Where to Start

On reading This Is Vegan Propaganda, the gap it left, and building a small tool to fill it.

There’s a moment in Ed Winters’ This Is Vegan Propaganda where he describes skipping a family event because the people there would be eating meat. He’s self-aware about the cost. He knows what he’s giving up, he knows how it looks, he’s made his peace with it. The honesty is disarming. It’s also, for me, a hard stop.

I listened to the book over a few days hiking in Taiwan and finished it convinced of the substantive arguments. The case for reducing (or eliminating) animal products on climate, welfare, and health grounds is strong and carefully made. Winters is a better writer than the “vegan propaganda” title suggests, and the anecdotes land. The story about Daisy, a cow who escaped a slaughterhouse and was cheered on by passersby who a few hours earlier had presumably been buying ground beef, stuck with me for days. Most of us are both the person who cheers and the person who eats, and we usually don’t have to notice.

I’m not going to rehearse the case for why any of this matters. I’ve written about that separately, and better writers than me have made the argument at length. What I want to write about is what happens after you’re convinced, which turns out to be the harder problem.

I finished the book persuaded and also a bit stuck. The family-event anecdote was honest, but it was also a version of the choice I wasn’t going to make. And the book’s broader framing, that vegan is the only serious endpoint, left me with no sense of where to start if I wasn’t going to go all the way.

The gap the book leaves

Winters makes a strong case that a vegan diet is the gold standard across the three axes that matter: climate, welfare, and health. I don’t want to poke holes in that. On the whole I think he’s right. But the book has an implicit message and an implicit response, and they don’t meet in the middle. The message is you should be vegan. The response, for most people with an average Western diet, is I’m not going to do that, so I’ll do nothing. That failure mode is where most of the available real-world impact gets lost.

I think the “vegan is the only serious option” framing is actively counterproductive. It makes the destination so distant that the journey doesn’t start. It treats a 90% reduction in meat consumption as a moral failure rather than an enormous win. Anywhere to the right of the average Western diet is better than average, and the further right you go, the better. A person who cuts beef and lamb has done something real. A person who goes pescatarian has done something bigger. None of them are vegan, and none of them should feel like they’ve failed the assignment.

What I wanted was a roadmap. If I’m going to change one thing, which thing? If two, which two? Where does the effort start to outweigh the gain? The book doesn’t answer this, because the book isn’t really trying to.

I should say upfront: I’m learning as I go and I don’t have all the answers. What follows is my attempt to build a tool that helps me think, not a prescription for anyone else.

Effort is not one number

The other thing missing from the book, for me, is a serious treatment of the social cost. Winters acknowledges it. The family-event story is exactly that acknowledgement. But the book’s answer is essentially yes, it’s hard, do it anyway. I understand the answer. I’m not going to follow it.

Part of the reason is that “effort” isn’t really one thing. It’s at least three:

• Cognitive effort: knowing what to order, reading labels, meal planning
• Social effort: restaurants, travel, dinners with other people, being a guest in someone’s home
• Adherence risk: the probability you fall off over a year

These move differently across dietary shifts. Pescatarian is low cognitive but can be high social at some dinner tables. A fully flexible “vegan at home, vegetarian when out” position is high cognitive at home but lower social than strict vegetarianism, because you’ve built in a pressure valve. Collapsing these into one number hides the most interesting part of the picture.

The framework

I built a simple nine-step ladder from an average Western diet to strict vegan, and scored each step on six dimensions. The steps are:

1. Baseline Western
2. Weekday vegetarian (~50% meat reduction)
3. No beef or lamb
4. No red meat (drop pork)
5. Chicken and fish only
6. Pescatarian
7. Vegetarian
8. Vegan-at-home (flexible)
9. Low-dairy vegetarian
10. Strict vegan

Each step gets three impact scores (climate, health, welfare) and three effort scores (cognitive, social, adherence risk), on a 0 to 100 scale, where 0 is baseline and 100 is strict vegan. Aggregate impact is a weighted average of the three impact axes. Average effort is the mean of the three effort axes.

The key metric is marginal gain per unit effort: the change in aggregate impact divided by the change in average effort, step to step. It tells you how much you get for what you pay at each rung of the ladder.

One note worth flagging: vegan health doesn’t score 100. The plant-based cardiometabolic evidence is strong, but a population-level health score has to account for real-world adherence and nutrient adequacy (B12, iron, omega-3, iodine). A careful vegan does great. An average one does less well.

The dashboard

The ladder is ordered by effort, so the curve reads as an efficient frontier. One feature worth flagging: flexible vegan sits at lower effort than strict vegetarianism, which is the surprise the next section unpacks.

The qualitative shape of the curve is surprisingly robust to reweighting. Unless you zero out one of the three axes entirely, the same steps come out as high-value and the same ones come out as expensive. That’s itself a useful finding. The disagreements are at the margin, not in the overall shape.

Three things I didn’t expect

Cutting beef and lamb is nearly free. It captures most of the climate story at almost zero social cost in any normal Western context. If I could give one piece of advice to someone who wanted to start, it would be this step, and only this step, for six months.

Flexible vegan is easier than strict vegetarian. This was the one that genuinely surprised me. The adherence risk is lower because the flexibility is a built-in pressure valve. “Vegan at home, vegetarian when eating out” gives you most of the climate and welfare gain of veganism, without the restaurant problem that strict vegetarianism creates and without the family-event problem strict veganism creates.

Strict vegan has a marginal gain per effort of 0.09. The last roughly 8% of impact costs about as much effort as everything before it combined. This is where the cost-benefit argument runs out and identity or moral commitment takes over. That’s a legitimate reason to do it, and I’m not dismissing it, but it’s a different kind of argument, and the numbers point at a real inflection.

Three quick caveats. The scores are estimates, generated with Claude Opus 4.7 and then adjusted by me. Some will argue that using an AI for this is unscientific and I should be reading the primary literature instead. That’s a fair objection, and at some point I probably will. But this is a practical guide rather than hard science, and the model’s estimates are grounded in the aggregate of what’s on the internet plus its own reasoning, which is roughly the same input I’d be working from if I did the reading myself. The ordinal structure is defensible; the exact numbers are not. And the effort scores are anchored to urban UK life. Flexible vegan in London is a different animal from flexible vegan in rural France.

What I’m actually doing

Honestly, not what the numbers point to. The numbers point to flexible vegan. I’m somewhere around step 5. Chicken and fish at home, try to hold that line when eating out, but I’ll have steak occasionally if I really want it. Dairy is the open question. I love yogurt, and I’m not ready to give it up. Low-dairy-ish is probably where I land after some more experimentation. On eggs, we’re organic-only at home, which the ladder doesn’t capture. It treats egg sourcing as fixed, but the welfare delta between caged and organic is large enough that it’s worth naming.

That’s a real gap between what the model says and what I’m doing, and I don’t want to paper over it. What the framework gives me isn’t a prescription, it’s a map. I can see where I am, I can see what it would cost to move one step further, and I can decide with open eyes. That’s more useful to me than being told what I should do.

The broader point I’d make to anyone reading this: don’t try to get to the far right of the curve in one move. Start with a nudge. See how it goes and how it feels. If it sticks, nudge again. A 90% reduction in meat that you can sustain for the rest of your life is worth vastly more than a 100% reduction you abandon in three months. This is about being practical and making changes that stick, not about arriving at an ethically pristine endpoint.

Close

The book gave me the conviction. The framework gave me a map, and also surfaced, pretty clearly, where I’m not following it. That’s fine. I’m learning as I go.

A follow-up post is coming on what the framework doesn’t capture: the limits of this kind of analysis, and what policies, defaults, and social norms could shift the curve itself. Most of the effort cost in these decisions isn’t really personal. It’s structural. That’s worth its own piece.

I still eat meat occasionally. The framework and I are both fine with that.