TOP 2014-09-09

A Network-Enabled Gift Economy in a Post-Capitalist Society

I believe there are many different ways that a post-revolutionary, post-capitalist, stateless society could operate. Here I’d like to present just a sketch of a peer-to-peer network for organising an economy without money, and place it in the context of a libertarian communist vision.

At least since Marcel Mauss’ short work The Gift, anthropologists have known that gift economies used to be common in pre-industrial societies. Recently, time banks and demurrage-equipped local currencies emulate gift economies to a certain extent, but still involve a form of accounting, or money. As verbosely posited in my other article, Nothing Costs Money, money is always going to pose problems related to lack of credit or debt, which get in the way of the central goal of meeting the needs of everyone in society. More deeply, I don’t think it’s ever possible to put a value on anybody’s work, or to put a value on the goods and services that people provide, simply because of the chain of dependencies between each person’s work, and between our work and the natural resources that we need in order to carry it out. The system I’ll describe here does justice to these interdependencies by instantiating them directly in the form of a social network.

Like any network, there are nodes (individual people) and links between them which represent transactions. A transaction has a “provider” or group of providers, a “recipient” or group of recipients, and an activity which the provider carries out on behalf of the recipient - in other words, each transaction is a gift. Every gift is logged by the system. The resulting network could be diagrammed using directional links to show the flow of gifts/transactions between people. Work contracts, gifts, transactions - whatever you want to call them - are never assigned a value. In our society, if you have 100 units of money then you ‘deserve’ 100 units of stuff, regardless of how you got hold of that money, whether ruthlessly or painstakingly, and regardless of how much or little you actually need and therefore whether 100 is actually enough to cover it. In this model, on the other hand, transactions are never assigned a value, nor are they even counted up, since each transaction could be of widely differing scale and importance and there’s no uncontroversial way to compare them. You don’t need ‘credit’ to initiate a transaction and you can’t get into debt.

Everybody has an individual account on the system with which they can post a request, outlining what they need. The request can be tagged, indicating the industries it subsumes and the skills that it requires. It could also be directed at specific people who can repost the request so that it gets greater visibility and is ultimately passed on to a suitable ‘provider’. All the functionality you’d expect from a status network like Twitter could be incorporated, with the ability to follow people to monitor their requests, either with a view to fulfilling them yourself, or to pass them on. Requests can be anything from fixing a computer, mending the roof, baking a cake or getting Spanish lessons, to building a house, printing a book, transporting an item of furniture or manufacturing toothpaste. Complex tasks like construction work would spawn a whole host of smaller tasks which can be taken up by different people. People can comment on requests and offer to do the work; once the work is done, both parties confirm the transaction and a link is established between them in the network. We would also find the inverse of the request, namely the offer: while a request is a gift without a giver, an offer is a gift without a recipient. In either case, the network matches the two people up, who form a link in the network once the transaction is confirmed as complete. Like in other social networking services people would also associate voluntarily in groups - community groups, industry groups, user groups or anything else. Requests can then target groups instead of individuals: for example, a request to build an extension on a community learning centre could be submitted by an individual member of the community, but the activity doesn’t only benefit that person - it benefits the entire “Fictionville learning centre users’ group”. Therefore links would be drawn to every member of this group.

And yes, I envisage the network as being entirely public. This is, in effect, an ‘open source society’. If widely adopted it would guarantee a level of transparency in the economy that just isn’t possible today, effectively re-humanising the links of dependence that connect together complete strangers. Being public, every transaction can, potentially, be audited by the community at large. You can’t just request a nuclear bomb or offer to dump toxic waste into a river; you can’t just request an assassination or offer to strip mine a whole region. At least, not without being shot down by the community. This is similar to how open source software and open content like Wikipedia maintain the level of quality that they do - because mistakes, bugs and vulnerabilities are quickly spotted and corrected by the community. Even an inoccuous request could get incrementally modified with the suggestions of community members, a practice which might well be expected when community works are requested.

So why would people work essentially ‘for nothing’? Well, gifts are meant to circulate, and if you receive a gift from someone it is supposed to encourage you to give a gift to somebody else, not necessarily the same person, but with everyone connected and interdependent, the gift ‘comes back’ to the giver. What this system does is simply make the connections between people more visible, so that you can see how your work benefits others and how the work of others, directly or indirectly, benefits you. Alice might think there’s no point fulfilling Bob’s request, because they don’t know each other and Bob isn’t offering anything in return - but then she sees that Bob has a link to Charlie, and Charlie has a link to Alice. If Alice knows and trusts Charlie, and Charlie knows and trusts Bob, then this link of provision also becomes a link of trust, even though Bob is still a stranger to Alice. We can make the level of trust a bit more precise. Although I said that no values would be put on the work performed, there are a couple of quantifications which I think could be particularly helpful, the most important of which is the length of the shortest (directed) path between two nodes in the network (that is, between two people). When considering somebody’s request, you can find out the shortest path between you and that person, and this, I argue, is a measure of how ‘trusted’ they are. You could instantly see, for example, that a stranger who is asking for a new window pane is connected to you by a path of length three - specifically, they knitted a scarf for the person who looked after the cat of the person who gave you piano lessons.

Another potentially useful measure is the number of ‘introducers’ - people who are directly connected to you as providers and simultaneously the endpoint of a chain of provision starting with that person. In the previous example, the person who gave you piano lessons is an ‘introducer’ relative to person X - they might be the only one, or there might be a more circuitous chain of provision which goes around the world and comes back to you through the person who did your plumbing. Together, these quantifications, as well as direct inspection of the network, are indications that the person you are helping out is not just an isolated vacuum of your labour power, but an active participant in society, a conduit of gifts, and someone who has ultimately facilitated your own receipt of them in the past. Most likely, every given pair of people in the world will be connected by some chain of provision, and it might be surprisingly short.

For the !Kung people, a group of hunter-gatherers in the Kalahari desert, being wealthy is not about having possessions, it’s about being able to channel lots of goods between people - in other words, being generous. (See here for discussion.) And it would be the same in the gift economy I’m sketching here. Being able to channel lots of gifts means having lots of short connections to lots of people, and the system can quantify this by calculating the average of the shortest paths between X and every other person that they are connected to: more precisely, for each P in the provision network of X, calculate the shortest path between P and X and average the results. Lower numbers are good. These are the people you want to get in with. Even if the shortest path between you and X is very long, fulfilling a request for them will make you instantly much ‘closer’ to the large number of other people that X provides for. Hence doing things for a ‘conduit’ in the network puts you within range of the trust of many other people, a fact which turns ‘conduits’ into attractive targets for gifts themselves. So there are dynamics here which encourage people to be generous, and dynamics which encourage helping people even if they’re ‘distant’. (I’ll leave it as an exercise to the reader to work out if performing this calculation for everyone in the network is computationally feasible.)

One interesting property of this system is that no particularly special mechanisms are required to allow for provision to people who are unable to or unexpected to work, so long as they have carers. If a link is established from the non-worker to their carer (or indeed anybody), then they instantly ‘inherit’ their carer’s entire provision network, albeit displaced from every node by one extra link. Fulfilling a request for a non-working person is also doing something for their carer, who in turn is able to give something back to the network. Of course this same schema could be used or ‘abused’ by people who don’t want to work: it’s OK to be lazy so long as you’ve got generous friends! In fact, I don’t think ‘free riders’ are likely to be a problem. Part of my reasoning comes from knowledge of the open source software (OSS) movement. The vast majority of OSS is produced by people working voluntarily in their spare time and for no monetary compensation at all. Projects are released for free on the internet and anyone is free to use them and modify them as they see fit. Some OSS developers put enormous amounts of time and effort into their projects. Others dip in to various projects and contribute small patches here and there. And still others don’t develop at all - they just take the finished products and use them. Nowhere in the OSS movement is there any ill feeling about end users being ‘free riders’ who get something for nothing and never contribute. One reason for this is that developers enjoy coding as a hobby - it’s something they’d do anyway. So the gift economy also functions best when work is enjoyable and comfortable, so that people see it as an end in itself, an activity that they’d undertake regardless of any expectation of a return. Of course, there are qualitative differences between digital and non-digital content which make emulating the success of open source potentially quite difficult. There is, however, an independent reason to believe that free riders might be rare anyway, simply because some tasks might not fall under the remit of the gift economy at all.

The Gift Economy and the Planned Economy

It seems to me that the usefulness of the gift economy is restricted to a certain scope, and outside of that scope we may need something else. In another article, I describe a kind of democratic planned economy, where communities join together in a political and economic federation. A structure of popular assemblies is used to discuss people’s needs and to co-ordinate members’ activity towards the fulfilment of those needs. Suppose that the two types of economy co-exist in the same society. (And there’s no need for the gift economy to be network-enabled in the way I described above. It could have a number of different forms, but the use of a social network just seems appropriate for our times.) The gift economy would probably be most effective for those things that people currently use time banks, mutual credit and personal favours for - little jobs that can be done by one individual for the benefit of another. The usefulness of the system for more complex tasks in manufacturing and construction remains to be tested. It is probably least suitable for work that is of wide social benefit, like maintaining transport networks or the sewage system. These tasks benefit practically everybody practically all of the time, and require a fixed and regular input of labour rather than sporadic individually tailored requests. There is no argument here that we need markets or money to cut the slack. On the contrary, the facts suggest that these tasks require a co-ordinated community effort to make sure that they get carried out - in other words, planning. Food production would probably also fall into the remit of planning, because of how essential it is to the survival of a community.

We’re now in a position to address another issue with the free-for-all gift economy described above - namely, resource management. Who owns and controls the raw materials needed for industrial activity? How do we ensure that there are enough resources to fulfill all of the requests? In the event of scarcity, who decides how the resources get used? How do we simultaneously ensure that resource use is sustainable? First of all, there are a few dynamics of the system which have some mileage in addressing these vital concerns. The open nature of the network enables a healthy amount of community censure, which - with the right amount of vigilance on the part of the community - could go a long way in preventing unsustainable or environmentally damaging activities. In addition, the bias given towards ‘generous’ or ‘conduit’ nodes is a fascinating way to give an inherent advantage to projects which are socially useful. If a request targets a community group or the user group of a community asset, like schools and hospitals, then anybody fulfilling the request stands to gain a large number of trust connections with the community in question - it decreases your average shortest path. And the software would allow you to order requests according to the shortest path metric between you and the requester. But are these measures enough? I’m certainly not going to start claiming that gift networks are inherently capable of solving all problems and righting all wrongs as free market zealots try to claim.

The planned economy and its federal structure provides the much-needed antidote: communism. In such a society, resources and means of production - mines, forests, fields and factories - are society’s common heritage. Even with a gift economy, private property would only perpetuate class divisions and the exploitation of the have-nots by the haves. With resources owned in common, however, everyone is able to exercise their skills and participate in the economy. If you’re a carpenter and you see a request for an item of furniture, you go to the community workshop and use its resources to fulfil that request. With private property, the owners of the workshop, the tools and the raw materials could claim a local monopoly on furniture requests, and you’d either have to submit yourself to their authority, or else let your talent go to waste. And while the communist solution allows a degree of autonomy (based on the concept of usufruct) over resource use, there is still a necessary accountability to the community. Decisions made at popular assemblies would establish guidelines about what sort of requests or offers in the gift economy are acceptable, regulating resource use so that it is fair and sustainable.

Just as our current economic system is a hybrid of a market economy and an (authoritarian) planned economy (as attested by Coase), this is also a hybrid, of a gift economy and a planned economy in a libertarian communist context. And similarly, the balance between these economies could vary from place to place, resulting in a pluralism of anarchist societies. In some societies, the majority of work would be planned by the federation, leaving the gift economy for odd jobs performed between individuals. In others, the federation would get everyone working a few hours a day in a key industry, rotating people between food production, transport and utilities, and leave everything else to the voluntary initiative of the gift economy. For these and other reasons we can see the potential for a diverse political landscape in a post-capitalist world.