Rothbard vs the Copyleft
[posted by Leonard]
First post!
I’d like to thank Jim for the opportunity to help out around here, perhaps spreading the word about anarchy to them as need it told. My one beef with guestbloggers is when I go to a blog expecting a known voice, then start reading, only to realize that I’ve been reading something attributing it wrong. It matters, who’s saying something. For this reason I commend the leftulicious Nell just downblog, for making clear it’s her writing a post, and not Jim. It is also for this reason I prefaced this post with my name.
Anyway. On to the show. Jim promised you anarchy, and anarchy you shall have!
At the Lew Rockwell site, there’s a polemic by the late anarchocapitalist Murray Rothbard against those “other” anarchists - the “anarcho-communists” as he calls them. The folks called “anarchists” by most of the world, where anarchocapitalism is unknown. (Perhaps the Net has changed this. I hope it has.)
If there is one thing, for example, that anarcho-communism hates and reviles more than the State it is the rights of private property; as a matter of fact, the major reason that anarcho-communists oppose the State is because they wrongly believe that it is the creator and protector of private property, and therefore that the only route toward abolition of property is by destruction of the State apparatus.
They totally fail to realize that the State has always been the great enemy and invader of the rights of private property.
As a polemic it is only mildly interesting. Rothbard makes his points, but you tend to want a bit more proof that anarchocommunists really believe what he says. But I enjoyed reading it from a different perspective: anarchocommunism is present in the modern world, in the domain of the Net, particularly in the Free Software movement.
Open source code is private property in a very narrow technical sense, but in any practical sense, it is communal property. Anyone can get it, modify it, use it, and even sell it. You can share it with your friends completely above board. Open source is also very slightly scarce (it does require time, at least, to download and install), but it is about as close as any good is likely to be to being abundant. But free software didn’t exist in 1970, when this essay was written. Does it hold up?
Read the essay substituting “the Open Source movement” for “anarchocommunism”.
Comment by Andrew Edwards —
June 5, 2006 @ 9:57 am
they wrongly believe that it is the creator and protector of private property
Not to start a needless mudfight, but aren’t the anarcho-communists basically right about this one?
I.e. that the main reason someone can’t walk in my apartment and take all my stuff while I’m at work is that the police will come and arrest them for it?
Comment by Jon H —
June 5, 2006 @ 9:59 am
FSF/GNU-style free software and content is not really free because it puts serious constraints on the user. There is a sacrifice of ”free-as-in-speech” freedom for a developer who uses GNU licensed code, in favor of downstream users who get more ”free-as-in-beer” code. As such I’m not sure how well it falls under anarchic values.
BSD-style licensing provides more freedom to the user, because the user is under fewer constraints and obligations as to how they use the licensed material.
Comment by Neel Krishnaswami —
June 5, 2006 @ 10:27 am
Economically, a good is property if it is rival, excludable, and alienable. ”Rival” means that if one person uses the good, another can’t — two people can’t drive the same car at the same time. ”Excludable” means that you can easily prevent a non-owner from using the property — for example, you can put a fence around a piece of land. ”Alienable” means that you can transfer ownership between people — your heart is inalienable, because you’ll die if you give it up.
Different goods are different combinations of these three. Air on a spaceship is rival, since two people will use up the oxygen faster than one, but it’s not excludable or alienable, because two people in the same chamber can’t prevent the other from breathing. Your heart is rival and excludable — while it pumps blood for you, it can’t pump blood for anyone else, and you can’t be made to share without opening up your chest — but it’s not alienable, since you can’t transfer your heart to another without dying. A car is rival, excludable, and alienable, since two people can’t drive a car at once, you can put locks on the car, and you can sell the car with no difficulty. (It’s also the most obviously ”property-like” of these three examples.)
Now, we can take the usual Coasian step and observe that all of these attributes exist on a continuum, and that when someone says a good is ”excludable”, they really mean that it’s very cheap to exclude, and when they say it’s ”inexcludable”, they likewise mean it’s expensive to exclude.
Software has a close-to-zero cost of reproduction. This means that it’s not really rival, since if I make a copy then both of us can use the software at the same time. Likewise, the ease of copying makes excludability very difficult to achieve, since you have to prevent any copies from leaking out to the net, ever. Alienability is easy, though — you can tranfer title to software very easily.
So software is not property-like, on two counts out of three. It shares these characteristics with other forms of data, like novels and music, and it’s not really a surprise that ”intellectual property” has never had a really satisfactory status in the natural law tradition.
Incidentally, free software is also a really excellent example of why the mere existence of public goods doesn’t necessarily entail government intervention. Naively, you might think that since the benefits of software production can’t be fully captured, it will be an underproduced public good. However, this neglects the fact that software is a complement to many other products, and so the producer of a complement can make software and distribute it to increase demand for what they’re really selling. IBM, for example, sells consulting services, and supports Linux programmers to reduce the cost of software to increase demand for consultants.
Comment by Leonard —
June 5, 2006 @ 10:53 am
Andrew, what you’re getting at is that the state does vindicate property rights. That is, to an extent it does act as their protector. Also, it creates property rights that probably would not happen in anarchy, such as near-infinite term ”intellectual property”. This counts both for and against it - certainly all anarchists dislike the idea that someone can fence off an idea that I physically cannot ”unknow” and claim as their own.
More importantly, the state does not permit you full property rights in various things that ought to be yours. Rather it taxes them, and is thus a much larger violator of your property than any thief. You may think the tradeoff worthwhile, and thus sort of consent to it; but we’ve seen what happens when taxes become voluntary; they quickly cease to exist. So this counts as a very important area where the state is directly opposed to the rights of property.
However, private property exists prior to, and outside of the state, in several important ways.
First, human nature - all humans have an intuitive grasp of ”mine”, including a sense of violation and injustice if ”theirs” is taken from them. The sense exists for everyone, from occupants of high capitalism to savages in huts. Indeed, if we did not have this sense, it is doubtful that we’d be having this discussion.
Second, in a defacto sense, we are each owned by ourself. Not because of any law, or human nature (of our minds, that is). But because we control ourselves; I can make myself dance almost effortlessly, whereas to make you dance against your will, I’d have to employ a team of goons, and the result would be lousy as dance. (I might also use threatened force to gain your unwilling cooperation; less costly and more effective, but still orders of magnitude more expensive than doing it myself.)
One aspect of this inalienable self-control is that only an individual can know his own thoughts. That is, secrets are possible. We can exploit that, creating ”private property” by information advantage. The classic illustration here is: the gun is ”mine”, regardless of what anyone else says, because only I know where I hid it.
Third, as a historical reality, long before the state had a standing army of police, there was private property. (Police are a modern invention.) The state existed at that time, and monopolized adjudication, but enforcement of property was largely a private affair.
Finally, as an ideological matter, a fundamental leg of anarchocapitalism is the idea that property rights could be enforced entirely privately. People can work together outside of the state to do this; monopoly of force is not necessary.
Comment by Andrew Edwards —
June 5, 2006 @ 12:40 pm
Leonard:
Thanks for your thoughtful reply. I’m a big-government liberal, so I obviously don’t fully agree with all your views. But I have an honest curiousity about a philosophy that’s somewhat alien to me, and so I appreciate your honest response.
In that tone, I’d like to probe how you view the difference between ’people working together to protect property rights’ and ’government’.
We’re getting kind of 18th-century-state-of-nature here, but I’ve always viewed government as an inevitable formalization of that idea of shared protection. I take it you don’t and I’m interested to see, procedurally, how you think we can design a process to limit a Seven Samurai style let’s-get-together-and-protect-the-village regime from blooming into a de facto (and then de jure) government.
Comment by Jim Henley —
June 5, 2006 @ 8:40 pm
Interestingly, I’ve been learning that freehold tenure - what we think of when we think of ”property rights,” in land anyway - is historically a creation of government, and there’s a further irony: government fostered freehold tenure to simplify and optimize taxation. Land tenure comes with a bundle of possible rights - domiciling; cultivation; grazing; scavenging (e.g. mast or firewood); mining etc - and premodern communities allocated (and allocate) those rights in what might seem to us to be bewildering combinations.
Problem is, then the King and his agents can’t be quite sure whom to tax. Even the Duke may not be entirely sure. But if you rationalize the bundles of rights into unitary title to every given parcel, you know exactly who’s on the hook for rendering unto Caesar.
It also simplifies commerce, since now you can buy, sell and trade plattes without having to inspect the property. There are all kinds of pragmatic, nontax benefits that arise. But ”private property” as we think of it today begins in State convenience. [Leonard, I recommend James Scott’s Seeing Like a State, which is about misadventures in State planning (famines and Brasilia among other things) but has some excellent historical background. Scott’s intro to the book suffers from a certain liberal anxiety - he’s desperately concerned to convince you that, while his book is about bad things that states do, he’s not one of Those People. But the book itself is full of fascinating background on how States made the modern world to serve their own interests (everything from ”scientific” forestry to surnames).]
Comment by Neel Krishnaswami —
June 5, 2006 @ 10:13 pm
Leonard is substantially right, too. ”Property-ness” is defined by how easy it is to achieve excludability and alienability, and while government institutions can profoundly shape those conditions, they don’t and can’t possibly determine it entirely. Observe that the US government — ie, the men with the most guns anybody has ever had in all of history — can’t stop teenagers from downloading MP3s.
That’s one of Hernando de Soto’s main arguments. Capital formation is necessary for high intensity economic activity (ie, for a country to industrialize), and that can happen if and only if the informal notions of property line up with the formal, legal definitions. Otherwise, wealth gets tied up in ”dark matter” which can’t be used as collateral in the capital markets, and which serves as a nexus for organized crime and government corruption.
One of the saddest things that I ever heard about was from some dairy farmers near my parents’ home in India. Over there, when you sell a piece of land you have to pay a stamp tax of IIRC 14% to officially transfer the title. Since poor people can’t pay that, they buy and sell land informally, with the official title resting with people who sold it years or decades ago, and not with the people who paid for it. Now, these farmers were being shaken down by local gangsters, and they couldn’t turn to the police for help, because they didn’t ”really” own the land their farm was on.
Comment by Leonard —
June 5, 2006 @ 11:20 pm
Andrew, what you’re asking for probably requires a lot of words to get right. So, let me beg off for now, with a promise to post something in future (if not at UO here in sweeps week, at Unruled). For now, just let me recommend my archives. At a point in 2002 I got interested in writing down some thinking points on anarchy, so, those might help to give you an idea of where I am coming from. Try this one for a morality argument for anarchy. Look here for a discussion of efficiency and anarchy. (This is a thin reed, but still I hope interesting.) Then read this post for a lockean ”start in the state of nature” approach. Then a comment on the costs of government. Finally, a caveat: anarchism isn’t about designing a system as much as describing a system. See this post for more on that.
In all of that, I still haven’t addressed your key question: how do we stop a ”nice” anarchic setup from evolving the state? And there’s a good reason for that: history has few examples to guide us, and plenty of bad examples to serve as warnings. It’s a point I think we can overcome with ideology (and one of the few places I see ideology as a necessity for anarchy), but this is by no means a well accepted conclusion.
Comment by Steve —
June 6, 2006 @ 9:29 am
Leonard, towards your original question, I think that there’s two different ways to look at the question of whether free software/open source software fits Rothbard’s argument.
I’m sure this is all pretty basic to anyone who would formulate the question, but to recap, one of the great schisms of the movement (such as it is) is over licenses. RMS, Richard Stallman, essentially the founder of the open source movement, is a very vocal advocate of his GNU Public License; the GPL offers the threat of regulatory mechanisms (as use of the code constitutes acceptance of the licensing terms, which mandate that any derivitive code also be subject to the GPL) to make sure that open source stays open source, and other less restrictive licenses, such as the BSD and MIT licenses. (I chose the MIT license for Jumpcut, a piece of free software I’m working on, for a variety of reasons, of which the less-restrictive nature was among the more important.) How do you classify voluntary contractual agreements backed by the implicit threat of state intervention? Did Cisco open source their router’s firmware to ward off the supposed men with guns, or to assure community good will? I don’t know that Rothbard’s critique really applies one way or the other, and some of the big players — IBM, Red Hat, Sun — clearly don’t give a toss about the ideology of free software but instead view it as a market mechanism to unsettle Microsoft.
But because of the nature of software — I’d argue that it fails all three of Neel’s qualitative tests — and a very concerted effort by a number of people, rooted in the ethos of MIT and Berkeley, open source software really is a gift economy for the majority of developers. People contribute vast amounts of time and potentially valuable effort to the open source movement in exchange for street cred and aesthetic satisfaction. (The fallout between JWZ and RMS over XEmacs may be a useful example for figuring out what drives it.) That strikes me as an left anarchist realization of what will drive people to work, rather than an right anarchist/libertarian one. (Some of the institutional models for open source development also remind me of Rockerian syndicalism, and Rothbard’s caricature of anarchism strikes me as reminiscent of some of Stallman’s detractors if you do your suggested terminology swap.)
Comment by Leonard —
June 6, 2006 @ 10:25 am
Steve, thanks for that comment. As you say, the licensing issue has been, and to some extent still is, a big question in the free software community. (Personally I prefer the BSD.)
I’d suggest that the ”enforcement” as it were of the copyleft is largely reputation (that is, voluntary). And for that matter, there’s probably lots of ”theft” of copylefted code; it would be very hard to determine that if the ”thief” was reasonably smart. (It’d be pretty easy to write a tool to rewrite code, guaranteeing functional equivalence but making any simple textual comparison of source and object differ.) That said, there certainly is an element of threatened force to copyleft enforcement. As far as I am concerned, more power to them.
Is that consistent with anarchism? I think so. Intellectual property is one of the edge cases of libertarian theory. It’s clearly useful (hence its inclusion in the Constitution), but at the same time, hard to square with an absolutist understanding of property. How can I ”own” something sitting in your head? Fortunately, anarchism can contain such contradictions without complaint, since it is not ideological. It just is. Some companies will offer intellectual property enforcement. Most won’t. My guess is that most corporations in anarchy would end up under IP laws. Most people would end up under weak IP laws; some would have none. (The answer to how I own your thought then becomes: because we implicitly negotiated it and you agreed to it.)
As Jon says (upcomments), the different licenses place power in different places. The copyleft emriches future users and empowers future developers, but restricts current developers. He questions whether this is anarchic, but I think that misses the point. I see the whole domain of software (free and not) as quite anarchic, including the rich profusion of licenses. Let 1000 flowers bloom. It’s perfectly OK to have some software copylefted, other software BSD’ed, other software public domain, and even fully proprietary software. Each license has strengths and weaknesses. None is outright superior for every use.
As for the ”gift economy” of free software, I agree that it fits well with the left anarchist vision. But it is also just peachy with this libertarian right anarchist. There’s plenty of room in our philosophy for the idea that people will work for things other than money. The challenge to the left is to create a popular ”Free Toilet Cleaning” movement, or to affirm that payment is necessary to movivate a lot of socially useful activities.
I totally agree with your observation re: Rothbard’s piece; I’ve seen the same kind of arguments deployed against the copyleft.
Comment by Steve —
June 6, 2006 @ 1:37 pm
copyleft emriches future users and empowers future developers, but restricts current developers. He questions whether this is anarchic, but I think that misses the point. I see the whole domain of software (free and not) as quite anarchic, including the rich profusion of licenses. Let 1000 flowers bloom. It’s perfectly OK to have some software copylefted, other software BSD’ed, other software public domain, and even fully proprietary software. Each license has strengths and weaknesses. None is outright superior for every use.
And yet, just to play Devil’s Advocate, that doesn’t seem to be how RMS feels about it. (He is probably an edge case in terms of open source ideology, but he’s an interesting one.) He very clearly has stated that he doesn’t believe in proprietary software and thinks the GPL is manifestly superior in all cases because it prevents people from slipping away from that model. And while I tend to agree with you about GPL compliance largely being a reputational rather than a legal issue for most companies, that doesn’t seem to be what RMS thinks; the XEmacs/Emacs fork happened not only because of aesthetic considerations but because the XEmacs people refused to transfer copyright to RMS, the better to allow state action to enforce the license against transgressors.
To the extent that political ideology maps on to open source ideology at all, I’d guess that the best fit would be people like Bob Black and Peter Lamborn Wilson, who have at least given a nod to the ideas of a post-scarcity economy. (Being neither an anarchist nor a libertarian, I may have missed less, uhh, out there political thinkers who’ve addressed these issues.)
Comment by Ray —
June 6, 2006 @ 2:54 pm
”The challenge to the left is to create a popular ”Free Toilet Cleaning” movement, or to affirm that payment is necessary to movivate a lot of socially useful activities.”
As soon as I meet some anarchists who think people will clean toilets for fun, I’ll be sure to send them over to you, and you can set them straight. Will they need to book in advance, do you think?
Comment by buermann —
June 6, 2006 @ 8:40 pm
I’ve never run across any anarchocommies who do not abide property defined under some variation of occupancy and use, so they all have property rights in the credo, but if you hire an employee they’re using your capital/land, so get property rights to your capital/land too, and all the tangle of theories that follow follow from how to manage that.
Also: ”BSD-style licensing provides more freedom to the user” - the singular on ”user” is the problem for RMS and his ilk. It doesn’t preserve the freedoms of the users. There’s a case to be made for BSD commercial licenses in development terms, like there’s a case to be made for patents in development terms, as was made to justify US patent/copyright law. But it’s ultimately a restrictive covenant, useful only in so far as it provides an incentive to work.
Comment by David T —
June 8, 2006 @ 12:59 pm
To me it seems obvious enough that ”private property rights could not exist (at least could not be enforceable) without the state” and ”the state is the greatest violator of private property rights” are both true, and in no way inconsistent with each other. But I’m not an anarchist of any sort, so what do I know?