Drug trafficking-related violence is becoming more extreme and brutal, while simultaneously the consequences of criminal convictions remain minimal. Although the Marengo trial has led to justified convictions of guilty people, aggressive drug crime itself, the entire network on which that trade rests, is largely unaffected.

It was even more interesting to recently attend a lecture at the IAS by PhD researcher Frederieke Oetker, who is affiliated with the Computational Science Lab of the University of Amsterdam. In her lecture, she discusses three so-called agent-based models of cocaine networks, which are computer models in which each player, or each 'agent' in a cocaine network, is represented as a ball in a kind of three-dimensional cloud. Once each player has been assigned a place in that cloud, you can then see what the consequences of certain interventions could be. You then run simulations with the system and see what happens to the network if, for example, you arrest the collectors of cocaine in the port containers. Or if you remove the bribed customs officials.

The underlying idea of ​​this approach is that it is not about catching as much drugs as possible or about securing as many convictions as possible (something that the Public Prosecution Service and the police often understandably focus on). No, the wish is that a better understanding of the operation of the entire network of cocaine trafficking, the so-called 'value chain', can allow you to study how that chain can be effectively and permanently disrupted. In short: you are trying to overthrow the entire system.

Oetker shows on her slides how all players are first given a place in the computer model she built. Just like in a normal company, there are different jobs. There are dealers, cutters, security guards, (bribed) customs officers, (often young) 'collectors' who remove the drugs from the port containers. There are drivers, coordinators, 'brokers' who arrange the elimination of enemies and, unfortunately, murderers. Collectors and dealers are common jobs, but the number of killers is limited. Oetker then does computer simulations in which players are removed from the networks each time. You can then see how the network reacts to this. How quickly will the radar be replaced by another?

In police television series it always goes like this: the officers are after the boss of a criminal organization. The idea is that you must catch small fish to ultimately put that one big fish, the top executive, behind bars. Once that boss is arrested, the entire network will collapse, that is the idea behind it. However, this appears not to be the case.

Twelve years ago, Peter Sloot, professor of complex systems at the University of Amsterdam and founder of the Institute for Advanced Study, questioned this assumption. At the request of the police, he made an agent-based model of illegal cannabis cultivation in the Netherlands. The aim was to find out how the network could be permanently brought out of balance.

Mapping such a system is difficult because, with every disruption in the system, everyone responds to everyone else. So the systems are not only complex but also adaptive systems that respond immediately to their environment after every disruption. This can result in a disabled member of a criminal organization being replaced at breakneck speed. The police busted up thousands of weed farms every year, but the networks still existed. How could this be?

According to Sloot, it had to do with an incorrect perception of the structure of such organizations. It was always thought that a criminal network was a top-down organization controlled by one big boss, the so-called kingpin. Think of the 'Godfather'. By removing the top executive, the rest of the network would collapse on its own. But that usually turned out not to be the case in practice. The structure of a hemp network turned out not to have a pyramid shape at all, but rather to have its own dynamics, with many freelancers and changes of positions, in short, a company that had to be decisive and efficient. In a system like this, if you remove someone high up in the tree, there is a good chance that another person in an adjacent subnetwork is ready to immediately take that one kingpin's place. In fact, this person is sometimes even more able to take on that role. In this way, removing a kingpin can even lead to the strengthening of a network rather than its destruction.

In the VPRO program Labyrint, Sloot explained in 2013 that it would be smarter to arrest people who do specialist work; people who are difficult to find and replace. For example, someone who insulates the areas used for cannabis cultivation or someone who installs irrigation systems. Or the electrician. If those people are removed from the system, you will see the efficiency drop, according to Sloot.

After Frederieke Oetker's lecture, I discuss this further with Han van der Maas, professor of psychology at the University of Amsterdam and a researcher at the IAS. He is sceptical about such 'disruption research', research into the possibilities of network disruption. He knows about the example of the electrician from Sloot, but 'I have not been able to find the long-term effect of removing that electrician in the studies.'

'There are forty or fifty of these types of cocaine networks in the Netherlands. If you remove one, another network will immediately take over," says Van der Maas. He believes we can learn a lot from the network approach to cocaine networks, but that the most important lesson is 'how super robust they are'. They are so-called 'small world' networks, he explains: 'A type of company that employs an average of about fifty people. The earnings in this company are so high that any disruption is resolved very quickly. They are networks with a limited number of connections, but in those networks, everyone is indirectly connected to everyone.'

The consequence of this structure is that major drug seizures or arrests not only make no difference to the networks themselves, but that disruption of the networks can even increase the amount of cocaine on the market. “After every drug seizure or arrest, the network has the opportunity to become more efficient.”

It is reminiscent of the undefeatable Hydra, the many-headed monster from Greek mythology. Every time you cut off one of its heads, two new heads grow back in that spot. For this reason, Van der Maas thinks that the only way to really stop these types of networks is to legalize and regulate the cocaine trade. 'There is largely consensus on this within the scientific world.'

In January, the mayor of Amsterdam Femke Halsema wrote in The Guardian that the Netherlands is at risk of becoming a narco-state. The Netherlands is not properly preparing for the expansion of the global drug trade, according to Halsema. She also advocates for regulation but with a focus on addiction care. In January she spoke about this during a conference of the municipality of Amsterdam. Van der Maas remembers an impressive speech by a former police officer from England. 'After a seven-month undercover operation, an entire network had been arrested, the largest network in that city. The entire operation managed to frustrate the cocaine trade on the streets for a total of two hours. After that, trade had already recovered to its previous level.'

Naturally, the great fear is that regulation will lead to an increase in drug use. Van der Maas thinks that the crucial factor here is whether you leave the cocaine trade to commercial parties. 'If you leave it to them, they will go wild with marketing, and you may get a real increase. That is why you should not only legalize but also regulate. So have the trade controlled by the government.'

Van der Maas finds it frustrating that a lot of money goes to disruption research. 'For example, they map out the money flows and then look at how you can disrupt them.' In the meantime, 'not a single dime' is available for research into sensible regulation. 'Look, of course, it is interesting to see how those networks work. But we've known that for a long time. It is one large network, a so-called “giant component”. This means that there are islands, but that there are connections everywhere. Such a huge network is actually impossible to disrupt. That might have worked in the past. But now there is so much money involved, and the players are therefore so easily replaceable that network research mainly shows that disruption does not work.'

The fact that there is money for disruption research and not for research into regulated legalization is also a political problem, Van der Maas thinks. 'It's a taboo. Crime is a major problem, and the only solution seems to be more law and order.' He notices that the real opponents of regulation are on the right side of the political spectrum.

But at the other end of the spectrum, very little is happening in the meantime: 'The left is just sleeping a bit. Of course crime is traditionally not a left-wing subject. It amazes me how few people know anything about this. We often know a lot about the addiction side, from a public health perspective. But nearly all scientists who have studied these networks well argue for controlled regulation.' 

Published in De Groene Amsterdammer on April 13, 2024. Translated by the Institute for Advanced Study.