Careful, you're contagious

Six Degrees - Uncanny Networks
October 31, 2003

A few days after September 11 2001, many New Yorkers were feeling guilty. They were uneasy because they had suffered too little inconvenience and disruption. A mile or two away, thousands of people had been murdered in the destruction of the city's most visible buildings. But for the survivors, the systems that kept the lights on, delivered food to shops and took away the sewage were undamaged. Fast-forward six months, and the picture was quite different. New York was in trouble. Businesses were closing, hotels and restaurants were empty, and skilled workers were losing their jobs.

New York is above all a network of people and organisations. The aftermath of September 11, as Duncan Watts points out in Six Degrees , shows networks in action at their best and worst. They can give strength but they can also multiply weaknesses.

As these two books show in very different ways, networks are being talked about as the underlying principle of a wide range of living and non-living systems. The worldwide web has attracted attention to networks because of its unplanned global spread. Human societies are being looked at in terms of networks, and network thinking is being applied to events such as the spread of the Sars virus.

As with any emerging discipline of wide application, there is a lot of loose talk, and Watts' book is a good antidote. Some of the new discussion about networks codifies things that sound too obvious to need stating. For example, you already know your friends. So the way to gain access to a new sphere of influence is likely to be via someone you know slightly or have just been introduced to. Watts is a frontline networks researcher at Columbia University in New York. He shows that away from all the breezy talk about a new knowledge paradigm, solid science is going on. For instance, the idea that weak links between groups - people who are in contact with more than one group without being core members of any - have disproportionate influence, has been proven formally rather than simply being observed. Data are being gathered, methodology developed and results demonstrated. Later, he says, the user's handbooks will be written. The real question is whether the network science being developed is as universal as he would like to think. Like other new fields before it, the subject has been bedevilled by physicists, including Watts, who believe that they are uniquely placed to tell everyone else about the universe.

After all, he claims, at least his subject has more to show than economics when it comes to useful outputs. At one point, he and a colleague give themselves four months to uncover the basic facts about human social networks, and make a point of not reading a word of the voluminous literature on the subject for fear of distracting themselves.

Watts has a sense of humour however and the book is good at explaining how scientists in this new field work. He does not oversimplify, and the book does contain some complicated arguments but brings rich insights. An example he explores is the 2001 outbreak of foot-and-mouth disease in the UK. Here, network thinking shows that the approach of cutting links was the right one. Bans on movement of cattle and on livestock sales were the right approach, and issues such as vaccination were a sideshow.

The science shows that power and influence in networks are not distributed along some sort of bell curve with a central hump. Instead, like personal wealth, they obey a power law in which a few people have a disproportionate share. It is well known that the internet depends on a few large computers whose disabling would cause the net to collapse worldwide. If a little money is available, these are the parts of the system worth reinforcing, protecting and backing up.

The same insight tells us something fundamental about how to attack a network. Critics of the US-led "war on terror" point out that al-Qaida is not an organisation whose office can be bombed, or a country that can be occupied. True, but as a network it must have some vital connections - perhaps key communications links - whose removal would be damaging to its capacity.

Six Degrees (called after the now-famous "Six Degrees of Kevin Bacon" game, which uses the actor to illustrate Hollywood's network connections) may also attract readers anxious to know how networks might help them to turn a profit. The vital concept, Watts suggests, is the "cascade effect" whereby once a few key people talk about something, everybody else does too. Watts points to the unstoppable rise of Harry Potter. J. K. Rowling may not be the best children's author - there is no way to tell - but her work set up a cascade.

Uncanny Networks by Geert Lovink is ostensibly on the same subject as Six Degrees , but comes at the topic from an angle so different that the overlap is negligible. Its publishers seem to lack confidence in the book, supplying it with both a foreword and a series foreword that rival each other in pomposity, but not bothering to provide an index.

The main text consists of interviews between Lovink, a media and net theorist, and a variety of people who have studied new media and cultural networks. Coming to his book from Six Degrees , one is struck by how little his interviewees know about the wider growth in network thinking. Some of the content is blatantly unattached to the subject - for example, a general discourse on Albanian politics. On the plus side, this book has a welcome emphasis on continental Europe rather than the US, and on under-reported minority media.

Uncanny Networks is an easy book to dislike. Its overblown language means that almost any page has a quote that invites mockery. A more serious objection is that few of Lovink's interviewees seem to do real research on where media come from, how they are used or what the effects are. Concepts such as a proper statistical analysis of how a weighted sample of people makes use of new media are not their thing. Their modus operandi is to fly into a new city, visit some cutting-edge clubs, and produce portentous conclusions about the future on the basis of the conversations about new media that they might be able to recall the next morning. In addition, some of Lovink's subjects believe that the change they are living through in today's network society is so huge that it is not worth knowing anything about the past. Dietmar Kamper says that until the 18th century, wars were "a kind of game", which nobody who has read any history would regard as plausible.

Despite these problems, Uncanny Networks is full of insights into new media thinking and will be used as a mine of handy quotes for students and thinkers in the area. Norbert Bolz, for example, points out that Nietzsche was the first to discuss the various uses of "medium", from technical ones to the world of telepathy. Arthur Kroker is interesting because he places the media revolution in its industrial context. Also praiseworthy is Saskia Sassen, who discusses the politics of who does and does not get access to the communications bandwidth that is the key to new media.

Uncanny Networks is an insight into debates whose participants attach little value to clarity but who are working on vital issues. It is a pity that Lovink likes his subjects too much to go for the throat very often. He gets tough with only one interviewee, Jonathan Peizer, who set up the Soros Foundation's internet programme in Eastern Europe. Peizer is one of the few people in the book who is doing something useful instead of criticising what everyone else is doing.

Martin Ince is contributing editor, The THES .

Six Degrees: The Science of a Connected Age

Author - Duncan J. Watts
Publisher - Heinemann
Pages - 368
Price - £20.00
ISBN - 0 434 00908 3

You've reached your article limit.

Register to continue

Registration is free and only takes a moment. Once registered you can read a total of 3 articles each month, plus:

  • Sign up for the editor's highlights
  • Receive World University Rankings news first
  • Get job alerts, shortlist jobs and save job searches
  • Participate in reader discussions and post comments

Have your say

Log in or register to post comments