Rational democracy and futarchy

This article explains how rational democracy and futarchy, summarized as “voting on values, betting on beliefs”, is a promising solution to the central problem in politics and ethics: moral and empirical uncertainty.

Rational ethics can be summarized with the slogan: “accurate in beliefs, effective in means, coherent in ends”. We start with the ends: our moral values. These values can be expressed in ethical rules or principles, such as the principle to maximize well-being. Our ends are coherent if they do not contain unwanted arbitrariness. Unwanted arbitrariness means making a choice whereby the consequences are unwanted by at least one individual (i.e. they cannot be consistently preferred by everyone) and the justification of that choice is not based on a rule. The latter condition means the choice is arbitrary. Examples of unwanted arbitrariness are discrimination between individuals, inconsistencies between ethical principles and ambiguities in moral values.

After determining the ends, we need effective means to reach those ends, and in order to find those means, we need accurate beliefs about the world. Now we face the central problem in ethics and politics: how to deal with moral and empirical uncertainty? Moral uncertainty is uncertainty about moral values: which ends or values are the correct ones? Empirical uncertainty is uncertainty about empirical facts: which beliefs about the world are the correct ones? The solution to this problem of moral and empirical uncertainty in politics is rational democracy.

In our current democratic system, political parties are characterized by political ideologies which contain a mixture of moral values and beliefs about empirical facts. This makes the choice or vote for most preferred moral values, most effective means and most reliable beliefs almost impossible, and it increases the risk of politicians being irrational and biased due to their identification with ideologies that distort their judgments about policies. So we first have to disentangle the moral values from the empirical facts.


Voting on values (the parliamentary model for moral uncertainty)

There are many possible coherent ethical systems, such as a deontological rights ethic, a consequentialist utilitarian welfare ethic, a libertarian ethic or pluralist ethics that combine several ethical principles (see this example). We have a moral uncertainty about which ethical system is correct. In a sense, all coherent systems are equally valid: I cannot give reasons why my coherent ethical system would be better than yours.

Nick Bostrom proposed a Parliamentary Model to deal with this kind of moral uncertainty. Here I take this idea literally: political parties should be primarily defined by their ethical systems. A political party corresponds with a cluster of similar ethical principles. For example, a consequentialist utilitarian party clusters consequentialist utilitarian ethical systems. A party member or eligible candidate can have his or her own preferred ethical system that is broadly in line with the position of the party.

An ethical committee, consisting of experts in moral philosophy, controls the coherence of the ethical systems held by all parties and party members. People who have incoherent ethical systems (i.e. with inconsistencies or unwanted arbitrariness) are not allowed to participate during election. For example: a party with a racist ideology is not allowed, because racism is a kind of unwanted arbitrariness.

Each voting citizen has 10 demivotes to vote on parties or eligible candidates. For example: if you have 90% confidence in a prioritarian welfare ethic and 10% confidence in a libertarian rights ethic, you can give 9 demivotes for the candidate who is closest to your prioritarian ethic and 1 demivote for the candidate of the libertarian party.

All the elected officials debate in the parliament about the most preferred mixture of moral values. The resulting consensus view is a kind of weighted average of all ethical systems, weighted according to preference or credence. The most important task of the members of parliament is the determination of measurable indicators for the following legislature. Analogous to life cycle impact assessments, we can make a distinction between midpoint and endpoint indicators. Possible example of midpoint indicators are: GDP, the Gini index for income inequality, lifespan, life satisfaction, depression rates, crime rates, the level of greenhouse gas emissions, measures of progress in scientific research,… These midpoint indicators can be aggregated into a limited number of endpoint indicators such as the human development index. These endpoints measure for example economic prosperity, environmental sustainability or general happiness. During the election, each eligible candidate can present his or her preferred midpoint and endpoint indicators, so all voters can have a clear picture of what the candidates find important.

The ethical committee checks if the resulting parliamentary consensus and chosen indicators do not contain unwanted arbitrariness. An independent bureau of statistics has the task to collect all the data to calculate the chosen indicators.


Betting on beliefs (the prediction market model for empirical uncertainty)

After determining the moral values or ends measured by the chosen midpoint and endpoint indicators, we now have to find the most effective means to reach those ends. These means are the policies and laws. To find those means, we need accurate beliefs. However, most people are biased and a lot of politicians have hidden agendas or personal (e.g. financial) interests. These biases generate inaccurate beliefs, resulting in ineffective means.

The question becomes: what is the best institution to find the most effective means? No institution is perfect: we do not have a completely unbiased, impartial institution with perfect scientific knowledge about economics and other relevant disciplines, that can determine the most effective policies. But this does not mean we cannot look for the least bad institution.

One interesting proposal that deserves more research, is Robin Hanson’s futarchy, which he describes with the slogan: “voting on values, betting on beliefs.” Voting on values was a least bad solution to moral uncertainty, and perhaps betting on beliefs is the least bad solution to empirical uncertainty. In a futarchy, prediction markets (speculative markets trading in idea futures) are used to determine the most effective policies. A prediction market is probably one of the most reliable and effective institutions to gain crucial information about e.g. the likelihood that a certain policy has a positive effect (measured as increases in the chosen indicators).

In a prediction market, people can trade in conditional bets. For example, if the chosen indicator is GDP and the proposed policy is a certain trade agreement, I can sell you a conditional bet that pays you $1 if the policy is adopted and GDP increases after a certain amount of time and $0 if the policy is adopted and GDP decreases. The bet is annulled if the policy is not adopted. If you have 70% confidence that GDP will increase if the policy is adopted, you (as a rational agent) are willing to pay at most $0.7. The maximum price you are willing to pay corresponds with your subjective degree of confidence (your subjective probability). Similarly, the minimum price I am willing to sell this bet corresponds with my degree of confidence. According to some research mentioned by Hanson, if there are many speculators, chances increase that the market prices of those conditional bets become reliable estimates of the probabilities of the effectiveness of the policies.

Prediction markets cannot only give probability estimates for the effectiveness of policies, but also probability estimates for future indicators chosen by future governments. This is important, because taking our moral uncertainty into consideration means taking into account that our future moral values might be different. So we have to be aware that in the future people might have other preferences for their moral values, or that new insights and technologies allow for the adoption of other, better indicators in the future.

In futarchy, a policy is adopted (and competing policies are rejected) if two conditions are met: 1) the price of the conditional bet for that policy and for the currently chosen indicator (i.e. conditional on that policy being adopted and the indicator being chosen) is clearly higher than the prices of the conditional bets for the competing policies (i.e. conditional on the other policies being adopted), and 2) the price of the conditional bet for that policy and for the most likely future indicator (i.e. conditional on that policy being adopted and the future indicator chosen) is not clearly lower than the prices of the conditional bets for the competing policies with that future indicator. The second condition guarantees that we will not regret our policy decision when our moral values (and the corresponding indicators) change in the future. For example, if chances are high that in the future another indicator than GDP will be chosen, and if the prediction market clearly predicts that the proposed trade agreement worsens that future indicator, the proposed agreement does not become law.

For a lot of policy choices, reliable information about empirical facts is highly important, and prediction markets are a good source of information (they can effectively aggregate information). Such information has a lot of value, because a lot is at stake. Therefore, prediction markets that predict the effectiveness of promising policies should be subsidized in order to attract enough speculators. This subsidy reflects the financial value of the information. The details of futarchy, as well as the rebuttals of some criticism, are discussed in Hanson’s paper.

Futarchy is just one promising proposal. There are many other possible solutions to the problem of moral and empirical uncertainty in politics. We cannot tell in advance whether futarchy or another proposal works well, but these promising proposals deserve more research and experimentation.


Geplaatst in Blog, English texts | Tags: , | 1 reactie

Ineffective actions and campaigns that can backfire

Effective altruism is not only about looking for actions that do the most good, but also about avoiding ineffective actions. Here I will give four examples of campaigns or actions that can backfire in the sense that they can do more harm than good. I cover three areas – environmental pollution, animal suffering and social injustice – and one general strategy – fundraising. The objective of this article is to let us think more critically about helping others and become more effective in doing good.

Environmental pollution

A recent ineffective campaign from the environmental movement was the campaign to ban the herbicide glyphosate, which is primarily based on the World Health Organization International Agency for Research on Cancer (WHO IARC) 2015 evaluation that in terms of hazard (i.e. whether the substance is capable of causing an effect), glyphosate is probably carcinogenic. First of all, there does not seem to be a scientific consensus on the carcinogenicity of glyphosate. For example, the European Chemical Agency (ECHA) says available scientific evidence did not meet the criteria to classify glyphosate as a carcinogen. Second, even if glyphosate may be a carcinogenic hazard, there seems to be a scientific consensus that it poses no carcinogenic risk (i.e. the actual exposure to the substance – for example through diet – is too low to show a significant effect). The European Food Safety Authority (EFSA), the UN Food and Agriculture Organization (FAO) and the WHO agree that glyphosate is unlikely to be genotoxic at anticipated dietary exposures and unlikely to pose a carcinogenic risk to humans from exposure through the diet.

It happens that glyphosate is one of the safest herbicides. The Environmental Impact Quotient (EIQ) of glyphosate is 15, lower than many other herbicides that were used instead of glyphosate, such as imazethapyr (EIQ 20), trifluralin (EIQ 19) and pendimethalin (EIQ 30). The human toxicity potential of glyphosate is 0,7 gram 1,4-DCB-equivalent per kg substance, almost 40.000 times lower than the average herbicide. The soil ecotoxicity and freshwater ecotoxicity potentials of glyphosate (per kg substance) are respectively 800 and 300 times lower than the average herbicide. For individuals, using acetic acid, salt or liquid bleach (sodium hypochlorite), or other common herbicides such as dicamba, is probably more polluting than using glyphosate. And they are more toxic for mammals. The LD50 lethal dose of glyphosate (for rats) is 5 grams per kg body weight, compared to 3 g/kg for acetic acid and salt, 1 g/kg for dicamba and 0,2 g/kg for sodium hypochlorite.

So what would happen if glyphosate were banned? It is not clear if that would lower the environmental impact overall. If other herbicides are not banned, farmers and individuals might switch to more toxic herbicides. It is also not clear that the alternatives for herbicides, such as ploughing, are better for the environment: they can decrease soil quality and increase fuel consumption and soil erosion. Evidence that herbicide-free organic farming is better for the environment, is lacking.

Given this lack of evidence that alternatives to glyphosate use are substantially better for human and environmental health, urging for a ban on glyphosate might be too early. We first need to know the counterfactuals: what would be used if glyphosate was banned? And we need more scientific studies about the environmental impacts of those alternatives.

What are more effective campaigns at this moment? A first, more straightforward campaign would be the ban on the most dangerous pesticides on the market and the promotion of integrated pest management. But above all, the most effective campaign for the environment, is a decrease of animal production (e.g. a campaign for a high tax on animal products or for the promotion of animal-free products). The production of meat requires on average 4 times more cereals and soy (used for animal feed) than the production of plant-based protein-rich products. This means meat requires 4 times more herbicides than vegan alternatives. Animal-free, vegan food offers many other human, environmental and animal benefits as well. Besides, according to the IARC, red meat is like glyphosate probably carcinogenic (category 2A) and processed meat such as bacon is carcinogenic (category 1). However, pointing this out, results in another problem, as we will see next.

Animal suffering

Red meat has a higher environmental and health impact than other protein-rich foods. So a focus on the environmental and health consequences of diet might result in people eating more chicken meat and eggs as a replacement for red meat. The same goes for people who are concerned about animal welfare and are in favor of mammals. Campaigns that focus on the suffering of pigs and cows might increase the consumption of chicken meat, and vegetarians might increase their consumption of eggs and egg-containing products.

However, in terms of number of animals used and killed or hours of animal suffering per kilogram product, chicken meat and eggs are about 20 times worse than red meat. (Concerning eggs: male chicks and less productive layer hens are killed.)

Even a replacement of red meat by a vegetarian (non-vegan) alternative that contains egg-protein might be more harmful to animals. Suppose a vegetarian sausage contains 4% chicken egg protein. To produce 1 kg of egg protein, one needs 14 kg of eggs (i.e. 230 eggs). However, eggs not only contain protein but also other by-products with a market value, so we need to multiply the amount of eggs with an economic allocation factor of 0,45. That means 1 kg of vegetarian products containing egg-protein requires 0,25 kg eggs (i.e. 4 eggs). According to this calculation, egg-protein containing vegetarian products are 5 times worse than red meat in terms of animals killed and hours suffering. Reducetarians and flexitarians who replace red meat by such vegetarian alternatives, increase animal suffering.

What is a more effective strategy? First of all, we should avoid single issue campaigns that merely ask for a reduction of red meat consumption without mentioning chickens and eggs. Hence, we should promote vegan alternatives. In particular, red meat can be replaced with vegan protein-rich foods and chicken meat and eggs can be replaced with nuts, seeds and vegetables. Vegetables are better for the environment and human health than chicken meat and eggs, so this message is compatible with a concern for the environment and human health. Second, in terms of reducetarian or flexitarian campaigns, we should focus on reducing the consumption of chicken meat, eggs and farmed fish, because these animal products have the highest moral footprints. Finally, consumers can put pressure on producers of vegetarian products to eliminate eggs in their products.

Social injustice

A third example of an ineffective measure is a consumer boycott of products made with low wage labor or relatively bad working conditions (e.g. sweatshops). As long as there is no involuntary slavery involved, such boycotts might easily backfire: a boycott might result in those workers losing their jobs, and hence they often become worse off because they no longer generate an income or they move to another job that is less favored (i.e. with worse working conditions).

Buying Fairtrade is not always the answer either. The higher prices for Fairtrade gives the producers an incentive to produce more, which can result in overproduction and consequentially a decrease in price of the non-fairtrade products. The producers who were not able to get a Fairtrade certification can end up being worse off, with lower prices for their products and hence lower incomes.

So what is the answer? What is more effective to improve social justice? The general answer are campaigns against unearned income (income gained not through labor or entrepreneurship but through ownership of land and other monopoly), economic rent (a surplus profit above normal profit, received for non-produced inputs) or rent seeking (seeking to increase one’s share of existing wealth without creating new wealth). More specifically, a tax shift is possible: taxing economic rent (e.g. natural resources) instead of labor. Other related examples are a global resources dividend (an idea from Thomas Pogge) and a clean hands trust and clean trade in natural resources (an idea from Leif Wenar).

For individual consumers, a more effective alternative than boycotting sweatshops and buying fair trade is donating money to organizations that give unconditional cash transfers, such as GiveDirectly or Eight. In fact, one can argue that we have a duty to donate money to those charities.


If we measure the cost-effectiveness of measurable interventions (e.g. in terms of numbers of lives saved, loss of quality adjusted life years avoided, kilograms of toxics avoided, hours of animal suffering avoided, levels of income increased or levels of crime decreased per dollar invested), we see a very skewed (often log-normal or fat-tailed) distribution. A minority of interventions is far more effective than the vast majority, doing a lot more good per dollar. Most interventions have an effectiveness below the averages, because the small minority of highly effective interventions drives up this average. This is just like the global income or wealth distribution with a small number of very rich people. We can expect that the immeasurable interventions (whose cost-effectiveness we are not able to measure yet) have a similar skewed distribution.

This has important implications for fundraising. We can consider three types of organizations. First there are the big, multiple-issue organizations that do a lot of campaigns, projects and interventions (e.g. Greenpeace, Peta, Unicef). If their campaigns are randomly distributed, these large organizations have an average cost-effectiveness. Next, there are the single-issue organizations focusing on specific problems or specific interventions. Most of those single-issue organizations have an effectiveness below average, because most likely they focus on low cost-effective interventions. Examples are local environmental organizations, animal shelters and organizations that focus on minority groups, poverty and diseases in rich countries. A third group of organizations are the minority of highly effective single-issue organizations (e.g. the organizations recommended by GiveWell and Animal Charity Evaluators). They have an effectiveness above average.

Now suppose that you go fundraising for an organization. As a result, the number of donations and the amount of money that people donate might increase a bit. But we also see a shift between organizations: people start to donate more to your charity and less to other charities. This means there is a shift away from a group of charities with an average effectiveness. If your charity has an average effectiveness (e.g. it is a multiple-issue organization), this shift is neutral. But if your charity has an effectiveness below average, your fundraising might actually do more harm than good. The average effectiveness might be an order of magnitude (a power of ten) higher than the median effectiveness (i.e. than the effectiveness of most single-issue organizations). That means if your fundraising causes a shift towards a charity with below-average effectiveness, the amount of dollar donated would have to increase with an order of magnitude in order to compensate for the loss of effectiveness by the shift between charities. A world where you are not fundraising for that charity might be a world where more good is done. So even fundraising for a charity can sometimes be a harmful job.


Geplaatst in Artikels, Blog, English texts | Tags: , , , , , , | 1 reactie

De Heetste Week: kies je goede doelen goed

Onderstaand opiniestuk over effectief altruïsme, verschenen in De Tijd, werd geschreven naar aanleiding van De Warmste Week – Music for Life. Uit de lijst van Warmste Week goede doelen denk ik dat de volgende organisaties het meest effectief zijn:

Voor menselijk welzijn (armoedebestrijding, gezondheidszorg):

Eight: deze organisatie geeft mensen in Oeganda een onvoorwaardelijk basisinkomen en is vergelijkbaar met GiveDirectly, een topaanbeveling van GiveWell.

Unicef: deze organisatie doet onder andere preventiecampagnes tegen tropische infectieziektes. Dat zijn heel effectieve projecten volgens GiveWell. Maar omdat Unicef reeds een grote organisatie is (die reeds veel donaties krijgt), en veel verschillende projecten doet (waarvan er waarschijnlijk ook een groot deel minder effectief zijn), zijn de topaanbevelingen van GiveWell waarschijnlijk effectiever dan Unicef omdat die topaanbevelingen meer ruimte hebben voor extra financiering en focussen op de interventies met de hoogste impact.

Voor de dieren

Bite Back en Animal Rights: deze organisaties komen qua werking het best overeen met Animal Equality, een topaanbeveling van Animal Charity Evaluators.

EVA: deze organisatie promoot plantaardige voeding, een heel effectieve maatregel voor de dieren volgens Animal Charity Evaluators.

Kies je goede doelen goed

De meest zingevende gebeurtenis van het jaar

Vraag aan brandweerlieden die het afgelopen jaar een hond of een kind uit een brandend huis hebben gered: wat was uw meest zingevende ervaring van 2017? Iemand helpen of redden is zo’n fantastische ervaring, dat we bereid zijn er veel voor op te offeren. Stel je springt in een rivier om een drenkeling te redden, maar je verliest daarbij je portefeuille. Zou je dat erg vinden, of zou je blij zijn dat je een kind hebt gered? De meesten zijn bereid een paar duizend euro te verliezen als ze daarmee een leven kunnen redden. Iemand helpen is belangrijker dan een vakantie of een nieuwe smartphone.

Diegenen die zich inzetten voor een goed doel van De Warmste Week, zullen deze week waarschijnlijk de meest wezenlijke, zingevende keuze van het jaar maken. Misschien heb je niet hetzelfde euforische gevoel als die brandweerlieden, maar wat is voor jou het belangrijkste: een gevoel van euforie en trots, of het welzijn van de persoon die je hebt geholpen? Doe je het voor jezelf of de ander? Als je het voor de ander doet, zou je je inzet deze week dan niet maximaal zingevend willen maken? Wil je de effectiefste goede doelen steunen? Dan zijn onderstaande tips voor jou.

Denk als een investeerder

Net zoals een investeerder een maximaal financieel rendement wil, zo kun je bij je keuze van goed doel streven naar een maximaal sociaal rendement waarbij anderen zo goed mogelijk geholpen worden. Dan moet je wel je verstand gebruiken, kritisch denken en niet zomaar je buikgevoelens volgen. Investeerders luisteren ook niet zomaar naar mooie praatjes van bedrijven, dus wees waakzaam voor mooie verhaaltjes van goede doelen.

Verbreed je blik en kies waar, wanneer en wie je helpt

Drie factoren bepalen de effectiviteit van een goed doel. Ten eerste de ernst: hoe groot is het probleem waar het goede doel een oplossing voor wil bieden? Eén persoon met een verstuikte enkel of duizend kinderen met een dodelijke ziekte? Stel jezelf drie vragen.

Waar wil je helpen? Als de plaats niet uitmaakt: de armoede in het Zuiden is veel groter dan de armoede in je buurt. Verminder de extreemste armoede, vooral door de preventie van grote en verwaarloosde tropische ziektes.

Wie wil je helpen? Als het uiterlijk of de soort niet uitmaakt: het aantal dieren die lijden in stallen en slachthuizen is veel groter dan het aantal lijdende mensen. Verminder de veeteelt, vooral door de promotie van diervrije voeding.

Wanneer wil je resultaten van je hulp? Als dat eender is: in de verre toekomst staan veel meer levens op het spel dan vandaag. Verminder de extreemste bedreigingen die de hele samenleving kunnen uitroeien.

De effectiefste goede doelen helpen veel personen met weinig geld. Als je kunt weten welk individu je hebt geholpen met je gift, dan behoort dat goede doel waarschijnlijk tot de minder effectieve, omdat er dan meestal veel middelen geïnvesteerd worden in weinig personen.

Gebruik de wetenschap en kijk naar de opportuniteitskost van je gift

Een tweede factor is de reduceerbaarheid: hoe eenvoudig is het probleem aan te pakken en hoe sterk kan het goede doel het probleem verminderen? Met wetenschappelijk onderzoek kunnen we nagaan hoe doeltreffend een goed doel is. Als we met experimenten de effectiviteit van projecten meten, bijvoorbeeld in termen van geredde levens, gestegen inkomens, betere schoolresultaten, gedaalde criminaliteit of gezonder leefmilieu, dan zien we telkens dat de meeste projecten weinig of niet effectief zijn en een kleine minderheid supereffectief is. Waarschijnlijk is dus ook een kleine minderheid van Warmste Week goede doelen veel doeltreffender dan de meerderheid. Bijvoorbeeld: met het bedrag om een blindengeleidehond op te leiden om één blinde 10 jaar te helpen, kan men in arme landen oogontstekingen behandelen en zo voorkomen dat 1000 kinderen blind worden. Of het bedrag om een hond in een asiel te helpen kan het leed besparen van meer dan 1000 veedieren. De keuze voor het ene goede doel heeft een opportuniteitskost: er is geen geld meer om aan een ander project te geven dat mogelijks meer goeds realiseert.

Zoek de niche en kijk naar wat anderen niet doen

De derde factor meet de verwaarloosdheid: hoeveel andere personen doen al iets aan het probleem? Net zoals slimme investeerders rekening houden met wat andere investeerders doen en op zoek gaan naar het gat in de markt, zo kunnen we op zoek gaan naar goede doelen die een verwaarloosd probleem oplossen. Mediagenieke organisaties krijgen al veel geld, waardoor je eigen bijdrage minder impact heeft dan bij organisaties die weinig steun krijgen en ruimte hebben voor meer financiering.

Als we deze tips van het Effectief Altruïsme gebruiken, dan realiseren we een veel hoger sociaal rendement en wordt het de Heetste Week.

Stijn Bruers is auteur van Morele Illusies en mede-oprichter van Effectief Altruïsme Vlaanderen

Geplaatst in Artikels, Blog | Tags: , | Een reactie plaatsen

Why naturalness is irrational and harmful

In the environmental and natural health movements, people value naturalness. However, from a rational-philosophical perspective, this notion of naturalness doesn’t make much sense: it is arbitrary and not well-defined. Second, from an ethical perspective, preferring naturalness is often harmful: it decreases the well-being of other people. In this sense, a preference for naturalness is a perfect example of a moral illusion: a persistent erroneous moral judgment that distracts us away from a rational ethic.

Why naturalness is irrational

Naturalness is a very vague concept. What does it really mean to say that a process or product is natural?

  • Does it mean that it occurs without human influence? That is arbitrary, because why would human influence make something unnatural and e.g. insect influence or mammal influence not? Mammals are part of the natural world, humans are a subgroup of mammals, so humans are also part of the natural world. Besides, what does “mammal influence” actually mean? If it has no meaning, then why should “human influence” have any meaning?
  • Does it mean that it is safe? No, there is absolutely no correlation between naturalness and safeness. Some processes and products that are considered natural (erupting volcanoes, parasitism, toxic mushrooms) are more dangerous than processes and products that are considered unnatural or synthetic (inflating bicycle tires, using medicines, wearing protective gear).
  • Does it mean that it is not invented? No, organic farming and natural health practices are invented, but considered natural.
  • Does it mean that it has high biodiversity? No, with genetic manipulation we could highly increase biodiversity and create a large number of new species, but that is considered unnatural.
  • Does it mean that it is ‘old’ or does it refer to a certain state of nature in the past? That is arbitrary, because at what time was nature most natural? Is a modern-day ecosystem that looks like an ecosystem 100 years ago less natural than a modern-day ecosystem that looks like an ecosystem 100.000 years ago? Is a health practice that was developed 20 years ago less natural than a centuries old health practice?

If you reflect on this notion of naturalness, you find it impossible to make it clear, well-defined and non-arbitrary. But the most worrying is that it is often harmful.

Why naturalness is harmful

Here are more than 10 examples of harm as a result of a belief in naturalness.

  • An anti GMO attitude. Genetically modified organisms are considered as unnatural. However, there is a scientific consensus that GMOs are generally safe (not riskier than so called natural plant breeds used in e.g. organic farming). And GMOs can offer many benefits: less pesticide use[1], higher incomes for poor farmers and higher nutrient values. An example is the resistance against golden rise, a GMO rice that produces pro-vitamin A and could save 30.000 lives a year.[2] Another example is the resistance against Bt-eggplant, a GMO eggplant that produces an insecticide that is also used in organic farming and hence no longer requires the application of insecticides by farmers, resulting in higher yields, higher biodiversity levels on the farms[3] and higher incomes of the poor farmers in South-East Asia.
  • An anti vaccine attitude. A lot of people are worried about vaccinations, thinking vaccines cause diseases such as autism. Vaccination is considered as an unnatural health practice. However, the scientific consensus and evidence is very strong: vaccines are highly effective, save millions of lives each year and the risks are very very small. If parents refuse to vaccinate their children, their own children and other children with compromised immune systems are at increased risk, herd immunity gets lost, which could result in many deaths.
  • An anti E-numbers and chemical additives attitude. In Europe, some substances that are permitted to be used as food additives (because they have evidence of safety), have E-numbers. However, many of those E-numbers are produced synthetically and hence are considered unnatural. A worrying example is the use of methyl cellulose in some vegan food products. Methyl cellulose has E-number E461 and is used as an egg-replacer. It is perfectly safe, not toxic and not allergenic, but a producer of vegetarian products decided to replace methyl cellulose with egg-proteins, because eggs are considered more natural. As a result, those vegetarian products are harmful to chickens. As a comparison, the production of 1 kg of eggs involves more than 10 times more hours of animal suffering and killings of animals than the production of 1 kg of red meat.[4] Another example is the avoidance of preservatives (E-numbers E200-E299): chemicals that prevent undesirable chemical changes and decomposition by microbial growth. This results in increased food waste. As there are sustainability challenges with feeding the world, food waste can be considered as being harmful.
  • An anti vitamin supplements attitude. The consumption of animal products harms animals and future generations (due to climate change). Vegans avoid this harm, but a healthy vegan diet requires a vitamin B12 supplementation (either by using chewing tablets or eating products enriched with B12). Some people consider this as unnatural and therefore keep on eating animal products, harming animals. Ironically, they buy products from modern day livestock farming, which is far from natural because those animals get a lot of vitamin supplements and antibiotics. The amount of B12 that goes to livestock is sufficient for almost 40 billion vegans.
  • An anti supplementation attitude in organic products. Some restrictive organic food regulations make organic foods less healthy. In particular enrichment with vitamins is not allowed in e.g. organic soy milk. Non-organic soy milk enriched with calcium and vitamins B12 and D can be healthier than non-enriched organic soy milk as well as cow milk. Hence, promoting organic soy milk can be harmful.
  • An anti antibiotic attitude in organic livestock farming. The over use of antibiotics in livestock farming poses a serious threat to human health. Organic farmers try to avoid antibiotics, but when their animals get microbial diseases, they often rather use homeopathic means (or reiki) that have no demonstrable health benefit for the animals and are definitely less effective than antibiotics. Avoiding antibiotics in this case causes unnecessary animal suffering because the animals are not cured effectively.
  • An anti synthetic pesticides attitude. Organic agriculture avoids synthetic pesticides but uses natural, organic pesticides instead. However, some of those organic pesticides are more dangerous (toxic) than some synthetic pesticides used in conventional farming. For example copper sulphate, sometimes used in organic farming, is very persistent and more than 10 times more toxic than alternative synthetic fungicides (measured in LD50 doses). Other organic pesticides are particularly harmful to non-target invertebrates such as bees (e.g. pyrethrine, azadirachtin, rotenone, eucalyptus oil, neem oil). According to one study for soybeans, organic pesticides were less effective in controlling aphids, were as toxic or more toxic for non-target invertebrates and had higher Environmental Impact Quotients than synthetic pesticides.[5]
  • An anti synthetic fertilizer attitude. Synthetic fertilizers are considered unnatural, so therefore a lot of animal manure is used in organic farming. However, due to the application of animal manure (that is not treated with non-organic radiation or antibiotic means to kill the bacteria), organic crops can have a higher risk of contamination with dangerous E.coli bacteria.[6] Furthermore, the animal manure can be more harmful to aquatic life than synthetic fertilizer: per kilogram of product, organic products have higher eutrophication levels than conventional products, resulting in the suffocation of more fish.[7]
  • An anti synthetic fibers attitude. Looking at greenhouse gas emissions, land use, water use, pollution (human toxicity, aquatic toxicity) and many other indicators, the production of synthetic fibers is in many ways much better for the environment and human health than natural fibers such as cotton or animal fibers such as leather and wool.[8] Cotton has human toxicity and ecotoxicity levels more than twice as high as synthetic fibers such as PE.[9] Animal leather has a carbon footprint twice as high[10] and a water footprint 100 times as high[11] as synthetic leather from polyurethane. Shoes from cow leather have a three times higher carbon footprint than shoes from synthetic rubber.[12] Synthetic wool (fleece) from recycled materials has a much lower footprint than wool from sheep: no land use, lower water ecotoxicity and 100 times lower greenhouse gas emissions.[13] Using animal products instead of synthetic fibers harms animals: the sheep and cows for their wool and skin, and aquatic animals due to increased water pollution (water ecotoxicity).
  • An anti plastic attitude. Plastic bags have a much lower carbon footprint than paper, cotton and compostable starch bags.[14] A paper bag should be reused at least three times and a cotton bag at least 170 times before they become better for the environment than a disposable plastic bag. If we take into account toxicity, water use and land use, a cotton bag should be reused 500 times and a paper bag 30 times before they become better than a single use plastic bag. Reusable plastic bags are of course better still.
  • An anti clean meat attitude. Clean meat is meat produced without the animal. It is also called lab meat or cultured meat because it is made in a lab using stem cell cultures. It will be available in the supermarkets in a few years. Some meat eaters are reluctant to eat meat produced in a lab, because this appears to be unnatural. They say they would continue eating meat from animals, which requires killing and harming animals.
  • An anti intervention in nature attitude. There is a lot of wild animal suffering due to predation, parasitism, diseases, starvation,… Environmentalists are reluctant to intervene in nature to improve the well-being of wild animals. Such intervention are considered unnatural, “playing God” or human arrogance. Those environmentalists believe that we should leave nature alone, we should not interfere, in order to preserve its naturalness.
  • An anti exotic species attitude. Some animals (e.g. rabbits) are introduced to new ecosystems by humans. As humans are involved in the spreading of these newly arrived animals, their presence in the host ecosystems is considered as unnatural. These exotic species can sometimes endanger local fauna and flora. For example herbivorous exotic animals might eat rare local plant species. Some environmentalists consider these herbivorous exotic animals as pests and want to control them. The culling of those herbivorous exotic animals harms those animals.


Our preference for naturalness causes many victims: poor people dying from vitamin deficiencies, children with compromised immune systems dying from viruses, layer hens and other animals suffering in factory farms, sick animals in organic livestock farms, bees dying from organic pesticides, futures generations harmed by climate change, sheep and cows used for their wool and skins, aquatic animals harmed by water pollution, wild animals suffering in nature and herbivorous exotic animals culled to protect plant species.

If we can give more than 10 examples where a preference for naturalness is harmful to other beings (decreasing their well-being), it is time to let go of this preference. This preference is merely our own preference. Nature itself doesn’t care about naturalness. And the many victims don’t care about naturalness, or if they did, they still value well-being above naturalness. If people are willing to harm other beings because they value naturalness, they give a stronger preference to their own values than to the values of their victims. This is a kind of arrogance or egoism.

A preference for naturalness is comparable to an esthetic preference for art. Just like naturalness, beauty is a very vague concept. Who decides what is beautiful and how much value beauty has? And a strong preference for beauty can be harmful. Imagine a burning art museum, and you can save either a child or a painting. The painting itself doesn’t value its beauty and doesn’t care about the flames. The child does not want to sacrifice itself in the flames in order for you to save the painting. The child values well-being more than the beauty of the painting. If you save the painting instead of the child, you let your own preference for beauty overtrump the much stronger preference of the child to avoid the flames. We should never let our own vague and arbitrary preferences surpass the stronger preferences of others.


[1] Klümper W. & Qaim M. (2014). A Meta-Analysis of the Impacts of Genetically Modified Crops. PLoS ONE 9(11): e111629.

Brookes G & Barfoot P. (2013) Key environmental impacts of global genetically modified (GM) crop use 1996–2011, GM Crops & Food: Biotechnology in Agriculture and the Food Chain, 4:2, 109-119.

[2] Stein A, Sachdev H.P.S. & Qaim M. (2006). Potential impact and cost-effectiveness of Golden Rice. Nature Biotechnology 24, 1200 – 1201.

Wesseler J. & Zilberman D. (2014). The economic power of the Golden Rice opposition. Environment & Development Economics 19(6):724-742.

[3] Marvier M, McCreedy C, Regetz J, Kareiva P. (2007) A meta-analysis of effects of Bt cotton and maize on nontarget invertebrates. Science 316(5830):1475–7.

[4] Saja. K (2012). The moral footprint of animal products. Agriculture and Human Values. 30:193-202.

[5] Bahlai, C., Xue, Y., McCreary, C., Schaafsma, A., & Hallett, R. 2010, Choosing Organic Pesticides over Synthetic Pesticides May Not Effectively Mitigate Environmental Risk in Soybeans PLoS ONE, 5:6.

[6] Mukherjee A, Speh D, Dyck E, & Diez-Gonzalez F 2004, Preharvest evaluation of coliforms, Escherichia coli, Salmonella, and Escherichia coli O157:H7 in organic and conventional produce grown by Minnesota farmers. Journal of food protection, 67(5), 894-900.

[7] Clark M. & Tilman D. 2017, Comparative analysis of environmental impacts of agricultural production systems, agricultural input efficiency, and food choice. Environmental Research Letters12:6.

[8] CE Delft (2015) Milieu-informatie textiel

CE Delft, 2011. The Environmental Impact of Mink Fur Production.

Rastogi, S.K. et al., 2007. Occupational Cancers in Leather Tanning Industries: a Short Review. Indian J. Occup. Environ. Med., 11 (1), p. 3-5.

Stockholm Environment Institute, 2005. Ecological Footprint and Water Analysis of Cotton, Hemp and Polyester.

WFN (2017) Water Footprint Assessment of polyester and viscose. C&A Foundation

Mekonnen, M.M. and Hoekstra, A.Y. (2011) The green, blue and grey water footprint of crops and derived crop products, Hydrology and Earth System Sciences, 15(5): 1577-1600.

[9] CE Delft, 2011. The Environmental Impact of Mink Fur Production.

[10] Canals, M. e.a. (2002). Use of Life Cycle Assessment in the Procedure for Establishment of Environmental Criteria in the Catalan Eco-label of Leather. The International Journal of Life Cycle Assessment, 7(1), pp. 39-46.

Swiss Centre for Life Cycle Inventories, 2009. EcoInvent 2.0.

ESU Services, 2010. Carbon Footprint Polyurethane.

[11] Mekonnen, M.M. and Hoekstra, A.Y. (2012) A global assessment of the water footprint of farm animal products, Ecosystems, 15(3): 401–415.

WFN (2017) Water Footprint Assessment of polyester and viscose. C&A Foundation.

URS (2012). Review of Data on Embodied Water in Clothing Summary Report.  Waste & Resources Action Programme.

[12] CE Delft, 2011. The Environmental Impact of Mink Fur Production.

[13] Wiedemann e.a. (2016) Resource use and greenhouse gas emissions from three wool production regions in Australia. Journal of Cleaner Production Volume 122, Pages 121-132

Brock e.a. (2013). Greenhouse gas emissions profile for 1 kg of wool produced in the Yass Region, New South Wales: A Life Cycle Assessment approach.

CE Delft, 2011. The Environmental Impact of Mink Fur Production.

[14] Environment Agency (2011), Life cycle assessment of supermarket carrier bags: a review of the bags available in 2006.

Geplaatst in Artikels, Blog, English texts | Tags: , , , , , , , , , | Een reactie plaatsen

One day vegan…

One day veganWhat is the positive impact of eating a plant-based (vegan) diet for one day? A person with an omnivorous diet spends just a few minutes a day consuming animal products. But those few minutes cause a lot of harm. Replacing those animal products with animal-free, plant-based, vegan alternatives avoids those harms. So I have made some calculations of the harms avoided when an average omnivore (in Western-Europe) becomes a vegan for 1 day and replaces red meat with protein-rich soy beans and leguminous vegetable products, fish with omega-3 rich linseed, walnuts and seaweeds, eggs and poultry with vegetables and seeds, cheese with vegan nut cheese and milk with soy-milk. Replacing these animal products with their alternatives results in an optimal healthy diet (if supplemented or enriched with vitamins B12 and D).

How much harm is caused by just those few minutes of consuming animal products in an omnivore’s day?  What is avoided each day by being vegan? Below are the results. Adding together the avoided harm over 365 days in a year, several years of your life, the total positive impact becomes enormous.

Short summary: a vegan day saves 1 week of animal suffering in captivity, 1,5 hours of a consumer’s life due to less chronic diseases (cardiovascular diseases, cancers and diabetes), 0,5 hours of someone else’s life due to less health impact from global warming, malnutrition, antibiotic-resistant bacteria and new infectious diseases, and 10 hours of an average species lifespan.

Note 1: these results assume a high elasticity close to 1, which means a decrease of demand with one unit results in the decrease of production quantity with one unit. For animal products, more realistic elasticity values are around 0,7, because a decrease in demand by one person results in a decrease in price which results in a small increase in consumption by other people. Taking this lower elasticity into account, the results should be reduced with 30% .

Note 2: for each result I also give my epistemic status, i.e. my level of confidence in the results.

Harm to the animals

The death of 1 sentient animal (vertebrates, crustaceans and cephalopods, including sea animals killed as bycatch or used as fish meal). [Epistemic status: low] This corresponds with 1 week of animal suffering in captivity. [Epistemic status: moderate]

Harm to the environment

The emissions of 2 kg CO2 and other greenhouse gases, the equivalent of 1 minute traveling by plane or 15 minutes driving by car. [Epistemic status: high] These emissions contribute to climate change and generate a health cost on future generations (diarrhea, malnutrition due to harvesting loss, cardiovascular disease due to heat waves, malaria due to the spread of mosquitoes by higher temperatures and floods due to extreme weather events and sea level rise), resulting in an expected 4 minutes shortening of someone else’s life in the near future. [Epistemic status: very low]

The use of 1 m³ fresh water, as much as 2 hours non-stop showering. [Epistemic status: high]

The occupation of 700 m² highly fertile land for one day, used as cropland to grow animal feed. [Epistemic status: moderate] Combined with other environmental impacts that harm biodiversity, this results in an expected 10 hours shortening of the natural lifespan of an average species. [Epistemic status: very low]

Harm to the human population

The conversion of 600 grams of edible crops into inedible manure, which means the waste of one meal. [Epistemic status: moderate]

The malnutrition of 1 person for 5 hours. (5 people eating vegan means 1 person no longer malnourished). This also corresponds with an expected 7 minutes shortening of someone else’s life.  [Epistemic status: very low]

The use of 20 mg antibiotics, which increases the risk of microbial resistance, resulting in an increased mortality and hence an expected 4 minutes shortening of someone’s life in the near future. [Epistemic status: very low]

The shortening of someone else’s life with 15 minutes due to increased mortality from new infectious diseases (e.g. viruses such as avian flu, swine flu,…). [Epistemic status: very low]

Harm to your health

The consumption of 20 grams saturated fat. [Epistemic status: high] Combined with other harmful chemicals in animal products this amounts to a 10% higher risk of dying prematurely, or an expected 1,5 hours shortening of your life due to chronic diseases (cardiovascular diseases, cancers and diabetes). [Epistemic status: low]

Harm to the economy

The loss of 8 euro economic wealth due to environmental and health costs (costs of climate change, loss of labour activities and extra health care costs due to chronic diseases). If everyone adopts a vegan diet, the increase in economic wealth corresponds to an increase in income of 8 euro per person per day. [Epistemic status: very low]

Calculations and sources

-Consumption levels, kg per day (average Belgian person):

FOD Economie bevoorradingsbalans vlees 2013,

FAO Food Balance Sheets 2013 Belgium (www.fao.org/faostat/en/#data/FBS),

Flemish food survey.

-Animals killed and hours suffering per kg product:

Saja. K (2012). The moral footprint of animal products. Agriculture and Human Values. 30:193-202.

Fish killed for fish meal: FAO Fishstat, FishCount (fishcount.org.uk/) and Counting Animals (www.countinganimals.com/how-many-animals-does-a-vegetarian-save/)

-Greenhouse gas emissions per kg product:

Nederlandse consumptie van eiwitrijke producten. Gevolgen van vervanging van dierlijke eiwitten anno 2008. Blonk Milieu Advies, Gouda.

CE Delft (2011). Life Cycle Impacts of Protein- rich Foods for Superwijzer. Delft.

CE Delft (2010). Milieuanalyses voedsel en voedselverliezen. Delft.

Springmann M. e.a. (2016). Analysis and valuation of the health and climate change cobenefits of dietary change. Proc Natl Acad Sci. 113(15):4146-51.

-Health cost due to climate change, per unit CO2 emitted: 3,5 DALYs (disability adjusted life years) per 1000 ton CO2 (according to egalitarian perspective)

Goedkoop M. e.a. (2009). ReCiPe 2008. A life cycle impact assessment method which comprises harmonised category indicators at the midpoint and the endpoint level. Report I: Characterisation. Ministry of Housing, Spatial Planning and Environment, the Netherlands.

-Fresh water use per kg product:

Mekonnen, M.M. and Hoekstra, A.Y. (2010) The green, blue and grey water footprint of crops and derived crop products, Value of Water Research Report Series No.47, UNESCO-IHE, Delft, the Netherlands.

Mekonnen, M.M. and Hoekstra, A.Y. (2010) The green, blue and grey water footprint of farm animals and animal products, Value of Water Research Report Series No.48, UNESCO-IHE, Delft, the Netherlands.

Pahlow M e.a. (2015). Increasing pressure on freshwater resources due to terrestrial feed ingredients for aquaculture production. Science of the Total Environment 536 (2015) 847–857.

-Occupation of land (agricultural cropland footprint per kg product):

Global Footprint Network (2015) National Footprint Accounts

-Shortening of lifespan of species:

De Vos J. e.a. (2015) Estimating the normal background rate of species extinction. Conservation Biology 29(2):452–462

Stehfest E. e.a. (2008). Vleesconsumptie en klimaatbeleid. Nederlands Planbureau voor de Leefomgeving (PBL).

Each year an expected 1 in 10.000 species go extinct, which is 1000 times higher than the background extinction rate. There are about 10 million species, which means 1000 extinctions per year. A species has an average life expectancy of 10 million years, which means the current premature extinction of a species causes a shortening of 5 million years of a species lifespan. About 1/3 of the biodiversity loss is due to livestock farming (in particular due to the higher land occupation of an omnivorous diet compared to a vegan diet), which means the worldwide consumption of animal products instead of plant-based alternatives causes 1 species extinction per day, or the loss of 5 million species years. An average Western-European person consumes a share of 1 in 3 billion of the total livestock production (per capita consumption of animal products in Europe is twice as high as the world average). This means the omnivorous diet of an average Western-European causes the loss of 10 hours of a species lifespan, relative to a vegan diet.

-Conversion of edible crops (grains, soy and other edible products used as animal feed per kg product):

Global Footprint Network (2015) National Footprint Accounts


Lusk J & Norwood B. (2009) Some Economic Benefits and Costs of Vegetarianism. Agricultural and Resource Economics Review 38/2:109–124.

1 day vegan reduces demand of edible crops (mostly grains) with 600 gram. Total world cereal production is 7 Mton. A 1% decrease in production of cereal (corn) implies a 2% decrease in price (Lusk & Norwood 2009). A 2% decrease in cereal prices correlates with a 3,2% decrease in number of people malnourished (between 2009 and 2016 cereal prices dropped with 14% and number of malnourished people dropped with 22% from 1020 million ton 790 million). Hence, 1 day vegan means 0,2 people less malnourished for one day, or 1 person less malnourished for 5 hours.

If a vegan day means 0,2 people less malnourished for one day and 11% of people are malnourished, a vegan day saves 7 minutes of life because 0,56% of global deaths are due to protein-energy malnutrition (vizhub.healthdata.org/gbd-compare/), there are 3 deaths per 100.000 persons per day and an early death means an average of 2 million minutes of life lost.

-Antibiotic use

Center for Global Development (2017) A Global Treaty to Reduce Antimicrobial Use in Livestock, CGD Policy Paper 102

Each year an expected 1 in 10.000 people die due to infections from antibiotic resistant bacteria. More than 70% of antibiotics are used in livestock in developed countries. One death corresponds with an average 40 years of life lost. Multiplying these factors results in 4 minutes shortening of someone’s life.

-New infectious diseases

Jones e.a. (2008) Global trends in emerging infectious diseases. Nature 451, 990–993

15% of new (infectious) diseases come from global livestock (Jones e.a. 2008). Per capita consumption of livestock products in Europe is twice as high as world average. Assume that the mortality rate of new infectious diseases equals mortality rate of diarrhea, lower respiratory and other common infectious diseases. These diseases contribute 9% to total deaths. Total mortality rate is 1% per year (1 in 100 people die per year). One death means the loss of 40 years of life (21 million minutes).  Multiplying these factors (15% x 2 x 9% x 1% x 21 million minutes / 365 days per year) results in 15 minutes shortening of someone’s life.

-Saturated fats per kg product:


-Hours loss of healthy life per kg product:

Spiegelhalter D. (2012). Using speed of ageing and “microlives” to communicate the effects of lifetime habits and environment, Britisch Medical Journal, 345:e8676.

Springmann M. e.a. (2016). Analysis and valuation of the health and climate change cobenefits of dietary change. Proc Natl Acad Sci. 113(15):4146-51.

Replacing 1 kg of red meat with 1 kg of plant-based protein sources increases expected lifespan with 5 hours (Spiegelhalter, 2012).  Also the replacement of 1 kg of cheese, eggs and poultry meat with 1 kg of vegetables saves 5 hours. At the daily consumption levels of a European person, this means 1,5 hours of life saved per day.

This estimate corresponds with two other estimates: a vegan world reduces early mortality (causes of death) with 10% (for a vegan diet according to Springmann e.a. 2016) to 16% (for a diet high in fruits, vegetables, legumes, whole grains, nuts and seeds and low in red and processed meat according to vizhub.healthdata.org/gbd-compare/). At an average mortality rate of 1% per year and an average loss of 40 years or 350.000 hours per early death, this results in 1 – 1,5 hours of life saved per day.

-Loss of economic wealth

Springmann M. e.a. (2016). Analysis and valuation of the health and climate change cobenefits of dietary change. Proc Natl Acad Sci. 113(15):4146-51.

Total extra costs in 2050 in the scenario where everyone has an omnivorous diet rich in animal products, compared to the scenario where everyone eats vegan is 31 trillion dollars (27 trillion euro), including environmental costs (social cost of carbon), health costs and willingness to pay for mortality reductions. This is divided by 9 billion people in 2050.

Geplaatst in Blog, English texts | Tags: , , , , , , | Een reactie plaatsen

It’s time to take the Red Pill and change your mind

Are you afraid of changing your mind about a deeply held belief, about an emotionally strong conviction that you have? A few years ago, I changed – as an environmentalist – my mind about GMOs. Since then, and due to my contacts with the effective altruism movement, I changed my mind about many beliefs that I had:

In the political spectrum I consider myself as a progressive left liberal, meaning that I value social justice and I am against all kinds of unwanted arbitrariness such as discrimination (racism, sexism, speciesism,…). I am part of the left, criticizing the right. Criticizing the right is easy, because right wing ideologies contain a lot of irrationalities (such as unwanted arbitrariness that violate the moral golden rule) and pseudoscience (such as climate denialism). However, I start to realize that also in my left wing camp there are a considerable amount of irrationalities (such as the opposition against GMOs amongst leftist environmentalists, the criticism against vaccines or the strategies used by some animal rights activists, social justice warriors or people from the so called regressive left). As a rational ethicist, I not only want right winged people to become more rational, but I also want to improve rationality amongst leftist people. Therefore I also criticize irrational beliefs amongst left-wingers.

It is unlikely that all your beliefs are true. Even the beliefs that you strongly, emotionally care about may be wrong. I can say this, because that is what I experienced in my own life. Ten years ago I would have underestimated the amount of false beliefs that I strongly believed. I would have underestimated the number of moral mistakes I made. Now I realize that I should not trust my convictions based on emotions and gut feelings. So now I try to become less emotionally attached to my beliefs. When I am confronted with new evidence that contradicts my belief and I feel a strong emotional reaction that attempts to defend my belief, I become more alert and I try to suppress that emotional response, because I’ve learned that those emotional responses are unreliable. They have deceived me so many times. I should not have trusted them. These emotions generate all kinds of cognitive biases such as confirmation bias and desirability bias. As a consequence of avoiding emotional reactions, I became much more flexible to update my beliefs in the light of new evidence, ideas and arguments. And as a consequence, I changed my mind about many things. This strongly improved my rationality and my effective altruism.

So I want to create a culture where changing one’s mind is socially accepted and admirable. A culture where we dare to change our minds, to become more rational (meaning accurate in our beliefs, effective in our means and consistent in our ends). If you believe that all your currently strongly held beliefs are true, you are most likely wrong. If you believe that your strong emotions do not generate cognitive biases, you probably have a cognitive bias: the bias blind spot. If you want to become more effective in doing good, you will probably have to experience changing your mind about beliefs that you hold dear. You will probably have to swallow the red pill (as in the movie The Matrix).

Speaking about the red pill: I recently saw a documentary that I highly recommend: The Red Pill by Cassie Jaye. It is about a feminist’s journey into the men’s rights movement. The documentary is interesting because of two facts: first it tells about interesting facts and arguments made by men’s rights activists. Second, it follows the director Cassie Jaye in her struggle to change her mind about feminism and the men’s rights movement. She started as a feminist being very critical about this new movement that in her eyes was highly misogynist. But interviewing those men’s rights activists, it eventually resulted in Cassie Jaye saying that she no longer calls herself a feminist, even though she off course still shares the leftist values of gender equality and antidiscrimination (antisexism). Cassie Jaye is a prime example of a leftist person with good moral values, but who dared to change her mind about a topic that she held dear. She interviews people like Warren Farrell and Erin Pizzey, two persons who were deeply involved in the feminist movement but changed their minds about men’s rights issues (which resulted in receiving threats by feminists). In the documentary, we see emotionally strong reactions by feminists protesting against men’s rights activists. In a similar way, the documentary itself became highly controversial after its release, resulting in boycotts and feminist protests against its screening.

So, the documentary also changed my mind about gender issues. First, I believe that the feminist movement’s reaction against men’s rights issues is irrational, with feminists misrepresenting a lot of men’s rights activists as rape apologists. Second, I now no longer believe in something like a patriarchal system that systematically privileges men and suppresses women. The existence of a patriarchal system is a core belief in many feminist theories, so in that sense I no longer call myself that kind of feminist. The documentary gives a lot of examples that indicate that if there were such a thing as a patriarchal system, then that system is highly inconsistent. It becomes reasonable to doubt the existence of such an inconsistent system. Here are some examples.

-Child custody: if the judicial system is dominated by patriarchal, male judges, then why are children so often assigned to the mothers in cases of divorce, even when the fathers clearly state that they strongly prefer custody over the children? (It reminds me of the movie Mrs. Doubtfire that I recently saw.)

-Criminal sentencing: if the judicial system is dominated by patriarchal, male judges, then why do men receive 60% higher sentences than women for equal crimes? Arrested women are more likely to avoid convictions and are twice as likely to avoid incarceration if convicted. This is confirmed by other studies (and these studies were done by women, so no male privilege bias here). The latter research by Sigrid von Wingerden in the Netherlands indicates that when a woman kills a man the sentence is 1.6 years lower than when a man kills a man. And if a man kills a woman, the man gets a longer prison sentence than when a man kills a man. So when a man is murdered, it is apparently not as bad as when a woman is murdered, and if a man is the perpetrator, it is apparently worse than when a woman is the perpetrator?

-Health: if the scientific research and health systems are dominated by patriarchal, male researchers, and if the burden of disease, in terms of loss of healthy life years (DALY), premature deaths (mortality) and loss of health (morbidity), both globally and in the rich, western countries, is higher for men than for women, then why does breast cancer receive twice as much funding than prostate cancer?

-Mental health: if female suicide rate would have been 3 to 4 times higher than male suicide rate, feminists would have highlighted it, so why is it not highlighted in a patriarchal society that in the western world, male suicide rate is 3 to 4 times higher than female suicide rate?

-Military: if the military is dominated by patriarchal males, then why are men drafted? Why would those privileged men send men to die at the front? More than 95% of soldiers that die in war are men.

-Disasters: if there was a patriarchal system that privileges men, then why “women and children first” in case of a sinking ship?

-Dangers: if the man is in charge in the house, then why would the man risk his life to go downstairs at night when there is a burglar in the house? Why send men on dangerous exploration missions?

-Work: if the economy is dominated by patriarchal males, then why are more men doing the dangerous jobs? The death rate on the job is 11 times higher for men than for women. It is as if men are more expendable. Men are also doing some dirty jobs (sewer worker, garbage collector, miner).

-Education: if men want to have power over others, why would men allow more women to be in charge of education, risking their own children being indoctrinated with feminist ideas? There are more female school teachers. Wouldn’t it be better for men if men did the education part and women did the dangerous jobs?

-Media: if the news media is dominated by patriarchal male journalists, then why did the abduction of about 200 girls by Boko Haram in Nigeria receive more attention than the kidnapping and killing of thousands of boys by Boko Haram?

-Homelessness: if men are in charge of the social security system, then why are there about 4 times more men than women homeless?

-Interpersonal violence: if men create a patriarchal system, then why are more men than women victim of interpersonal violence (in terms of deaths and loss of healthy life years, both globally and in rich, Western countries)?

-Domestic violence: if politics is dominated by patriarchal male politicians, then why are there 2000 times more women shelters than men shelters in the US, for victims who want to escape from situations of domestic violence, even if domestic violence is close to gender symmetric? There are almost as much male victims and female perpetrators of non-reciprocal (no self defense) domestic violence. If the police were dominated by patriarchal male policemen, then why are women who assault their male partners more likely to avoid arrest than men who attack their female partners? When a woman calls the police to report domestic violence, the man is often arrested or ordered to leave the house, but if a man calls, the woman is almost never arrested or ordered out of the house, and even worse: the man who calls has a more than 10% probability of being arrested himself. Why would a man call the police if he risks being arrested himself?

All of this doesn’t make any sense in a patriarchal system that systematically privileges men. There are too many weird inconsistencies. And worse: if feminists target a patriarchal system, if the problem (patriarchy) is framed as being caused by men and the solution (feminism and women’s rights) refers to women, it might harm men even when those men are not the real problem. The real problem is gender roles that systematically disadvantage women in some ways and men in other ways. We should avoid a simplistic black-white male-female dichotomy where men are the privileged evil-doers. We should simply focus on eliminating all kinds of sexism and gender discrimination, of both men and women. And feminists should acknowledge that the men’s rights movement does not need to be silenced and that a lot of men’s rights activists raise valid concerns and are not rape apologists who hate women. Some but not all men’s rights activists hate women, but also some but not all women’s rights activists hate men.

Instead of calling myself a feminist, a women’s rights activist or a men’s right activist, I prefer to call myself an equal rights activist.

PS: if you think the above implies that I minimize the problem of women rights violations, you have a moral gravity bias. The above should not be interpreted as an endorsement of suppression of women, because that would be a logical fallacy.


Geplaatst in Blog, English texts | Tags: , , , | 11 reacties

De onderschattingsdenkfout

Wat hebben de volgende voorbeelden met elkaar gemeen?

  • In een interview beweerde de filosoof Maarten Boudry dat het jihadisme haatdragender en daarom in zeker opzicht gevaarlijker is dan het nazisme, omdat jihadisten geloven dat andersgelovigen en ongelovigen zoiets slechts doen (namelijk niet geloven in Allah) dat ze een straf van eeuwige extreme marteling in de hel verdienen en men dus best mag pronken met de gruwelijkste slachtpartijen van andersgelovigen, terwijl nazi’s naar joden kijken zoals een tuinman naar onkruid: iets dat efficiënt uitgeroeid maar niet oneindig gehaat moet worden. Vanuit antinazistische en antifascistische hoek kwam hierop kritiek, met de bewering dat Maarten Boudry aan negationisme (Holocaustontkenning) doet en daarom ontslagen moet worden van universiteit van Gent.
  • In een artikel maakte psychologe Roos Vonk een vergelijking tussen de vee-industrie en de Holocaust: in beide gevallen gaat het om een systematische grootschalige onderdrukking, opsluiting en doding van miljoenen slachtoffers. Ook hierop kwam kritiek vanuit antifascistische hoek, met onder andere de bewering dat dergelijke analogieën ongepast zijn voor de slachtoffers en nabestaanden van de Holocaust (ondanks het feit dat veel Holocaustoverlevenden zelf na de oorlog die analogie gingen maken; zie het boek Eternal Treblinka van Charles Patterson).
  • In de documentaire The Red Pill over de mannenrechtenbeweging presenteert de documentairemaakster Cassie Jaye enkele cijfers over hoe mannen in de westerse samenleving op een aantal vlakken systematisch benadeeld worden en er dus moeilijk sprake kan zijn van een patriarchaal systeem dat mannen stelselmatig bevoordeelt en vrouwen onderdrukt, zoals veel feministen geloven. Zo blijkt dat er in de VS meer dan 2000 vluchthuizen zijn voor vrouwen die willen wegvluchten van situaties van huiselijk geweld door hun partners, terwijl er slechts 1 vluchthuis is voor mannen. Dat terwijl het huishoudelijk geweld bijna gendersymmetrisch is: er zijn volgens statistieken van de Centers for Disease Control and Prevention (CDC) bijna evenveel mannelijke als vrouwelijke slachtoffers van huiselijk geweld en bijna evenveel mannelijke als vrouwelijke daders (het geweld van een man is wel iets schadelijker voor het slachtoffer). Vanuit feministische en antiseksistische hoek kwam zware kritiek op deze film, met onder meer boycots, verstoringen van filmvoorstellingen en het zwart maken van Cassie Jaye in de media.
  • In een tweet schreef de beroemde bioloog Richard Dawkins: “Date rape is bad, stranger rape at knifepoint is worse. If you think that’s an endorsement of date rape, go away and learn how to think.” Daarop werd Richard Dawkins door feministen en antiseksisten aangevallen met de kritiek dat men verkrachtingen niet met elkaar mag vergelijken, dat de schade van date rape voor de slachtoffers niet te onderschatten is en dat dergelijke uitspraken ongepast zijn voor die slachtoffers van date rape.
  • Als dierenrechtenactivist maak ik regelmatig de vergelijking tussen niet-menselijke dieren enerzijds en mentaal gehandicapte weeskinderen anderzijds, in de zin dat beiden rechten verdienen omdat ze voelende wezens zijn, ondanks hun gebrekkig rationeel-moreel-zelfbewust denkvermogen. Daarop werd ik door Social Justice Warriors (SJW’s) beschuldigd van ableisme: discriminatie op basis van handicap, alsof ik gehandicapte (disabled) mensen discrimineer ten opzichte van ongehandicapte (abled) mensen.

Wat deze voorbeelden met elkaar gemeen hebben, is wat ik noem de onderschattingsdenkfout of  ‘moral gravity bias’. De vergelijkingen tussen bijvoorbeeld jihadisten en nazi’s of tussen varkens en mentaal gehandicapten raken telkens een gevoelige snaar bij de critici en wekken bij hen hevige emoties op. Die emoties zorgen ervoor dat de critici als het ware blind zijn voor hun denkfout. Ze denken spontaan maar ten onrechte dat de bekritiseerde personen in kwestie de ernst onderschatten van bijvoorbeeld het nazisme, de Holocaust, huiselijk geweld door mannen, date rape of rechtenschendingen van mentaal gehandicapten.

De bekritiseerde personen – Maarten Boudry, Roos Vonk, Cassie Jaye, Richard Dawkins en ik – maken een vergelijking tussen A en B en de critici denken dat de bekritiseerde personen daarmee A trivialiseren en de ernst van A onderschatten. Maar een vergelijking maken tussen A en B zegt nog niets over hoe erg A is. In plaats van denken dat de bekritiseerde personen de ernst van A onderschatten, kan men net zo goed denken dat de critici de ernst van B onderschatten. Dat is logisch gezien evengoed mogelijk. Maarten Boudry kan dan stellen dat hij het nazisme even verwerpelijk vindt als zijn antifascistische critici maar dat zijn critici de ernst van het jihadisme ernstig onderschatten. Roos Vonk kan repliceren dat ze evengoed tegen de Holocaust is maar dat haar critici de ernst van de veeteelt onderschatten. Cassie Jaye kan stellen dat de feministen de schendingen van mannenrechten onderschatten en dat er niet minder vluchthuizen voor vrouwen maar wel meer vluchthuizen voor mannen moeten komen. Richard Dawkins kan dan weer argumenteren dat zijn antiseksistische critici de ernst van verkrachting door een vreemde met mes op de keel onderschatten en dat de kritiek van die antiseksisten ongepast is voor de slachtoffers van dergelijke verkrachtingen. En ik kan zeggen dat ik niet de morele waarde van mentaal gehandicapten onderschat, maar dat mijn critici (die zelf betrokken zijn bij dierenrechtenschendingen door bv. vleesconsumptie) wel de morele waarde van niet-menselijke dieren onderschatten.

Waarom wijs ik op deze denkfout? Ik ben tegen nazisme, tegen verkrachtingen, tegen huiselijk geweld, tegen onderdrukking van vrouwen, tegen sociaal onrecht en tegen alle vormen van discriminatie en dus ook tegen ableisme, racisme en seksisme. Ik deel dus de linkse progressieve waarden van de antifascisten, feministen en social justice warriors. Wijzen op de talrijke irrationele opvattingen en denkfouten van het rechtse conservatieve kamp is eenvoudig. Dat conservatieve kamp promoot vele vormen van discriminatie en sociaal onrecht. Maar ook in het linkse, humanistische, liberale, progressieve kamp circuleren soms nog irrationele opvattingen en denkfouten. Als rationeel ethicus wil ik het progressieve kamp sterker maken door het op vlak van rationaliteit te versterken. Daarom uit ik relatief veel kritiek op linkse activisten die denkfouten maken, ook al zit ik samen met hen (en samen met Maarten Boudry, Roos Vonk, Cassie Jaye en Richard Dawkins) in hetzelfde progressieve kamp dat strijdt tegen allerlei vormen van discriminatie.

Geplaatst in Artikels, Blog | Tags: , , | 18 reacties