Bryan Caplan  

EconLog Book Club: For a New Liberty, Chapter 5

Amar Bhide Monday... Marginal Tax Rates Will Rise...
This chapter begins with a premise shared by almost everyone - that "involuntary servitude" is impermissible:
The libertarian, therefore, is totally opposed to slavery.  An academic question nowadays, one might object?  But is it really? For what is slavery but (a) forcing people to work at tasks the slavemaster wishes, and (b) paying them either pure subsistence or, at any rate, less than the slave would have accepted voluntarily.
Rothbard then decries the neglected continuation of involuntary servitude in such forms as the draft, the military's rules against "desertion," taxation, subpoenas, jury duty, and psychiatric commitment.

Critical Comments
This is the least radical chapter of the book.  Its lead example, the draft, is long gone, and the War on Terror has not revived it as a serious possibility.  While the abolition of subpoenas might seem like a big deal, Rothbard has an incentive-compatible alternative: 
The most the court should be able to do, then, is to notify the defendant that he is going to be tried, and invite him or his lawyer to attend; otherwise, if they choose not to, the trial will proceed in absentia. Then, of course, the defendant will not enjoy the best presentation of his case.
The only truly radical proposal in this chapter, the abolition of taxation (!), strangely gets just one full paragraph:
In a sense, the entire system of taxation is a form of involuntary servitude. Take, in particular, the income tax. The high levels of income tax mean that all of us work a large part of the year--several months-- for nothing for Uncle Sam before being allowed to enjoy our incomes on the market. Part of the essence of slavery, after all, is forced work for someone at little or no pay. But the income tax means that we sweat and earn income, only to see the government extract a large chunk of it by coercion for its own purposes. What is this but forced labor at no pay?
I suspect that many readers will fail to notice the enormity of what Rothbard is saying here; and since he doesn't discuss radical privatization until much later in the book, the readers who do notice the enormity will dismiss him as a crazy utopian.

The chapter's analysis of desertion is interesting:
It might be objected that, in the case of enlistees, the soldier or officer has voluntarily agreed to serve for a certain term, and he is therefore obligated to continue in service for that term of years. But the whole concept of "term of service" is part of the problem. Suppose, for example, that an engineer signs a contract with ARAMCO to serve for three years in Saudi Arabia. After a few months he decides that the life is not for him and he quits. This may well be a moral default on his part--a breach of moral obligation. But is it a legally enforceable obligation? ...If so, that would be forced labor and enslavement. For while it is true that he made a promise of future work, his body continues, in a free society, to be owned by himself alone...

But if this is true of ARAMCO, or of any other occupation or job in private life, why should it be different in the army? If a man signs up for seven years and then quits, he should be allowed to leave. He will lose pension rights, he will be morally criticized, he may be blacklisted from similar occupations, but he cannot, as a self-owner, be enslaved against his will.
My question: Whatever happened to the standard libertarian argument that if you own something, you have a right to sell it?  Yes, I know Rothbard elsewhere works out a whole doctrine of the "inalienability," but it just doesn't seem convincing.  In any case, there is a loophole big enough to drive a truck through, because in a Rothbardian regime, financial penalty clauses are fully enforceable.  So all ARAMCO would have to do to eliminate early termination is include a $1B penalty clause in the contract.  And don't forget that according to Rothbard, in a free society, "there are no bankruptcy laws and defaulting borrowers are considered criminal." (Man, Economy, and State)

The high point of chapter 5 appears in the discussion of anti-strike laws.  Rothbard is no friend of unions, but he points out that without government assistance, they're just a minor nuisance.  Instead of imposing regulations to limit the abuse of union power, why not just take away union privileges and let market forces sort things out?
It is characteristic of our statist trend that, when general indignation against unions led to the Taft-Hartley Act of 1947, the government did not repeal any of these special privileges. Instead, it added special restrictions upon unions to limit the power which the government itself had created. Given a choice, the natural tendency of the State is to add to its power, not to cut it down; and so we have the peculiar situation of the government first building up unions and then howling for restrictions against their power. This is reminiscent of the American farm programs, in which one branch of the Department of Agriculture pays farmers to restrict their production, while another branch of the same agency pays them to increase their productivity.

Comments and Sharing

Twitter: Bryan Caplan @bryan_caplan

COMMENTS (18 to date)
dearieme writes:

"what is slavery but (a) forcing people to work at tasks the slavemaster wishes, and (b) paying them either pure subsistence or..." Golly, what a hopelessly deficient definition.

Jason writes:

I thought this was a well laid out chapter.

He should have expanded further on taxation, but I am assuming (though I do not know) he expands on it in later chapters. When you tell people they work through Q1 for the government (longer if you count sales, corporate, dividend, capital gains, etc), it really tends to have an effect.

Corey S. writes:

I hate to derail the comment thread, but I'm working on a project and I thought this might be a good place to come.

I'm trying to compile a list of movies or documentaries from the last five or so years with libertarian themes. They don't have to be American, and they don't have to be major budget pictures.

Any help anyone can provide would be great.


Hume writes:

V for Vendetta
Lord of the Rings

guthrie writes:

'300' has some good 'freedom' themes. And 'Braveheart' has it's moments.

Bill R writes:

Here's a fun list of movies from the Mises Institute. Many are older than 5 years though:

Troy Camplin writes:

The only one that has given negative consequences has been the move against psychiatric commitment. Rules making that almost impossible have resulted in the high levels of homelessness we have experienced since the 80's and a large number of people not getting the help they really need. I have a friend whose sister desperately needs to be committed, but legally cannot be committed. She would likely have been cured had she been committed a long time ago. Instead, she just gets worse and worse. What sense does it make to give people who by definition cannot make good decisions for themselves, who do not live in reality, the ability to decide whether they should receive help? Their reality is so skewed that they cannot know what is best for them.

This is one of the few places where I am not a full-scale libertarian (though I would not have these decisions made by the government, I would equally wish they didn't prevent people from being committed when they need to be). The other one is vaccines to wipe out disease. I don't think you get to decide if something like smallpox gets carried on into the next generation.

Arare Litus writes:


Destination #1: Involuntary Servitude. Overall this chapter outlines some of the coercive mechanisms of society is a clear and straightforward manner. The conscription component of this chapter likely has less impact on reader now than previously, as many countries currently do not use forced conscription. But the idea is not totally off the table, and we can expect increased consideration of this in more insecure times (e.g. consider Israel and the rumblings in the US regarding reinstating draft boards during the height of the Iraq conflict, and “stop loss” programs). Rothbard asks “Why can’t they hire defenders as well?”. Well, one reason could be bias: A wrongly (underestimated) perceived danger could lead to less security being bought than really justified, and as it is easier to destroy than create, this could lead to very poor outcomes. I don’t think this warrants conscription, just on technical grounds alone it seems a poor argument - after all a few specialized security firms would have highly skilled operators that would be much more effective than essentially untrained conscripted. Although this technical argument would weaken in a protracted conflict. In terms of moral grounds some might use this bias, plus the implication that the current state is so ideal that removal from it is effectively death, as legitimate reasons. But, once again, Rothbard does not seriously consider alternative views and stays away from issues that might weaken his presentation.

An interesting aspect that Rothbard does not clearly discuss is punishment of criminals. This goes to the concept of “implicit moral contract”: that the criminal is to be punished for breaking the rules of the land, which they are implicitly bound to. Libertarian thought, as laid out by Rothbard so far, requires a residual moral state – the common acceptance of the legitimacy of libertarian principles and the ownership that flows from them, from which the legitimacy of punishment flows. This residual state is minimal, but implicitly required. As such, where do we draw the line on the limits of states? Rothbard says “none”, but his project (seems to, as described so far) rests on some sort of diminutive state like entity. Sure, there can be technical reasons for not liking large states, but unless Rothbard can somehow argue such that this “moral function of the state” is not required isn’t he just stating a preference on size? Some people prefer big states. Some prefer little ones. Extremely large clearly leads to problems, as any critical thinker knows, but is a minimal but larger than zero state truly evil? I hope Rothbard addresses this more clearly in the chapters to come – is Rothbard simply arguing to a limit, which breaks down and no longer makes sense in the extreme? Or is he laying out clean pure truth? So far, it is more statement of beliefs, overly simplified argument, and picking easy targets. Pick up the pace Rothbard, lets get to the meat of the issue. I thought you were constructionalist. Lets see more careful argument about the base, the weakness, the strong points. Yes, it is fun so far, and you have lots of nice tidbits, but lets get some honest, deep, and useful presentation.

Arare Litus writes:

"The other one is vaccines to wipe out disease."

A good point, which is a nice example of possible under consumption effects:
"A wrongly (underestimated) perceived danger could lead to less security being bought than really justified, and as it is easier to destroy than create, this could lead to very poor outcomes."

As for homelessness*, this is a good example of what the state is ideally for - helping, providing a safety net, basic security. I hope Rothbard addresses issues such as this, vaccinations, etc. (i.e. insurance issues when risks are small, but effects are huge and bad) that speak directly to real issues that are not as easy pickings for libertarian rah-rah. Without actually discussing hard cases, how can you promote a radical change?

[* Dr. Szasz has a point, but he seems to be arguing from disliking the outcome to disbelieving the prior - the idea of the brain being effected and this affecting rational behaviour, is this surprising and hard to believe somehow? It seems the libertarian affliction is disliking one extreme, of large state and uncontrolled/arbitrary power, and then retreating in the opposite direction without careful reflection on the weaknesses of the new stance of no state. I see socialists of falling into this same ideological trap, except going to the other extreme.]

Something I really hope to see in the coming chapters is a discussion of the trade off between transaction costs and monopoly inefficiencies, as opposed to rent seeking and coercion complaints. It is easy to slam the latter two, and many socialists likely also are against these. But does not a state and public goods reduce transaction costs? Having a public street is nice, in that I do not have to find a road I can afford, or negotiate some sort of contract. Yes, with technology these transaction costs become lower - but they exist, and completely ignoring them is foolish. Is it not conceivable that the monopoly inefficiencies are the lesser evil than transaction costs, at least in some cases? We have firms for this reason, at some level is not the state the "firm of people in geographical location X"?

Also, does not the fact that no libertarian states exist tell you something? Perhaps they are not stable, not functional, or not desirable. Is Rothbard ignoring this on purpose? Does he not see it as relevant for some reason?

If anyone can recommend some solid libertarian books that I can read after this one I am all ears; I've got some ideas already from this discussion, but the more the merrier.

Jeremy, Alabama writes:

Good chapter. I find myself more reactionary in the article of income taxes than many - having been pursued with great prejudice by an out-of-control IRS agent (he lost his case but it cost me 30k in lawyers).

He made a point somewhere about debt defaulters being criminal (could not dig out the quote right now) - but surely bankruptcy action is contractual just like everything else. The justification as to why bankruptcy laws tend to be communal is because a debtor usually owes several creditors, and it is not clear how precedence would be handled with a mish-mash of contracts.

I disagree with his ideas about courts. Some institutions get their power from their great veneration - and even if Rothbard's idea is better, surely screwing around with them "resets them to zero", thus creating the opportunity for political shenanigans very soon after. This opinion reveals my inner conservative.

Overall, I am not greatly excited by the prospect of a pure libertarian society, but am I convinced that we are a long way from it, this is bad, and we should move toward it. Almost any of the suggestions in this chapter (except the courts) would be an improvement.

Corey S. writes:

Thanks to everyone who posted suggestions.

Grant writes:

Arare Litus,

Vaccinations and disease are interesting, and are classic "public goods" arguments. I really think its more effective to address all positive externalities at once rather than pick specifics, but I find disease interesting because it has a counterpart that exists in the anarcho-capitalist environment of the Internet: computer viruses.

Of course there are differences between computer and biological viruses, but they both cause harm and both spread via agents who aren't protected against them. However, the mostly-free market of computer software and the Internet does not seem to have a large problem with viruses. Of course, the public goods argument is that virus protection is still under-provided, but from a practical standpoint its not clear to me that the government could improve upon things. This may be because the environment where viruses thrive (computer networks) is also an environment that greatly lowers transaction costs and thus allows markets to provide public goods more easily.

I'm not sure that the distinction is really between monopoly vs. transaction costs as much as it is bundled goods vs. transaction costs. Because it is costly to sign a contract with a road provider, a mower of lawns, and a security service, people join gated communities. Bundles reduce transaction costs, but they aren't necessarily monopolies.

A coercive monopoly can obviously reduce transaction costs, but isn't total cost what we should be talking about? If the "customer" does not like what the monopoly is selling, he's forced to pay. If he does like it, then his costs may have been reduced. We know markets take transaction costs into account when providing public goods (and I believe this is a feature, not a bug), but behavior of governments in providing things people actually want is less clear to me.

E. Priest writes:

Troy Camplin,

While I certainly understand your position regarding the mentally ill, I'd like to offer an alternate point of view. I've worked in the mental health field for a number of years. In my experience, a patient who cannot or will not voluntarily engage in the healing process cannot be "cured." Moreover, even those who do sometimes can never really be "cured." The question you have to ask yourself is, "How can I know the quality of another person's existence?" Ultimately, I think we have to admit we can't. We simply cannot know if a person is happier living on the street where they are exposed to a great deal of danger (but are nonetheless free) than they would be in the safer (but clearly less free) environs of a psychiatric facility. For some of the seriously mentally ill (as we call them), there is no cure. These individuals can be medicated to the point where they are no longer capable of being a danger to themselves or others, but in many such cases these persons seemingly become the living dead. They do little but sleep or wander about in a drug-induced stupor. While it may seem humane to protect such people from the elements or those who might harm them, we really cannot know the effect this has on the person's quality of life. Persons who aren't considered mentally ill often choose behaviors detrimental to themselves. Yet, we allow them to do so because we respect the free will of individuals. In my opinion, as hard as it can be to do, we must extend the same respect to all persons.

Arare Litus writes:


"I find disease interesting because it has a counterpart that exists in the anarcho-capitalist environment of the Internet: computer viruses."
"However, the mostly-free market of computer software and the Internet does not seem to have a large problem with viruses."

True, however the risk and costs are much less - computer viruses do not irreversibly destroy machines, or even data (since backup is so cheap now and can be seamlessly done). Human viruses lead to real losses - death, handicaps, loss of human potential and wealth. There is also much more variety in computational systems, with the easy to attack windows machines not making up the backbone of the internet.

The "vaccines" that government could promote, using unix and opensource, is already undertaken by the community without any outside influence. We do not have to worry about poor perception of risk leading to bad outcomes, both due to less downside and decent perception and decisions being the norm.

I'm not sure what fraction of people don't "believe" in vaccinations or discount the gains, but it is at least conceivable that many would not be willing to directly pay for them if not offered "free" in an accessible manner. This is clearly an issue where the rich would have an incentive to pay for the poor, even on pure selfish grounds, and where taxes could be a reasonable means of funding without free rider or underfunding issues.

But I have to agree - this is a great example. What I wrote may suggest I am in favor of involuntary vaccinations. I do not. I am in favor of promotion, suggestion, making it easy and cheap.

On a related note: Does anyone know if Dr. Szasz considers the possibility of "mind viruses" as causes of some mental conditions, similar to how software viruses (not "real" physical viruses) can effect computers?

" bundled goods vs. transaction costs"

Excellent point, I'll have to think about this. All I'll say for now is monopoly was a poor choice of word - I should have said something like "public infrastructure skeleton". Consider that even gated communities are connected with public "commons". Imagine if you lived on property that could become inaccessible due to sale of surrounding access properties (i.e. the private roads are sold to someone who turns them into his private gardens and denies access). Without some sort of public commons and access private properties seem to be fragile to issues like this. Yes these issues can be addressed by the market, but perhaps government is more efficient here.

"A coercive monopoly can obviously reduce transaction costs, but isn't total cost what we should be talking about?"

True, the devil is in the detail. Often the reduction in transaction costs will be much less than the losses incurred, but I'm sure the opposite case can occur also. The trade off is real, which way it favors will depend on details, which will change as technology and ideas etc change. Monopoly was a poor choice of word, I mean some sort of base public commons and infrastructure - private competition likely would be beneficial in ensuring quality public options in many (most? all?) cases.

I think a coercive "sale" of these services via taxes would be required, and since property rights are communally given it is not unreasonable to take a percentage of accrued profit from property. We will see how my views change reading and thinking and discussing this book... Apparently Rothbard has some radical ideas coming up!

Jeremy, Alabama

30K - ouch!


Has anyone studied how bankruptcy laws have evolved? It appears to be, overall, a good thing - in that it allows for the liquidation and reuse of resources in a somewhat tractable manner. It also likely slightly increases risk taking, while keeping prudence in mind - both excellent traits for the market. Most "free-market" proponents also seem to really like the idea of bankruptcy as a pragmatic tool.


I agree, common law rests on it being common. Sure we could have private institutions supplying justice, but it seems to rest on a common framework and in that sense is a "state" function. I have no problems with governments outsourcing road building (I prefer this), likewise with courts (though I have not seriously thought about this).

"Overall, I am not greatly excited by the prospect of a pure libertarian society, but am I convinced that we are a long way from it, this is bad, and we should move toward it."

I'm with you there, I'll really learning a lot reading and discussing this book.

E. Priest

A good post, thank you. The only case were forcible treatment would be justified is for short term treatment that would have real impact; sadly I don't think we really have the knowledge and ability to provide this in most cases. If I suffered a blow on the head, drank way too much, or similar I would be thankful if others restrained me in "my best interests", if I really was undertaking dangerous actions. Most people would agree. Unfortunately we really cannot provide this service to most.

Larry Peoples, Sr. writes:

Didn't Bob Dylan have a song called "You got to serve somebody"? Servitude and autonomy aren't necessarily mutually exclusive. Ask anybody who’s had a long-term, loving marriage.

b. writes:

Arare Litus writes:
"If anyone can recommend some solid libertarian books that I can read after this one I am all ears"

The Machinery of Freedom, by David Friedman is good for a pragmatic and utilitarian take on anarcho-capitalism (and it is a very easy read).

Anarchy, State, and Utopia, by Robert Nozick is a classic in the minimalist state libertarian vein.

Some more law-focused libertarian readings:

The Structure of Liberty, by Randy Barnett
The Enterprise of Law, by Bruce L. Benson
are both great anarchocapitalist works.

Minimal state readings would be:
Skepticism and Freedom, by Richard Epstein, defends classical liberalism from some more recent modern (and postmodern) critiques. His earlier Simple Rules for Complex World and Principles for a Free Society: Reconciling Individual Liberty with the Common Good are also worth reading.

And of course Hayek's Law, Legislation and Liberty

I'm sure I'm missing some obvious ones.

Hopefully Bryan will continue the book club after FNAL, though I'm guessing we won't be doing Hayek.

Arare Litus writes:


Thank you. I've had many recommend The Machinery of Freedom, I'll be reading this soon (picking up a copy today). I'll look into the others also.


Bill writes:

I'm glad that the military has rules against desertion, particularly in wartime. Being Medical Corps, they didn't teach me how to fight (I'm considered a noncombatant when deployed). I know that the Marines who protected us in 2004 and 2007 did so largely from a sense of personal and corporate honor. Nonetheless, my garrison experience in CONUS tells me that desertion can occur. I view punishment of desertion as a good insurance policy. It protects everyone when the going gets tough.

I guess this means I'm not a libertarian. Oh well...

Comments for this entry have been closed
Return to top