The extreme cost-effectiveness of cell-based meat R&D

In previous articles, I argued that supporting research and development of cell-based meat technologies could be perhaps the most important strategy to protect animal rights and improve animal welfare (with a possible exception of research in welfare biology to improve wild animal welfare). Here I want to do a very rough back-of-the-envelope Fermi-estimate calculation of the cost-effectiveness of cell-based meat R&D, and compare it with traditional animal rights and vegan advocacy campaigns. I only estimate the orders of magnitude, in powers of ten. The results are presented in the table below. The three measures are:

  • The number of vertebrate animal saved per euro, which includes all fish, birds and mammals that are no longer killed by humans for food (i.e. excluding invertebrates and animals not directly killed by humans).
  • The number of vertebrate land animals spared per euro, which includes all farm animals that are no longer bred in captivity.
  • Ton CO2e emissions avoided, which includes all anthropogenic greenhouse gases that are no longer emitted, measured in CO2-equivalents (excluding the carbon capture and storage capacity of reforested farmland).

Cell-based meat R&D calculations

There are 1011 vertebrate land animals used (i.e. bred and killed) per year by humans. Assume that this number is constant until cell-based meat enters the market. The number of vertebrate animals directly killed by humans for food is an order of magnitude higher: 1012. The human population counts 1010 humans, also assumed to be constant, which means an average human uses 10 vertebrate land animals per year and kills 100 vertebrate animals per year. Hence, eating vegan for one year spares 10 animals and also saves 100=1 ton CO2-equivalent greenhouse gas emissions.

Global funding for cell-based meat is 108 euro per year. This corresponds with 102 cell-based meat companies and research units at universities, employing on average 10 employees per organization and 105 euro per employee per year.

Assume in the business-as-usual scenario (where you do not contribute) the same amount of money is funded (by other people) every year until cell-based meat becomes cost-competitive with animal-based meat on the market. In other words, if 108 euro were not invested in cell-based meat this year, the arrival of cell-based meat on the market would be delayed by one year. If 1 euro were invested this year, the arrival on the market will be advanced with 10-8 year.

Also, assume that the probability that cell-based meat will eliminate animal-based meat and animal farming, is 1/10 (or cell-based meat is guaranteed to take 10% of the meat market in the future). This is probably a low estimate.

The above estimates measure the scale (1011 animals used per year), the solvability (1/10 probability of eliminating animal farming) and neglectedness (10-8 years faster elimination per extra euro funding). Now the number of animals spared per extra euro donated to cell-based meat R&D can be calculated as the product of scale, solvability and neglectedness: 1011x10-1x10-8=102. This means one euro extra funding spares 100 vertebrate land animals. Including captured and aquaculture fish (also fish used for fish meal for farm animals), the number becomes an order 10 higher: 1000 vertebrate animals saved per euro.

As sparing 1 farm animal corresponds with reducing 0,1 ton CO2e, this one euro funding also means a reduction of 10 ton CO2e, the same order of magnitude as the emission by an average human in one year. Used as carbon offsetting, cell-based meat R&D has a price around 0,1 euro per ton CO2e averted. This is much lower than most other carbon offsetting mechanisms.

Note: the basic (in my opinion realisitic) assumption in the above calculation is that other people invest in cell-based meat R&D anyway, and that in the business-as-usual scenario (where you do not fund anything) no other strategy (technology, intervention, vegan outreach campaign,…) will be able (even with more funding) to abolish animal farming before cell-based meat enters the market at competitive prices. Suppose cell-based meat arrives within a few decades and eliminates animal farming in say 50 years, whereas another, next best strategy would eliminate animal farming in 100 years. Suppose that this other strategy was less costly, for example requiring only 10 million euro funding per year over a period of 100 years to abolish animal farming, whereas cell-based meat would require 100 million euro funding over 50 years. And suppose that other strategy was more neglected, for example receiving only 10 million euro funding per year, compared to 100 million for cell-based meat. Even then, extra funding for that other strategy would not be effective when it is impossible to speed it up such that it will eliminate animal farming within 50 years. When that other strategy takes more than 50 years anyway, it will become obsolete anyway in the business-as-usual scenario where cell-based meat arrives earlier and eliminates animal farming earlier. A global coordination such that all cell-based meat funding goes to that other, less costly strategy, is not effective (not so feasible). Hence, the most effective thing to do for us, is to accelerate that cell-based meat research, such that it enters the market one year earlier. That saves an extra year of animal suffering and greenhouse gas emissions. If other strategies received more funding, there is a likelihood that they make cell-based meat obsolete, and this consideration is included in the estimated 10% probability of cell-based meat eliminating animal farming.

The above is a low estimate of the impact of cell-based meat R&D. A higher estimate can be obtained as follows. Suppose it takes 102 years of research at 108 euro of funding per year before cell-based meat becomes competitive with animal-based meat. Suppose 90% of the funding are investments that will eventually be payed back by consumers who buy cell-based meat. The remaining 10% has no return on investment and hence counts as real costs. Hence, the amount of funding costs is 107 euro per year. Suppose without cell-based meat, humans will use farm animals for another 10.000 years at 1011 animals per year. The probability that cell-based meat will eliminate animal farming is again 10-1. In this scenario, contributing 1 euro of funding has an impact of 104 years times 1011 animals per year times 10-1 probability divided by 102 years times 107 euro per year, which equals 105 vertebrate land animals spared per euro. This sparing of farm animals is again accompanied by avoided greenhouse gas emissions, but most of those avoided emissions would have happened in the far future. Considering only the short term emission reduction for a time period of 10 years, this again comes down to a carbon offsetting price of around 0,1 euro per ton CO2e averted.

Note that the neglectedness is important. Consider for example investments in plant-based meat, which is an order of magnitude larger than investments in cell-based meat, i.e. 10 times less neglected. Suppose plant-based meat also has a probability of 10% of eliminating the animal-meat market (or reducing animal farming by 10%). Then the effectiveness of investments in plant-based meat is an order of magnitude lower than the investments in cell-based meat. Of course, both plant-based and cell-based meat can mutually reinforce each other (i.e. they can be complementary instead substitutable strategies), and from a risk perspective, it is useful to invest in a diverse portfolio of strategies.

Vegan advocacy campaigns calculations

The above impact estimates of cell-based meat R&D can be compared to other measures to reduce animal farming.

Animal Charity Evaluators estimates a cost-effectiveness of around 10 farm animals spared per euro donated to its top recommended charities. This is an order 10 lower than cell-based meat R&D.

Vegan outreach leafletting has an estimated impact of 1 animal spared per euro. I did a personal leafletting study (at the Belgian animal rights organization Bite Back) whereby the leaflets included a survey that asks questions about the reduced consumption of animal products due to the leaflet. Based only on the responses of non-vegans who answered that they reduced their animal product consumption, it requires roughly 1000 leaflets for one equivalent conversion to veganism. This was measured in vegan-equivalents, i.e. in terms of the equivalent reduction of the number of animals used. For example, two meat-eaters who reduce their consumption by 50% count as one vegan. Assume that a respondent remains vegan or sticks to his reduced animal product consumption for 10 years. One vegan-equivalent spares around 10 farm animals per year and one leaflet costs 0,1 euro. That means a cost-effectiveness of 1 spared animal per euro (i.e 10 animals per vegan year times 10 years divided by 1000 leaflets times 0,1 euro per leaflet). This is in the same order of magnitude of other cost-effectiveness estimates of leafletting.

Vegan education (giving presentations about veganism) also has a cost-effectiveness of 1 spared farm animal per euro: 10 participants of a lecture times 1% probability of a participant becoming vegan (based on a small personal study that surveys high school students who participated my vegan education lectures) times 10 years of remaining vegan times 10 animals spared per vegan year divided by 10 euro costs per lecture (if I were to be paid an hourly wage of 10 euro).

We can also estimate the overall cost-effectiveness of animal advocacy campaigns. The US population has an order of magnitude 108 people. Suppose meat consumption is decreased by 10% due to people becoming reducetarians, vegetarians or vegans. Suppose 10% of this reduction is due to animal advocacy campaigning. Then the number of US vegan-equivalents for animal welfare reasons is 106. The two largest animal advocacy organizations (HSUS and Peta) have a yearly budget of 108 euro. If their campaigns caused the reduction in meat consumption, we get a cost-effectiveness of 0,1 farm animals spared per euro donated to those animal charities (106 vegans times 10 animals spared per vegan per year divided by 108 euro funding per year). This means cell-based meat R&D is about 1000 times more effective than average animal advocacy.

As I do not expect that the traditional vegan outreach campaigns are more likely to eliminate animal farming sooner than cell-based meat in a business-as-usual scenario, a high estimate calculation similar to the cell-based meat high estimate is not possible.

The case for cell-based, clean meat R&D can be compared to the case for clean energy R&D, as argued here. Clean energy R&D funding is estimated to be more effective than e.g. regulatory climate measures, cutting fossil fuel subsidies and environmental behavioral change campaigns. The latter are analogous to animal farming regulations, cutting animal farming subsidies and vegan consumption campaigns.

Conclusion

Cell-based meat research and development is roughly 10 times more cost-effective than top recommended effective altruist animal charities and 1000 times more cost-effective than average animal advocacy and vegan campaigning. One euro funding for cell-based meat R&D could spare the lives of 100 farm animals, save the lives of 1000 vertebrate animals and avoid 10 ton CO2-equivalent emissions. That makes cell-based meat R&D probably the most effective measure to reduce anthropogenic animal suffering and greenhouse gas emissions.

You can support cell-based meat R&D by donating to New Harvest.

For a further discussion, including another estimate of the cost-effectiveness of cell-based meat (with a roughly same result but a different method), see the comments section here.

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

Veganmodernism: the end of veganism?

Just like the environmental movement gave birth to ecomodernism, the vegan movement can give birth to veganmodernism. Ecomodernism focuses on technological innovations (e.g. clean energy, genetically modified organisms,…) to decrease our environmental impact, rather than consumer behavioral change campaigns or corporate pressure campaigns to persuade consumers and producers to go green. Veganmodernism does the same: instead of persuading consumers to go vegan, it focuses on the development of animal-free versions of animal products, such as cultivated (cell-based) meat, leather and milk without cows, and egg-proteins without chickens.

Veganmodernism focuses in particular on research and development of cell-based meat technologies. This is probably one of the most effective things we can do in the short term (e.g. the next two decades) to make the world better.

Focus on big problems

Veganmodernism helps to solve some of the biggest problems.

  1. Anthropogenic suffering. Most anthropogenic (human-caused) suffering is due to meat production (animal farming and fishing). The number of humans killed is much smaller than the number of farm animals killed for meat. The number of humans in extreme poverty is much smaller than the number of farm animals who are likely to have net negative welfare levels. The number of animals kept in captivity for experimentation, fur production or entertainment is much smaller than the number of farm animals. The number of animals used for meat is larger than the number of animals used for eggs and dairy.
  2. Climate change. Combining the greenhouse gas emissions and the carbon opportunity costs, animal farming is probably the human activity with the largest climate impact. Hence, the replacement of animal meat by animal-free (cell-based) meat is a very effective climate measure.
  3. Pandemics. Animal farming is one of the leading causes of infectious zoonotic diseases that could become pandemics (e.g. bird flu, swine flu, coronaviruses,…)

Avoiding many problems

Veganmodernism avoids or attenuates many problems and discussions (see also here and here).

  1. Avoiding the meat-eater problem. In most cases, economic development and saving human lives causes increased meat consumption and hence increased animal suffering and environmental impact. Animal farming increases human health risks (e.g. infectious zoonotic diseases), uses a lot of resources and contributes to climate change. Hence, replacing animal products with cheaper, healthier and cleaner alternatives improves the economic welfare and health of humans without generating extra animal suffering and environmental impact.
  2. Avoiding the welfarist-abolitionist debate. Welfarist animal charities and advocates want to improve the living conditions of farm animals, whereas abolitionists want to eliminate animal farming. The abolitionists strongly value animal rights such as the right not to be used as merely a means, and this is not compatible with animal farming. These abolitionists criticize welfarists, claiming that it is difficult to know what improves the welfare of farm animals, that most proposals of welfare improvements can have negative side-effects (e.g. creating extra animal health risks or environmental impacts) and that welfare improvements can increase meat consumption (because they soothe the conscience of consumers) and hence the number of animals being used and killed. The production of cell-based meat avoids using animals and hence avoids animal rights violations and welfarist negative side-effects. The meat is produced without the sentient animals. Cell-based meat promotion is compatible with both utilitarian (welfarist) animal welfare and deontological (abolitionist) animal rights views.
  3. Avoiding backfire effects. In contrast with corporate pressure campaigns, developments of animal-free products is not expected to provoke a lot of backlash from the animal industry. For example, after a release of undercover investigations of factory farms, the animal industry pushes back by advertising for more meat consumption. However, even some meat processing companies are investing in plant-based and cell-based meats, and even some researchers in animal production who have strong ties with the animal industry, are doing research in cell-based meat. None of those people and companies were supporting veganism.
  4. Avoiding psychological and sociological uncertainties. Psychologists are studying what causes people to change their behavior. Sociologists are studying what causes societies to change their cultural norms (values and systems). They do research on nudging (changing the choice environment such that people are automatically inclined to perform more of the preferred behavior), motivational interviewing, persuasion, effective communication, social protest movements,… But these areas of research are still full of uncertainties, and silver bullets or simple effective solutions have still not been found and progress is very slow. When cheap, high quality animal-free products are available and marketed by competitive firms, no knowledge about behavioral change (e.g. nudging) and cultural change (e.g. effective protest movements) is necessary. People can still eat the same products and meals, only the production processes differ (the new processes exclude the use of animals).
  5. Attenuating the wild animal suffering problem and predation problem. Decreasing animal farming could free agricultural land for reforestation. More nature also means more wild animals, and this can increase wild animal suffering. However, more forests also means more carbon capture and storage, hence less climate change and less animal suffering from climate change. In this sense, animal-free agriculture is one of the most effective strategies to decrease climate change. But in the long run, cell-based meat can also be part of the solution to the predation problem: carnivorous predator animals can eat animal-free meat instead of animal meat. Solving the predation problem could drastically decrease wild animal suffering.

Plant-based versus cell-based meat

To eliminate the market for animal meat, there are four approaches. There are two markets: for animal products and animal-free alternatives. Each market has two sides: supply and demand. Hence, we can target either the demand side or the supply side, by changing respectively the behavior of individual consumers or the choices of food producers. For each target, we can apply either push or pull strategies: making animal products less attractive (pushing the economy away from the animal product market) or making animal-free foods more attractive (pulling the economy towards the animal-free alternatives market). Of the four possible market strategies, I argued that the supply side pull strategy has the best prospects, because the other three have shown poor track records over the past decades.

The supply side pull strategy consists in the development of plant-based and cell-based meat. Based on the Importance-Tractability-Neglectedness (ITN) framework, I will argue that priority should be given to research and development of cell-based meat above plant-based meat (elsewhere, I applied the same ITN-framework to argue that charities that support cell-based and plant-based meat developments are highly effective; here I argue why in particular cell-based meat could be prioritized).

Considering importance or scale, cell-based meat is expected to have a bigger market than plant-based meat. Not only humans can eat cell-based meat, but cell-based meat can also be beneficial for carnivorous animals under human care (e.g. pets and rescued wildlife animals), and in the long run other wild animals. There are many predators in nature. People can doubt whether plant-based meat is healthy (sufficiently high in quality) for e.g. cats, but cell-based meat is the same product as animal-based meat and hence has the same food quality for carnivorous animals as animal-based meat.

Considering neglectedness, in 2019 there were 55 cultivated meat and seafood industry startups globally, receiving $77 million of venture capital investments. In contrast, in 2019, in the US alone there were 143 plant-based meat, dairy and eggs companies, receiving $460 million of venture capital investments. There are no cell-based meat retail sales and no cell-based meat companies on the stock market. In contrast, in the US alone, plant-based meat retail sales were $900 million in 2019, and the sector received $290 million in net new public share offerings. Hence, cell-based meat has a much smaller industry than plant-based meat, which means it is more neglected. As a comparison, in the US, animal and environmental charities received almost $12 billion donations in 2018. This is much more than the global venture capital investments in cell-based meat. As processed cell-based meat at competitive retail prices is not expected on the market within 10 years, and unprocessed (whole tissue) cell-based meat is not expected on the market within 20 or 30 years, we can expect that cell-based meat will remain relatively neglected the next two decades.

Cell-based meat research is still in its infancy, requiring a lot of fundamental innovative research. This kind of research is undersupplied in a competitive free market, due to a market failure (knowledge about cell-based meat production processes has the characteristics of a public good). Therefore, cell-based meat is expected to have higher long-run impact research opportunities compared to plant-based meat for the coming years.

Due to the relative neglectedness, the value of information of the potential cell-based meat impact is relatively high. We do not yet have a lot of information about the potential impact of cell-based meat, e.g. how fast the production costs will decrease, how fast bottle-necks will be solved, how fast consumers will accept it, how fast it will resemble animal-based meat. From all four market strategies (the abovementioned push and pull, demand and supply strategies), the effectiveness of a supply side pull strategy remains most uncertain. Investing in cell-based meat technologies now allows us to quickly gain new valuable information about the effectiveness of cell-based meat with regard to eliminating animal farming.

Considering tractability or solvability, research and development of new technologies has a long track record of high impact. This also goes for new food and bio-engineering technologies. Hence, it is very likely that extra funding for cell-based meat R&D will be productive. This can be contrasted with traditional veganism strategies that primarily focus on behavioral change. It is unlikely that the next two decades will generate a lot of new knowledge about effective psychological persuasion strategies to persuade people to go vegan. Effective communication or changing the choice architecture (nudging) have limited impact and no good track record of improvements. Scientific evidence about the effectiveness of e.g. leafleting or online ads remains very limited, with small effect sizes and a lot of statistically insignificant results.

The tractability of cell-based meat R&D is not lower than plant-based meat R&D. It is unlikely that plant-based meat can replace all kinds of unprocessed meats and seafood. With cell-based meat, on the other hand, meat eaters can still eat their preferred ribs, beef stew, pork tenderloins and bacon, all cell-based. Hence, it can be expected that cell-based meat is more appealing to traditional meat eaters than plant-based meat. Traditional meat eaters are conservative in the sense that they are reluctant to change their behavior or identity. Hence, messages such as “eating vegan” (i.e. changing behavior) and “going/becoming vegan” (i.e. changing identity) are less effective for them. With cell-based meat, they can eat the same product, only the production process is different: cell-based meat requires cells, animal-based meat requires whole animals. As the product is exactly the same, no behavioral change (change in consumption choices) is required. Furthermore, as cell-based meat is the same product as animal-based meat, it can have the same name. The difference between cell-based and animal-based meat is the production process (one involving cells, the other animals), but the name of a non-trademarked product category such as ‘meat’ or ‘milk’ does not depend on the production process.

I expect that cell-based meat is more limited in the number of possible cost-effective production technologies than plant-based meat (i.e. there are more different ways to produce plant-based meats than cell-based meat), and that cell-based meat production will be more technology intensive than plant-based meat production (i.e. cell-based meat is more high-tech than plant-based meat). That means cell-based meat production technologies are more susceptible to patenting and market monopoly power. To avoid problems with market monopolies and intellectual property rights, open source research becomes more important. This kind of research requires more independent funding instead of private investments. Both cell-based and plant-based meat will benefit from private (venture capital) investors who invest in start-ups, but for the short term I expect that cell-based meat will also be relatively more benefited from donors (governments, philanthropists, animal advocates) who finance fundamental open source research in cell-based meat technologies. Plant-based meat will benefit less from philanthropic donor funding, due to the already high levels of private investments and the lower risks of market monopoly powers related to intellectual property rights.

The end of veganism?

As mentioned above, cell-based meat allows for traditional meat eaters to eat the same products that they used to eat, but without using animals. Combined with animal-free dairy, eggs, leather, wool and other products that used to be derived from animals, veganism becomes redundant. No behavioral or identity change are required. Messages like “eat vegan” and “go vegan” as well as vegan cookbooks, vegan cooking workshops, vegan potlucks, vegan recipes, vegan festivals and vegan outreach become superfluous.

The advent of the mass-produced cars in the 1920’s resulted in an almost complete elimination of the use of draft horses for carriage within four decades. In the film industry, real animals (e.g. a real orang-oetan in the 1978 movie Every Which Way but Loose with Clint Eastwood) are replaced by computer animated animals (e.g. a CGI-created dog in the 2020 movie The Call of the Wild with Harrison Ford). Plenty of other examples (messenger pigeons, whale oil,…) demonstrate that new technologies replaced the use of many animals, without much animal activist pressure campaigns or consumerist behavioral change campaigns. These campaigns became obsolete. When cell-based meat enters the market, the same is likely to happen for vegan consumer and corporate outreach campaigns. Instead of vegan organizations, cell-based meat companies will do the marketing for animal-free products.

In fact, all of this means that we can eliminate animal farming, without the need of the word ‘veganism’. People do not have to call themselves ‘vegan’, traditional meat eaters do not have to know what veganism is. Compare it with the hypothetical ‘automobilism’, the ideology that we should not use horses for transport and use horse-free vehicles such as cars instead. One could have started ‘go auto’ or ‘drive auto’ campaigns to persuade people to stop using horses. One could do research on the most effective, convincing strategies that persuade people to go auto. One could do pressure campaigns targeting draft horse companies and horse breeders. One could inform the public about all the problems with draft horses: animal suffering (exhaustion, whipping, captivity), pollution (horse manure in the streets), inefficient use of resources (land area for horse feed),… But all of this would have become superfluous when the efficiency and usability of cars increased and their prices dropped drastically due to new car mass production technologies (e.g. Ford’s Model T). Just like an automobilism ideology became unnecessary, a veganism ideology can become unnecessary when cheap, high quality cell-based meat enters the market and outcompetes animal-based meats due to its better production process.

Here we can draw again the analogy between veganmodernism and ecomodernism. The traditional environmental movement is reluctant towards ecomodernism, because ecomodernism makes traditional environmentalist value systems such as ‘localism’ (e.g. deglobalization, degrowth, bioregionalism, anticorporation, small scale production) and ‘naturalism’ (e.g. organic agriculture, non-synthetic products, low-tech production) obsolete.[1] Ecomodernism focuses on high-tech solutions to decrease our environmental impact, instead of a drastic behavioral change (austerity). In the past, new technologies allowed for fast and drastic reductions in environmental impact (e.g. LED-lights that use renewable and nuclear power), which could not be achieved by less effective austerity campaigns.

Veganmodernism  and cell-based meat (and dairy, eggs, leather,…) could be the final strategy for meat abolition, could be the end of animal farming, but could in a sense also be the end of veganism in the animal rights movement, just like ecomodernism could mean the end of localism and naturalism in the environmental movement.

Hence, animal rights activists and advocates can shift their strategies and tactics: instead of spending time and money doing traditional veganism behavioral change and corporate pressure campaigns, they can look for opportunities to raise, earn and donate more money to an organization like New Harvest, that supports open source research and development of new cell-based meat technologies. It could be the case that, just like this analysis for effective climate change policies, clean meat R&D is more effective than e.g. a meat tax or cutting livestock subsidies (see the table of climate policies ranked according to a combined importance, neglectedness and tractability score, with clean energy R&D at the top, carbon taxes at position 5 and cutting fossil fuel subsidies at 9).

Even if vegan advocacy and corporate pressure campaigns become obsolete when all animal products are replaced by the same products that do not use animals in the production processes, campaigning for antispeciesism and moral circle expansion towards all sentient beings remains relevant. In fact, when humans no longer use animals for food or clothing, moral circle expansion becomes easier, because humans will have less cognitive dissonance when they no longer use animals.


[1] There is a crucial difference between the localist and naturalist value systems in the environmental movement, and the veganist value system in the animal rights movement. Localism and naturalism can be seriously harmful or counterproductive, whereas veganism is not counterproductive.

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

The very complex welfare impact of fishing

Does fishing worsen the state of the oceans? For animal rights activists and most environmentalists, the answer seems evidently ‘yes’. But thinking more carefully, matters become very complicated.

The easy, philosophical problem: measuring aggregate welfare

First we need to solve the easy problem: the philosophical or moral question how to measure the state of the oceans. As aquatic ecosystems have no consciousness and hence are not concerned about e.g. their biodiversity, integrity, stability, natural beauty or some other ecocentric value, I argue that we should focus on those values and instead consider the welfare of sentient beings in the oceans. Sentient beings do care about their own well-being and preference satisfaction, and they do want to avoid negative experiences.

When valuing the welfare of fish, the causes of welfare loss should not influence the valuation. Fish can have a preference for not being killed, but they most likely do not make a distinction between death caused by predatory fish versus death caused by humans. If we want to avoid speciesist arbitrariness, we should not make a distinction between harm (welfare loss) caused by humans and harm caused by non-human animals such as predators.

Deciding that only the welfare of sentient beings is important, is not enough. We need to decide how to aggregate all the welfare levels of all sentient beings. This is the problem area of population ethics. Here I propose the total view (total utilitarianism): the state of the ocean should be measured in terms of the total sum of welfare levels of all sentient beings in the oceans. This can be contrasted with e.g. the aggregate view, whereby the average of all welfare levels is considered.

If we focus on the aggregate welfare in the oceans, fishing becomes very important, because the number of vertebrate aquatic animals killed in fisheries and aquaculture (more than 1 trillion per year) is an order of magnitude larger than the number of vertebrate land animals killed in livestock farming and hunting (less than 100 billion per year, excluding invertebrates). Hence, the potential welfare impact of fishing is huge.

The hard, scientific problem: studying complex food webs

Having solved the philosophical problem, we now need to solve the next question: what are the welfare consequences of fishing? This is a scientific problem, and is very difficult.

The aquatic food web is very complex. To simplify, consider a linear food chain: phytoplankton (1st trophic level), zooplankton such as copepods (2nd level), planktivorous fish such as anchovy (3rd level), piscivorous fish such as mackerel (4th level) and apex predators such as tuna (5th level). What happens if you catch fish at trophic level N? To simplify further, let’s only consider linear influences (no ecological side effects based on non-linear ecological processes). Catching fish at trophic level N results in a decrease of the population at level N (as well as decreases of populations at higher levels), which results in an increase of the population at level N-1, which again results in a decrease of the population at level N-2, and so on.

To understand the welfare effects of this food chain dynamics, there are three crucial questions: which populations of aquatic animals are sentient? How high are their levels of welfare? And how bad is dying (being killed and eaten) for those sentient animals?

Suppose first that phyto- and zooplankton (levels 1 and 2) are not sentient and hence have no well-being, whereas all fish at levels 3, 4 and 5 have a positive welfare. Suppose also that dying causes negligible suffering, which means killing fish only has a negative effect on aggregate welfare because of it lowering the population sizes (hence fewer animals with positive welfare, which decreases welfare according to a total utilitarian perspective). In that case, catching planktivorous fish (level 3) is bad, because welfare decreases. Planktivorous fish are innocent in the sense that they do not harm anyone else, because zooplankton was supposed to be non-sentient.

Piscivorous fish also catch planktivorous fish, which means they do something bad. Hence, if the welfare loss of catching piscivorous fish is relatively small (when killing causes negligible suffering and population sizes of piscivorous fish are much smaller than those of planktivorous fish, which is usually the case in food chains), catching piscivorous fish can become good: as the population of piscivorous fish decreases, there will be less predation on planktivorous fish. One piscivorous fish harms many other, innocent sentient beings: the planktivorous fish. A piscivorous fish is like a serial killer, and killing serial killers could reduce overall killing.

So if we catch piscivorous fish (level 4), the total amount of fish harm (which is proportional to the total amount of innocent sentient fish at level 3 captured by both humans and piscivorous fish) can decrease. By the same reasoning, catching apex predators (level 5) can become bad, because those apex predators catch many harmful, non-innocent piscivorous fish. The apex predators are the serial killers of the serial killers.

We conclude that if the welfare loss of killing fish is relatively small and zooplankton is non-sentient, catching fish at an odd trophic level is bad, whereas catching fish at an even level is good. However, this result completely turns around if zooplankton was sentient and had a positive well-being. In that case, planktivorous fish are no longer innocent: they harm a lot of sentient beings. Catching planktivorous fish becomes good because it saves many lives of innocent sentient beings (the zooplankton). Catching piscivorous fish becomes bad, catching apex predators becomes good.

However, this result again completely turns around if the well-being of a trophic level becomes net negative, i.e. when the animals have more and stronger negative than positive experiences. Lives with net negative welfare are generally not worth living: one would prefer non-existence above such a life. It is not known whether fish have net negative welfare, but the likelihood increases for many aquatic animals who have high reproduction rates. If one fish gives birth to thousands of offspring, only one of those offspring on average can survive long enough to reproduce. All the other offspring die prematurely. We can expect a positive correlation between the length of a life and the net welfare level of that life: the shorter the life, the more brutal it is, with predominantly experiences of suffering from hunger, parasites and diseases.

Now suppose the lives of zooplankton are in general not worth living: the vast majority of zooplankton animals have a net negative well-being (short lives with experiences of hunger and diseases). In that case it would be good to decrease the population of zooplankton. Planktivorous fish are doing a good job. Hence, catching piscivorous fish becomes good, because that increases the population of planktivorous fish and decreases the population of zooplankton.

In summary: if the suffering caused by killing is negligible, catching fish at an odd trophic level will be good if the lowest trophic level at which sentience occurs is even and if well-being is positive, or if the lowest trophic level at which sentience occurs is odd and if well-being is negative. It is bad otherwise. And the reverse is true for catching fish at an even trophic level.

Matters become even more complicated when the suffering caused by killing is not negligible. Catching fish causes suffering and hence decreases aggregate welfare. Catching predatory fish who catch fish might decrease overall fish killings and hence increase aggregate welfare, depending on the trophic levels. This is in line with the discussion above. But we also have to consider the fishing intensity.

The number of fish being killed in fishing, is the product of the fishing intensity (the probability of a fish being captured) and the fish population size. It is possible to increase the fishing intensity beyond the level of maximum yield: overfishing reduces the fish populations to such a degree, that there are almost no fish left to be captured. This counterintuitively means that a very high fishing intensity can result in very low fish captures. With a huge fishing fleet, it is possible that only a few fish die, because there are no fish left in the ocean. Hence, the welfare loss due to the suffering of captured fish is a non-linear function of the fishing intensity. As a lot of fish populations in the ocean are being overfished, decreasing the fishing intensity can increase the number of fish being killed.

Conclusion: should we stop fishing?

Given the fact that we catch huge amounts of fish, catching fish will be either very good or very bad, depending on the trophic level of the captured fish, the trophic levels that contain sentient animals, the positive or negative welfare status of the trophic levels, and the fishing intensity. The goodness switches if the trophic level of the captured fish is changed, if the lowest trophic level at which sentience occurs is changed, if the welfare level switches from positive to negative or if the fishing intensity switches from below to above the maximum yield level. It becomes extremely difficult to estimate the welfare impact of fishing. And it becomes even more complex in more realistic situations with non-linear aquatic food webs and non-linear ecological processes and trophic cascades (side-effects).

Given the fact that we catch many fish, knowing the sentience and welfare levels of aquatic animals and the full dynamics of aquatic food webs becomes very important. A lot is at stake.

In the appendix of this document I describe a purely theoretical approach to the problem of the welfare impact of fishing. I simulated coundefinedmplex food webs with and without fishing, and calculated a welfare function that measures the total sum of welfare levels of the sentient animals minus the welfare losses due to dying. The results are very sensitive to the choice of parameters in the theoretical food web model and the welfare function. Overall, the conclusion is that fishing is slightly more likely to decrease aggregate fish welfare at low fishing intensities, with the exception of situations where most fish populations have negative net welfare levels and fish mortality is not the dominant contributor of welfare loss. Let’s say that the probability that fishing decreases aggregate welfare, is something like 51%.

What should we do with fishing as long as the important scientific knowledge is lacking? We are in a situation of risk, where we risk doing a lot of bad when fishing, but we may also do a lot of good. If a lot is at stake, most people become risk averse and prefer the status quo of non-intervention. That is what we would choose when humans instead of fish were involved. In order to avoid speciesist arbitrariness, we can ask ourselves the question what we would do if all aquatic animals were large and small swimming humans (making up a complete food web, with cannibalistic humans). Then we would not simply go fishing humans, because fishing would be too bold. We would rather do scientific research and study the situation more carefully before we intervene. Furthermore, we have one certainty: catching fish always causes some harm to the captured fish. So fishing implies a certain welfare loss plus an uncertain very high positive or negative impact on welfare. Even if that means that the probability that fishing has a negative effect is only 51%, i.e. slightly more likely than flipping a coin, we would abstain from fishing, because so many fish are involved. Fishing is impermissible, until we have robust scientific evidence that fishing is the only means to improve well-being and decrease harm.

The above considerations are also relevant when it comes to problem prioritization. The uncertainty about the welfare effects of fishing means that priority could be given to abolishing aquaculture and livestock farming first instead of abolishing fishing first. Even if the number of livestock animals killed per year is an order of magnitude lower than the number of wild fish captured per year, the negative welfare impact of aquaculture and livestock farming is more certain (as their food chains are simpler) than the welfare impact of fishing. The animals in captivity have most likely net negative welfare levels, which means it is very likely that breeding and slaughtering such animals decreases aggregate welfare. Hence, priority could be given to decreasing livestock farming and aquaculture. Besides, as a lot of livestock animals and farmed fish eat fish meal, decreasing livestock farming also decreases fishing.

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

Asymmetric altruism

In effective altruism, we have to prioritize the most effective ways to do good. But there are different notions of altruism that influence our prioritization. Altruism has to do with helping others. But the tricky question becomes: helping who exactly? And what is helping? I will argue that we have to make a distinction between positive versus negative altruism, and that this distinction becomes important in effective altruistic prioritization.

To start, consider a person who is about to undergo a surgical operation. At time 1, before the operation, the person is fully conscious and has mental state P1. We can choose between two options A and B. At time 2, during the operation, the person can either have anesthesia (option A), or not (option B). This can be described with two possible worlds. In the world A where we choose for the anesthesia, the anesthetized person is unconscious, having an empty mental state P2A=0 (i.e. no subjective experiences and preferences). In the second world, the patient does not get the anesthesia and will be in extreme agony, with mental state P2B. Altruistically speaking, it is better to choose option A, because this is helping the patient. Giving the anesthesia is something the patient wants.

In the case of the surgical operation, it is clear who is being helped. We can consider the mental states P1, P2A and P2B as belonging to the same person, because those mental states are related to each other. In particular, the person at time 1 with mental state P1 is concerned about his/her own future and hence identifies him/herself with the future mental states P2A and P2B. Similarly, the person with mental state P2B can acknowledge that he/she is the same person as P1 as well as P2A. P2A is basically P2B’s alter ego in the other possible world. A slightly tricky issue arises when we consider P2A, who is unconscious and hence not able to feel a personal identity with neither P1 nor P2A. P2A has no beliefs, and hence no belief that he/she is the same person as P1. Still, given the beliefs of P1 and P2B, we can consider P1, P2A and P2B as the same person, and the anesthesia helps that person.

Is veganism altruistic for animals?

Giving anesthesia to the patient is a clear example of altruism: it helps the other. But what about veganism? Animal farming causes animal suffering. Almost all farm animals have very negative experiences. We can avoid this suffering, by eating vegan. But that means those farm animals would not be born and hence not exist.

Consider at time 1 a bunch of atoms and molecules floating around. This group of molecules has an empty mental state P1=0. Then we have a choice to eat vegan (option A) or eat meat (option B). Option A means those atoms will keep floating around, having again an empty mental state P2A=0. Only in option B will those atoms rearrange themselves to create a mental state P2B in an animal brain. P2B has unwanted negative experiences.

If we choose option A, are we helping the animal? Which animal? The animal does not exist in option A: the mental state P2A was empty. P1 also is an empty mental state, which means no identification with neither P2A nor P2B. And it is very unlikely that animal P2B can identify him/herself with the non-existing animals (i.e. the bunch of molecules) P1 and P2A. Hence, P1, P2A and P2B cannot be considered as the same person. So, are we really helping an animal when we choose a situation where the animal does not exist?

Is saving the future altruistic?

Next, we can consider existential risks: situations that lead to the extinction of intelligent or sentient life. At time 1, future generations are not born yet, and hence they can be represented by a bunch of atoms having empty mental states P1=0. Then we can choose between two options: either we do not avoid the existential catastrophe, which means those atoms will have a future empty state P2A=0. Or we prevent the extinction, which means those atoms will rearrange themselves and future people will be born, having mental states P2B.

If we choose option B, are we helping those future people? Yes, because those people will exist in world B. But if we choose option A, are we harming those people? No, because those people will never exist in world A.

Positive versus negative altruism

It is time to consider two kinds of altruism. Positive altruism means: choosing what someone else wants. Negative altruism, on the other hand, means: not choosing what someone else does not want. This is a bit related to the two versions of the golden rule: “Treat others in ways that you want to be treated”, versus “Do not treat others in ways that you do not want to be treated.”

By choosing the anesthesia, we are altruistic in both positive and negative senses. We choose what person P1 wants (the anesthesia), and we do not choose what person P2B does not want (the suffering). By choosing veganism, we are only being negatively altruistic: we do not choose what person P2B does not want. And by choosing to avoid the existential risk, we are only being positively altruistic: we choose what people with mental states P2B want.

When we have to prioritize between different ways to do good, the question is whether double altruism (i.e. both positive and negative altruism) is more valuable than single altruism, and whether single positive altruism is more valuable than single negative altruism. How can we tell which is most important?

It can be argued that double altruism is twice as good as single altruism, in the sense that double altruism takes into account the preferences of two mental states P1 and P2B, whereas single altruism only considers P2B. Hence, when choosing between double and single altruism, double altruism can be prioritized (all else equal, hence assuming the preferences or wants are equally strong in the different situations).

But suppose we have to choose between single positive and single negative altruism. For example: should we prioritize veganism or safeguarding the future (assuming that an equal amount of animals and potential future beings are involved, with equally strong preferences for option B)? We see a lot of asymmetries in ethics (e.g. killing someone is worse than not saving someone, and causing the existence of someone who constantly suffers is always bad whereas causing the existence of someone who is always happy is not always good). Some asymmetries can be defended (see e.g. here), and I tend to believe that negative altruism is more valuable than positive altruism. If negative altruism is considered very important, then veganism becomes more important.

In theory, we can solve this issue by being altruistic: let the others decide. In particular: ask the farm animals and the future generations whether they prioritize negative altruism above positive altruism. But that is of course unfeasible. How to weigh positive versus negative altruism is a question I will leave for further investigations.

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

Black Lives Matter: racism at unexpected places

Recently, in the wake of the Black Lives Matter protests, I stumbled upon a graph about interracial violent crime incidents in the US in 2018. That graph was spread by conservative, right-wing and racist people. The original source, containing the underlying data of the graph, was the US Department of Justice. The data were surprising to me, and demonstrate how complex the issue of racism is. I am not an expert on this topic, I am not a criminologist or sociologist and haven’t read many scientific studies on this topic. Nevertheless, I think this topic is a good exercise in how views can be influenced by briefly looking at some studies, statistics and graphs. This example of interracial violence shows that there are plenty of hypotheses and surprising conclusions that one could draw from only a small graph. It is also a good exercise in rational, critical thinking.

As I will show, police violence and violent crime statistics do reveal racism at several places, but not at the places that I or many other people would expect. This implies that the causes of racism could be different or more complex than what one would naively guess. At least, the situation is more complex than I would have guessed.  

To be clear, I focus on statistical racism: the shares of crime rates by ethnicity compared to what one would expect based on population shares. For example, 60% of the US population are non-Hispanic whites, 13% are black and 17% are Hispanic. Hence, without statistical racism, one would expect 13% of victims and 13% of offenders of crimes to be black, and 1,7% (i.e. 13% times 13%) of crimes to be with black offenders on black victims. Statistical racism could be explained by, but should not be confused with e.g. ideological racism, institutional racism or systemic racism. If people have racist beliefs (ideological racism), that could influence the crime statistics and generate statistical racism, but statistical racism is possible even if people do not have explicit racist ideologies or implicit (unconscious) racist attitudes.

Let’s move to the graph that was spread by the racist people. That graph selected only the interracial violent crime rates of three ethnic groups: whites, blacks and Hispanics. Here I present the full graph, that also shows the intraracial crime rates of those three ethnic groups. The left, blue bars represent the expected number of crimes if there was no statistical racism. The right, orange bars are the real crime rates.

Some findings.

  • There are more interracial crimes by black offenders on white victims (‘black on white’) than the reverse (‘white on black’), and these crime rates are far from what one would expect based on population shares. For example the crimes by white offenders on black victims is 7 times lower than black on white, and 10 times lower than what one would expect. This was the point made by the right-wing, conservative and racist people spreading the graph. One could conclude that, when it comes to interracial violence, there is no sign of white supremacy or antiblack racism by white people. White and Hispanic people are the least dangerous people for other ethnic groups.
  • Blacks are statistically speaking more violent than whites or Hispanics. The total number of crimes by black offenders (the sum of the three left orange bars) is almost twice as much as what one would expect based on population shares (the three left blue bars). Whites and Hispanics are 10% less violent than what one would expect. The higher crime rates by black people could be due to misreporting (e.g. when people are more likely to report a crime when the offender is black) or a racist bias (e.g. police officers misreporting the ethnicity of the crime suspects). However, mere misreporting or bias could not explain the discrepancy between crime rates for different ethnic group victims. Looking at the graph: why would the racist misreporting be so much larger when the victims are black (look at the discrepancy between expected and real ‘black on black’ and ‘white on black’ rates), compared to when the victims are white? The ratio of real versus expected crime numbers for white on white is almost the same as for black on white, whereas those ratios are extremely different for white on black and black on black. Hence, something more than mere misreporting must be going on. I don’t expect that correcting for misreporting would significantly change the conclusions we can draw.
  • Intraracial violence (within the same ethnic group) is higher than interracial violence (between groups), and this is true for all ethnic groups. For example crimes by black offenders on black victims are higher than what one would expect. There could be several reasons for this, such as the formation of ethnic communities. For example, black people tend to live in neighborhoods with many other black people, hence encountering more black than white people in their daily lives, and hence encountering more black offenders and victims.
  • Black intraracial violence (‘black on black’) is almost 5 times higher than what one would expect, whereas white intraracial violence is only 10% higher. I don’t know why this is the case, but based on my economics knowledge, I can formulate some hypotheses. For example the black community can have relatively high levels of intragroup competition. The labor force of a minority group can be relatively more homogeneous, for example comprising of low-skilled workers, which means jobs are less complementary, which means more intragroup competition on the labor market, which means lower wages and incomes, which means dire economic situations, which means more resort to intragroup violent crimes against people considered as competitors.  
  • The total crimes on minority groups (black people and Hispanics) is 17% to 19% lower than what one would expect, whereas the total crimes on whites is 9% higher than expected. This means minority groups are relatively safer against crimes. Even with a very high black on black crime rate, black people are less likely to be victims of crimes compared to a situation without statistical racism. (This conclusion may no longer be true if crimes against blacks happen to be underreported in the data. Perhaps the number of white on black crimes in the data is underestimated, but then one would expect that the number of black on black crimes is also an underestimate, which means in reality the intraracial black crime rate is even higher than the already very high level presented in the data.)

Now we can look at a measure of statistical anti-black racism. I define this as a ratio of two ratios. The first ratio is the real number of crimes with black victims to the real total number of crimes (i.e. summed over all victims). The second ratio is the expected number of crimes with black victims to the expected total number of crimes (expected based on population shares). When this ratio of ratios is larger than 1, there is statistical racism. This statistical anti-black racism measure corrects for e.g. the facts that some ethnic groups are larger than others or more violent than others. Here we see really surprising results.

  • The statistical anti-black racism by the whole population is 0,75, i.e. lower than 1. Hence, the whole population does not have an anti-black racism bias when it comes to violent crimes. This can also be compared with the statistical anti-white racism by the whole population, which equals 1,15. This value is higher than one, so in general there is rather an anti-white bias. Whites are relatively worse-off than blacks when it comes to statistical racism in violent crimes.
  • The statistical anti-black racism by whites (i.e. considering white offenders) is even much lower: 0,17 (consistent with a higher than 1 anti-white racism by white people of 1,33).
  • Most surprisingly, the statistical anti-black racism by blacks is 2,7. This is extremely high. It is also higher than the statistical anti-white racism by whites (1,33). This means that the abovementioned ethnic community effect (blacks living among blacks, whites living among whites) cannot explain this anti-black racism by black people. When it comes to violent crimes, the real racism against black people comes from black people. I suspect that there is a socio-economic cause underlying these high levels of statistical anti-black racism by blacks and black intraracial violence. The lower socio-economic status of black people could partially explain the statistical racism, and could itself be the result of structural racism (e.g. structural racism on the labor market), but below I present another interesting hypothesis that is not related to racism (or only very indirectly).

The Black Lives Matter movement also focuses on police killings. Here the data and studies are very clear: of the people shot to death by police officers in the US in 2019, 31% were black (totaling 235 black people killed), 48% were white, 21% were Hispanic (neglecting the other ethnic groups for simplicity). 31% is off course higher than the population share of blacks (13%), but also higher than the share of violent crimes by blacks (25%, considering only crimes by blacks, whites and Hispanics). Even if black people commit more violent crimes, the police force is even more violent against blacks than that.

The statistical anti-black racism of the police force cannot be explained by the ethnic composition of the police force: roughly 13% of police officers are black, which equals the population share of blacks in the US. So the next question is whether white police officers are more anti-black racist than black officers. I could not find convincing data for a higher anti-black racism among white officers. One study says that “white officers appear to be no more likely to use lethal force against minorities than nonwhite officers”. Another study and aftermath discussion indicates that… the matter is very complicated.

At this moment, I personally don’t think that white officers are significantly more racist. The statistical racism of the police force is not at the level of the individual officers. It is more hidden and structural. One hypothesis (that I didn’t check) that could explain the statistical anti-black racism, is that the police patrols more in black neighborhoods, for example because they expect higher crime rates there. Or the police can target more black than white suspects. Also (and perhaps especially so) black police officers are send to the black communities. That means police officers (including black officers) are more likely being confronted with black offenders and black suspects on the streets.

The good news is: general crime rates, and rates of police killings are declining. People and police officers, be they black or white or Hispanic, become less violent. There are some interesting hypotheses that could explain not only this declining trend in violence and crimes, but also the higher crime rate by black people, why their decline is lagging behind the decline of the crime rate by white people. One of my favorite hypotheses is the lead-crime hypothesis (also discussed here and here). The idea is that black people live at places with higher levels of lead pollution. Lead uptake in the body by young children can cause a decrease in IQ and learning abilities (hence a decrease in socio-economic status later in life) and a decrease in impulse control. This makes black people more vulnerable to violent crimes. Lead pollution is decreasing, so we see a decreasing crime rate by black people.

The lead-crime hypothesis is an interesting example of why we might need to look for far deeper, less trivial causes of violence and crimes than the usual antiracism rhetoric and ‘easy’ explanations (e.g. about white supremacy or white privilege; these are notions that could be useful in other contexts, but not in the context of violent crimes). If true, hypotheses like the lead-crime hypothesis offer much clearer and effective solutions to decrease crime rates and hence police killings.

What other solutions could be effective? A possible explanation for the high anti-black killing rate by police officers, is the higher police patrol rate in black communities. Perhaps we should send police forces more to white neighborhoods instead of black neighborhoods? However, this might increase the crime rate in black neighborhoods to even higher levels than they already are. (I am not familiar with the scientific literature, so I don’t know whether this is true.) A better option might be to have more police officers. When there are only a few officers, they have to work long shifts, hence they become more tired, more easily frustrated and less able to deal with conflictual situations in non-violent ways. Campaign Zero proposes more measures to reduce police violence.

Perhaps most importantly, in terms of effectiveness: we need a criminal justice reform, especially in the US. Chloe Cockburn and the Open Philanthropy make some interesting recommendations (see also here and here for discussions about effective charities to reduce systemic racial injustice and police violence).

Geplaatst in Artikels, English texts | Tags: , | 6 reacties

What anti-vaxxers truly believe (or how reframing improves critical thinking)

Vaccines save millions of lives every year. However, the anti-vaccination movement causes a decline in vaccination rates, which results in extra diseases and deaths. Anti-vaxxers are people who have doubts about the safety and effectiveness of vaccines that are strongly recommended by health organizations. As a consequence, they want to refuse vaccinating their children against diseases such as polio and measles.

The anti-vaccination movement is harmful, so we need effective strategies to fight this movement. To do this, it might be interesting to understand what anti-vaxxers truly believe, by reframing the issue of vaccination. Reframing is an important philosophical technique that helps to detect fallacies or inconsistencies in beliefs. Here I present a reframing, using a thought-experiment, to understand exactly what anti-vaxxers stand for.

Suppose we have a vaccine, but unlike a classic injection vaccine, this is a self-producing vaccine that can spread through the air. Injection vaccines are produced in labs, whereas the self-producing vaccine is produced in the bodies of vaccinated people. Those bodies become vaccine factories. The exhaled air of the vaccinated people carries invisible vaccine particles to other people, which means that they are therefore automatically vaccinated. So you get this vaccine by inhalation instead of injection. With this vaccine in circulation, it is almost impossible to avoid vaccination, because then you should avoid all contact with vaccinated persons, wear gas masks or stop breathing.

The interesting property of this vaccine is that it is costless for health authorities: no more syringes and doctors required to administer the vaccine. Another benefit, in the eyes of people who dislike pharmaceutical companies, is that companies do not earn anything from this vaccine, because the vaccine produces itself using the bodies of vaccinated people. For people who don’t trust large pharmaceutical corporations: the vaccine was not designed by ‘Big Pharma’. In fact, no commercial interests were involved in the development of the vaccine.

The self-producing vaccine has some disadvantages, though. First, the vaccine has not undergone safety testing and evaluation by independent health scientists and regulatory agencies. Second, the vaccine production lacks transparency: it is not sold with a package leaflet that describes possible side-effects and safety instructions. No-one knows how the vaccine was initially developed. Third, the dose given to a person cannot be controlled as with an injection vaccine. That means some people get very high doses of the vaccine, even without knowing it. Fourth, and most importantly, the vaccine contains two dangerous additives, i.e. chemical ingredients that improve the self-production process. The first additive is a protein that allows for the vaccine particles to penetrate deeply into the cells of the body. The second additive is ribonucleic acid. This is a group of chemicals that can alter the genetic expression of cells. The cells become genetically modified to produce more vaccine particles. Unfortunately, this results in cell death. As a consequence, the self-producing vaccine can have some very serious side-effects. Many vaccinated people become seriously ill and can even die from the vaccination.

If you had to choose, which vaccine do you prefer: the injection vaccine or the self-producing inhalation vaccine? This is not merely a theoretical thought-experiment. The reader has already understood that the self-producing vaccine actually exists: it is called ‘measles’ (or ‘polio’, ‘mumps’,…).

Now there are people, called anti-vaxxers, who oppose the traditional injection vaccines. As a consequence, they are in favor of the self-producing vaccine, even if the self-producing vaccine turns out to be more harmful to health. Those people have several reasons to oppose the injection vaccines.

First, they absolutely do not want large pharmaceutical companies to earn money. As traditional injection vaccines have to be produced by pharmaceutical companies, they oppose those injection vaccines and turn to the self-producing vaccines. Hence, those anti-vaxxers think it is more important that some companies earn nothing than that people are healthy.

Second, the anti-vaxxers have a restrictive notion of freedom and autonomy. They believe that a person’s freedom or autonomy is violated when a vaccination is intentionally forced upon that person by people (doctors, ministers of health,…) who care about the health of that person. Here the intention is important. As people cannot escape the self-producing vaccine, this vaccine is forced upon them, but not intentionally. Nature (in particular the laws of chemistry) forces the vaccine upon them, and nature does not have intentions. Nature doesn’t care about the health of people like doctors do. Hence, the anti-vaxxers think intention is more important than health: it is more important to avoid intentional restrictions of freedom than unintentional restrictions of freedom, even when the latter cause more harm.

The anti-vaxxers claim to be against the injection vaccine because they believe that the vaccine causes diseases (such as autism), that the production lacks transparency, that the vaccine lacks safety testing, that certain additives are too dangerous,… However, as they chose the self-producing vaccine that has deadly side-effects, has no package leaflet, was not tested on safety, and definitely includes dangerous additives (that can deeply penetrate cells and genetically modify them), all those arguments are invalid.

The expression ‘self-producing vaccine’ was merely a reframing, intended to make clear that the beliefs of anti-vaxxers are inconsistent. The above thought-experiment is a good exercise in critical thinking. Many ethical issues can be tackled by reframing them.

Consider another belief of anti-vaxxers: that injection vaccines are harmful because they weaken our immune system (e.g. the immune system becomes too lazy not having to deal anymore with the real viruses). The anti-vaxxers believe the self-producing vaccine strengthens the immune system (at least for those people who survive the disease). As increasing the injection vaccination rate decreases the self-producing vaccination rate, the anti-vaxxers oppose the injection vaccines. But if decreasing the self-producing vaccination rate is bad (in terms of weakening the immune system), what about increasing the self-producing vaccination rate? Would creating new self-producing vaccines (i.e. new infectious viral diseases) be good for our immune system? Here I reframed the issue: instead of considering a decrease, we can consider an increase. As anti-vaxxers are against creating new diseases, neither decreasing nor increasing the self-producing vaccination rate is a good idea, according to anti-vaxxers. Hence, they believe that the self-producing vaccination rate (at a zero injection vaccination rate) is the optimal rate, but they cannot explain why this should be the case. This is an example of a status quo bias, tackled by a reframing technique called the ‘reversal test’.

It appears that the anti-vaccination gained strength during the Covid-19 crisis. Interestingly, during this crisis, many anti-vaxxers were against protective measures such as quarantines, lockdowns and obligations to wear facemasks. As these measures do not generate profits for pharmaceutical companies, the distrust in Big Pharma is not at play here. On the contrary: many anti-vaxxers are in favor of using chloroquine against Covid-19, believing that this drug has an antiviral effect. Here again we see a dangerous twist: at the time when anti-vaxxers proposed chloroquine, there was no evidence that chloroquine is safe and effective for treatment of covid-19 patients. To be clear: chloroquine is a drug that is produced by pharmaceutical companies. And if this drug will be used by thousands of covid-19 patients, it will be produced and sold at a large scale, which can only be done by Big Pharma. It again demonstrates that Big Pharma is not the real issue for those anti-vaxxers. The same goes for the worry about conflicts of interest. Anti-vaxxers distrust scientists who claim that vaccines are safe and effective, because they suspect financial conflicts of interests between those scientists and pharmaceutical companies. However, researchers with a high reputation in the anti-vaccination movement, such as Andrew Wakefield and Romain Gherardi, had some serious, undisclosed conflicts of interests. Hence, conflicts of interest is also not the real issue for anti-vaxxers.

Anti-vaxxers claim to value human health, children’s lives and personal freedoms. However, their opposition to effective injection vaccines is irrational and counterproductive. By choosing a more dangerous self-producing vaccine that is enforced upon everyone, they cause more deaths and loss of health, welfare and freedoms. Not only is the self-producing vaccine forced upon everyone, but patients also lose all their freedom when they die. The anti-vaccination movement is one of the most striking examples of irrationality, where good people, with the right moral values, can be turned towards bad choices, with harmful consequences.

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

Our final strategy for meat abolition?

After a long reflection, here are my thoughts on how to most effectively decrease or end the biggest kind of human-caused suffering: animal farming (and fishing). I argue that the best strategy for animal rights advocates to abolish meat, is supporting open access scientific research and development of cellular agriculture.

How can we eliminate the animal product market? There are two interlinked markets: for animal products and animal-free alternatives. We need to shift the system from the market of animal products to the market of animal-free products, either by pushing the system away from the first market, or pulling it towards the second market. Like pulling on one side of a rope is more effective than pushing on the other side, I will argue that a pulling strategy is more effective than a pushing strategy.

Both markets have a demand side of consumers and a supply side of producers. Hence, we can distinguish four (two by two) strategies. The two push strategies influence the animal product market: they make animal products less attractive, either by decreasing demand (a leftward shift of the demand curve) or increasing production costs (an upward shift of the supply curve). The pull strategies make animal-free products more attractive, either by increasing its demand (a rightward shift of the demand curve) or decreasing its production costs (a downward shift of the supply curve). Hence, we can analyze the effectiveness of these two strategies at the two market sides.

Demand (upward slope) and supply (downward slope) curves for the markets of animal products and animal-free alternatives. Push strategies act on the market for animal products, decreasing demand or increasing prpduction costs. Pull strategies act on the market for alternatives.

Demand side strategies

Let’s start with the demand side. Many animal advocates focus on individual behavior change with consumerist vegan and vegetarian outreach campaigns. Despite all the outreach campaigns of the past five decades of activism, the number of vegetarians and vegans did not increase very much and stays below 10% of the population.

The demand side push strategy consists of presenting moral arguments against animal production, informing people about the horrors of animal farming, showing undercover investigations of factory farming, producing documentaries or talking about human health costs and environmental problems of animal products. This strategy often faces a pushback from the animal industry: they increase their marketing campaigns to counteract the negative information spread by animal rights activists. It becomes a kind of arm-wrestling: the harder you push, the harder the opponent pushes back. At best, the strategy makes animal products a bit more expensive, because the industry needs to pay for its extra marketing campaigns, and this can decrease the demand a little.

The demand side pull strategy consists in making veganism more attractive by organizing vegan cooking workshops and vegan community events, distributing vegan meal recipes or letting people taste traditional vegan products. This behavior change strategy has its limits as well, because it gives the impression that eating vegan requires a change in behavior (having to learn new things such as vegan recipes, making new choices) and a change in identity (considering oneself as a vegan, adopting a new, vegan ideology). People are reluctant to change their behavior and identity, so this pull strategy was not sufficient to convince the broad public.

Supply side strategies

Moving towards the supply side, animal rights organizations did some effective campaigns to change the system. The supply side push strategy consists in attacking the industry. Blockades, supply chain disruptions, governmental regulation, meat taxation, are all examples to make the production of animal products more costly (shifting the supply curve of animal products upwards). This can be moderately effective in two situations: to prevent further expansion of the industry and to prohibit a part of the industry.

First, the further expansion of the industry can be prevented, for example by a public non-violent direct action campaign that prevents the construction of a new slaughterhouse in a neighborhood. This can be effective, because preventing something that does not yet exist is easier than breaking down something that already exists and already has strong vested interests. However, the effectiveness is limited, because as long as demand is high, the slaughterhouse can be built elsewhere. At best, the strategy makes animal products a bit more expensive, because the industry needs to build its infrastructure at more expensive places.

Second, with the help of government regulation, a part of the industry can be prohibited. For example, animal rights organizations were successful in officially banning sports that involved animal cruelty and circuses that used wild animals in many countries. This can be effective, because influencing a (local or national) government is easier than directly attacking the industry. However, the effectiveness is limited, because the scope of the problem is relatively small: there were fewer wild animals in all circuses combined than livestock animals on a single, large factory farm.

The supply side push strategy works on the market for animal products and therefore also faces a pushback effect from the industry, reducing the overall effectiveness. In general, the past five decades did not show many big results with this strategy.

That leaves us with one final strategy, that is not much attempted yet for animal farming: the supply side pull strategy, that works on the market of animal-free alternatives. Basically, it comes down to developing new food technologies, such as cellular agriculture that can produce cell-based meat. It decreases the production costs of animal-free products that are equal or better in quality than animal products. This final strategy can be very effective.

New technologies that replaced animals

In the past, we have seen several examples of animals being replaced by new technologies, often without much resistance (almost no pushback from the industry or from adversarial government legislators) and even without animal rights campaigning. Some examples:

  1. draft horses for carriage were replaced by cars,
  2. oxen for plowing were replaced by tractors,
  3. whale oil was replaced by kerosene,
  4. messenger pigeons were replaced by telephones and telegraphs,
  5. wool was largely replaced by synthetic fibers such as nylon (this is one of the most important reasons why sheep agriculture in the US declined by almost 90%),
  6. beeswax for candles was replaced by light bulbs,
  7. pig and cow insulin for diabetes patients was replaced by biosynthetic human insulin from recombinant-DNA yeast,
  8. monoclinal antibodies from animals were replaced by antibodies from cultured cells,
  9. rabbit skin tests were replaced by cultured human skin cells (an example of a replacement of animal experiments by animal-free alternatives),
  10. movie animals are being replaced more and more by CGI digital animals.

Note that in many of those examples, a sector that used thousands or millions of animals was completely or almost completely abolished within only a few decades. The general reason behind these drastic transitions is that the animal-free new technologies were simply better in terms of quality, usability, reliability and production costs, such that market forces were sufficient to shift the economy. Only limited pressure from the public or the government was needed.

We can expect that gradually all animal technologies will be replaced by animal-free technologies. This can be understood by looking at the technology space: the abstract space of all physically possible technologies that we could ever invent. This is a huge space, and only a small island in this space consist of technologies that use animals. At this moment, animal farming is the technology that uses the most animals. Animal farming is on the island of animal technologies.

Our human history can be understood as an exploration of technology space. With our first technological inventions, we were dropped in technology space. Because we met a lot of animals in our daily lives, as it happens, we landed close to the island of animal technologies. Therefore, we explored this island and hence a large part of our first technological inventions involved animals. That is why we started to use more and more animals, inventing new ways to use them. But as we explore technology space further, we are expanding the scope far beyond the small island of animal technologies. With this exploration, the probability that we discover animal-free technologies that are in all aspects better than animal technologies, increases.

New food technologies allow for the production of animal-free foods that become better in all aspects (tastier, safer, healthier, cheaper, environmentally friendlier,…) than animal foods. Hence, the final strategy of pulling the supply side towards animal-free food production, is likely to be very effective. Especially the replacement of animal meat (slaughtered meat) with animal-free meat is important. When it comes to animal-free meat, we can make a distinction between plant-based meat and cell-based meat. A prospective timeline estimates that this decade, we will see large improvements in new plant-based versions of processed animal products such as burgers and sausages. In the 2030’s, we will see processed cell-based meat products, animal-free dairy and eggs and pet food on the market. And by 2050, unprocessed, whole tissue cell-based meat is expected to enter the market.

Animal rights activists can help speed up this process of animal-free meat entering the market, by supporting research and development, assisting the marketing and legislative process, and influencing the distribution networks and supply chains for the new food technologies. Two organizations are of prime importance in this area: the Good Food Institute and New Harvest.

The case for cell-based meat support

When choosing between supporting R&D of plant-based meat versus cell-based meat, the latter could be more effective. First of all, new start-up companies are already bringing new plant-based meats to the market, backed by large investors (see for example the very successful stock market launch of Beyond Meat in 2019). This means there is already a lot of investment in this area of plant-based meats. Cell-based meat is not yet on the market and there are not yet cell-based meat companies selling shares on the stock market or marketing cell-based meats. Hence cell-based meat is more benefitted by extra support that enhances market introduction.

Second, the timeline to influence the developments of new plant-based meat is short: they are expected to capture a large market share of the meat sector already this decade. Whole tissue cell-based meat, on the other hand, is expected to enter the market over a few decades. When a solution is further away in the future, efforts to shorten the timeline become more important. When a solution takes ten times longer to develop, speeding up developments with 1% has ten times more future impact (as explained with a graph here).

Third, the demand of plant-based meat is likely more limited. It might remain difficult to persuade die-hard meat-eaters to switch to plant-based alternatives, because those alternatives can only imitate processed animal meat. Whole tissue cell-based meat, on the other hand, would taste and feel just like unprocessed animal meat. If cell-based meat becomes indistinguishable from animal meat, but cheaper than animal meat, it is likely to attract more meat-eaters. Furthermore, cell-based meat is likely more appropriate for non-human carnivorous consumers as well, such as pets and animals in wildlife rescue centers. In theory, in the far future, demand for cell-based meat could extend to all carnivorous wild animals.

Fourth, private companies are already investing in applied R&D for plant-based meat, whereas cell-based meat still requires much more initial or fundamental R&D that is more neglected by private companies. Hence, donors have more opportunities for financially supporting this more fundamental research.

Fifth, there is a risk that cell-based meat awaits the same fate as GMOs. GMOs can improve the food system, but they received a public backlash, largely due to GMOs becoming associated with corporate secrecy and large corporations controlling intellectual property. When a large corporation would appropriate all intellectual property for the production of cell-based meat, it gains a monopoly power, which means that cell-based meat will be sold at high prices and other companies cannot easily develop new cell-based meats. This reduces supply and demand of cell-based meats. Especially cell-based meat technologies are vulnerable for patenting. To avoid patenting of cell-based meat by large corporations as much as possible, we need to increase support for open access research into cellular agriculture as much as possible. That is why an organization like New Harvest is so important.

As cell-based meat will be better in all aspects than animal meat, it is likely that it can replace animal meat just like the ten technology examples given above replaced animals. Nevertheless, it is possible to give counterexamples of better technologies that were not able to replace worse technologies. The best counterexample is probably bottled water: tap water is equally healthy but more than hundred times cheaper and better for the environment, as well as easier in use (no need to buy and carry heavy bottles from a shop). Still, a lot of consumers buy a lot of bottled water. The crucial question is whether cell-based meat is comparable to tap water, or rather to a cheaper and higher quality bottled water. The reason why people still buy bottled water, is mostly marketing based. Bottled water is more marketed than tap water, because companies want to sell it. We can expect that companies selling cell-based meat will advertise their cell-based meat. So cell-based meat should rather be compared with cheaper and higher quality bottled water than with tap water. And when cell-based meat is better for public health and the environment, governments can more easily prohibit animal meat, just like governments can easily prohibit the most environmentally destructive type of bottled water packaging.

Conclusion: towards veganmodernism

In the past, animal rights advocates tried several strategies to decrease or end the biggest kind of human-caused suffering, namely animal farming. All of them failed so far, but one strategy is not yet attempted much: a supply side pull strategy towards animal-free food production. Within this strategy, the most effective tactic could be the financial support for open access scientific research and development of cell-based meat. Hence, animal rights activists and advocates can shift their strategies and tactics: instead of spending time and money doing traditional behavioral change and corporate pressure campaigns, they can look for opportunities to raise, earn and donate more money to an organization like New Harvest. This implies the animal rights movement should shift more towards veganmodernism, where technology is the solution, just like the environmental movement gave birth to ecomodernism.

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

Het meest verwaarloosde leed

Di artikel verscheen ook op Kwintessens – Humanistisch verbond.

Het is een mooie lenteavond; tijd voor een filosofische wandeling in het bos. Onderweg stellen we ons de vraag: wat zijn nu de allerbelangrijkste humanistische waarden? We komen uit op een shortlist, met bovenaan: het vermijden van ongewenst, onnodig, extreem leed. Goed, als dat belangrijk is, dan volgt een tweede, uitdagendere vraag: wat is de meest verwaarloosde vorm van leed, waar we te weinig aandacht voor hebben? Bij zo’n aangename lentewandeling in de natuur is het natuurlijk niet eenvoudig om na te gaan welk leed we het meest over het hoofd zien, want het leed in de wereld lijkt dan ver weg. Of net niet?

We genieten van een fluitende vogel. De vogels in dit bos lijken wel tevreden, net als wij. Maar we zijn misleid. We weten dat de moeder van die vogel pakweg een tiental eieren heeft gelegd. Die vogel heeft dus tien broertjes en zusjes. Waar zijn die dan? Die zijn gestorven, want als alle vogels zouden overleven en zich voortplanten, dan hebben we elke generatie tien keer meer vogels. Dus voor elke fluitende vogel zijn er tientallen kuikentjes die we niet te zien krijgen omdat ze op jonge leeftijd gestorven zijn. Verreweg de meeste pasgeboren dieren in de natuur hebben korte levens met veel negatieve ervaringen, gevolgd door een pijnlijke doodstrijd. Als jij zou reïncarneren tot een willekeurig dier, zul je zeer waarschijnlijk een kort en naar leven hebben, met honger, dorst, vrieskou, gevechten, ziektes, ongevallen, parasieten, roofdieraanvallen enzovoort. De dieren die we in dit bos zien, zijn de geluksvogels, en we maken een denkfout als we daaruit concluderen dat de meeste dieren wel gelukkig en gezond zijn. De vele pechvogels zagen we niet, want die zijn gestorven en opgegeten.

In extreme armoede sterft ongeveer een kind op zes. Hoe erg moet het leed van dieren in de natuur dan niet zijn, waar meer dan negen op de tien pasgeborenen vroegtijdig sterft? Sommigen zijn bezorgd over menselijke overbevolking: een sterke stijging van de bevolking die leidt tot milieuproblemen en een plotse populatiecrash. Maar de menselijke populatie is waarschijnlijk de eerste en tot nu enige dierenpopulatie die overbevolking net vermijdt: zowat elk geboren kind heeft een lange levensverwachting en krijgt gemiddeld een eigen kind dat ook weer volwassen kan worden. In de natuur zien we daarentegen elk jaar een overpopulatiecrisis. De bevolking van wilde dieren groeit plots met meer dan een factor tien, en omdat de natuur dat niet aankan, sterven meer dan 90% van de pasgeborenen. Een factor tien daling, dat is pas een populatiecrash. Je zou kunnen zeggen dat de natuur een failed state is: ze slaagt er niet in het welzijn van haar inwoners te bevorderen.

Het terrein van het wilde-dierenleed is ideaal voor een filosofische lentewandeling, omdat het een mijnenveld is van denkfouten. Waarom is dat probleem zo sterk verwaarloosd, zelfs door dierenactivisten?

Misschien twijfel je dat wilde dieren leedervaringen hebben? We zien een bij op een bloem; heeft die een bewustzijn? Er zijn aanwijzingen dat bijen gemoedstoestanden hebben. Een bij kan leren dat een waterdruppel op een verticale markering lekker suiker bevat, en een waterdruppel op een horizontaal streepje wansmakelijke, bittere quinine bevat. Hoe zal ze reageren op een twijfelgeval: een schuin streepje? Gaat ze proeven van het water, in de optimistische overtuiging dat het suiker bevat? Nadat een bij wordt geschud (wat de aanval van een honingdas simuleert), is ze minder snel geneigd om het water op die schuine markering te proeven. Dit is een pessimismeneiging. Die angstige bij heeft lagere niveaus van de gelukshormonen dopamine en serotonine, en ze wordt optimistischer als ze antidepressiva krijgt. Als wij depressief of angstig zijn, dan worden wij ook pessimistischer in onzekere situaties, en we worden optimistisch als we gelukkig zijn. Deze judgment bias hebben we ook al waargenomen bij onder andere honden, ratten, kippen, koeien, varkens, schapen en spreeuwen.

Bijen kunnen ook rekening houden met hun mate van onzekerheid. Als de proef met het schuine streepje te moeilijk is, kan de bij leren om naar eenvoudigere situaties te vliegen, waar de streepjes duidelijk horizontaal of verticaal zijn. Dit vereist metacognitie, wat je kan interpreteren als een soort zelfbewustzijn van het eigen gevoel van onzekerheid. En bijen hebben net zoals onder meer mensen en kippen zelfcontrole en een tijdsbesef: ze verkiezen een grotere maar uitgestelde beloning boven een onmiddellijke maar kleine beloning.

Ontkennen dat wilde dieren leed ervaren, heeft dus geen zin. En we lijden hier aan omvangverwaarlozing (scope neglect): doordat het probleem zo groot is, met triljarden lijdende wilde dieren, wordt ons brein overbelast. We kunnen geen empathie meer voelen met zoveel slachtoffers. Als het bos afbrandt en je moet snel kiezen tussen het redden van een egel of twee egels, kies je waarschijnlijk voor het grootste aantal. Maar tussen het redden van 537523 en 537524 dieren ben je onverschillig geworden.

Op onze wandeling hebben we nog even tijd voor hardnekkigere denkfouten. We zien een roofvogel. Ik zeg je dat al die roofdieren veel leed veroorzaken en dat er dus beter minder roofdieren zijn. Jij antwoordt dat dat leidt tot andere problemen, zoals overpopulaties van prooidieren die dan sterven van de honger. Minder roofdieren is meer leed? Goed dan, laten we dan het aantal roofdieren verhogen. Laten we in dit bos extra vossen vrijlaten, en tijgers, slangen, genetisch gemanipuleerde superarenden, tyrannosaurussen. Die verhoogde predatie zou dan toch het leed in het bos moeten verminderen? Of geloof je dat het huidige niveau van predatie toevallig optimaal is voor het dierenwelzijn? Nee, de natuur is blind en is niet begaan met dierenwelzijn, dus is er geen reden om te geloven dat het aantal roofdieren in een natuurlijk evenwicht toevallig overeenkomt met het optimum welzijn. Als je op een landkaart een willekeurige plek aanduidt, is de kans ook klein dat je een bergtop hebt gekozen. Als je denkt dat het leed in dit bos minimaal is wanneer het een natuurlijk evenwicht kent van dierenpopulaties, dan maak je een status quo denkfout.

Verwant hiermee is de naturalistische denkfout. Dat dierenleed is natuurlijk, want niet veroorzaakt door mensen. Is het daarom minder erg? Dat is een vorm van ongewenste willekeur, een soort van discriminatie op basis van soort, dus speciesisme. Wat maakt het uit of een mens al dan niet de oorzaak is van leed? Waarom zou enkel het leed veroorzaakt door mensen onverantwoord zijn, en niet bijvoorbeeld enkel het leed veroorzaakt door vrouwen, door zwarten, door primaten, door zoogdieren? Voor het slachtoffer, het wilde dier, maakt het niet uit of diens leed veroorzaakt werd door een mens of iets anders. Dat slachtoffer wil gewoon geen leed.

Een heel oneerbiedige denkfout is de rechtvaardige wereld denkfout (just world hypothesis): het geloof dat de wereld goed is en dat de slachtoffers in feite zelf schuldig zijn, alsof de wereld een onzichtbare morele kracht heeft die het morele evenwicht herstelt. Een dader van geweld gelooft al snel dat het slachtoffer het verdiend heeft, het zelf gezocht heeft. Iemand die bezorgd is voor overbevolking, gelooft al snel dat een nieuw pandemisch virus de eigen schuld van de mensheid is. Hetzelfde zagen we bij de prooidieren. We begonnen te denken dat die prooien zelf schuldig zijn aan hun eigen leed: ze moesten zich maar niet zo snel voortplanten. Zebra’s zouden de savanne overbegrazen en dus zelf leed door hongersnoden veroorzaken, als er geen leeuwen waren.

Als afsluiter is er de autonomiedenkfout. Natuurgezinde mensen beweren dat we de natuur zoveel mogelijk haar gang moeten laten gaan en wilde dieren dus niet veel mogen helpen. Ze geloven in een soort van autonomie, natuurlijkheid, integriteit of ongereptheid van de natuur. Maar in feite schenden ze zo de autonomie van anderen. Ze leggen namelijk hun eigen waarden (dat ongereptheid goed is, dat we niet voor God mogen spelen) op aan de slachtoffers, de wilde dieren, op een manier die de slachtoffers niet willen. Ik kan waarde toekennen aan de integriteit of ongereptheid van een natuurgebied, maar de natuur heeft geen enkel besef van haar eigen integriteit en interesseert zich niet in zulke ambigue abstracte waarden. De slachtoffers interesseren zich daarentegen wel in hulp. Als ik dan waarde toeken aan het welzijn van een wild dier, respecteer ik diens autonomie, want dat dier waardeert zelf haar eigen welzijn. Ik speel niet voor God als ik hulp aanbied en dus doe wat de ander, het dier, wil. Ik speel wel voor God als ik mijn eigen waarden voor ongereptheid, of mijn eigen esthetische voorkeur voor natuurlijke schoonheid, opleg aan alle wezens in de natuur, terwijl noch de natuur, noch een dier deze waarden en voorkeuren deelt. De eigen esthetische voorkeur laten primeren boven de voorkeur van een ander om geholpen te worden, is egocentrisch.

Het is nog niet duidelijk hoe we veilig en doeltreffend de natuur kunnen helpen bij het bevorderen van het welzijn van wilde dieren. Maar daar komt verandering in. Nieuwe organisaties zoals Wild Animal Initiative en Animal Ethics zijn volop actief in het oprichten van de kersverse wetenschappelijke onderzoeksdiscipline ‘welzijnsbiologie’, de studie naar het welzijn van dieren in de natuur. Zo krijgt het wilde-dierenleed iets meer aandacht die het verdient.

Geplaatst in Blog | Tags: | Een reactie plaatsen

A game theoretic solution to population ethics

As mentioned previously, population ethics is probably the most important area in moral philosophy for effective altruists who want to do the most good. There are some very serious problems in population ethics that relate to crucial considerations about doing the most good. Here I explain an elegant solution to the population ethical problems, which I called variable critical level utilitarianism, with the help of a game theoretic analogy. This new theory solves for example the ‘very repugnant conclusion’ of future generations and the ‘happy meat’ problem.

For a mathematical description and a discussion of population ethical theories, see here.

 

The population ethics game

Variable critical level utilitarianism can be described as a strategic game. Consider a building with a lot of rooms. Outside the building are the players waiting. Before the players may enter the building, the game master explains the set-up. Only when a player’s name is written on the door of a room, the player is allowed in that room. In each room, a player (who is allowed to enter that room) can expect a room specific payoff[1]: a reward (the player receives an amount of money from the game master) or a punishment (the player has to pay an amount of money to the game master). Before they enter the building, all players are fully informed about the payoffs they and the other players can get in all the rooms in which they are allowed.

For each of the permissible rooms, a player can declare his or her avoidance amount: the willingness to pay to avoid that room. Players cannot declare an avoidance amount for the rooms in which they are not allowed. With these declared avoidance amounts, and the payoffs of the players in each room, the game master computes a net welfare value for each room: the sum of the payoffs of all allowed players in the room minus the sum of the declared avoidance amounts of those players.

The room with the highest net welfare level is selected (in case of a tie, one of the rooms is randomly selected). The game master opens this room and the players whose names are written on that door must enter that room and receive their reward or punishment. (As they do not avoid that room, they do not have to pay their avoidance amounts.)

However, there is a catch. When players declare infinitely high avoidance amounts, it could be the case that all rooms receive a negative infinite net welfare, and this makes room selection impossible. To avoid this problem, the game master sets an upper bound on the sum of the avoidance amounts, which is calculated as follows. Some players are allowed in all the rooms, i.e. their names are on all the doors. These are the ‘necessary players’, because they necessarily receive a payoff. The other players are the ‘contingent players’, because they only receive a payoff based on the contingent fact that their name is written on a door. For each room, the game master calculates the sum of the positive payoffs (rewards) of all the contingent players of that room. The upper bound on the sum of avoidance amounts is given by the maximum of the sums of positive payoffs of contingent players, where the maximum is taken over all the rooms. If there are no rooms with contingent players having positive payoffs, the maximum is simply set to zero.

The analogy with population ethics is as follows. The building corresponds with the choice set in the population ethical problem. Each room corresponds with a possible situation that can be chosen. Who exists in the future depends on our current choices. The rooms where player  is not allowed correspond with the situations where person  does not exist. The players who are allowed in all the rooms, correspond with the necessary people: they exist in all possible situations. These people include currently living people. The other players correspond with contingent people: their existence depends on the choice of situation. The payoffs are the utilities. A reward in room s corresponds with a positive utility, i.e. a life worth living in situation s. A punishment corresponds with a negative utility, e.g. a life with more negative than positive experiences. The declared avoidance amount of a player corresponds with the critical level chosen by a person.

 

Avoiding the very repugnant conclusion

This population ethics game is a strategic interaction between players, because the payoffs of the players depend on the strategies played by the other players. Each player can choose a strategy which consists of his or her declared avoidance amounts for the permissible rooms. As an example, consider a building with three rooms, two necessary players (An and Ben) and thousand contingent players. These contingent players are only allowed in room 3, where they each get a positive but minimal payoff of 1. In room 1, An receives a reward of 300, Ben receives 100. In room 2, the payoffs are reversed: An receives 100, Ben receives 300. In room 3, An and Ben receive a punishment of -100 each. The upper bound on the sum of avoidance amounts is given by the maximum of the sum of positive payoffs of the contingent players, which is 1000. In room 1, Ben can choose a positive avoidance amount of 1000 to maximally influence the selection of room 2, which is his favorite. Similarly, An can choose an avoidance amount of 1000 in room 2. In room 3, An and Ben are worst-off, so they set an infinitely high avoidance amount in order to avoid that situation. However, when they do that, the upper bound of 1000 will be used to calculate the net welfare. Hence, the net welfare values of the rooms 1 and 2 are 400 (the total payoffs of An and Ben) minus 1000 (the total avoidance amount), and for room 3 it is 800 (the total payoffs of An, Ben and the contingent players) minus 1000 (the maximum avoidance amount). As room 3 has the highest net welfare, this room will be chosen. But this room is strongly disliked by An and Ben.

In population ethics, this is called the very repugnant conclusion: a situation where everyone (An and Ben in the example) is very happy can be worse than a situation where all those people are very miserable, with extreme suffering, if the second situation contains a huge number of extra people with lives barely worth living. Total utilitarianism faces this very repugnant conclusion, because the total utility of all the barely happy extra people in the second situation can trump the extreme misery of the extremely miserable.

However, this conclusion can be avoided in variable critical level utilitarianism: if An and Ben chose a zero avoidance amount, they could manage the selection of their more preferred rooms 1 or 2. By choosing between two strategies, i.e. ‘zero avoidance amount’ and ‘maximum avoidance amount’, An and Ben are in fact playing a strategic game called ‘chicken’ or the ‘hawk-dove game’. This game contains two pure and one mixed Nash equilibria. Both An and Ben playing the strategy ‘maximum avoidance amount’, which results in the very repugnant conclusion, is not a Nash equilibrium.

We can add an extension to the game, such that the necessary players are forced to cooperate to avoid the repugnant conclusion, by putting the necessary players behind a veil of ignorance. Those players know the distributions of payoffs of all the necessary people in all the rooms, they know that they are necessary players (allowed in all the rooms), but they do not know which of those necessary people they are going to be. Those players will then prefer the room that maximizes the expected payoff of the necessary players.

 

The rules of variable critical level utilitarianism

Someone’s relative utility in situation S is his or her utility in that situation minus his or her declared critical level. The social welfare value of a situation S is the sum of the relative utilities of all people existing in S. If a person does not exist in situation s, both its utility and critical level are zero. The social optimum situation is the one with the maximum social welfare value of all possible situations.

A full variable critical level theory allows the individuals to be free to set their own critical levels. This maximally respects autonomy of individuals. Someone may choose different critical levels in different situations and when the set of possible situations changes (e.g. situations are no longer possible or new situations become possible), people may change their critical levels. However, there are three restrictions.

First, if a person does not exist in a situation, that non-existing person is not allowed (or able) to choose a critical level for that situation.

Second, an individual can only choose a non-negative critical level. This is a rationality constraint: if a person would choose a negative critical level, that person kind of acknowledges that his or her existence can improve the social welfare, even if that person would have a negative utility. Or in other words: an individual should be willing to accept a life with a utility equal to the chosen critical level, and no-one could reasonably accept a life with negative utility.

Third, the total critical level, i.e. the sum of all critical levels set by the individuals in a situation, has an upper bound, given by the maximum over all possible situations of the sum of positive utilities of the people who exist in that situation but do not exist in all possible situations. This restriction is required to avoid that people choose infinite critical levels.

 

Dynamic inconsistency

Variable critical level utilitarianism faces the possibility of dynamic inconsistency. Consider a choice set with three situations. Situation s contains N very happy people, with a high average per capita utility level UN(s). Situation s’ contains the same N people, who are slightly happier (utility UN(s’)>UN(s)) plus M extra people with low happiness (positive utility UM(s’)<UN(s)). Situation s’’ contains the same N people, slightly less happy than in situation s (utility UN(s’’)<UN(s)), and the extra M people who are very happy (utility UM(s’’)>UM(s’)).

The game consists of two choices or stages. The first choice involves the addition of the extra M people. The second stage occurs once the M people are chosen to be added, and involves choosing low or high utilities for those M people (i.e. situations s’ or s’’). This game can be solved with backward induction, where we first consider the final subgame, i.e. the stage when the M people are chosen to be added. Although situation s’ is the best for the N people, the M people in that situation can complain and prefer situation s’’, such that they choose maximum critical levels totaling M.UM(s’’). In situation s’’, the N people can complain and set maximum critical level also totaling M.UM(s’’), to turn the balance again in favor of situation s’. The critical levels cancel, so the situation with the highest total utility will be chosen. Suppose N.UN(s’’)+M.UM(s’’)>N.UN(s’)+M.UM(s’), then situation s’’ is chosen. However, this solution for the subgame is not an equilibrium in the complete game, because situation s is preferred to situation s’’ by the N people. In other words: the N people cannot accept the existence of the M people, because if they did so, they know that the end result will be a situation s’’ that has a lower payoff than the situation s without the M people. If the N people choose a maximum critical level in s, then they know the selected situation will be s’’, which they do not prefer. Therefore, they can choose to set a low or critical level in s, which means in the first stage of the game situation s will be selected. What is optimal in a subgame where the choice set consists of s’ and s’’, becomes suboptimal in the complete game with choice set {s,s’,s’’}. The optimal choice depends on the stage in the game. This is known as dynamic inconsistency.

Here we see again that variable critical level utilitarianism is choice set dependent. If situation s’’ was not possible (i.e. was not an element of the choice set), situation s’ could become better than s (because the M people in s’ can no longer complain that situation s’’ should have been chosen). The value of adding extra people depends on the possible situations that contain those people.

 

Examples of dynamic inconsistency

The dynamic inconsistency of variable critical level utilitarianism is rather a virtue than a vice, because it avoids an old problem concerning animal farming or slavery, called the ‘Logic of the larder’. Consider the case of ‘happy meat’, i.e. meat from a livestock animal that had a life worth living (a net-positive life with more positive than negative experiences). In situation s, meat consumption and livestock farming are not allowed and N humans need to eat vegan food. In situation s’, animals are raised at happy farms (no factory farms) where they have net-positive welfare, but they are killed prematurely so that humans become a little happier by enjoying the taste of meat. In situation s’’, those animals are not killed prematurely, but can live long happy lives at farm animal sanctuaries. Their happiness increases a lot, but now humans can no longer eat meat, and they have to take care of the animals (e.g. feeding them), which bears an extra cost. In this situation, humans get the lowest welfare (lower than in situation S), but still positive.

Henry Salt (1914) argued that eating happy meat (situation s’) is not allowed, by comparing the situation with human slavery: we are not allowed to breed human slaves, even if those slaves would have net-positive lives. It is better that those happy human slaves are not born, so Salt prefers situation S. This is also the outcome of variable critical level utilitarianism, due to the dynamic inconsistency.

Suppose the happy livestock animals or happy human slaves had such positive lives, that they prefer existence (as meat animals or slaves) above non-existence. When situation s’’ is part of the choice set, those animals or slaves could complain once they exist in situation s’. However, if they would complain, the already existing N humans would decide not to breed those people, because they want to avoid situation s’’. However, if it would be possible to exclude situation s’’ from the choice set, situation s’ could be chosen (by choosing lower critical values in s’). In games with dynamic inconsistency, this can be done with a commitment device. Suppose for example that we can genetically modify a cow such that the cow will die at the age of two years (when he normally gets slaughtered in situation s’). The cow can be raised on a farm sanctuary and is not killed, but after the cow dies, he can be eaten.

Another example of dynamic inconsistency is climate change. In situation s, the current generation (N people) invest enough in climate policies and clean energy such that harmful climate change is avoided and the next generation (L people) have very happy lives. In situation s’, the current generation does nothing about climate change, they are happier because they can consume more and worry less, but their decision to travel a lot with cars and airplanes influences the exact timing of fertilization of their future children. Having sex a second later, and a son instead of a daughter is born. As a consequence, the next generation is not the L people, but other people are born. These M people do not exist in situation s. Suppose the M people in situation s’ have to deal with the consequences of dangerous climate change, but they still have slightly positive lives. These people prefer a third situation s’’, where they exist and get huge compensation fees from the N people who caused climate change. In s’’, the M people are happier, but due to the compensation payments, the N people become worse off in s’’ than in situation s. In this case, variable critical level utilitarianism could pick situation s.

This consideration also influences the discount rate that is used in cost-benefit analyses of climate policies. If situation s is chosen, it implies that the welfare of the next generation should not be discounted much: it is better that the next generation is very happy (the L people in situation S) instead of slightly happy (the M people in situation s’). However, the welfare of generations in the more distant future can be strongly discounted according to variable critical level utilitarianism. For the more distant future, this population ethical theory can resemble the asymmetric person-affecting theories. This is because in the more distant future, the current generation no longer exists and hence is no longer able to pay compensation fees to e.g. the Q people of the fifth generation. Suppose those Q people had low but still positive welfare levels due to climate change. They cannot complain against the N people (the current generation), because if the N people chose policies to avoid climate change, the Q people would not be born. In other words, a situation analogous to s’’ for the Q people in the more distant future is impossible. That means the welfare of further generations in the more distant future can be strongly discounted (at least when they still have positive utilities: when they get a negative utility due to climate change, their negative relative utilities strongly decrease the welfare function because their critical levels cannot go below zero).

Conclusion

Variable critical level utilitarianism is a theory in population ethics that uses a welfare function composed of the sum of the relative utilities of all existing people, whereby a relative utility is the actual utility of a person in a situation, minus a critical level. These critical levels are variable: people are free to choose their own critical levels (up to a well-chosen maximum), and so these critical levels can differ between situations and can even depend on the choice sets of possible situations. Traditional population ethical theories are limiting cases of variable critical level utilitarianism, with constraints on the critical levels. Due to these restrictions of the critical levels, those traditional theories face counterintuitive implications such as the very repugnant conclusion.

The flexibility of variable critical level utilitarianism allows to avoid the population ethical problems. The fact that people can choose their own critical level and take into account the choices of other existing people, creates a strategic game. Variable critical level utilitarianism has a game theoretic dynamic inconsistency. Some examples (consuming meat from happy livestock animals, breeding happy human slaves, causing climate change) demonstrate that this dynamic inconsistency is a virtue rather than an vice: it can explain when and why breeding happy livestock animals or happy human slaves is not allowed, why we have to prevent climate change and why we should not strongly discount the welfare of at least the next few generations.

[1] A player’s payoff consists of the received reward or punishment, but can also include other considerations that are valued by the player. For example, if a player values equality of rewards, the player may prefer a room with a lower personal reward, if all players in that room receive the same reward. This can be compared with a situation in ethics. Suppose you can choose between two situations. In the first situation, you are very happy (a high utility level), but everyone else is miserable. In the second situation, everyone else becomes extremely happy, at the cost of a slightly lower happiness for you. With some altruistic inclination, you might prefer the second situation, even if you get a lower personal utility.

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

Doing good now or later, by yourself or by others?

How can we do the most good: by helping others now or later? By helping others directly by ourselves, with our own hands, or indirectly through the work of others? These are important questions studied in the effective altruism community (see e.g. here and here). In this article, I want to structure the debate and present my personal conclusions and recommendations.

Doing good directly (by yourself)

Most people have a preference for having a personal, direct impact. The impact or amount of good that you do by yourself is the product of two factors: your personal productivity (good done per hour) and your time (hours spend doing good). As your time is limited and you can use the same hour only once, your time is costly. For example you can increase your personal productivity by learning new skills, but that is costly because it takes time to learn new skills.

As most people do not face very interesting opportunities to do much good personally (for example you do not have an opportunity to save a drowning child in front of your eyes right now), doing good by yourself is most often not so effective.

Probably the best, most cost-effective option we have for doing good by ourselves, is going vegan. Of all our personal behavior choices, veganism is most likely the most welfare improving (it prevents a lot of animal suffering and has many cobenefits for the environment and public health) and has an almost negligible time cost (for example vegan food is generally not more expensive, so veganism does not require spending more hours to earn extra money to buy food).

Doing good indirectly (by others)

There are many good opportunities to do good indirectly, through the work of others. Doing good by others can be effective, for example because of a multiplier effect: I can go vegan by myself, which means only one person eats vegan, but I can also do campaigns, support interventions or develop new vegan food that result in many more people going vegan.

As someone’s impact is productivity times time, we can increase our indirect impact by increasing the personal productivity or the amount of time of other people doing good. For each of these improvements, there are several strategies involving research, entrepreneurship, policy and advocacy.

  • Increasing the productivity of other people
    • By developing new knowledge or technologies, through research. With philosophical research we learn what are the most important problem areas, with economic research we discover how to do good more effectively and with technological research we can invent new technologies that increase our productivity. For example, we can develop new food technologies like cell-based and plant-based meat, that facilitate veganism.
    • By exploiting current best available knowledge or technologies, through entrepreneurship and policy. We can invest in organizations that use the best available technologies to improve the world, or influence policy makers to make better choices. For example, we can do corporate vegan outreach to change the food industry.
  • Increasing the time of other people
    • With monetary incentives (payments). We can donate money to charities or universities such that they can hire more employees or researchers, which means more hours spend on the cause or on the research. One strategy with a potential multiplication effect, is earning to give: instead of doing direct work at a charity, you can look for a high-paying job, earn a high income, and donate a lot of money to charities who can hire more people. This can increase the total number of hours of good being done in the world.
    • Without monetary incentives. We can do campaigns that stimulate people to spend more time doing good, we can do advocacy work and influence politicians such that governments subsidize organizations to hire more people to do more good, or we can support effective altruism movement building such that people become more involved and spend more time doing effective altruistic things.

When it comes to indirectly doing good, the two most important resources are information (knowledge) and money. Both resources are necessary, and they can mutually enhance each other. On the one hand, information can increase the money resource, because knowledge is the main driver of economic growth that generates higher incomes. Knowledge about cost-effective interventions also means that money can be used more productively or effectively. On the other hand, money can increase the information resource, because money can be used to do research that delivers new information. Both knowledge and money are especially important in the long run, i.e. for doing good later (see below).

Doing good now

In many cases, being a patient philanthropist by doing good later is more effective than being impatient and doing good now. However, one important consideration implies prioritizing doing good now: if  good opportunities to have a high impact are expected to dry up soon. The two main factors that dry up an opportunity are crowdedness and redundancy. If these factors are expected to become present, then doing good now can be preferred.

  • The solution or cause area is expected to become crowded soon. Due to a decreasing marginal productivity, using an extra unit of resource does less and less good if there are more and more investments made in the problem area. Probably the most important example of an effective solution that is important in the short term, is the development of new vegan food technologies such as cell-based agriculture and precision fermentation. At the moment, this area is still highly underfunded, so an extra unit of investment can do a lot of good. But in the future, when global meat consumption is in decline and there are billions of dollars more invested in many new animal-free food companies, adding an extra dollar will generate less impact.
  • The solution or cause area is expected to become redundant soon. This can be the case when an important event arises, such as the arrival of a new, completely different but more effective technology. For example, campaigns to improve the living standards of farm animals can become redundant when people switch to the novel vegan food products. Probably the most important example of events that make basically everything redundant, are the existential risks (X-risks): catastrophic events that can end humanity or civilization. An X-risk can kill all people who are able to effectively do good, or wipe out all important knowledge that we gained. An X-risk event dries up all opportunities to do good in the future. Hence, next to the development of new vegan food technologies, we have to prevent X-risks. The two most important cause areas are:
    • Biosecurity: preventing engineered pandemic risks.
    • AI-safety: preventing misaligned artificial superintelligence risks.

Doing good later

Doing good later is generally much more effective than doing good now, because we will have more resources later. We will have more money when money is invested, and better knowledge about effective interventions when research is done.

  • Increasing the money resource through savings and investments. Most people are impatient in the sense that they have a high personal discount rate for the future and don’t consider benefits in the far future. This results in a high interest rate (a high return on investments), probably higher than the rate at which doing good becomes more costly (the economic growth rate can be used as an estimate of the doing good cost growth rate). We have an opportunity to let our money grow exponentially at a high rate, such that our future impact, even at higher costs to do good, can be much bigger than if we choose to spend our money to do good now. When it comes to doing the most good, an effective altruist should be patient, because choosing a high time discount rate is incompatible with intergenerational impartiality. Simple economic models (e.g. the Ramsey model of intergenerational welfare economics) or tools for patient philanthropists, most often give very high estimates of the optimum rate of savings, often higher than 90%. That means patient philanthropists should save and invest more than 90% of the money they intended to give to charities, and donate the money plus the interests much later (probably even after they die). These results are robust against many extensions and nuances of the models.
  • Increasing the knowledge resource through scientific research. Most people are selfish in the sense that they have an incentive to free ride and don’t consider positive externalities of investments in public goods. Free-riding is a market failure, where people do not pay for a public good but still reap the benefits of that public good. One of the most important public goods that allow us to do good, is knowledge. Even altruists can have an incentive to free-ride, by letting others pay for scientific research, and then use those new technologies and research results to do good oneself. This means research on effective interventions can be underfunded, even by altruists. Research and development are very important, for example because they are the main drivers of economic growth. The rate of return on R&D for industries and commercial firms is higher than the rate of return on capital or the interest rate, so I expect that the social return on investment in R&D for charities is also very high. Especially highly neglected and extremely underfunded research areas such as welfare biology offer interesting opportunities for discovering new effective ways to do good.

Recommendations

Based on the above considerations, I recommend the following choices.

Mistakes that I made

I will end with a personal note. I have been a social justice, animal rights and environmental activist for twenty years. However, for most of those years, I made a lot of mistakes that made me less effective in doing good. Two important kinds of mistakes were not saving/investing enough money and not spending enough time thinking about important problem areas and effective strategies. As a result, my impact was very low. I donated a lot of money to less effective charities, and spend a lot of time doing less effective actions. If instead I saved my money and spend more time thinking about effective altruism, I would have been in a much better position to do more good and my overall lifetime impact would have been much higher.

As argued above, there are strong market economy arguments why doing good later is more effective (in particular the high interest rate and the importance of R&D), but I have also a personal argument: my high rate of updating my beliefs (i.e. changing my mind) about effective strategies. Given that past belief update rate, it is likely that I will change my mind on important topics again in the future, and probably the same goes for other effective altruists. A lot of important, mind-changing insights gained by effective altruists, are very new (discovered the past two decades), which makes it likely that we will discover more new ideas about crucial considerations. Hence, it is a safe strategy to not spend too much resources now, while we still have limited knowledge. We can save more resources for later, or spend them now on doing research to know how to better spend our resources later.

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