Things I've changed my mind on recently

New year new beliefs

In the spirit of some end-of-year reflection, I’ve been thinking about important issues I’ve changed my mind on recently (e.g in the last two years). It’s quite an eclectic range of things: from artificial intelligence to insect suffering to disruptive protest. I would love for others to do something similar, so please do feel free to copy this format or leave some comments below on things you’ve changed your mind on!

I’m not going to reference all the things that caused me to change my mind, as often it was a bunch of conversations and small things, as well as more substantial single data points. Instead, I’ll try to reference the most influential 1-2 pieces of evidence. I changed my mind across a broad range of topics, so I’ve clumped them together by section, in case you only want to read certain bits. The topics I’ve included below are:

• Social movements

• Animal advocacy

• General doing good in the world (or Effective Altruism-related topics)

• Climate change 

Social Movements

I was pretty sceptical that Just Stop Oil’s most publicly disruptive tactics, e.g. blocking the M25, would lead to any positive public opinion outcomes for the climate movement

• What I thought before: Things like blocking the M25, damaging fuel pumps and other very publicly disruptive things would have some observable negative consequences on public opinion.

• What I think now: The climate movement, in the UK at least, can take fairly disruptive action with little negative public opinion consequences. Obviously I don’t know how far this goes, but I thought M25 blockades might be past this line!

• Why did I change my mind? The two surveys we did, one around the M25 protests and one for their oil depot-focused protests. In both cases, we found no statistically significant loss in public support for climate policies, and instead found some pretty clear positives:

  • An increase in support for more moderate climate organisations, such as Friends of the Earth (for the November M25 protests)

  • An increase in reported likelihood to take part in climate activism (for the April oil depot protests)

  • I also found Professor Colin Davis’ experimental findings along the same lines quite interesting, showing no loss in support for climate policies despite disruptive protests.

I’m more worried about increased government repression towards a social movement as a result of more radical tactics

• What I thought before: I didn’t think this was a huge concern, mainly hoping that democratic safeguards would stop anti-protest legislation, but this was clearly wrong (at least for the UK).

• What I think now: A serious long-term harm from radical tactics, including nonviolent ones, could be increased repression towards a social movement. As a result, they might face less people willing to take part, leading to lower numbers and overall smaller chances of success.

• Why did I change my mind? 

  • The passing of new anti-protest legislation in the UK, which was more or less directly attributed to Extinction Rebellion and Insulate Britain.

  • This article by the New Statesman somewhat confirming what I thought - the UK police is cracking down much more on climate protestors recently

  • Tompkins (2015) who finds decreased mobilisation (but no harm on success) for armed radical flanks. Some inference is necessary to extrapolate for nonviolent radical flanks but I still think it’s interesting.

  • Other small bits of news e.g. 6-month prison sentences for peaceful climate activists, tens of people held in prison for 3-6+ months without a trial (on remand), and an Australian climate protestor initially sentenced to eight months in prison, before she was let out early on bail

You can get a lot of media attention with relatively little effort, using some well-designed direct action stunts (e.g. throwing soup). 

• What I thought before: To get media attention for your protest or campaign, you had to do increasingly complex disruptive blockades (e.g. blocking 13 Amazon sites on the same day, with lots of effort involved in making bamboo towers and lock-ons to use in the blockade).

• What I think now: Rather than getting increasingly sophisticated and slick, you could also just target things that are straight up bizarre or unexpected, with little effort, and get way more media coverage. I thought this before, but didn’t appreciate the full extent of how far things could go. I’m also not convinced that this attention is positive, but there’s also evidence to show that disliked tactics don’t harm support for a cause.

• Why did I change my mind? Just Stop Oil’s rise to infamy through their soup-throwing at a Van Gogh painting, disrupting a football game, as well as Animal Rebellion pouring milk on supermarket floors. 

Policymakers get public opinion signals from all sorts of weird places, and it seems fairly irrational or non-systematic

• What I thought before: Elected representatives (e.g. MPs) would be fairly on the pulse of public opinion polling, and would act vaguely in line with public opinion. Obviously all sorts of vested interests affect their decision-making, but I thought they would at least have a good sense of what true public opinion on a topic was.

• What I think now: Elected politicians get a lot of their public opinion “data” from non-trustworthy or non-representative sources, such as talking to random constituents, or traditional news media. As a result, they’re not actually in touch with public sentiment

• Why did I change my mind? Long story short, I talked with Stefaan Walgrave, a Professor of Political Science at Antwerp University, who studies how politicians interpret public opinion signals. Specifically, this paper finds that politicians are not very good at estimating public opinion, and when they interviewed almost 200 elected officials in Belgium, they found that MPs number one source of public opinion was talking to ordinary constituents. This seems crazy to me, that public opinion polling isn’t even in their top four sources of public opinion data! See more from our conversation in our conversation summary here.

Animal Advocacy

Insects can very likely feel pain, and we should do a lot more to help them

• What I thought before: Maybe there is a 20% chance they can feel pain

• What I think now: It’s more like 60-80% they can feel pain, or at least in some of the insects we currently farm the most (around 1 trillion per year currently, composed mainly of crickets, black soldier flies and mealworms).

• Why did I change my mind? Mainly this academic literature review, synthesised very nicely here, which looks at all the evidence whether insects can feel pain. Work by Rethink Priorities definitely also influenced my thoughts on this, who examined the cognitive and hedonic capabilities of many different animals, including bees, silk worms and flies.

An overview of the different criterions for pain reception, and how different orders of insects fare against them. For reference, here are what the orders correspond to in terms of species most of us might recognise: Blattodea (cockroaches, termites), Coleoptera (beetles), Diptera (flies, mosquitoes), Hymenoptera (bees, ants, wasps, sawflies), Lepidoptera (butterflies, moths), Orthoptera (crickets, katydids, grasshoppers). This table is an adapted version of Table 11 on page 49 of the original paper; figure credited to Sajedeh Sarlak.

Cultured meat might be a long way off, and maybe even never reach the same price as the cheapest animal products

• What I thought before: Cultivated (or clean) meat will be an essential tool in animal advocacy, to replace animals in our food system. We could have cost-competitive and widespread lab grown meat in the next 20-30 years (across all meat types except the cheapest chicken).

• What I think now: We will likely (e.g. 70% probability) not have cultivated meat displacing animal-based meat, in any significant proportion (i.e. over 10% of total meat production), over the next 30 years. We also might never reach cost-competitive prices with the cheapest meat products e.g. low-cost chicken. As a result, we should probably diversify our alternative protein portfolio such that plant-based meats and precision fermentation also get adequate funding and talent. In addition, we should probably also diversify to theories of change to help animals, such as legislative advocacy or more movement-building.

• Why did I change my mind? A techno-economic analysis published by David Humbird, as well as some reviews of this and other analyses by Joe Fassler and Rethink Priorities. Quotes like “it seems unlikely that we can do better than a lower bound of $3.87/kg.” which is already more expensive than the cheapest chicken and pork products, was quite worrying to read! Two other bits of research by Rethink Priorities also made me feel more sceptical, due to: 

  • Most cultured meat forecasts being too optimistic 

  • Forecasters estimate that cultured meat will only grow to small production volumes by 2050.

I still feel quite unsure about this, and think tanks and scientists have historically estimated the lowest possible cost of solar to be much higher than what it is now, so we could see a similar effect here (e.g. David Humbird underestimates technological progress). 

General doing good / Effective Altruism

Improving lives, to me, is more important than extending lives 

• What I thought before: Saving lives, e.g. from malaria, is one of the best ways to help humans alive today. 

• What I think now: Improving people’s quality of life is closer to what I care about. This is slightly different to the other things I changed my mind on, as it was less about facts, but more about realising what mattered to me. To note, I still think saving lives from infectious diseases or easily preventable ailments is extremely important! I just think my personal preference is more towards ensuring people alive today, particularly those living particularly tough lives (i.e. in extreme poverty), have a better life. If you want to read more about this moral quagmire, I would recommend checking out my previous blog post where I discuss this in more depth.

• Why did I change my mind? Research by Happier Lives Institute comparing a charity treating depression amongst low-income women in Uganda and Zambia, StrongMinds, relative to some global health ‘classics’ e.g. Against Malaria Foundation. Long story short, they think StrongMinds is a more cost-effective way to help people improve their well-being (not without controversy!) and I think this is reasonably compelling. Additionally, I found this research on the importance of philosophical considerations when thinking about extending vs improving lives quite interesting too.

Artificial intelligence (AI) is making even faster progress than I expected 

• What I thought before: AI is making rapid progress, and we could see superintelligent AI systems within our lifetime (e.g. maybe around 2060)

• What I think now: AI is making even more rapid progress than I expected, and we could see industry-transforming AI in the near future (e.g. over the next 3-5 years). For example, I mean some occupations or industries (e.g. copywriting or programming) might see widespread and significant changes in how work is done, as a result of AI systems.

• Why did I change my mind? DALL-E 2, ChatGPT and CoPilot. Also increasing investment in AI (see graph below) and training language models on ever larger data sets meaning we’ll probably see some more significant improvements in coming years.

Climate Change

I thought the UK wouldn’t do particularly well on climate, but I also didn’t expect them to open a new coal mine just after their COP presidency ended

• What I thought before: The UK’s policies aren’t close to actually meeting it’s Paris Agreement targets to stay below 1.5°C, but they won’t do egregious things like commission a new coal mine in 2022, just after hosting COP26 in Glasgow

• What I think now: The UK Conservative Party, especially in its current form, has basically no interest in meeting the UK’s climate goals. Not only that, vested interests (read: the fossil fuel industry) basically control policymaking on oil and gas exploration, such that public opinion or expert advice doesn’t even seem like a consideration.

• Why did I change my mind? The UK did indeed approve the Cumbrian coal mine, despite opposition from even Tory MPs and the UK steel industry saying that they would not use the majority of this coal. Additionally, there was condemnation from the UK Government’s climate change advisory group, and public opposition to the project.