Pirates and rebels

Mynttorget in Stockholm during the June 3, 2006 pro-piracy protest. Image: Jon Åslund.

The Pirate Party in Sweden codified activist organisational strategy and led the way to Extinction Rebellion's success.

The novelty of Swarmwise is the integration of the horizontal - a traditional hierarchical organisation, with the vertical - the swarm of self-actualised activists taking their own initiative.

“New ways of organising go beyond just breaking the old rules into downright shredding them — leaving executives in the dust, wondering how that band of poor, ragtag, disorganised activists could possibly have beaten their rich, well-structured organisation.”

The promise made by Rick Falkvinge in his 2013 book Swarmwise: The Tactical Manual to Changing the World is bold and seductive: “It is possible for one person to set out to change the world and succeed.” It should be taken seriously because of Falkvinge’s own remarkable success in launching the Pirate Party in Sweden, but also because Swarmwise is something of a foundational text for the strategy that delivered the incredible rise of Extinction Rebellion (XR) - while also presaging its equally impressive collapse. 

XR fundamentally shifted the debate on climate here in Britain, where it was initiated, and also around the world. It succeeded in its first aim in pressing the UK Government and other administrations to “declare an emergency”. It galvanised and reinvigorated a climate movement marooned in despair. It provided a “radical flank” using high impact direct action that threatened to make London ungovernable. But its two higher level “demands” - reducing emissions to net zero by 2025 and creating an all powerful people’s assembly - remain distant dreams.

Nonetheless, activists will obviously benefit from knowing how XR went from nowhere to everywhere, how it succeeded in organising mass protests, in delivering acres of news coverage, and in breaking the “Overton window” for the climate narrative for a generation. Not every activist in XR will know this, but its strategy was developed and devised over at least a decade before launch - and some vital aspects of its organisational technology were first codified by Falkvinge. Indeed, many of the distinctive features of XR can be found in Swarmwise. I disagree with much in this book, but will provide a critique in another article. Here I want to provide a summary of his system.


Falkvinge offers some rather vague autobiographical details in the book, asserting his credibility as an authority. He claims to have founded a company at 16, completed army officer’s training and picked up “a dash of management experience from the dot-com era”. The book - and we take everything it says on trust for this discussion - uses as evidence the success of the Pirate Party and “the twenty odd cases I have observed first-hand”. He does not claim to be an expert in sociology or psychology but promises “merely [to] share experiences and methods that have been proven to work in practice.” 

The novelty of Swarmwise is the integration of the horizontal - a traditional hierarchical organisation, with the vertical - the swarm of self-actualised activists taking their own initiative.

If we heed Falkvinge’s advice, it is because of his success with the Pirate Party. The story begins in 2006. “When I kick-started the swarm of the Swedish Pirate Party, I had posted a rough manifesto on a rather ugly website and mentioned the site just once in a chat channel of a file-sharing lobby.” The site got three million hits, out of a Swedish population of nine million. People joined in their thousands. Membership trebled in a week, and trebled again. “Getting 20,000 new colleagues and activists in a week isn’t a pipe dream”. Soon it had 42,000 members and 18,000 activists, claiming to be the third largest political party in the country. Within three years, on 7 June 2009 it received 225,915 votes and won seats in the European elections. It spent €50,000 compared to the €6 million blown by its rivals. But today the Pirate Party has all but disappeared.

This success must have been about the extraordinary circumstances of the times, the overwhelming demand in Sweden for a new political party, one that was radical and anti-establishment but also presented a concrete aim and benefit for the population: file sharing made culture available to the masses, regardless of the ability to pay. But people faced prison for pirating films, books and other resources. Falkvinge places the emphasis on his own methods when describing his success. Not enough emphasis is placed on the opportunity that happenstance offered for such an initiative. But even if he didn’t make the wave, he did successfully ride it and therefore he has some authority when teaching us how to surf. 

The novelty of Swarmwise is the integration of the horizontal - a traditional hierarchical organisation, with the vertical - the swarm of self-actualised activists taking their own initiative. This is the structure of the swarm, and what makes it stand out both from traditional NGOs or corporations and also from the flat but entirely blunt instrument that was Occupy Wall Street, or the spontaneous crowd. 


Falkvinge’s approach is distinctive from “traditional organizational theory” in proposing an incredibly high level of autonomy for its members, its front line activists. At the same time, it does not abandon structure entirely but rather advances that a strictly hierarchical “scaffold” must be built to retain the shape and efficiency of any organisation over time. Falkvinge claims the founder is instrumental in securing consistency, but that any three activists can permit any action on behalf of the “swarm”. The press team can say what it wants, as long as quotes are attributed to the founder. This duality is the “secret sauce”.

This integration of the opposites (dialectical sublation if you will) of autonomy and authority can be understood as a real world example of Rodrigo Nunes’ academic interrogation of political organisation that is, as per his book’s title, Neither Vertical nor Horizontal. Falkvinge explains: “A swarm organisation is a decentralized, collaborative effort of volunteers that looks like a hierarchical, traditional organisation from the outside. It is built by a small core of people that construct a scaffolding of go-to people, enabling a large number of volunteers to cooperate on a common goal in quantities of people not possible before the net was available.”

The prescription in Swarmwise might be particularly compelling for environmentalists as it implicitly claims, or assumes, that processes and systems found in nature can be taken as a model for humans organising within social contexts. It’s a kind of biomimicry, akin to that developed by the practitioners of Holacracy, for example. Falkvinge compares the organisation to “army ants in the Amazon rainforest”, refers to the “hivemind” and promises we will be “overpowering your opponents with sheer biomass through superior ability of organisation and ability to channel volunteer energy”. 

The ecological and evolutionary metaphor extends to a prescription for infinite variety - and natural selection. A number of swarms should be seeded, each with “a slightly different set of parameters from a different founder” and, therefore, “the result will be several solutions that are tried in parallel.” Everyone should welcome “competition between many overlapping swarms.” The selection process takes place because activists are free to move between the competing swarms. “Activists can float in and out of organisations, networks, and swarms that best match the change they want to see in the world,” he explains. Those with appealing and effective leadership, aims and processes will prosper and those deceased by insincerity of aims or infighting will shrink and die. The tasks within the organisation also evolve: “The workflow becomes an iterative, evolutionary process of trial and error, of constantly adapting and improving…”

We will now set out Falkvinge’s entire organisational structure. His swarm always begins with the founder, and the founder remains the keystone of the hierarchy throughout. The founder retains a singular and all-powerful position in the swarm. It is worth noting that XR in practice had multiple founders but did endow those activists with a special place within its governance. Falkvinge’s founder is implicitly self-selecting, and gains her position through the proof of merit that comes from initiating a good campaign that attracts activists, “a miracle-type event.” The founder brings the swarm into being: “The hardest steps you can take in a business is going from one person to two, as you recruit your first employee.” The founder retains special rights, but must delegate power. They also have special responsibilities, including taking care of themselves. 


The founder is a god-like figure. The goals of the swarm “come from you, the swarm’s founder”. And these must remain non-negotiable. “If the swarm were allowed to start discussing its purpose in life, then it would immediately lose its power to attract new people”. A key claim is that the founder sets the aim, values and culture of the organisation. “The scaffolding, the culture, and the goals of the swarm need to emanate from a founder.” The founder must therefore draft the values, the mission statement, the strategy in “write only documents”. The power of the founder extends to the practical. “You, in control of the swarm’s formal name and resources, allocate budgets to officers, who subdivide their budget in turn.”

A specific responsibility that has to be retained by the founder, and only the founder, is as the public face or “avatar” for the swarm as a whole. This means press statements should quote the founder, and the founder must attend all events in order to speak to the broadcast media. “Your name is the only one they know of,” he argues. We are told that “every swarm needs an avatar”, there must remain “one organisation, one face”, “write quotes and attribute them to you, the swarm leader or founder” and therefore “the avatar faces of the swarm, typically you, have to be at the most important events”. He explains: “A swarm organisation, the organisational culture cannot be communicated from person to person as the organisation grows — it must be actively communicated centrally, and repeatedly communicated as new people keep joining.”

The founder has specific responsibilities. The most important of which are providing “endless encouragement” and support for all members of the swarm. The leader must be consistent, to ensure the swarm remains focused. “Every degree of uncertainty leads to inaction at this stage” and therefore the founder has to “encourage and explain why these actions lead to positive results”. Further, “It is a vital part of the leadership role to personally train those who regard you as their leader.” 

The Swarmwise prescription is for a highly centralised and coordinated scaffold - or skeleton even - which is then surrounded by a swarm of highly autonomous, self-organised activists. The scaffold is there to serve the swarm, and not to direct or dictate to it. The founder is the head of the scaffold, the spine perhaps, who builds out the structure and provides the aim, the culture, the “assistance, support, or resources”, and continuously assures that the people who act to maintain the structure always delegate the activity of campaigning, the power and decision making, to the swarm. “The role of this scaffolding is not directing and controlling the masses..” This enables the Swarm’s “blinding speed of operation, its next-to-nothing operating costs, and its large number of very devoted volunteers.”

The founder must from the outset start to build out the structure, and therefore the constitution, of the new swarm. Falkvinge is incredibly precise and detailed in how this should be done. “The swarm’s very first task will be to self-organise, and it excels at such tasks. But it is you who must set the structure and explicitly give the swarm the task to self-organise.” As they are new, they will accept the norms and decisions of the founder at first. The scaffold is structured as a pyramid, with the founder at the top, working with a leadership team, and then cascading out to “officers” or subgroup leaders. 


Falkvinge estimates that the initial launch, especially with online campaigning, could deliver 300 new activists, many unknown to each other instantly. At first, everyone will be in the same chat-group. This will form an “initial and horizontal team of people that will gather in a chat channel or similar spot, probably titled ‘chat channel about everything related to the swarm’.” However, the founder is then responsible, throughout, for breaking the swarm down into subgroups. Falkvinge claims throughout that people work best in teams of seven, groups of 30 and supergroups of 150 people at the most. He has adopted these three group sizes – seven, thirty, and 150 — as part of my army officer’s training” and they “correspond to squad, platoon, and company size.”

The first tasks are: 1. Set up a web forum or Wiki; 2. Divide the activists into subgroups of 30, usually based on geography (e.g. the London branch); 3. “Select” a leader from each subgroup (he later adds: “Refrain from giving instructions as to how that leader should be selected.”); 4. Establish contact with the leaders, 5. Set up a sub-forum for each group; 6. Set up a forum for the founder and the leaders, but also: “Make sure that other people can read it. Don’t keep secrets.” 7. Recruit a leadership team of about seven people, to mediate between the founder and the subgroup leaders. He concludes: “Getting the basic structure in place is first priority, enabling further absorption of more activists into the swarm.”

As the organisation grows and matures, function subgroups of 30 people will need to be formed. The subgroups would cover, for example, “PR/media, for activism, for swarm-care, and for web, information, and infrastructure.” The function groups should adopt a leader, and each leader should have a deputy because people will drop out. The function groups, as with the scaffold in general, must support rather than direct activity. “The activism leader would not lead activism as such, but rather support it.” 

The swarm by now will need a public organisational chart, so people can take up roles and orient themselves. Indeed, the chart should presage where the swarm wants to be, with 99 per cent of the roles created before they are populated. Falkvinge goes as far as advertising that the founder “construct[s] the entire scaffolding at its finished size at the swarm’s get-go, providing space in the organisational chart for everyone from geography leaders down to the neighbourhood level.”

The structure should include two key features. The scaffold should be centralised. Falkvinge says it is both necessary to “centralise the administrative workload to one or a very few people” and also to “welcome new activists into the swarm and continually measure the overall health of it.” The subgroup leaders, or officers, must be empowered to recruit activists to officer level. The officer is responsible for maintaining their region of the organisational chart. Each person will ideally occupy one role on the organisation. 

The swarm needs to discover “officers and leaders who are comfortable with appointing other officers and delegating authority over resources and responsibilities.” Further: “They need to have the authority to do this independently, and they have to know that they have this authority and are expected to use it.” At the same time, officers and activists will need to be accountable. This will necessitate “mechanisms for conflict resolution, for decision making, and for reward culture.”


At the same time there must be an “activation ladder” to recruit new members into the swarm and then to promote activists towards the centre, and therefore up the hierarchy (but not to founder level). The process must be transparent and fair. The activation ladder is a deliberate process for taking new recruits from the fringes into the organisational core. “From the centre, where the people leading the swarm are located, the swarm looks like a flat mesa (with just one steep step to climb), but from the outside, it looks like a rounded hill (with many small steps). This is key to making it easy for people to move to the highly active centre of the swarm: as we want to activate people in the swarm, it’s important to understand that activation is a gradual process with many steps on the activation ladder.”

There are special considerations in relation to media work. The swarm will need a press centre. As earlier described, the founder will often be the face of the campaign and the person quoted in press releases, and therefore the media itself. But someone has to do the rest of the work. This should be the responsibility of a “media sub-swarm of thirty people at the most.” The press team will also be responsible for internal communications. Almost all comms will be transparent and open to the public. However, they would also need to be tailored and directed. Falkvinge proposes “heartbeat” messages that reinforce the aims and methods, and the dissemination of positive results. As with all subgroups, “this sub-swarm should be autonomous and have full authorisation to speak independently on behalf of the swarm.” One further comms prescription is the adoption of a distinctive identity for the swarm such as “wearing distinctive clothing — purple, crisp-looking short-sleeve shirts with our logo and the person’s last name printed on the back.”

The swarmwise approach is supposed to be a radical departure from “traditional hierarchical organisation” but thus far this has not been entirely evident. Its species differentia is that the scaffold or central structure must serve the autonomous swarm, and not attempt to control it. “Swarm leadership is not so much managerial as it is janitorial,” Falkvinge affirms. “Leaders and officers are not somebody’s boss.” The activists do not answer to the officers, there must be no financial or organisational advantage to being an officer, there is no central planning and no central control. Further, there is “no assignment or micro-supervision”. Indeed the very opposite must be the case. 

“People are allowed, encouraged, and expected to assume speaking and acting power for themselves in the swarm’s name.” Leaders, and aspiring leaders, must gain a following through “a swarm meritocracy”, running effective teams that attract effective activists. The swarm members will “gravitate by themselves to a subtask where they can help deliver the desired result.” Leaders maintain leadership through achievement. “A lot of people in general want to be on the winning team in most contexts and will adapt their behaviour to match it.”

The transformation in leadership necessitates the same change in follow-ship. Activists are required, as well as encouraged, to make decisions, to be courageous, to discuss and not promote. The success of the swarm is possible “when somebody is self-motivated and self-reliant.” Further, activists should leave subgroups that are dysfunctional, and empower “groups that form around accomplishing specific tasks”. People do not sit around voting for a group to take an action, they leave the talking shops and join the teams already engaged in taking action. This is “the law of two feet”. This kind of “spontaneous swarm work” is “the backbone of our activism”. 


This high autonomy approach avoids conflict and terminal negotiation. “We can easily observe that, in any organisation, it happens that one person wants to limit what another person in the organisation can do. This creates a conflict.” It also makes activists more accountable. “Asking permission, after all, is asking somebody else to take responsibility — no, accountability — for your actions.” Finally, it reinforces leadership that enables autonomy, in a positive feedback loop. “Influence is achieved by individual leadership and individual appreciation — if you think something needs to be done, you just do it, without asking anybody.”

Autonomy is most clearly pronounced in the “three activist rule,” perhaps the most distinctive feature of movements that have adopted some of the organisational technology of the Pirate Party (and why it is sometimes known as the three pirate rule). The rule is simply: “If three activists agree that something is good for the organisation, they have a green light to act in the organisation’s name.” The rule is based on the principle that activists “deserve people’s confidence and trust.” It also enables hyper scalability and acts against bottlenecks. “Named people should never be gatekeepers, as they can be unavailable for a myriad of reasons, and therefore bottlenecks.” The corollary is that “no one is allowed to empower himself or herself to restrict others”.

Indeed, the rule goes further, stipulating that the activists should “never ask permission”. The fact that three activists have agreed that an action will serve the aim of the swarm should be sufficient risk management. A “very basic sanity check was to have three people agree on an idea as good for the swarm.” This should be extended to press work and “three people in the media sub-swarm need to approve a press release before sending it.” Finally, three activists cannot agree to proscribe an action or statement from other swarm members. 

This decision-making process will be familiar to some XR activists and observers as the precondition for the debacle when three activists decided to barricade trains at the Canning Town station on the London Underground. The decision had been taken autonomously from the XR founders and central team, who then attempted, but failed, to keep to the non-criticism rule described below. 

Many will see this as proof of the failure of the three activist rule. However, Falkvinge should be credited with foreseeing exactly this kind of moment, if nothing else. “Many baulk at this,” he argued in 2013: “Letting activists run loose like this? Trusting them with your name and resources to this extent?” He prescribed that the swarm must never indulge in a “blame game”, adding: “If something goes wrong, the swarm deals with it after the fact and never spends time worrying in advance about what might go wrong.” The more important point is how the XR structure allowed the organisation to develop and advance from this mistake - not least developing a better understanding of class, culpability for climate breakdown, and what constitutes a legitimate target for disruption. 


The swarm must never, we noted, engage in blame. Falkvinge takes this further than most organisational theorists. The “golden rule” of swarmwise is that you should never contradict or criticise others, and by this creed “the swarm lives or dies.” If you do not like an initiative, you can simply initiate an equal and opposite initiative. “If you see something you don’t like, contribute with something you do like.” This is “absolutely paramount” and one of the few things the founder must “enforce”. Mistakes do not need to be interrogated or published. “Failures are expected, but with every failure comes a learning experience.” 

This approach empowers activists and creates the preconditions for creative work. “Peers in the swarm don’t fear other people being angry with the swarm, and punish the risk-taker.” This is, in part, because such controls and criticism rarely have any actual effect. He observes: “If this kind of rich organisation can make such monumental mistakes, then no amount of advance checking can safeguard against making mistakes.” It is especially true that the swarm should not heed to the criticism of opponents, or even outsiders. “When people in the swarm get criticised by the public and by influential people, that is a sign you’re on the right track.” The swarm must, nonetheless, listen to feedback. The magic happens when such feedback is not given or received as criticism, and when it can be quickly integrated into the action and culture of the swarm. “Failures are expected, but with every failure comes a learning experience. “To truly outrun the competition, you need to minimise the iteration cycle — the time from a failure to the next attempt at succeeding.”

A fundamental corollary to the need for autonomy is the need for radical transparency. The swarm members can only learn from experience and each other if activists are honest and open. “There are almost no secrets at all” and “conflicts are transparent for all to see.” Firstly, this is necessary to develop trust. “Distrust depends on information starvation”. He quotes the video game character Pravin Lal, arguing: “Beware of he who would deny you access to information, for in his heart, he dreams himself your master.” Another major advantage is that “work isn’t applied to keep some people away from information”. It also prevents misunderstanding, deliberate or otherwise. He notes that “traditional chain-of-command structures” risks “somebody in the middle … distort[ing] information passed up or down.” 

There are exceptions. Falkvinge explicitly states that the identity of donors can remain secret. His justification is that “people would consciously or subconsciously try to please the larger donors.” And he implicitly argues that founders’ actions in setting the DNA of the organisation can also be shrouded in secrecy. This includes the breakup of groups into subgroups: “Don’t announce this intent, as doing so would cause a distracting discussion about that action.”

A question that must be raised when developing an open and non-hierarchical organisation is, “How to protect your open source project from poisonous people.” The swarm does not come without constraints. Falkvinge discusses how to deal with disruptive individuals, and corrupted internal institutions. His most negative statements concern “attention junkies” and the “attention-craving maverick” who “crave and demand attention” and are attracted to the “visibility of the swarm”. Today we might call such people narcissistic. Such activists “demand attention from you personally”, will try and “hijack” the swarm, and want to attain and reinforce “rank and hierarchy.” This can get personal. “In the process of running the organisation, you will occasionally discover people who don’t feel they get enough attention from you personally for their ideas on how to run the swarm.” The primacy of the role given to the founder has its shadow in the character of the narcissist, it seems to me, and the narcissist might loom as the greatest threat to the high status founder. 


Falkvinge does provide sound advice on how to respond. He notes that “attention is reward”, and this does not have to be positive attention. “You cannot and should not try to keep these people out…[so] deny them the space and spotlights they crave… isolate the maverick, and the disturbance will lose critical mass.” Specifically, “If there are behaviours we don’t want to see growing, we should ideally pretend they aren’t even there — block them out from our conscious radar, and spend time rewarding other kinds of behaviour.” In other words, shine attention where it is deserved. “Shatter this identity by recognising good contributors in the group”. Don’t argue with the narcissistic activist. “You will never be able to convince the maverick that he or she has bad ideas.” The first level principle of openness cannot be extended to the narcissist entirely, and avoiding conflict is a bad idea. “If you do cave in to get rid of the disturbance, you will teach the entire organisation that creating loud disturbances is a very effective way of getting influence.”

The swarm, as with any other activist group, will have to deal with other kinds of difficult people. The movement cannot afford not to respond to negative behaviour. “Do not show tolerance when somebody shows disrespect.” There are also the “wannabe ‘fixers’” including “the MBA-type people”, the NGO professionals that demand traditional structures, the technical people who want to focus on process, and the general malcontents. “There will be no shortage of people who want to reorganise — or even organise, as they will call it.” 

But perhaps a greater threat to the success of the swarm are the institutional challenges that will inevitably arise and become increasingly critical over time. The main problem, ironically, can be success. The Pirate Party after all won Parliamentary elections, with individuals suddenly able to establish high profile, well renumbered careers. There is a “predictable but treacherous mechanism” when success arrives, as some “people will start seeing inevitable money and resources everywhere”, will covet everything from the “lavish jobs to expensive toys to personal visibility” and “will start fighting for the five-hundred-or-so jobs that would be the outcome of such an election result.” This is the death knell for the swarm. “The swarm’s success will collapse in months.”

A surprising claim is that democracy is not so much a cure for corruption and internal atrophy, but instead another existential risk. Falkvinge goes as far as to argue that “‘democratic legitimacy’ is a contradiction in terms in a swarm organisation.” This is the most controversial claim in Swarmwise, but is worthy of consideration. A campaign will have limited budgets, equipment and human labour time. “Some people will insist on ‘democratic control’ over these resources.” But this, he claims, is a mistake. Falkvinge objects both to the 50 per cent plus one democracy familiar to many state elections and progressive organisations, such as trade unions. He argues that “the process of voting creates losers.” The tension of elections can drain resources and morale. “People who anticipate a voting process prepare themselves for the possibility of losing — so they become motivated by fear of losing personally.” Democracy involves “limiting the minority”, “limiting diversity” while also being “highly demoralising”. He concludes that swarms should “avoid voting to the extreme and only use it as a very last resort: voting creates losers.”

However, Falkvinge does concede that there are times when the swarm as a whole must come together to achieve consensus. This, one can infer, would include times when the whole wishes to deviate from the original strategy or plan of the founder. Or where a risk poses an existential threat to the work of everyone engaged. Here he prescribes a General Assembly which acts as a “consensus circle” with a veto for every attendee. 


“A final decision must not be proposed until it appears absolutely certain that the group will accept it.” Moreover, “if just one person blocks the final decision, the issue may not be discussed any more that day.” The veto ensures that the attendees find a solution to which all members of the swarm can adhere to, in the same way they each responded to the original objectives of the campaign. Further, “a decision that makes harmfully large portions of the organisation upset about the decision in itself should be rescinded.” Those with voting rights can nominate others to attend in their place, presumably to address accessibility concerns. This he terms “liquid democracy”. The consensus circle must be a very rare exception, nonetheless, with normal activity depending on the structure of the scaffolding and the autonomy of the swarm.

The rapid rise of the Pirate Party in Sweden, and the success in getting activists elected to parliament, meant that its influence spread through campaign groups across Europe, including the UK. Specifically, Swarmwise is namechecked by the Momentum Community organisers in a series of video tutorials about setting up social movements engaged in direct action based in the US. Momentum Community in turn influenced Roger Hallam and other founding members of XR. The strategy and structure of XR is clearly based on the prescriptions found in Swarmwise, as modified and assimilated by Momentum Community. This includes an anticipation that XR once established would be battleworthy for up to six years, before inevitable internal contradictions and external pressures would tear it apart.

The fact that both the Pirate Party and XR ultimately failed in their high end demands should not deter us from learning the lessons from these movements, both of which were the products of highly thought out strategies that engaged hundreds of thousands of people. The fact is, carbon emissions continue to rise, thus so far no person, institution or method has been shown to be ultimately successful. 

However, there are some aspects of the swarmwise methodology that are self-evidently inadequate, or at least irreconcilable with the values and objectives of justice campaigns (not least the power bestowed in an individual founder). The work that needs to be done is to salvage all that is useful from the wreckage of the Pirate Party, and from XR, so that our next attempt at building a just and sustainable world is as effective as possible. “Those who cannot remember the past are condemned to repeat it.”

This Author

Brendan Montague is editor of The Ecologist.

More from this author