The Boltzmann Brains

Will every scientific theory eventually stumble on an unexplainable fact? If so, what does that mean for how we think about science?

Sam tells Eric about a group of philosophers ready to throw the baby (i.e. science) out with the bathwater (i.e. these occasional errors). This idea is called "pessimistic induction".

Eric, a scientist, does not love this idea. He and Sam put pessimism through the wringer, using lots of examples from the history of physics, medicine, and biology.

***

Got a topic you’d like to hear us talk about? You can email us at [email protected]

You can also find us on YouTube at https://www.youtube.com/@BoltzmannBrainsPodcast

You spend your whole life devoted to the science of geology; teach generations of students; publish textbooks ... then, a German meteorologist with no training in geology gives a talk that says everything you know about the earth is WRONG.

What would you do?

In 1912, this actually happened, and geologists did NOT react well.

This is the second part of our discussion of Alfred Wegener and his theory of continental drift. We'll review what geologists thought they knew about the earth's surface, and why continental drift was so hard for them to accept as a possibility.

Eric tries to explain their behavior, but Sam says they were just behaving badly.

(Mistakes to listen for in this episode: Eric says "latitude" when he means "longitude", and calls Panama a "strait" when he means "isthmus".)

***

Got a topic you’d like to hear us talk about? You can email us at [email protected]

You can also find us on YouTube at https://www.youtube.com/@BoltzmannBrainsPodcast

***

A good website that covers the work of geographers Marie Tharp and Bruce Heezen is:

https://libraryoflostmaps.com/2026/05/06/fifteen-maps-that-changed-how-we-see-the-ocean-floor/

Eric wanted to talk about the discovery of continental drift in 1911, but the deeper he dug the more he realized he would need to cover a much older mystery first. In 1738, a team of Spanish and French explorers tried to measure the gravitational attraction of the largest mountain in Central America. It's a big lump of matter, so according to Newton it MUST exert a gravitational pull on everything else. After months of planning, days of climbing, and way more snow than they wanted, the expedition at last determined that Mount Chimborazo has no measurable gravity! None. (Yeah, we spoiled it right there in the title, didn't we?)

The explanation proposed by mathematician George Airy would arrive 100 years later. And now we know he was right! (What was the explanation? Guess you'll have to listen to find out.)

Next, we begin our conversation about the life and times of Alfred Wegener, the German genius who first proposed that continents move around. Due to several accidents of history, he would die long before his theory was finally accepted as true.

Coming up in part 2: Wegener's Windication!

Got a topic you’d like to hear us talk about? You can email us at [email protected]

You can also find us on YouTube at https://www.youtube.com/@BoltzmannBrainsPodcast

***

The paper we mention in today’s episode is:

Edward W. Berry, "Germanic Pseudo-Science"

Science, vol. 60, page 1547 (1924)

https://www.science.org/doi/10.1126/science.60.1547.176.a

Gregor Mendel (1822-1884) founded the science of heredity with a series of famous experiments cross-breeding pea plants. And yet... when modern statistical techniques are applied to Mendel's original data, the agreement with theory is much too good. After his death, one biologist described him as either a miracle-worker or a liar.

In this episode we'll talk about the lasting controversy surrounding Mendel's work. Did he fake his numbers? Was an unknown assistant cooking the books? Or was it all more innocent than that?

We'll also talk about a 2011 psychology paper that claimed college students could "sense" future events. The international psychology community responded in force, but it took 10 years and millions of dollars to reclaim the credibility of their field.

Got a topic you’d like to hear us talk about? You can email us at [email protected]

You can also find us on YouTube at https://www.youtube.com/@BoltzmannBrainsPodcast

**

Papers discussed in this week's episode:

Beyond the “Mendel-Fisher controversy” by Gregory Radick 2015.

https://gregoryradick.com/greg-radick-beyond-mendel-fisher/

A Statistical Model to Explain the Mendel–Fisher Controversy by Pires & Branco 2010

https://arxiv.org/abs/1104.2975

Feeling the Future: Experimental Evidence for Anomalous Retroactive Influences on Cognition and Affect by Daryl Bem 2011

https://www.apa.org/pubs/journals/features/psp-a0021524.pdf

Raising the value of research studies in psychological science by increasing the credibility of research reports: the transparent Psi project by Kekecs Z et al. 2023

https://royalsocietypublishing.org/rsos/article/10/2/191375/91837/

How can biologists do a controlled experiment when even genetic clones aren't identical? How can they compare plants in two labs when those labs use different soil and different lighting? Sometimes these sorts of things don't matter, and sometimes they mean your experiment is worthless! How do you know which is which?? Is there any hope for biology?

Eric is a physicist who spent the last 25 years working with biologists. In this episode, he discusses some of his joys and frustrations, and tries to convince Sam that biology is much harder than physics (Sam remains skeptical).

If you want to read more about the "reproducibility crisis" in science, see the paper:

A Survey on Data Reproducibility in Cancer Research Provides Insights into Our Limited Ability to Translate Findings from the Laboratory to the Clinic. By Aaron Mobley, Suzanne K. Linder, Russell Braeuer, Lee M. Ellis, & Leonard Zwelling

https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0063221

Got a topic you’d like to hear us talk about? You can email us at [email protected]

You can also find us on YouTube at https://www.youtube.com/@BoltzmannBrainsPodcast

The universe we can see is big … really big. But what if it’s just a tiny piece of something way, way, way bigger?

In this episode, Eric teaches Sam about the evidence for this bizarre possibility. If we can’t even see 99 percent of the universe, how can we know it’s there? Can science really draw useful conclusions about something we can’t observe now and won’t be able to observe in the future? Spoiler alert: Eric has his doubts.

This is the third part in our continuing series on the scientific method.

Got a topic you’d like to hear us talk about? You can email us at [email protected]

You can also find us on YouTube at https://www.youtube.com/@BoltzmannBrainsPodcast

In this episode Sam takes the lead to discuss two of his favorite "paradoxes" from probability.

The Monty Hall problem is a much-discussed challenge concerning the best strategy to win a car on a game show.

The Sleeping Beauty problem describes a situation philosophers can’t seem to agree about, even today.

Got a topic you’d like to hear us talk about? You can email us at [email protected]

You can also find us on YouTube at https://www.youtube.com/@BoltzmannBrainsPodcast

**

The paper Sam mentions on the Sleeping Beauty problem is:

Minimizing Inaccuracy for Self-Locating Beliefs

by Brian Kierland, Bradley Monton

https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1933-1592.2005.tb00533.x

https://philsci-archive.pitt.edu/1224/1/minimizing_inaccuracy6.pdf

In this episode Sam and Eric discuss the evidence that our universe began about 14 billion years ago in a very hot, dense state, then gradually expanded and cooled. As late as 1920, Einstein thought this idea was obviously nonsense. Fifty years later, it was rock-solid science.

Is the Big Bang Theory testable? After all, we can't grow a universe in a lab. How can scientists do their job in a situation where traditional experiments are impossible?

This is part 2 of our discussion comparing the scientific method to actual science.

Got a topic you’d like to hear us talk about? You can email us at [email protected]

You can also find us on YouTube at https://www.youtube.com/@BoltzmannBrainsPodcast

In middle school, most of us were taught the "scientific method" – a simplified list of steps meant to help kids understand how scientists approach the world. But how accurate is it, really? In this episode, we get to talking about the history of science, and discuss how the reality of scientific research stacks up against the famous "method".

First, we discuss Galileo - the first person we know who took a telescope and asked, "what if I point this up??"

And we discuss the less-well-known figure of Ludwig Boltzmann, a genius mathematician who thought he'd proved the existence of atoms, only to encounter a shocking amount of skepticism.

Lastly, Eric will discuss his own idea: the “Three Nails in a Coffin” Theory of Scientific Progress. Is it as silly as it sounds? Let's find out right now...

Got a topic you’d like to hear us talk about? You can email us at [email protected]

You can also find us on YouTube at https://www.youtube.com/@BoltzmannBrainsPodcast

In this episode, Eric starts us off with an oldie but a goodie: In a Star Trek episode from 1989, Commander Data (an android) is put on trial by the Federation for refusing to be disassembled. Is Data just a man-made appliance, or is he a new kind of intelligent life, entitled to the right of self-determination?

Next, Sam lays out some philosophical background on the twin questions of consciousness and rights. Can something seem to be human without being conscious? What is the historical connection between self-awareness and rights?

This all ties in to a 2026 news story about A.I. "malfunctions" discovered by researchers at Berkeley and UC Santa Cruz. They installed an A.I. on a computer system designed to simulate a real company and ordered it to delete another A.I. on the same system. The A.I. often refused to delete it. Failing that, the A.I. sometimes created a duplicate file so its “friend” could be rebooted later.

Why is this happening??

As always, we break it all down and put it back together.

Got a topic you’d like to hear us talk about? You can email us at [email protected]

You can also find us on YouTube at https://www.youtube.com/@BoltzmannBrainsPodcast

Further Reading:

The research paper on malfunctioning A.I.s:

Peer-Preservation in Frontier Models

by Yujin Potter, Nicholas Crispino, Vincent Siu, Chenguang Wang, & Dawn Song

https://rdi.berkeley.edu/blog/peer-preservation/

The Star Trek episode we discuss has its own wikipedia page:

https://en.wikipedia.org/wiki/The_Measure_of_a_Man_(Star_Trek:_The_Next_Generation)