Why do social networks work?

by John Q on May 30, 2006

Yochai Benkler’s The Wealth of Networks: How Social Production transforms Markets and Freedom is full of interesting things to discuss, but the point that interests me most is the question of why people contribute to social production and what economic and political implications it has, for states as well as for markets and freedom. Benkler previously discussed the same question in Sharing Nicely, and I’ll talk a bit about this as well.

To the extent that there is a conventional wisdom about these things, it’s Eric Raymond’s idea of a gift economy, derived from his participation in the open source software movement. Raymond focuses exclusively on reputation as a motive for contributing to social production, seeking to assimilate all other motives (such as craft values) to reputation. It’s precisely this kind of totalising logic, I’ll argue, that is absent from social production.

On the contrary, there are a wide variety of motives which might lead people to contribute to networked social capital, for example by participating in various aspects of blogging (make posts and comments, linking and blogrolling, improving software, various kinds of metablogging). Possible motives include altruism, self-expression, advocacy of particular political or social views, display of technical expertise, and a desire for social interaction. Particularly in relation to a collective, and largely anonymous, product like Wikipedia, Raymond’s central focus on reputation does not fit the facts.

As against these various motives, there are the two standard methods that have been relied on to deliver most information production and innovation for the past 150 years: markets and bureaucracies.

Benkler discusses markets, and the associated profit motives, at some length, making the point that narrowly economic motives tend to crowd out alternative forms of motivation. He mentions the classic work of Titmuss on blood donations and some other examples to show that monetary and social-psychological motives are likely to conflict, rather than reinforcing each other. By contrast, different social-psychological motives are usually complementary or at least mutually consistent.

Why is this? At a superficial level, it’s obvious that people act differently, and are expected to act differently, in the context of relationships mediated by money than in other contexts. Behavior that would be regarded favorably in a non-monetary context is regarded as foolish or even reprehensible in a monetary context.

One of the most important general differences relates to rationality and reciprocity. In a non-market context, careful calculation of costs and benefits and an insistence on exact reciprocity is generally deprecated. By contrast, in market contexts, the first rule is never to give more than you get.

This rule applies in market contexts but not in social contexts, where such careful calculation is, as Benkler notes, generally deprecated, because markets create opportunities for systematic arbitrage that do not apply in other contexts. In an environment where exchanges are not carefully calculated, a trader who consistently gives slightly short weight can amass substantial profits. If trading partners assume honourable behavior, none will suffer enough to notice, but eventually arbitragers will drive out their less calculating trading partners.

Similar points can be made about other motives. There are a whole range of sales tricks designed to exploit altruism, friendship, desire for self-expression and so on. Hence, to prosper, or even survive, in a market context, it is necessary to adopt a view that ‘business is business’, and to (consciously or otherwise) play a role as a participant in the market economy that is quite distinct from what might be conceived as one’s ‘real self’. This is a prime example of what Goffman calls an obligatory role.

The crucial feature of economic motives in a money economy is not that they are less noble or desirable than alternatives such as desire for fame, but that a money economy provides a total system of rationality, from which most of the motives associated with social production are excluded.

Markets are not the only total system of rationality that operate in a modern society. bureaucracy and the state have a logic of their own. For most of the 20th century, the central issue of politics and economics was the question of where the boundary between markets and bureaucracies (public and private) should be drawn.

Benkler largely ignores the state. As he says (p. 16) ‘In much of [my discussion], the state plays no role or is perceived as playing a primarily negative role, in a way that is alien to the progressive branches of liberal political thought’. But this position overlooks the critical fact that both the Internet and the World Wide Web were developed primarily by state agencies or state-funded institutions (DARPA, NSF, the university sector, CERN, NCSA and so on). Yet this outcome was not the product of rational bureaucratic planning. Rather, like Topsy, the Net and the Web ‘just growed’.

Like market rationality, bureaucratic rationality implies a complete specification of behavior When dealing with a representative of a bureaucracy, we (mostly) expect consistent application of rules, rather than an adherence to standard social norms such as ‘look after your family/mates before others”. Similarly, and more crucially, bureaucracies are supposed to allocate their resources to the achievement of specified goals, not to do things because they would be fun, or even socially beneficial. All of this seems to leave little room for social production. So how did the state come to give us the Internet? More significantly for our present purposes, what kinds of public policy will facilitate the further growth of social production and the wider distribution of its benefits?

I don’t have a complete answer to either question. However, some obvious implications run counter to current developments in policy.

First, if monetary returns are weakly, or even negatively correlated with the value of social production, there’s no reason to expect capital markets to do a good job in allocating resources to supporting innovation. It follows that the policy orientation of the past thirty years, in which increasing reliance has been placed on capital markets, is going in the wrong direction. The examples considered by Benkler are illustrative. Both blogs and wikis have their roots in the late 1990s, a time when capital markets were splashing hundreds of billions of dollars around on Internet-related projects. Most of these projects came to nothing, while blogs and wikis developed with little or no venture capital money to help them along.

A second implication is that the policies of New Public Management, which attempt to tighten bureaucratic accountability and focus on competitive disciplines and measurable outcomes may be misguided. Rather than seeking to drive people harder in the search for increased productivity, government macro-economic policy should be oriented towards making room for creativity and facilitating its expression. Similarly, while competition has its place, public policy should be at least as much concerned with promoting co-operation. The assumptions we have had about the competitive nature of innovation are, therefore, undersupported in the new environment in which we find ourselves. If governments want to encourage the maximum amount of innovation in social production then they need to de-emphasize competition and emphasize creativity and cooperation.

{ 2 trackbacks }

orgtheory.net » Blog Archive » undertheorizing networks?
05.30.06 at 11:07 am
Crooked Timber » » Introduction: The Wealth of Networks seminar
05.30.06 at 12:12 pm

{ 15 comments }

1

john c. halasz 05.30.06 at 12:41 pm

Just to add a point about the motives of social production, hasn’t a prime driver for cooperative endeavors on the internet to date been economic: namely, rebellion against monopoloid production and the subversion of the forced consumption of its products. Certainly, a prime motive of the open-source movement in software was the perceived deficiency of commercial/proprietary software, egged on by the exclusionary monopolistic practices of Microsoft, which blocked off avenues of innovation, no? Technical oneupmanship and pride in craft were backed up by frustration and resentment at the inefficiency and uninteroperability of software and the stultification of work conditions. Similarly, hasn’t much blogospheric activity aimed at the subversion and Umfunktionierung of the products of the “consensual” MSM and of commercialized mass culture? The “authoritative” subservience of corporate media to a centralized “consensus” of power and the imposed standarization of cultural signifiers of expression and satisfaction into a mediocre and passively experienced “reality” provoke the need to reach out into the virtual “reality” of the internet in the hopes of an unpredictable growth of alternatives. It’s a strange marriage of techne with praxis. But, then, doesn’t “social production” on the internet remain “virtual”. How does it exit the “virtual” conditions of its application to re-enter intersticially the larger institutional orchestractions of public reality and effect actual changes and alternatives in that reality?

2

pdf23ds 05.30.06 at 3:29 pm

The online sphere isn’t any more or less “real” than other spheres of social and political action and interaction. It’s simply less connected to certain other spheres than yet other spheres are. The internet, cultural attitudes, mass media, the publishing industry, political power and speech are all different spheres who have varying degrees of cross influence. The internet simply doesn’t have as much influence on other spheres as they do on each other—it’s not as integrated as it could potentially be. I think this framing of matters is much more useful than any talk of “real” or “virtual”.

3

Tracy W 05.30.06 at 8:55 pm

In a non-market context, careful calculation of costs and benefits and an insistence on exact reciprocity is generally deprecated.

In an environment where exchanges are not carefully calculated, a trader who consistently gives slightly short weight can amass substantial profits. If trading partners assume honourable behavior, none will suffer enough to notice, but eventually arbitragers will drive out their less calculating trading partners.

Isn’t there substantial evidence that a chunk of our brain is devoted to spotting cheaters? And that it doesn’t get turned off in social situations?

In social environments, I notice an effort amongst people to appear to be more generous than is strictly necessary in the hope of avoiding being regarded as someone who is taking advantage of minor discreprenancies to win out. Eg a common exchange amongst friends/workmates/etc is for one person to lend another a small sum of money, say $1 or $2, for a bus fare. The person borrowing the money will promise to repay, the person lending it will wave their hand and say it doesn’t matter. $1 or $2 is very insignificant in the terms of weekly salary of everyone involved, but still the person borrowing will almost certainly repay it. Another one is fights over who will pay the bill with the winner winning. And letters to advice columns frequently mention concerns about a “friend” who appears to be breaching social reciprocation.

There are some exceptions, eg between different generations of a family, between dating couples when a man is trying to impress women with his earning power, but then many a law firm’s reception and meeting rooms strike me as trying equally to impress the visitor with the firm’s financial status.

And it is not the first rule in market environments to never give more than you get. If that was the rule, pharmacists would never voluntarily sell life-saving antibiotics. What does the customer get? Perhaps another 50, 60 years of life? What does the pharmacist get? Even accounting for different levels of government subsidies, probably far less than $10 per sale. Markets work in the long run because both parties get more than they give.

I am also rather doubtful about your claim that “If governments want to encourage the maximum amount of innovation in social production then they need to de-emphasize competition and emphasize creativity and cooperation.” I was on the Internet from the mid-1990s on and I recall a fair bit of competition as well as cooperation. Even non-VC people were developing various technologies, some of which turned out to be useful and some not. People on the internet can chose from a vast number of different websites and technologies, and chose the ones that best fit their needs.

And on the other side, markets have never excluded cooperation. That’s how limited liability companies came about. That’s how companies came about in the first place. Sports leagues manage to be very competitive between teams, while seeing lots of cooperation between teams. At the extreme, wars have sadly seen massive innovation in military hardware, as the result of cutthroat competition between groups and cooperation between various people within a group. The Manhattan Project involved massive cooperation between physicsts, engineers, technicians, the armed forces, etc, at a time of deadly competition between the Axis and the Allies. Implying a trade-off between cooperation and competition is a false dichotomy.

4

John Quiggin 05.30.06 at 9:18 pm

“Markets work in the long run because both parties get more than they give.”

Which is exactly consistent with the rule, “never give more than you get”.

5

John Quiggin 05.30.06 at 10:06 pm

To clarify the comment above, gains from trade make it possible for both parties to get more than they give, contrary to the pre-market intuition that gains for one party must be losses for the other.

But that doesn’t rule out the possibility of exchanges in which one party gains at the expense of the other, and there are always economic actors seeking out such arbitrage opportunities. To avoid being exploited by such agents, you need to be sure never to give more than you get.

6

john c. halasz 05.30.06 at 11:27 pm

pdf23ds:

Perhaps one could make a distinction between media of communications and institutions, organizations, communities, in which practices, actions, relations are embedded with consequences and their dysfunctions and conflicts, wherein norms are generated, recognized and applied, effecting actual distributions of resources, power, goods and statuses. The facelessness of the internet and the absence of any actual relations to which consequences apply,- (inspite of any blogospheric pissing matches),- are what legitimate the epithet “virtual”, though that’s no more pejorative per se than calling memory “virtual”. In fact, many of those other “spheres” that you mentioned, qua media of communication, would be no less “virtual” in their own way. (Indeed, they are often enough deployed to block off any awareness of really embedded consequences, dysfunctions and conflicts.) But if the new media of the internet present a great potential for aggregating and collating information and generating or altering social knowledges or understandings, the latter must still spill-over onto and be applied in really embedded circumstances. There may, indeed, be a great potential for the distributed networks of the internet to generate a new kind of social “surplus value”, but it’s by no means clear that that would occur “autonomously”, nor in an automatic or self-regulating fashion.

7

john c. halasz 05.31.06 at 2:09 am

Tracy W.:

If there is an inefficient distribution of resources, then a series of “Pareto-optimal” exchanges will move the entire system closer to the “Pareto frontier” to everyone’s benefit, so the theory goes. But if markets are continuously in operation, there would be few, if any, sequences of exchanges that would advance the system closer to the “Pareto frontier”, which is to say, enhance collective well-being, while improving the efficiency of distributions. The upshot is that markets unto themselves, beyond a certain point, do not produce economic surpluses. Rather productivity increases due to technological, organizational and logistical innovations produce the economic surpluses that market exchanges then distribute, which distributions are affected by prior distributions and the structural/positional effects of the foregoing organizational factors. And further, whatever the claim to reciprocity, qua voluntariness and mutual benefit, market exchanges are asymmetrical, since producers/consumers do not directly exchange with producers/consumers, but rather producers positionally exchange with consumers and consumers positionally exchange with producers, which is precisely the mediating function of markets. So there’s no reason to claim that markets necessarily result in an optimalization of mutual benefits, let alone that that benefit is continuous or lasting. To the contrary, the functionning of markets all but inevitably result in variable concentrations of market power, which, in turn, result in asymmetrical, unequal, perhaps only notionally voluntary distributions of productive surpluses, which, in turn, may or may not be “justified” by increased productive efficiencies, but equally ensure variable effects of relative wealth and poverty and their variable reproduction. But markets are quasi-institutions, such that participants enter into them notionally in terms of certain norms of competition, which, provided there is an adequate legal framework of regulations, result in an overall cooperative effect. And those norms of competitive cooperation are precisely those involving a determination not to be cheated, as J.Q. says, because of the asymmetries and resource limits that market distributions impose. The upshot here is that there is a quotient of ideological illusion in the very “rationalism” of your account, since markets unto themselves do not necessarily optimalize general welfare, do not embody “complete”, but only highly constrained voluntariness, do not extend ideally to all risks, needs and factors, tend toward the reproduction of inequities, and tend by the very logic of their operation toward oligoplistic concentrations of market power, that exclude competition, coerce “cooperation”, and “capture” other markets. (And I’m not even talking about the evolutionary psycho-babble, which itself imports illicitly economistic criteria).

By contrast, in the social “economy” of gifts, the gift itself represents a surplus, even and especially if it has no immediate utility, since it marks the acknowledgement of a relationship and the site of an obligation. It weaves together a nexus of cooperative social relations precisely because it foregoes immediate advantage in favor of serial distributions. It certainly does not abolish egotism, rather than anchoring it. To the contrary, it might well give rise to an agonistic competition of gift-giving, as in the potlatch ceremonies that secured social status. But the “destructiveness” of non-advantageous gift-giving is an investment in collectively shared potential. It might seem odd to figure the internet, with its technological hyper-modernity, facelessness, relational distantiation and sheer abstraction in terms of a “primitive” gift-giving “economy”. But perhaps that just marks the site of a displacement and alienation of needs, from without the universal abstraction and reification of pervasive market exchanges. But if, indeed, distributed internet networks do contain the potential for the generation of new kinds of social surpluses, then the first order of business is to prevent the capture of those surpluses by oligopolistic interests and organizations, in order to preserve the possibility of their re-entry and re-distribution to actual, really embedded social sites where needs and relationships are formed and counter-hegemonic strategies against a collapsed public sphere and against social dislocations and deprivations can take shape, in the face of the organized powers-that-be that draw their sustanence from them.

8

joejoejoe 05.31.06 at 5:37 am

John Quiggen: Rather than seeking to drive people harder in the search for increased productivity, government macro-economic policy should be oriented towards making room for creativity and facilitating its expression.

There was an excellent popular psychology book on this topic a few years back called ‘Hare Brain, Tortoise Mind: How Intelligence Increases When You Think Less’ by Guy Claxton. Here’s an excerpt related to this discussion:

“In one study, Carl Viesti asked his subjects to try and detect which of three complicated patterns was the odd one out, and looked at the extent to which their performance improved over a series of such tests. Although they were given plenty of time to examine each set, those subjects who were offered significant rewards for the correct detection performed worse, and learnt less, than those who were given only a token payment. Viesti concludes that ‘regardless of their size, monetary utilities [sic] do not appreciably increase performance on insight learning tasks, rather, their presence may interfere with such performance’.” pg 80, Claxton [Viesti, ‘Effect of monetary rewards on insight learning task’, Psychonomic Science, Vol. 23 (1971), pp. 181-3]

One political example of the success and failure of the New Public Management vs. Wealth of Networks would be the accurate intelligence forecasts of the tiny US State Dept. Bureau of Intelligence and Research (INR, see link) vs. the massive CIA. Here’s the Washington Monthly on the INR:

“‘The difference between the State Department and the CIA is that the CIA pays for information, and State doesn’t’, explained former CIA officer Baer, ‘and the reliability of paid-for information is often questionable’.”

I’ll say.

9

joejoejoe 05.31.06 at 5:38 am

Quiggin. My apologies.

10

mkl 05.31.06 at 9:25 am

John, a few quibbles:

a. The “never give more than you get” rule does not apply to speculative markets, e.g. venture capital or drug development, where the median outcome is a loss. Contributions to social capital can be viewed in this framework where participants are seeking aggregate gain in their social circumstances.

b. There is a whole lot of indirect compensation being derived from nonprofit social capital building enterprises on the net. I’d venture that Glenn Reynolds has done more with his hobby to build the reputation of the U.Tenn law school than his, or any of his colleagues, day jobs over the past 5 years.

c. Arbitrage is not short-weighting your customers, it’s providing capital and financial structure to reduce market inefficiencies, and the profit derived is the return on that capital and labor. Gotta defend my day job, too.

d. There is a very high degree of competition among cooperative social capital building schemes on the net. Look at all the social networking sites, blog communities, etc. that succed or wither in different niches. This dynamic is not too different than the success or failure of products on a grocer’s shelves.

Societies are networks, and people have not been materially changed by electrifying the network.

11

pdf23ds 05.31.06 at 9:49 am

John C. Halasz,

though that’s no more pejorative per se than calling memory “virtual”

To be clear, I’m not objecting to the “virtual” terminology because of its pejorative connotations.

Perhaps one could make a distinction between media of communications and institutions, organizations, communities, in which practices, actions, relations are embedded with consequences and their dysfunctions and conflicts, wherein norms are generated, recognized and applied, effecting actual distributions of resources, power, goods and statuses.

Goods can take different forms in different contexts. Goods exist on the internet, in actuality, in the form of sought-after words, images, and other kinds of bit strings.

Power and status only exist in the context of social relations, and social relations on the internet aren’t any less real (though limited) than physical ones, so in that sense power and status on the internet are just as real.

It seems to me that you’re trying to distinguish between those spheres affecting (did you mean affecting, or effecting?) “actual” goods, and those only affecting … virtual ones? And, I suppose, actual power and resources and statuses would be those relating to actual goods. But in this case, you’ve just pushed back the question, to that of which goods are actual and which are virtual.

One could make the case that goods necessary for the sustenence of life and health deserve a conceptual primacy, and we could perhaps call them “real” and others “virtual”, to the degree that they’re separated from the real ones.

Power on the internet (presently) has less potential to affect primary, or real, goods than do other forms of power, so it too could be said to be more virtual in this sense. (Though online marketers like Amazon show that the power isn’t zero, by any means.)

But the class of goods that affect “the sustenence of life and health” is actually not so easy to delineate. We can readily include various medications to treat life-threatening conditions, but what about medications for mental problems? Cognitive therapy? Self-help books? Internet support groups?

In short, while I think the distinction can be made, it’s difficult, and it doesn’t appear to be that fruitful. Rather than have some “real” realm of things, that different social spheres (governments, the internet) can affect in different capacities, I think we should analyze those spheres in terms of their effect on individual people.

The facelessness of the internet and the absence of any actual relations to which consequences apply […]

Again, you’ve begged the question of how to distinguish real (which has consequences) from virtual (which doesn’t have “real” consequences).

And in any case, consequences do follow from internet-mediated relations. If an online business fails, its employees are out of work. It’s true that there are few (but growing) political consequences following from actions in the blog sphere, but this observation fits more neatly into my framework.

12

John Quiggin 05.31.06 at 3:39 pm

“Societies are networks, and people have not been materially changed by electrifying the network.”

I agree. My point is rather than aspects of people’s ordinary behavior that are laregely suppressed in market and bureaucratic situations are more relevant in a network context.

A minor point is that the term “arbitrage” is used rather differently by economists from the way it is used in finance markets. So I don’t mean to denigrate your day job.

13

mkl 05.31.06 at 5:02 pm

JQ, I think even strictly, among the non-practicing, arbitrage is not worse than taking advantage of riskless, intermarket price disparities. So where the arbitraged counterparty may have been able to receive a better price had he known of and had access to the other market, arbitrage is not fraud. The job provides sufficient opportunities for denigration and opprobrium, yet prosecution’s a bit more than called for.

My thought is that the aspects of behavior you reference in respect to networks are more common and less suppressed in market environments than indicated, and that market practices do substantially impinge on networks, for better and worse. Viz, the substantial difficulties in maintaining open blog comments of late, which are appreciated.

14

Mary Kay 05.31.06 at 6:01 pm

A tangential point. I’m not sure Eric’s gift economy idea came from the open source software community. Eric is also involved in the science fiction fandom community where the gift economy, with payment in egoboo, has been a staple for going on 70 years.

MKK

15

John Quiggin 06.01.06 at 1:03 am

“market practices do substantially impinge on networks, for better and worse. Viz, the substantial difficulties in maintaining open blog comments of late, which are appreciated.”

This is an important point, which I think is consistent with my theme of conflict between market rationality and co-operation. Spam is a fine example of socially destructive arbitrage.

Comments on this entry are closed.