7 May 2018

This article draws on a post by Agnete Strøm on the website  of Kvinnefronten, a Norwegian feminist organisation. We are indebted to her work.

“Reading the report on Norway, we question the whole foundation of Amnesty. It seems inconceivable to advocate for women’s rights and gender equality, as Amnesty does, and on the other hand designate a whole category of persons for whom a life of humiliation, attacks on their dignity, and violence is acceptable.” – Agnete Strøm


Amnesty International conducted research  in Norway during the development of its policy on prostitution. There were many irregularities  in how they went about developing and consulting on this policy. Key among these was (a) that the policy was originally proposed by Douglas Fox, a pimp who has a powerful vested interest in a thriving prostitution industry; and (b) that Amnesty’s International Secretariat had decided  the policy approach in 2013, before it commissioned the research and consulted with members.

This should set your alarm bells ringing. It certainly goes some way to explaining the design of the research.

At the time, opinion within the wider Amnesty community was polarised between two different approaches:

  • The Nordic Model, which decriminalises prostituted individuals and provides high-quality services to help them exit the trade, and makes buying sex a criminal offence – with the key aims of changing social norms and reducing the demand for prostitution that drives sex trafficking. This approach also cracks down on profiteers (pimps, brothel keepers, and procurers).
  • The full decriminalisation of the entire sex trade, including profiteers and punters (sex buyers). This was the approach that was favoured by Douglas Fox and Amnesty’s International Secretariat. In this article we call this approach “full decriminalisation.”

In these circumstances, a principled approach would have been to conduct research in places that had implemented these two different approaches so that comparisons could be made.

But instead Amnesty chose to conduct research in Papua New Guinea, Argentina, and Hong Kong, which all have prohibition, and in Norway, which introduced the Nordic Model in 2009. There was no significant support within the Amnesty community for prohibition, in which all parties are criminalised, so there was no obvious purpose for this amount of effort and research being conducted in countries with this system.

However, this meant that the results of the research in Norway would inevitably be compared with the results of the prohibitionist approach in the other three countries where prostituted individuals suffer violence and extortion at the hands of the police and other authorities, because selling sex is illegal and there is widespread corruption.

It is particularly reprehensible that Amnesty didn’t commission any research anywhere that has the full decriminalisation approach, which they were determined to see as policy and now recommend to governments all over the world.

This may explain why Amnesty didn’t restrict the research to laws that directly relate to prostitution but widened it to include tenancy, vagrancy, and immigration policy. If the research had simply looked at the impact of the prostitution laws, the difference between the results in Norway and in the prohibitionist countries would have shown that the Nordic Model is in fact an effective approach.

Most European countries have brutal and unfair immigration laws that are disproportionately applied to the most vulnerable. Moreover, racism is a problem all over Europe, including in Norway. This is all unacceptable, but they are separate issues from the prostitution-specific legislation and need analysing and tackling separately and in their own right.

If Amnesty had been sincere in researching the effectiveness of the Nordic Model, it would have conducted research in Sweden, which pioneered the approach and has had the greatest political will to implement all that’s needed to make the system successful, including: public information campaigns; education in schools; investment in exit services; training for the police and prosecutors; and employment and training opportunities for women.

Instead, Amnesty chose Norway, which passed the law ten years after Sweden, and only five years before the research was carried out.

Agnete Strøm suggests Amnesty chose Norway because prominent Norwegian Amnesty members openly opposed the Nordic Model, and in 2013 the incoming Norwegian government made it a manifesto promise to repeal the Nordic Model law. Amnesty wasn’t to know then that the incoming government would in fact abandon these plans when they realised how well the law was working in practice.

The Nordic Model involves a profound paradigm shift that challenges men’s historic entitlement to sexual access to women and girls. Many people, particularly men, resist the changes involved. This means it takes time and political will for the approach to bed in, and training and education are required. Moreover, there can be many opportunities for the spirit and implementation of the approach to be sabotaged – on a national, regional, or local level. There is no evidence that Amnesty understood this.


The research methodology Amnesty used falls short of research norms and good practice. Under standard research protocols we would expect a detailed account of research aims and rationale; how participants were selected; profiles and demographics of participants; research methods used, including a schedule of interview questions; and how the data was analysed. Most of this is missing.

The aims of the research are described like this:

“This report is part of a series of research reports undertaken by Amnesty International to document human rights abuses experienced by sex workers in four countries and to explore the role which criminal laws on sex work and penalization of sex workers play in relation to these abuses.”

This reveals an ideological position that implicitly accepts prostitution as unproblematic and views the human rights abuses to which the people involved are subjected as extraneous to the institution of prostitution itself. This casts doubts on the credibility of the research findings.

Unlike the UK, Norway has ratified the 1949 UN Convention  on the Suppression of the Trafficking in Persons and of the Exploitation of the Prostitution of Others. This recognises prostitution as incompatible with the dignity and worth of the human person enshrined in the Universal Declaration of Human Rights, and places an obligation on ratifying states to crack down on profiteers and those who let premises for prostitution.

Amnesty didn’t mention this convention once in the entire report, not even in the list of relevant human rights treaties on page 19; nor did the report acknowledge that Norway’s approach has widespread international support, is recognised as a legitimate approach within the UN human rights frameworks, and is recommended  by the EU Parliament. This is not the even-handed approach one expects from an internationally renowned human rights organisation.

On page 15, there is a statement that the research is qualitative rather than quantitative. While qualitative research is an accepted methodology, it is recognised that generalisations cannot be made with any confidence to a wider context than the one studied, and that the researcher’s presence and viewpoint inevitably have a profound effect on the results. This, along with the bias against the Nordic Model approach that we have documented, means that this research cannot reasonably be taken to show that “the Nordic Model doesn’t work” as is often claimed.

Interviews with women involved in prostitution

Interviews were conducted in Oslo over three weeks between November 2014 and February 2015. There is no record of the interview questions and, worryingly, interviewees were not asked to sign consent forms, but were merely requested to provide “verbal consent.”

We are told that Amnesty “spoke with” 30 women who had experience selling sex, but not who in Amnesty spoke with them, whether they were male or female, or what relevant training they had had. Nor are we told where the interviews took place, or who else was present, or what measures were taken to help the interviewees feel comfortable and to ensure their safety. There is no information about which languages the interviews were conducted in, whether interpreters were needed, and if so, whether they were male or female.

Of the 30 women, we know the following: about one third had exited prostitution; three were transgender; three described being trafficked; and 23 were migrants, many with only three-month tourist permits.

The women were identified by three organisations, one of which, PION, is a member of the Global Network of Sex Work Projects (NSWP), whose members all endorse the view that prostitution is work, and are ideologically opposed to the Nordic Model. It is not clear what steps, if any, Amnesty took to ensure that interviewees, particularly those introduced by PION, were able to speak freely. Similarly, we are not told whether the interviews were recorded or whether notes were taken, and if so, by whom.

We are not told how old the women were (at the time of interview or when they entered prostitution), how long they were involved in prostitution, what had led to their involvement, or what the nature of their involvement was. Nor are we told what they most needed help with, how the Nordic Model had affected them, or whether it had helped them in any way (for example, whether they had taken advantage of the exiting services).

Of the 30 women, only 16 are quoted in the report. All of the quotes support Amnesty’s position. However, there is evidence that some of the women interviewed disagreed with that position. For example, Agnete Strøm reports that a Nigerian woman spoke out in the Norwegian media shortly after the publication of the report to say that she was interviewed as part of the research. She said she’d told the interviewer that she supports the Nordic Model and that the police had helped her escape her pimp. However, Amnesty did not quote her or include her testimony.

In most cases, pimping meets the elements of sex trafficking, which is a serious human rights abuse. That the Nordic Model enabled this woman to escape her pimp was therefore pertinent to the stated aims of the research. That her contribution wasn’t included in the report begs the question whether it was left out because it reflected well on the Nordic Model and Amnesty was determined to show it in a bad light. We can’t help wondering what else has been left out.

This is further evidence why Amnesty’s research should not be used to make a generalised judgement about the effectiveness of the Nordic Model.


“A significant proportion of the women interviewed by Amnesty International said that they had encountered violence, in some cases severe and life threatening, while selling sex in Oslo in recent years.”

This is how the report’s section on violence against the women begins. Harrowing examples follow along with the observation that the most common perpetrators are punters.

This section of the report relies very heavily on Ulla Bjørndahl’s 2012 Dangerous Liaison report on research commissioned by Pro Sentret, a Norwegian organisation that promotes a “harm reduction” approach to prostitution. Bjørndahl claims there was an increase in violence against prostituted women after the introduction of the Nordic Model. The report is referenced by Amnesty a total of 29 times.

However, not all is what it seems. Agnete Strøm explains what is not obvious to English speakers:

“Difficult for any reader of Amnesty’s report on Norway today to know that in June 2012, two days after the release of Dangerous Liaison, the leader of Pro Sentret, Bjørg Norli, had to admit publicly to the media that the statistical foundation for the conclusions was very questionable and did not give any foundation for claiming that violence has increased; on the contrary it is possible that the opposite was correct; violence may have decreased.

But this is not the end of this story, one year later in April 2013, Norli made the report available in English and distributed it internationally and once again the false statistics popped up on the front pages, and was used by the pro-prostitution lobby in Northern Ireland, in France and is still used globally. You find it on the home page of Pro Sentret.” [Our emphasis]

Feminist researcher Samantha Berg analysed the Dangerous Liaison report and found that the research does not show an increase in violence; on the contrary, it shows that all serious violence is down, as shown in the following chart. Bjørndahl’s claim that violence had increased was based on manipulation of the data. We recommend reading Berg’s article in full.

The Norwegian government commissioned an evaluation  of the Nordic Model in 2014 from social research organisation Vista Analyse. This confirmed that there was no clear evidence of an increase in violence:

“This analysis finds no clear evidence of more violence against women in the street market after the introduction of the law. It is the customer that engages in illegal action and thus has the most to fear if reported to the police by a prostitute. The police have no indications on more violence following the ban on purchasing sexual services.”

Amnesty described the level of violence the women reported as serious human rights abuses and they lamented the lack of effective police protection against these acts of violence. This is oxymoronic in its refusal to face the reality: that prostitution is inherently violent  and nothing can make it safe.

The only effective way of preventing the violence that’s inherent to prostitution is to eradicate the institution itself. This is the ultimate aim of the Nordic Model.


The law in Norway prohibits knowingly letting premises for the use of prostitution, as required by the 1949 UN Convention mentioned earlier. The penalty is a fine or imprisonment for up to six years and is part of the law against promoting prostitution.

Amnesty is vexed by this law and dedicates about twice as many pages to matters relating to evictions than to the violence against the women. There is considerable detail of the police operation called “Operasjon Husløs” (Operation Homeless), even though it began two years before the introduction of the Nordic Model and ended several years before Amnesty conducted its research.

Unfortunately the full Vista Analyse report is in Norwegian and only a summary is available in English. But Agnete Strøm explains that the full report is clear that women weren’t losing their homes en masse. She translates the relevant text as follows:

“The Oslo police have been criticized for causing foreign women selling sex to be thrown out of the apartment they are living in, and they are thus losing their home. This can happen when the owner has been informed that the apartment has been used for selling sex. But the police point out that it is not correct to say that the police are taking the homes from the women, when 90 % of the market are women travelling in prostitution, and the home is thus a working place for a short period, from one day up to 14 days or more, but nevertheless of short duration. The negative effect of being thrown out of the apartment might be that the women lose the paid deposit, a deposit the owner of the apartment in some cases have refused to pay back.”

Amnesty does not make it clear that many of the evictions it refers to were from short-term lets on the basis of breach of contract. But a close inspection suggests this was indeed the case. Of the five women affected, four were Nigerian and one was of “African origin.”

The Executive Summary (on page 7) starts like this:

“‘Last year was really hell for me.’ That’s how Mercy, a Nigerian sex worker, described her experiences of discrimination, social exclusion and human rights abuses in Norway. Her story culminated in her being forcibly evicted from her home in 2014. The fact that Mercy was made homeless, without any due process or notice, constitutes a human rights violation under international law.”

It’s not for another 38 pages that we learn Mercy is “a young Nigerian woman who travels to Norway to sell sex. She has residency in the Schengen area and can, therefore, stay in Norway for three month periods.”

So Mercy was a temporary resident on a tourist permit and the rental terms of the property from which she was evicted must therefore have been short-term and not more permanent as the report misleadingly suggests. This does not make what happened acceptable, but it is not evidence of widespread evictions of prostituted women from permanent homes.

Sweden also has a law against letting premises for the use of prostitution with the aim of discouraging the sex trade. Research  shows that in practice, the Swedish police do not pursue prostituted women under this law, but use it against pimps and traffickers. It would seem that there are lessons that Norway can learn from Sweden on this.

A number of the women said they wanted to be able to operate from premises with other prostituted women, for safety’s sake. The idea that women are safer working together is a common argument for the decriminalisation of small brothels and at first sight it can seem persuasive. However, when you look more deeply, it becomes clear that things are not as straightforward as they might at first seem. See The problem with “safety in numbers”  for more on this.

Amnesty recommends that the Norwegian law against the promotion of prostitution is replaced by one that only bans exploitative practices, on the basis that there is nothing wrong with “consensual sex work.” This brings us to the core of the different world views behind the two opposing approaches: the Nordic Model and full decriminalisation.

At the heart of the difference of opinion is the question of whether prostitution can ever be compatible with human rights and whether a society based on equality between the sexes is possible if males regard it as their right to buy (predominantly young, poor, ethnic minority) females for sexual use.

Norway believes that this is not possible and therefore aims to abolish the institution of prostitution. The aim is abolition, which means a new social consensus that recognises the harm and violence intrinsic to prostitution, both to the individuals caught up in it and to wider society, and the adoption of concrete measures to bring about its eradication, while decriminalising those caught up in it and helping them to make a new life outside.

This is different from prohibition, which simply aims to suppress it by criminalising all the actors. Bizarrely, Amnesty refuses to acknowledge the profound difference between a prohibitionist approach and an abolitionist one, and the report frequently refers to the Nordic Model as prohibition, which is incorrect and misleading.

Intrinsic to the Nordic Model are high-quality services for those in prostitution, including the provision of genuine routes out. The aim is not to punish prostituted persons but to enable them to build a new life outside. Inexplicably, Amnesty left out this key aspect from its definition of the Nordic Model on page 18.

Some of the evidence Amnesty provides suggests that the services are not meeting the needs of all those caught up in prostitution, particularly migrant women, and that more needs to be invested in services and in ensuring that there are viable opportunities for women to make a living through other means.

This was recognised by a 2012 report from GRETA: European Convention on Trafficking of Human Beings, which read, “Norway should scale up their exit programs to meet the real need.” Amnesty did not mention this, perhaps because they believe that prostitution is work and there is nothing wrong with it, so women do not need help to exit or viable alternatives.

Sex trafficking

There’s something bordering on desperation in the urgency with which Amnesty claims that human trafficking and “consensual sex work” are entirely different, and must never ever be “conflated.” But they don’t explain how they are to be separated in practice. There is no separate market for trafficked women, and punters’ own words on punter forums  make it clear that many (perhaps most) don’t give a fig whether the women they buy in prostitution want to be there or not.

And Amnesty never mentions that the sex industry is a vast ruthless capitalist money-making machine in which the commodity on sale is sexual access to the young, vulnerable, and predominantly female.

“Human trafficking is insanely profitable. If you really think about it, you can sell a kilo of heroin once; you can sell a 13-year-old girl 20 times a night, 365 days a year.” – Daily Kos, 6 February 2018

Amnesty fails to mention it’s the money that punters pay that drives the entire industry. If they did, their virulent opposition to making buying sex a criminal offence would be revealed as irrational and unethical. This, and their lack of serious consideration of an effective and practical means of cracking down on the excesses of this brutal industry, leaves serious questions about their ethics and commitment to fighting for the rights of women and girls and for real equality between the sexes.

Amnesty suggests that, rather than implementing the Nordic Model, countries should make it a crime to buy sex from a trafficking victim, ignoring the fact that this has been tried in England and Finland and has been shown not to work.

There is clear evidence  that where prostitution is legalised or fully decriminalised, sex trafficking increases. It follows that to really tackle sex trafficking, there must be a zero-tolerance approach to pimping and brothel keeping, not least because they are prohibited by CEDAW, and pimping usually does satisfy the elements of sex trafficking, as defined in the Palermo Trafficking Protocol. A reader wouldn’t get to know this from reading Amnesty’s report.

It is as if Amnesty wilfully misinterprets these binding UN human rights treaties in order to justify its pre-decided policy.

We agree with Amnesty, however, on its criticism of the way migrant women in the sex trade are targeted for immigration offences in Norway and are often peremptorily deported. This is particularly egregious when the women are victims of sex trafficking. This should stop, along with all heavy-handed policing of migrant and prostituted women.

Other interviews and desk-based research

Page 8 of the report lists various other people that Amnesty spoke with in Norway in the course of its research. One of these was the Minister of Justice. Agnete Strøm explains that shortly after Amnesty’s report was published, he complained in the media about how Amnesty treated his contribution. He told Agnete:

“Amnesty gave me the opportunity to control my quotations, but after that I had no control over how they used the quotations, in what settings, what the focus was etc. I think Amnesty International had their ideology and politic already made up and used the quotations to suit their policy.”

The only pro-Nordic Model Norwegian academic they interviewed was treated in much the same way. This is how Agnete Strøm describes it:

“One researcher, Kotsadam, is the only academic doing research about the effect of the law who has a feminist view on prostitution, he is a young economist who looks into social change of young men, the demand, the market, the money and the criminal gangs, trafficking in the European countries and the laws on prostitution, and his research is interesting. Nothing of this is quoted, but Amnesty found in one of his articles that 8 months after the law in 2009, there was a slight increase in people who thought both buying and selling should be criminalised. When I asked him if he had any comments, he said: ‘There are definitively no mistakes in what they quote, even though they have chosen to focus on what suits them.’”

In contrast, three Norwegian academics who are ideologically opposed to the Nordic Model were quoted extensively. Bjørndahl we have already discussed. Amnesty referenced the other two, Skilbrei and Jahnsen, 42 and 24 times respectively.

When the European Parliament was discussing the Honeyball report,  which led to the recommendation that EU countries adopt the Nordic Model, Skilbrei and Jahnsen organised a petition against it, on the basis that the Nordic Model is ineffective and dangerous.

As mentioned earlier, the Norwegian government commissioned Vista Analyse to evaluate the operation of the Nordic Model law in 2014. The main findings were as follows:

“The ban on purchasing sexual services has reduced demand for sex and thus contribute to reduce the extent of prostitution in Norway. The enforcement of the law, in combination with the laws against trafficking and pimping, makes Norway a less attractive country for prostitution based trafficking than what would have been the case if the law had not been adopted. Furthermore, the economic conditions for prostitution in Norway are reduced following the implementation of the law. These effects are in line with the intentions of the law and are thus not considered as unintended side effects. This report does not find any evidence of more violence against prostitutes after the ban on buying sex entered into force.”

Skilbrei and Jahnsen heavily criticised this evaluation, and Amnesty quoted that criticism enthusiastically, as if Skilbrei and Jahnsen didn’t have a heavy ideological axe to grind.

If you think it can’t get any worse, you would be wrong, because Agnete Strøm found that much of the research that Amnesty references was conducted before the introduction of the Nordic Model in 2009.


This article does not provide a complete critique of Amnesty’s research in Norway. It aims merely to point out some of its more glaring shortcomings.

While we agree with Amnesty that prostituted individuals should not be discriminated against or be targeted because they are prostituted, we disagree with much of Amnesty’s analysis, in particular its commitment to the full decriminalisation of profiteers and punters.

We reject the terms “sex work” and “sex worker” because they confuse and obscure the reality and suggest that prostitution is work not unlike being, for example, a waitress, when nothing could be further from the truth.

We also reject what seems to be the underlying premise that “sex work” is more than an occupation and that being a “sex worker” is a form of identity. This is almost like saying that prostitution is something that people just ‘are’ and we absolutely refute that.

For the vast majority of those in prostitution in the world, it is a catastrophe and the result of misfortune and multiple intersecting structural inequalities. We believe that prostituted individuals deserve more than being used for men’s sexual gratification and ego massage.

Instead of fighting for the rights of men to buy sexual access to vulnerable people and for pimps and brothel owners to make fortunes off the suffering of those vulnerable people, we wish that Amnesty would fight to bring down the edifices of structural inequality that drive people into prostitution and for real alternatives for those who find themselves stuck in its nightmare.

Amnesty’s research in Norway was biased and of poor quality, and there is no justification for using it to claim that the “Nordic Model doesn’t work” or that the “Nordic Model endangers women in prostitution.”


Download a PDF version of this article.

Further reading