Weekend Reading & Selected Links
Happy weekend! Here are some links to things I've been reading or watching that you might also enjoy:
- My new podcast conversation, with Laura Deming. At the bottom of this email, I've included four excerpts from the conversation.
- Charles Mann on how running water systems work.
- 'Artificial General Intelligence's Five Hard National Security Problems', a recent RAND paper.
- Terry Tao streams himself using Github Copilot.
- How economics has changed.
- Remembering Alasdair MacIntyre.
- 'Artificial Wombs Will Save Lives Not Birth Rates', Lan Dao for Palladium.
Thanks, and have a great weekend,
Joe
Excerpts from my podcast with Laura Deming
1. Regulatory bottlenecks to longevity drugs
JOSEPH WALKER: Say you're appointed FDA Commissioner tomorrow. What are some of the first things you're doing to reduce regulatory bottlenecks to longevity drugs?
LAURA DEMING: I think I would lay the groundwork for ageing as an indication. I think there's a lot of regulatory work to be done to conceptualise what it could mean for a longevity drug to exist for that indication. So I put that groundwork in place.
But honestly the most important thing would be I would just find some way to shorten timelines for review cycles, while still being effective and thoughtful, from let's say six months to ideally a couple of weeks. Although, I don't know if that's actually plausible for the FDA. I’d want the FDA to do what it had to do to be safe and effective.
But I think that the six month review cycles for preclinical companies can be very... It's basically just like: every time you want to make a change, you have another six months of iteration, of waiting and just kind of like you submitted something and you're not really sure. And there's some parts of the FDA that are amazing and very collaborative and very helpful and will give you a lot of feedback so that this process works well. But I think just that uncertainty…
Again, it might be that it’s required for some kind of internal process. But just that is I think such a huge contributor to timeline uncertainty for companies. So anything that would help the FDA shorten those processes by functioning more quickly in some way — that's very helpful for companies.
2. A more humane transhumanism — and what it means to aspire to become something you can't yet comprehend
WALKER: I know that lately you've been searching for a more humane transhumanism. And I'm curious what it is specifically about the core framing of transhumanism that you're trying to substitute for.
DEMING: I think there's a couple of things, and I don't really understand this that well now, but I'm thinking about it a lot. I think one thing is... I was trying to understand at some point what the transhumanist manifesto, philosophy, was, because I'd been adjacent to this movement for a long time and hadn't really understood it.
So I went and read some stuff… If you ‘Control+F’ and search for, like, the word “love” in a lot of these manifestos, it's just not [there]. I think David Pearce might mention this more in his work. But a lot of the stuff that I've read is very oriented around, like, gaining power and just being really powerful and like this drive to survive.
And I don't think that's bad or even necessarily shouldn't be part of the future, but it feels pretty incomplete. It doesn't feel inspiring to me personally… And so that's one part.
The other part is it feels way too confident in the types of technologies and the types of ways that things could change. And I think the more that I think about this stuff, the more it's interesting to me to see populations change in ways that are hard to predict versus individuals change in ways that are easy to predict. And so I'm interested in versions of transhumanism that are more oriented towards the former and less centred around the latter, if that makes sense.
WALKER: So the core framing is far too focused on the individual as the unit of analysis?
DEMING: I'm not sure. I think the thing that I know for sure is that there's some version of this that we just have no idea what it even looks like yet. And I think the individual-versus-population view or like the determined-versus-emergent view are two axes where it feels like there's some push and pull. But yeah, it almost feels like in my head it's like receding from a totally different point in phase-space or something.
One interesting fact is that a lot of sci-fi authors, I think intentionally, don't write futures that have qualia very different from our own because that's very hard to relate to. And so their books would be very not popular.
Greg Egan mentions this. He's one of the most futuristic authors ever. He's like, “Past a certain point, I can't write narratives that are more than a given amount of sci-fi with regards to qualia because no one will relate to this book; you need to have characters that are relatable.”
And yeah, something about that also feels related to this lack of really new visions for what we would call transhumanism looks like.
WALKER: Yeah, on that, have you found any literature that does a satisfactory job of capturing a more humane transhumanist vision?
DEMING: Not that I really deeply… I think yes, but maybe the thing that for me does this right now and I think would make no sense to like most people is just Rilke's Duino Elegies, which I don't understand at all. Like, I understand .0005% of these poems. But I think they are trying to say something about something that feels relevant to this. That feels interesting to me.
WALKER: I see. I haven't read them, but can you describe them?
DEMING: So they’re these poems that…I'm trying to remember the opening line. I think it's something like: [“Who, if I cried out, would hear me among the angelic orders?”] Or something like this. I mean, it's almost impossible for me to begin to just understand personally or even describe what they're about. But it's about kind of an individual in some sense trying to connect with something that feels on a different level, I think of what you might call understanding or experience, that they only see very dimly. A theme that often comes up in the poem is angels. Not in even a Christian, or any specific kind of religious, sense. But just as a metaphor for something that is very dimly felt.
And this feels really interesting to me as an analogy for us as humans trying to... I think the interesting thing about the idea of transhumanism, or a related concept like transcendence, is that you're trying to become something that you don't understand. It's like Flatland, like Edwin Abbott. You're like a little 2D thing. And then you're trying to become a 3D object. But you can't even conceive of what that 3D object is in your current 2D form. And that to me is the most interesting, right now, part of transhumanism: is that act and what it means to try and do that.
WALKER: Right. I was trying to think about this question last night myself, about whether there's any good humane transhumanist literature. I couldn't really think of any examples. I thought maybe The Gentle Seduction — the short story by Mark Stiegler.
DEMING: Yeah, so it's really interesting that you... I was like, damn: one of the most interesting sets of questions. Yeah. A colleague of mine, Kat, showed me this story and it's amazing. And I think in a sense the way that our company thinks about technology is something that she's been thinking about a lot and feels really interesting for the company that I run with my co-founder.
But I think for me personally it's a bit of a different flavour… I think even in the story like the types of technologies they're talking about — it's so relatable to our current human experience. It kind of avoids the whole question of how strange the change actually will be. It just says that you should do it gently. Maybe the thing I could agree on with the story is that it's probably good to care about how people feel while you're making changes to what we would call the human experience over time. That makes sense.
3. On scientific awe
WALKER: Is it possible to train the ability to feel scientific awe?
DEMING: To feel scientific awe? I sure as hell hope so. Like, it'd be really sad if not. Yeah, I really hope it is, and I don't know if it is. Yeah.
WALKER: For people who are capable of feeling it, do you know whether it's possible to have it on demand?
DEMING: Yeah, for me at least. Yeah.
WALKER: Oh, that's so interesting... I'm curious that you might be able to conjure it, and I was hoping you could share more about that.
DEMING: Yeah, I have spent an enormous number of hours thinking about that.
WALKER: Oh, really?
DEMING: Yeah. So actually, there was a point in 2018 or 2019 where I had this experience of my friends were out of town for the weekend. I was alone by myself in our house. And I just had this intense absorption into an evolution question that I was thinking about. And I just got really into it, and I remember feeling like, “Oh, I can see the universe. I can see what you might call your conception of God,” or my personal feeling of just seeing being one with the universe. And I was so excited by it. I was like: “I can't forget.”
And so the thing that I did, which, you know, honestly, I'm terrified, might be, like, really bad for my personal health — but I wrote on my hand a number. And it was a number of hours that I wanted to be in that state of just intense awe and absorption. And then every couple of days when it rubbed off, I would rewrite the number on my hand so that I would remember. It was like the most personal kind of tattoo you could imagine, of trying to remember to be connected with this.
I've spent a lot of time iterating on different things. I actually have a page on my website that goes through the things that work for me to get into the state. So for me personally, it's really important to be in some kind of grassy, open environment; to have music playing; to have eaten sugar recently; to have some kind of set of mental objects that are developed enough that they feel like you can interact with them in the state.
I think cell biology is very good for this. But I think if you read Einstein's work and a lot about his early education in high school, I think there's actually this whole tradition in mathematics that Einstein was also exposed to of high school teachers who would ask their students to do very visceral things. I think I read about a math teacher who was telling his students to hold apples in one hand and, you know, some amount in one hand, some amount in the other, and then get an intuitive sense for quantity that way and then use that kind of somatic intuitive sense in their thinking. And I think Einstein's high school had some kind of tie to this kind of visceral... They had some very specific philosophy of education that was related to this.
And yeah, so for me there's just a set of things that work to get into the state. But also there's a lot of preparation, intellectually, to get objects that actually you can then manipulate once you're there.
WALKER: Would it be fair to say that that is the emotion you optimise for in life?
DEMING: Absolutely.
WALKER: Yeah, wow. I don't want to kill the vibe by trying to quantify it too much, but how many hours per week or per month do you think you would spend in that state?
DEMING: So 512 hours since 2018.
WALKER: That's amazing.
DEMING: Yeah.
WALKER: Wow. When you think about the role of scientific awe in doing science, would you characterise it more as a kind of behavioural thing where it's just an important way of helping you maintain motivation and persevere at very difficult long term projects? Or do you characterise the importance as more about helping you actually achieve better insights in the short term by putting you in a state where you're somehow more creative?
DEMING: ...Honestly, I struggle with this enormously. I think when I first encountered the state, I was like, “This is everything; you should be in this state all the time.” And I do optimise like most of my life around being in this state. And at the same time, over time it was like, yeah, like a lot of great scientists probably aren't in that state ever. It's not clear that being in that state solves all your problems…
I think a little bit is certainly very good, but I don't know how much being in that state is going to make you more likely to win a Nobel Prize or something. Probably most people who have won a Nobel Prize have been in that kind of state at some point, I would guess. But you know, honestly, over time it's just like I don't care at all how much...
I think part of the state does feel that it's tied to some idea of truth or something. And I think that feels important to me, the idea that being in this kind of state…
It’s interesting: I've had a lot of experiences that are more… I don't know if “non-dual” is correct, or I don't know if I've ever experienced that, but just like more meditative, let's say, where it's like, “Oh, I feel, you know, great, or I feel like some kind of calm, or I feel like peace.
And this feels different than that. I've always been confused about the difference between these two. This feels much more like there's some laws of the universe that are real and I get to talk to them and they like me or something, and they want to talk and they want to hang out. And I think I just like the state for itself. I don't like it for any functional reason.
WALKER: I see. It's like a hedonistic thing.
DEMING: Totally. Yeah, 100%.
WALKER: That's awesome. And so it sounds like quite a spiritual experience.
DEMING: Yeah. I think at some point reading descriptions of people having intense religious experiences, I was like, “Oh… For some reason my brain is wired so that I experience something that sounds like what they're describing, but in response to reading a physics textbook.”
4. Visualising the cell as an intuition pump for cryopreservation
DEMING: It's incredibly strange that you can cryopreserve and rewarm anything.
Because if you think about it, let's say that you took like a really complicated factory, and you stopped everyone in their tracks in the factory, and so they stopped moving, and then you spun them around and had them walk in a completely random direction and then had them, you know, go back to their normal walking speed — but everyone was walking around in directions — that's what cryopreservation is.
Basically the way that it works (and I'm not sure if it'll make sense to explain it), but it kind of just randomises the molecular motion of all the molecules after you go to very low temperatures. Which is super weird, right? And almost everything that we build at the human scale is incompatible with that kind of… We don't really have systems that are invariant to that kind of randomisation. But cells are for some reason.
And this was immediately obvious… Or it was pretty easy for me to guess why that might be quickly, personally, and I'm sure this has been described also from literature, but because I'd already been doing a lot of thought experiments in the cell, it's very clear when you do that, just like the cells run on passive diffusion — and how strange that is. Like, the cells run on molecules bouncing about in random directions. There's some phases, but it's just like… I don't know if I'm doing justice to the concept. But unlike a computer, where you kind of know that if you start current off here, then it might go here….
In a cell, you just have this bag of molecules that's being shaken all the time and that's how it runs. And nothing is guaranteed to be in a particular place unless it's very bonded to be in that place.
And so that's just so different from how we design objects ordinarily that then when you think about things from that perspective, it makes a lot more sense why cryopreservation is even possible. But if I hadn't spent so much time hanging out in the cell — or this was obviously pretty early on — but I think I wouldn't have really internalised how weird it is that the cells just run on passive diffusion. Does that make sense?
...And it's complicated because you could argue that, well, because we empirically know that you can cryopreserve things, what good is this toy model?
But I do feel like I understand cryopreserving better for seeing the link between this property and the fact that cells evolve to run with very high amounts of thermal noise and therefore have all these properties that make them so good at being invariant to this randomisation that we induce with cryopreservation. And how strange it would be that we would think that would be true for any other system, you know.