways. There would still be roads and highways, but they would not belong to the state. They would be owned by private individuals or by corporations. Presumably the owners would expect some compensation for the use of the roads. Though I can imagine how this would work for inter-city highways, or even for major thoroughfares, how, I wonder, would neighborhood streets be managed? The street in front of my house would be held as a monopoly - I literally could not leave my house without trespassing even if I walk. Of course, in the minimal state I would be trespassing on someone’s property from the time I leave my driveway until I come home at night.
- Many inter-city highways would be monopolies, as well. For example, it’s difficult to see how competition for the Portland-Boise route could be achieved. Wouldn’t the minimal state need to deal with monopoly situations if they interfere with commerce? And how could the minimal state do that? Nozick gets great mileage out of showing the difficulties of deciding the correct value of an exchange. His conclusion is that, for the most part, the value of a holding should be decided by negotiation between the holder and a prosopective buyer. But this breaks down in a monopoly situation and so libertarians are left with a choice. They can argue that since each step that led to the monopoly was just, the minimal state has no reason to interfere no matter what the consequences. Or they can argue that the monopoly situation in fact interferes with the rights of those who are forced to use the monopolized resource (the street or road) and so the minimal state must step in and require the monopoly holder to provide compensation to those whose rights have been violated. How would the minimal state decide the amount of such compensaation? And, since the monopoly situation is inherent in the structure of the minimal state, wouldn’t the minimal state have to expend an enormous amount of time deciding such cases?
- There would be no national parks, no state parks, and no city parks. There would still be parks, I think, but they would be privately held. You would pay admission to get in. This does not seem much different than the situation we have today, except that city parks are still free of charge. I imagine that the great scenic parks such as the Grand Canyon, Yosemite, and Yellowstone, would be very expensive to maintain and that park admissions alone could not compensate an owner enough to even perform basic maintenance of the facilities. Perhaps by building casinos, water parks, theme parks and the like the owners would be able to make enough profit to continue to maintain the scenic viewpoints.
- Some types of regulations would actually be strengthened under the minimal state. For example, the national electrical code would be more rigidly enforced, and retroactively applied, than it is today. Banks would insist on an evaluation of code compliance before lending money for a structure - the absence of government-paid inspectors would necessitate more direct intervention by banks.
- Air and water pollution would be treated as a tort, and polluters would find themselves embroiled in lawsuits, assuming that class action lawsuits would be possible under the minimal state. Come to think of it, why would the minimal state allow any lawsuits other than those brought by an individual or corporation?
- There would be no zoning laws or restrictions. This would almst certainly be compensated for by Codes, Covenants, and Restrictions - though I think we would see a kind of patchwork result in which small neighborhoods with strict CCRs would be adjacent to other neighborhoods with few or no restrictions. I should not like to live on the boundary between such neighborhoods. If I have submitted and agreed to CCRs that limit what I can do with my property, I would be very unhappy to be living next to, say, a brothel, an auto mechanic, or the studio of an up-and-coming rock band. But since my objections would be merely subjective it would be difficult for me to prove that I have experienced a wrong that should be compensated.
- Generalizing a bit, there would be far fewer outright prohibitions under the minimal state, but there would be a corresponding increase in the number of activities that would be judged under tort law. It is hard to see how this would be a good thing for liberty. Looking at American tort law over the past 20 years we see that, according to libertarians, the greatest threats to liberty have come from over-zealous juries. Why would this be different under the minimal state? Nozick’s presentation of how the minimal state could arise makes it sound as though just compensation is a simple matter that can be easily disposed of. But that is certainly not our common experience.
In all, a pretty mixed bag.
The Transition Problem
How would we make the transition from our non-minimal state to the minimal state?
If the state is to remove itself from the various ‘businesses’ that it is presently engaged in, it would need to divest itself of its holdings. National forests, parks, dams, highways, bridges, many office buildings, etc. would be sold to private interests. I imagine the process would take quite a long time, requiring a vast number of auctions. The proceeds from the auction would, after deduction for costs, have to be turned over to the original owners (that would be us, I guess). And how would that happen? Perhaps the distribution would be made on the basis of cumulative taxes paid.
The national pension plan (social security) would be abolished, of course. How would compensation be paid to those who have paid into that system? Perhaps we would argue that since payments into the program were always used, and intended to be used, to fund current outlays, therefore those who have paid into the system are entitled to no compensation whatsoever. Whether we argue so or not is largely irrelevant from a practical standpoint, since there is no way that compensation could be paid without taking money from one person in order to pay another - and that is precisely the act that the minimal state is designed to avoid.
Can we empirically verify any of Nozicks’s claims? In fact, does Nozick make any testable claims at all? His argument is primarily aimed at proving the justice of the minimal state (in opposition to the anarchist claim that all states are inherently unjust) and the injustice of any non-minimal state (in opposition to end-state theories of justice and redistributive schemes). It is only in the final section of the book that Nozick goes beyond this purely theoretical argument to speculate about the desirability of the minimal state.
There is now and historically a range of ‘non-minimalness’ in state power. America has transformed itself from something fairly close to the minimal state into something approaching a maximalist state. There is at present scarcely any human activity that is not governed to some extent by the American state. Should we not be able to detect some correlation between this consolidation of state power and the general happiness and well-being of at least some Americans? Wouldn’t Nozick claim that this infernal interference, these innumerable regulations and prohibitions, would cause a stifling of innovation and a flattening of the wealth distribution curve?
The America described by de Tocqueville in the 1830’s appears to be very close to the minimalist state. After the American civil war state power became much greater, and the federal government took on powers that it never had previously. Did this result in a diminution of liberty? Did it reduce the opportunities available to Americans? Again, beginning in March, 1933 the federal government began an expansion of powers that has continued undiminished to the present day. Is there any measurable effect on individual opportunity as a result? If we look simply at innovation and at wealth creation (that is, making money), then it is very hard to see that the growth of state power has had a negative effect. The wealthy became very wealthy after the civil war, and the wealthy have become even wealthier during the past 30 years - a time of great government intrusiveness. Is it perhaps even possible that the liberties denied us by the state have contributed to this increase in opportunities? I don’t know - and honestly I think the burden of proof is on libertarians to show an empirical relationship between the growth of state power and any decrease in individual opportunity.
The Open Society
I don’t doubt that the American state has become too intrusive. The principles of justice defined and described by Robert Nozick would go a long way towards removing that intrusiveness. I remain unconvinced, though, that principles of justice have any absolute, independent, standing. Why should we give absolute authority to justice, no matter how we define justice? Have we ever done so? I think that the appeal to justice is misplaced, to the extent that justice is seen as absolute, or first among all other values. If the pursuit of happiness, for example, means that justice must be partly set aside, then why not do that? Isaiah Berlin spent his life teasing apart the various ways in which basic, justifiable, values conflict, and argued that it is the selection of a single value that supercedes all others that leads to the great crimes committed by humanity.
In his book The Open Society and its Enemies Karl Popper traced the history of ‘utopian’ schemes from Plato through Hegel and Marx. He argued that there has been a continuous intellectual attack on the concept of the open society and that these attacks have often been accompanied by grandiose utopian schemes. For the utopian it always nothing or everything. Mere tinkering with the workings of society will not suffice: the entire edifice must be overthrown and we must begin anew on the basis of sound, provable principles.
Popper argues that principle-based utopias will inevitably fail and are, in any case, undesirable. The only sensible path is the pragmatic one: try something, and if that doesn’t work try something else. Make small changes, not large ones. Society is complex, and there is a multitude of competing, conflicting interests and values. Changes to social structures should be seen as engineering, not as architecture. If it works, keep it. If it doesn’t, throw it out.