Norms and Networks

_The Wealth of Networks_ is a very important book, not only for people involved in debates on information and technology policy, but for the left as a whole. While it clearly builds on work by Larry Lessig, James Boyle, Pamela Samuelson and others on intellectual property and the public space, the real contribution (as with some of these other writers) is to a broader tradition of thought; that of people like Jane Jacobs, James Scott, Richard Sennett and Iris Marion Young. Benkler’s vision of the good society is one in which people have a high degree of autonomy, so that they have the practical capabilities to pursue their own interests and their own forms of cultural expression. He argues that new communications technologies, if they’re left unhampered, might radically increase this practical autonomy. They can do this by making it far easier for individuals to become producers as well as consumers of culture, and to share their cultural products with each other, so that they can build collectively on each other’s work. On the one hand, technologies such as the Internet allow us to engage with each other in new ways, and to form networks of collaboration and of conversation, creating possibility conditions for the kinds of diversity and critical thinking that democratic theorists prize. On the other, these technologies do so in a relatively non-constraining way.

This is an important set of arguments for the left. It suggests that many of the values of the left can be achieved without either grand planning initiatives by the state, or the often stifling structures of community. Benkler is far from being a libertarian – he argues that the state should support education, health and other basic social goods as it has done in the past. More to the point, he argues that the new spaces of collaboration that he’s interested in are important precisely because they open up new possibilities for organizing production than free markets. But even if he’s not a libertarian, he’s interested in freedom; he offers a set of practical prescriptions that are intended to create spaces for the free play of diverse interests and sensibilities. His arguments about politics seem to me to have more in common with Elinor and Vincent Ostrom’s work on polyarchies than the claims of either liberal or communitarian political theorists. Benkler sees the state as ideally being an enabler of actions by individuals and loosely grouped communities of interest, rather than as a hands-on regulator. This is the source of his criticisms of the (US) state, which he sees as too eager to regulate these spaces, and as representing a set of interests (those of copyright holders) that are inimical to the spread of autonomy. The transition from an economy of cultural production that is based around a sharp distinction between producers and consumers, to one in which everyone is potentially a producer, will have clear distributional consequences. There’ll be winners and losers. If the potential losers have sufficient political clout, they can hamper, and perhaps even block change. There’s nothing inevitable about the emergence of the new economy of collaboration; depending on institutional configurations, it may be still-born.

If it’s not clear already, I’m broadly in agreement with Benkler’s normative claims and diagnosis. Still, I believe that there’s an important piece missing from his argument. Benkler is a lawyer, and while he’s concerned with institutions, his primary concern, naturally enough, is with formal institutions – that is government issued and government enforced laws and regulations, which can be interpreted by the courts. But the kinds of exchange that he’s interested in rely less on laws, than on informal institutions and norms are what anthropologists call “gift exchange” and “generalized reciprocity.” They don’t relay on formal rules so much as informal institutions and norms. Benkler doesn’t completely neglect this – he discusses it at several points in the narrative, but doesn’t say as much as he might about the internal dynamics of these informal rules and norms, and how they affect exchange. While we (by which I mean social scientists including anthropologists, sociologists, economists and political scientists) still don’t understand as much as we’d like to about the ways in which these informal institutions can change over time, exactly such change is an important problem for the kinds of decentralized order that Benkler is interested in. We can’t just look at changes in formal institutions, and how these may make technology-based cooperation easier or less easy. We also need to pay attention to the factors that might precipitate changes in informal norms. If informal norms change, then so too may we expect the kinds of cooperation that they support to change. And some of these changes may have negative repercussions for the kind and level of cooperation that we might expect.

We can see this if we look at the example of the blogosphere, which occupies a prominent place in Benkler’s book (and, indeed, in my own thoughts). The blogosphere, as I see it, works as well as it does because of three key norms – linking, attribution and authenticity. The first norm – linking – is that when one wants to criticize, agree with, or otherwise comment on a particular item of source material, one should link to the source material if it’s available online. The second norm – attribution – suggests that when one has come across a particular piece of source material thanks to another blogger, one should credit the blogger in question by providing a link to him/her (perhaps sending some traffic back in his/her direction). Sometimes both of these norms overlap – when the original source material is a post by another blogger, then linking to that post will satisfy the norms of both linking and attribution. But sometimes they don’t. To take a topical example: when I posted recently on the Wealth of Networks, Matthew Yglesias wrote a “response post”:http://yglesias.tpmcafe.com/node/28875 which both linked to the Wealth of Networks home-page (satisfying the norm of linking) and to my original post (satisfying the norm of attribution).

The final norm (or, more precisely, set of norms) is a bit more difficult to pin down – it’s a norm of authenticity. Roughly speaking, I take this norm to say that individual bloggers should represent their own points of view in an honest and straightforward fashion. The comparative advantage of bloggers vis-a-vis other kinds of pundits is that they have (or should have) a strong personal voice based on their internal beliefs. This distinguishes the blogosphere from many other spheres of publication, where individuals are expected to represent the positions of their institution, or their political party rather than their own personal position on the issue at hand. It also distinguishes blogging from genres of writing (op-eds, speeches, political autobiographies) where authorship is blurred and ghost-writing by others than the official author are considered to be perfectly acceptable. Bloggers who are perceived as not representing their own position on the issues, or as having their material written for them by others, are likely to have a hard time getting their writing accepted by other bloggers.

None of this is to say that these three norms are universally adhered to in practice; few norms, if any, are. But they are adhered to frequently enough to shape behaviour and expectations. The norms of linking and attribution provide a basis for both generalized reciprocity and for more specific patterns of gift exchange. Bloggers who don’t have many readers, but who feel that they have a post that might attract a large audience have an incentive to inform well known bloggers of their post, because this norm gives them a reasonable expectation that the well known blogger won’t simply steal their idea. Instead, it is likely that the well known blogger will provide a link back to the original post, perhaps providing a massive (albeit perhaps temporary) influx of readers. Over time, this may allow less well known bloggers to become well known themselves. For example, Glenn Greenwald, after a series of posts that were well received among prominent left wing bloggers, has himself attracted considerable readership and attention for his blog, and for a book that builds upon arguments first presented in his blog, which is #1 in sales at Amazon.com as I write. Furthermore, as Jack Balkin has argued, norms of linking provide at least a minimum of exchange of views across ideological boundaries. In order to criticize people who say things that you harshly disagree with, you have to link to these people in the first place. There are limits to the efficacy of this norm in promoting exchange of views. It’s not clear how many people actually follow cross-ideological links (I recall, although I can’t find it, a post by Mark Kleiman where he said that approx. 1% of Glenn Reynolds’ readers followed the link when Reynolds said something rude about him). Furthermore, many prominent bloggers on left and right link to other, less-well known bloggers critiques, rather than directly to the subject of the critiques. Even so, Balkin’s point stands – the structure and norms of linking allow a degree of cross-ideological exchange that are nearly impossible for other media (such as print) to emulate. Finally, the norm of authenticity is what gives the blogosphere its flavour. I may think that some of the bloggers whom I argue with and criticize are political hacks. I don’t usually believe that they’re bought-and-paid-for political hacks (and if I did, I’d probably criticize them in very different ways).

If norms of attribution and linking didn’t exist, then, many of the attractive features of the blogosphere would disappear. There would be much less bubbling up of ideas from small bloggers to big bloggers, and hence to a wider audience. There would also be less exchange of views across ideological boundaries. If the norm of authenticity didn’t exist, then the blogosphere would be (for me at least, but also, I suspect, for many others) a much less interesting place, something much closer in spirit to the empty and sterile back-and-forth of flacks and politicians on op-ed pages. Now it may well be, as Balkin argues, that “the customs of linking” (and perhaps, although Balkin doesn’t claim this, the norms of authenticity) are relatively robust, because they “make sense, given the way that the technologies work.” But it seems reasonable to me to at least entertain the possibility that they’re not robust in this sense, and that they could change if actors’ underlying incentives change, or if all actors don’t have the right incentives.

Two possible factors that might precipitate this change are money and invasion. First, money. Via Michael Froomkin, this argument by Edward Leamer seems worth exploring.

bq. We [academics] are part of a “Self-Organizing Collaborative Community” called the research universities of the United States and increasingly the rest of the world. Unlike contributors to Wikipedia and Linux, we get paid for our work, not by those who consume the fruits of our labor, but by taxpayers and by donors and by our students, all of whom we have convinced are better off by virtue of the research that we do. When it got started fifty years ago, this system worked great, but it isn’t working as well anymore. While we are doing plenty of worthwhile research we are also doing plenty that isn’t worthwhile, and the competition for research talent defined by the fads of the moment is driving up the cost of education to unaffordable levels. Adam Smith would have understood what’s wrong here. It takes sales for the invisible hand to do its magic. Begging in your work clothes when you aren’t working isn’t enough, even though the pastime may be lucrative. On the contrary, the more lucrative is the begging, the more likely is the conclusion that the work is worthwhile. But it takes accurate market prices to tell us what’s valuable and what’s not. Of course, good will and good intentions can carry a collaborative community productively for a while, but financial rewards relentlessly bend the system to their will, slowly perhaps, but inevitably. That’s the invisible hand at work. Thus, open-sourcing has the same problems and the same probable longevity as the communes of the 1960s — they worked great for a while, but the participants chose other ways to live once they got to know the people in the community.

There are two parts to Leamer’s argument, a normative claim and an empirical claim, which blend together in a confusing way in this paragraph. The normative claim is that the price mechanism is superior to other metrics in telling us “what’s valuable and what’s not.” This can be swiftly disposed of – it’s a common trope among economists, but one that is, at the very least, controversial as a general argument. The second claim is much more troubling: his suggestions that “good will and good intentions” are inevitably going to fold in the face of financial rewards. We don’t have to accept his normative claim, or indeed his suggestion that this is an ineluctable social process to agree that this may be a real problem. Even if we want to celebrate the possibilities of collaborative communities, we can realistically worry that they may be vulnerable to disruption by market forces, given historical examples of how decentralized systems of collaboration have been destabilized by markets in the past. To take one example, which. Benkler mentions in passing, amateur sports, in the nineteenth century sense, have given way to professionalization in very nearly every field of sporting endeavour as star athletes have gone after the money. The point here is not to celebrate amateurism uncritically (there was a lot that was shoddy about it – distinctions between players and gentlemen etc), but to use it as an example of how collaborative fields of endeavour may be disrupted by financial pulls from outside.

How could the norms structuring the blogosphere be disrupted by money? One plausible way might be as follows. Some bloggers earn quite substantial amounts from advertising, and have tried to capitalize on this through creating miniature publishing empires, such as the Denton publishing stable, the soft porn Suicide Girls blogsite and so on. These commercial enterprises have very different incentive structures than not-for profit blogs (or indeed blogs that only earn pocket money revenues from blogging). Specifically they have much weaker motivations to provide external links. They want their readers to stay with them (or to go to other blogs with the same owners, or to websites that are prepared to pay for external links). They’re in it for the money. This suggests that they have far less incentive to follow the norms of linking and attribution than other blogs do. All other things being equal, as time passes, we might expect that there will be less and less external links from these blogs to other parts of the blogosphere. We might also plausibly expect a weakening of the ethic of amateurism, and something much closer to the ethos of the traditional print media emerging. In short, we might expect an unwillingness on the part of for-profit blogs to acknowledge other blogs by linking back to them, and a consequent degradation of the dense networks of communication that allow interesting ideas to percolate rapidly through the blogosphere. They would become more like newspapers, which are notoriously unwilling to acknowledge the scoops of competitors.

Now the crucial clause in the argument above is “all other things being equal,” because other things aren’t equal, at least at the moment. We haven’t seen this become a major problem among political blogs, in part because nobody has really figured out a successful way to consistently make large amounts of money from blogging. So far, efforts to create blogging conglomerates such as Pajamas Media have been dismal failures. But this isn’t a structural necessity, and it may change in the future. More generally, the key point is that the success of the blogosphere as a network doesn’t just rest on the technological possibilities offered by hyperlinking. It also rests on a set of norms about when hyperlinks are appropriate, and there is no ex ante reason to believe that these norms are invulnerable to change in actors’ incentives, or that actors’ incentives mightn’t change in the future.

The second vulnerability that the blogosphere faces is that of invasion. Here, the norms most at risk are those surrounding authenticity. Blogs prize authenticity – but they don’t have any very good means of authentication. The blogosphere is highly open to newcomers, but this also renders it vulnerable to invasion by actors who don’t share the social commitments that most bloggers have. Indeed, we’ve seen evidence of something like this already, in the degradation of two aspects of the networked blogosphere. First has been the invasion of comments sections by spammers. I’m not one of the first generation of bloggers, or even the second, but I have been around long enough to remember when comments sections were open to all comers. Unfortunately, exactly this openness made blog comments sections vulnerable to spammers who wanted to increase websites’ Googlejuice by inserting bogus linkbacks into comments. As a result, open comments sections are relatively rare in the blogosphere today. Most blogs have at least a minimal set of hoops that commenters need to jump through before they’re able to leave comments; some have strong registration requirements. While there isn’t any ideal technical solution for comment spam, at least there are acceptable workarounds. The same, unfortunately isn’t true for the second aspect of blogging which has been degraded – the Trackback system. Trackback, which when it worked provided a real reflexivity to conversations, by allowing the readers of a blogpost to see which other blogs had linked to that post, and follow links back to them. Trackback is almost completely unusable thanks to spam links these days, and alternative workarounds (such as using Technorati or other services) are considerably inferior.

How might invasions of this sort threaten authenticity? Through a specific kind of invasion – astroturf blogs – which I suspect are going to pop up like mushrooms everywhere during the coming electoral cycle. The example of the Thune v. Daschle race, where individuals paid by the Thune campaign ran purportedly independent blogs, which apparently helped sway local media coverage, suggests that there’s a lot of scope for astroturf blogs – i.e. blogs that claim to represent authentic grassroots opinion, but are in fact written to order by paid operatives working on behalf of particular interests or political parties. As blogging becomes more closely integrated into campaign activity and public controversies, we’re likely to see more and more dubious use of blogs to inject slurs and falsehoods into political debate. It seems to me that prominent bloggers like Kos and Jerome Armstrong have negotiated the tricky boundaries between political consultancy and blogging with admirable honesty – but it’s not at all clear that they’re going to be the rule rather than the exception. None of this is to say that these problems are at all unique to blogs – but they’re likely to plague blogs in particular, because of the very low barriers to entry. I’d like to think that astroturf blogs will be less likely to succeed in winning attention than the genuine variety, but I wouldn’t want to bet my life on it.

None of these problems is necessarily insuperable. There’s no ineluctable historical reason why systems based on gift exchange and generalized reciprocity have to give way to markets. The two co-exist, and systems of informal trust and exchange are sometimes reproduced at the heart of capitalism (cf. the sociology of stock markets and of diamond merchants). As Benkler points out when he talks about Slashdot’s system of moderation, technology can help at least to some extent. Furthermore, more elaborate forms of self-regulation of the kind found on Wikipedia can help too. But in order to understand the problems, and how they might be solved, I think that we have to pay attention not only to the formal rules that might stymie free exchange of ideas and cultural products, but also to how best to support the informal norms and institutions that structure this exchange. These informal rules too are subject to change over time, and may, in the absence of considered responses, change in ways that make conversation and interplay less likely or less attractive. It may well be (and I don’t want to overstate my case here) that the forms of exchange that Benkler rightly celebrates are more vulnerable than they appear to be at first, even in settings where there isn’t formal regulatory interference.

Furthermore, to the extent that problems do arise, the best available solutions (which I take to be self-regulatory ones) also carry costs. Self regulatory solutions are likely to help reduce the risks of bad behaviour, but they also, by the same token, reduce spontaneity and reintroduce a certain level of hierarchy. The structures underlying Wikipedia seem to me to be becoming more rigid over time. I suspect that the same is going to be true of the blogosphere too over the next few years. Less freewheeling conversation; more codes of conduct, implicit guild-style self-regulation and development of technologies to inhibit bad behaviour. This seems to me likely to happen – but it’s hard not to lament what we would lose in the process.