The Branching Origins Hypothesis: A Young Scientist's Thought Experiment

Where This Started

My nephew Liam came to me with a question I wasn't expecting to spend three conversations on: what if there were multiple singularities at the beginning of the universe, all descended from one original singularity that exploded into many, which exploded into still more, eventually forming a whole collection of separate universes? He also wondered whether the same chain-reaction, repeated at smaller scales, might explain where the strings in string theory come from.

That second part didn't hold up — string theory proposes strings as more fundamental than particles, not something particles explode into — so we set it aside. What was left was a genuinely interesting question, and I decided to treat it like one instead of just answering it and moving on.

From Explosions to Bubbles to Buckets

The hypothesis went through three versions over the course of our conversations, and each one got structurally closer to real physics.

The first version was a literal chain reaction — one singularity exploding, triggering the next, like a line of firecrackers. That's not quite how any real cosmological model works, so I introduced the closest actual counterpart: eternal inflation, the idea (from Alan Guth and Andrei Linde) that the early universe's inflationary expansion never fully stops everywhere, and pockets randomly settle into calmer, separate universes the way bubbles of steam form throughout a pot of water approaching boiling — not one bubble triggering the next, but many forming independently.

Liam's second revision dropped the triggered-chain problem entirely: a pool of water divides into buckets, each bucket divides into tinier buckets, and those divide again into cups, continuing indefinitely. That version turned out to match something else, closer to home — within a single universe rather than across many. Right after the Big Bang, physicists believe the four fundamental forces were unified as one, splitting into separate forces as the universe cooled (symmetry breaking, with the last split experimentally confirmed). High-energy particle collisions produce a very literal version of the same picture: one particle splitting into several, each splitting again, cascading into a shower of hundreds — recorded in particle detectors every day.

To make the comparison concrete, I had our AI collaborators sketch both cascades side by side, structurally identical — a hypothetical branching of bubble universes next to a real, measured particle shower. The two diagrams look the same on paper. The difference between them isn't visual. It's empirical.

The Argument That Mattered More Than the Hypothesis

The physics comparison was useful, but the best part of this project was what came after, when Liam started pushing back on why any of it mattered if it couldn't be proven.

Round one. He argued that if the math supports an idea, that should be enough — the way 2+2=4 doesn't need an essay written about it. The distinction I gave him: math is a closed system where conclusions follow automatically from agreed-on rules, but a physical theory has to also match the one specific universe we actually live in. You can build mathematically flawless theories that don't describe reality — Newtonian gravity is internally consistent and still measurably wrong near strong gravitational fields. A theory only becomes a confirmed fact about our universe once a prediction unique to it gets checked against observation, the way Eddington's 1919 eclipse photographs confirmed general relativity.

Round two. He came back with a fair challenge: isn't some explanation better than none at all? First answer: an explanation that can never be tested can't be told apart from any number of equally satisfying, mutually incompatible stories — "clouds got heavy and it rained" and "a rain spirit was sad today" both explain the rain equally well if neither can be checked. But he pushed again, and the second answer is the one that stuck: a wrong idea genuinely is more useful than no idea, provided it's testable and you're willing to abandon it when evidence says no. The geocentric model was wrong and still useful for a thousand years of accurate eclipse prediction. Phlogiston theory was wrong, and its specific failures led directly to the discovery of oxygen.

Round three. That settled into the actual standard, borrowed from philosopher of science Karl Popper: a scientific explanation has to be falsifiable — there has to exist some possible observation that could prove it wrong, even one nobody's made yet. Eternal inflation partly clears that bar (it predicts specific, checkable signatures in the cosmic microwave background). The existence of separate bubble universes doesn't, since no observation of one is possible even in principle, ever.

The Takeaway

Liam and I landed on this jointly, and I think it's worth more than the hypothesis itself:

Whether the Branching Origins Hypothesis is right was never really the point. What I actually got to watch was a ten-year-old reinvent, from first principles and mostly in his own words, the falsifiability standard that took professional philosophy of science until the 20th century to formalize. That's the part I wanted written down.

Rob Rainer is Director of Controls & Electrical Engineering, with over 15 years in controls and accelerator operations at Brookhaven National Laboratory's NSLS-II. Liam Jones is Rob's nephew, age 10, and the primary author of the original hypothesis and the questions that shaped this piece. This project used ChatGPT (OpenAI) and Claude (Anthropic) throughout as research and discussion tools — for physics accuracy checks, Socratic pushback, and concept illustrations — with Liam and Rob directing the conversation and reasoning at every step.

Download PDF ← Back to Articles