First published December 29, 2022; minor update January 9, 2023
Draft
Simon Knutsson
Efforts to improve the world and help others effectively predate and occur outside of what is called ‘effective altruism’, of course. There are older and broader societal currents about evidence-based policy, impact evaluations, cost-effectiveness, the effectiveness of foreign aid, and more broadly which larger societal changes are desirable and how to achieve them. The following are two interesting examples, although many more examples could be given: The first example is Disease Control Priorities. Their website says:
In 1993, the World Bank published Disease Control Priorities in Developing Countries (DCP1), an attempt to systematically assess value for money (cost-effectiveness) of interventions that would address the major sources of disease burden in low-income and middle-income countries.
The other example is the Campbell Collaboration. Its website talks about “the creation of The Campbell Collaboration in 2000”, its vision statement is “Better evidence for a better world”, and its mission statement says that it “promotes positive social and economic change through the production and use of systematic reviews and other evidence synthesis for evidence-based policy and practice.”
Still, when GiveWell started in 2007, it provided a seemingly valuable service by trying to synthesise information and filling a gap for individual donors by giving recommendations about where to give.[1] Around 2008–2010, GiveWell recommended charities working on global health, early childhood care, and the like.[2]
In this area of attempted effective philanthropy, I recall that the Center for High Impact Philanthropy already existed. Its website says it was founded in 2006 and that it “is a trusted source of knowledge and education to help donors around the world do more good.” There were other organisations too, such as Philanthropedia. An article from 2009 says: “Fledgling organizations like GiveWell and Philanthropedia are attempting to come up with a new way to evaluate charities based on what they accomplish. Charity Navigator is also working on an effectiveness measure”. An archived version of the Philanthropedia website from 2009 says: “We aim to create a community of donors who act collectively to support top performing nonprofits”. I don’t think of the Center for High Impact Philanthropy and Philanthropedia as having been affiliated with effective altruism, as the phrase ‘effective altruism’ is currently used.
It seems things took a turn for the worse with the involvement of some moral philosophers and the start of what I will call ‘Oxford-style’ effective altruism, existential risk, and longtermism. This essay is about the historical and future impact of this Oxford style.
Oxford-style effective altruism
Giving What We Can was founded in Oxford in 2009.[3] It seemed to be more about outreach while GiveWell was more about analysis. My impression was that Giving What We Can tried to promote more giving (to effective organisations), which sounded fine, but it didn’t have any special edge when it came to evaluating where to give. Perhaps the early days of Giving What We Can were innocuous.
The Centre for Effective Altruism was founded in 2011 as an umbrella organisation for Giving What We Can and another organisation (80,000 Hours), according to this website. If we look at an old Centre for Effective Altruism web page from 2014, it says that the three trustees were the moral philosophers Nick Beckstead, William MacAskill, and Toby Ord. Other moral philosophers have also been involved in the Oxford-style circles, including Nick Bostrom and Hilary Greaves. They all seem to have similar, controversial views on moral philosophy. For example, Toby Ord, writes: “I am very sympathetic towards Utilitarianism, carefully construed.”[4] And they seemingly care very much about there being vast amounts of value, positive welfare, or the like in the far future, even at the expense of increased total extreme suffering and risks of astronomical amounts of suffering.
What is characteristic of this Oxford style of effective altruism is neither effectiveness nor altruism. Even if we grant that they are altruistic and aspire to be effective, which one can dispute, the range of people and organisations that could reasonably be called effective and altruistic is much wider. Rather, the following seems characteristic of Oxford-style effective altruism (the features by which it stands out):
- Traditional total utilitarianism or similar. Traditional total utilitarianism is essentially the moral view that one should just maximise the sum of well-being (positive well-being minus negative well-being).
- A focus on the far future, especially on space colonisation and bringing about vast amounts of value, welfare, or the like in the far future.
- A focus on expansion in terms of influence, networks, resources, and the like. For example, building a specific movement and community. Targeting students at universities to try to influence their values and priorities. Spreading the just-mentioned ideas about morality and the future (in a way that is not educational). Working on relations with rich and powerful people, working on the brand, media strategy, lobbying, and so on.
- An ends-justify-the-means mentality and associated unethical behaviour. For example, being extremely strategic, one-sided, and biased; a lack of honesty and transparency; biased Wikipedia editing; secrecy; etc. Some speak of the behaviour in terms of it being ‘corrupt’ and ‘manipulative’.[5]
All this arguably makes sense from that traditional utilitarian moral foundation. And, of course, from that moral perspective, there need not be anything wrong with any of the behaviours I just mentioned.
Without going into debates about whether the moral views, priorities, and behaviours are defensible, which I don’t think they are, we can ask what consequences this Oxford style of effective altruism, existential risk, and longtermism has had and will have.
To be clear, I am not talking about a regular person who works for a charity, wants to be altruistic and effective, likes effective altruism, and thinks of themselves as an effective altruist. They are not the topic of this text. I am talking about the Oxford style that we can find among the aforementioned philosophers and organisations such as the Centre for Effective Altruism, 80,000 Hours, Future of Humanity Institute, Global Priorities Institute, and, at least to some extent, Open Philanthropy.
What are the effects of Oxford-style effective altruism?
A defender of the Oxford style might say something like the following, among other things: Sure, there have been efforts to get people to care about and work on existential risk, space colonisation, longtermism, and the like. A way to get more work done and money spent on such things is to influence students, rich and powerful people, and so on. But these Oxford-related people and organisations have also done other things. They have raised money for poverty relief and the like, and they have made more people more altruistic. So, even from a more standard altruistic perspective (not only the utilitarianism-space-colonisation-stuff), these people and organisations have made the world better.
But that doesn’t seem obvious. It’s of course important to consider what would have happened if these philosophers and organisations had not done what they did. As I mentioned above, there were already trends of effective philanthropy, evidence-based social programmes, evaluation of foreign aid, and the like. And those trends would likely have continued regardless of the people and organisations in question. The idea that one should do more or donate more is also not new.
Here I will mainly just mention two possible effects of the Oxford style.[6]
The first possible effect is that the Oxford style of effective altruism, existential risk, and longtermism has led to there being, in counterfactual terms, fewer people and resources dedicated to more common-sense altruism, such as trying to prevent suffering and violence that occur nowadays. The mechanism might simply be that adherents of the Oxford style have convinced people to focus on certain issues like existential risk and bringing about astronomical amounts of value in the far future at the expense of these other issues. And, generally, altruistic people would have done altruistic work anyway and would have come into contact with information and debates on effectiveness and priorities. To oversimplify, in general, these people would have done effective altruistic work uninfluenced by the Oxford style, but now they instead spend their efforts and resources on ensuring that there will be astronomical amounts of value in the far future.
The second possible effect is that the Oxford style discourages effective altruistic work. The general idea of trying to make the world or the future better (or less bad) and doing so effectively is, of course, appealing. It’s something to be enthusiastic about; something one might want to work on or give money to. But the Oxford style makes it less appealing. The area of work that the Oxford style has influenced is now shady. Impactful altruistic efforts could have been (more) attractive. How many people who would have been interested in improving the world have been dissuaded due to the problematic behaviours in this style of effective altruism? How much of a difference to beings that really suffer would they have made?[7]
Notes
[1] According to GiveWell’s website, two people “started GiveWell in 2007 as a full-time project.”
[2] See https://www.givewell.org/charities/top-charities/2008, https://www.givewell.org/charities/top-charities/2009, and https://www.givewell.org/charities/top-charities/2010.
[3] According to https://www.centreforeffectivealtruism.org/history.
[4] For replies to Ord’s text, here is one by Magnus Vinding and here is one by me.
[5] I mention a fraction of the problems at https://www.simonknutsson.com/problems-in-effective-altruism-and-existential-risk-and-what-to-do-about-them/.
[6] I set aside effects such as the FTX and Alameda Research thing.
[7] I am grateful to Émile P. Torres and Magnus Vinding for comments on earlier drafts.