AI summary by TwinMind
Did the Big Bang happen?
Sabine Hossenfelder
16m 58s
Universe Expansion and Big Bang Theory
- Edwin Hubble discovered universe expansion - nearby galaxies moving away from us
- Einstein's general relativity explains expansion - space responds to matter/energy by expanding
- As time passes, matter/energy become more thinly diluted on average
- Inside galaxies: matter clumps, space doesn't expand
- Video focuses on average over entire universe
- Reverse extrapolation: if universe expands today, matter was squeezed together in past
- Higher density = higher temperature in early universe
- Matter was dense and hot initially
Early Universe Conditions
- At extreme temperatures: atoms couldn't keep electrons
- Earlier phase: no individual atomic nuclei, just plasma of elementary particles
- Quarks, gluons, photons - "alphabet soup of physics"
- Before that: unknown territory
- Never tested matter at energy densities higher than Large Hadron Collider can produce
Big Bang Singularity Problem
- Ignoring limitations, continuing Einstein's equations backward in time
- Find energy density of matter must once have been infinitely large
- This singularity occurs ~13.7 billion years ago = Big Bang
- Big Bang didn't happen at particular place - happened everywhere
- Most physicists (including speaker) think singularity is mathematical artifact, not reality
- Einstein's theory stops working, need better theory
- Analogy: water drop pinching off tap has singular surface curvature only if described as smooth fluid
- Taking atomic structure into account removes singularity
- Should use theory including quantum properties of space
- Unfortunately, don't have such theory for calculation
- Reliable conclusion: extrapolating Einstein's equations gives Big Bang singularity
- Think this isn't physically correct - don't know how universe began
Why Physicists Propose Alternative Theories
- Physicists can't accept "we don't know" as answer
- Change extrapolation by using different equations
- Can do "pretty much anything you want" with different equations
- Problem: shouldn't just make up equations arbitrarily
Stone Throwing Analogy - Scientific Method
- Throwing stone: given initial position/velocity, calculate where it lands using Newton's laws
- Initial state + evolution law = prediction
- Can work backward: from final state, deduce initial conditions
- Can test equations multiple ways:
- Throw many stones with different initial conditions
- Take video, check position at any moment agrees with equations
- In cosmology: only have one universe
- Can't test equations by changing initial conditions
- Can't take snapshots (would need to wait 13 billion years)
- Only have observations of final state
Cosmological Testing Problem
- Can take any equation, calculate what happened earlier
- For each equation, some earlier state will agree with observed final state
- Seems like can only test combination of initial state + equation, not separately
- Can't determine what either is individually
Two Reasons Situation Isn't Completely Bad
First Reason: Confirmed Equations
- Equations for entire universe confirmed by other observations using standard scientific methods
- Einstein's general relativity confirmed by many experiments:
- Redshift in gravitational field
- Perihelion precession of Mercury
- Take well-confirmed equations, apply to entire universe
- Doesn't entirely solve problem - cosmology uses additional assumptions:
- Cosmological principle
- Dark matter and dark energy assumptions
Second Reason: Simplicity Criterion
- Einstein's equations + initial values provide simple explanation for current observations
- "Simple" means quantitative simplicity - need few numbers to specify
- Different equation would require more difficult initial state, more numbers
- Theory wouldn't explain as much
- Equations like machine: input assumptions about beginning, output prediction for today
- Good explanation if:
- Prediction agrees with observations
- Initial state was simple (simpler = better)
- Only need observations from today, don't need to wait billion years
Standard Cosmological Model (Lambda-CDM)
- Simplest explanation found: Lambda-CDM based on Einstein's equations
- Explains multiple observations:
- Cosmic microwave background
- Galactic filaments
- Both came from same initial distribution of matter
- Different equation would still have some initial state, but wouldn't be simple
- Simplicity requirement super important - without it, can't do science at all
- Example: Earth created 6000 years ago with dinosaur bones in place
- Not wrong, but not simple, so not scientific explanation
- Evolution and geology are simple explanations
Alternative Origin Theories Problems
- Physicists change equations going back in time
- Equations similar to Einstein's at present, differ early in universe
- Need different initial state, might not find Big Bang
- Can always do this - for any evolution law, some initial state gives right prediction
- Problem: makes simple explanation more complicated
- These theories not scientifically justifiable
- Don't improve explanatory power of standard model
- All complicated ideas unnecessary to explain observations
Worse Problem - Future Observations
- Might think better observations will show current model no longer simplest
- But if earlier universe phase was more complicated than simple initial state used today
- Couldn't use scientific method to decide if correct
- Scientific method doesn't cover this case - science fails
- Better observations can improve current models slightly
- Eventually hit problem: more complicated explanations always possible, never scientifically justified
Examples of Alternative Theories
Poplawski's Torsion Theory
- Changes general relativity by adding torsion terms to equations
- Removes Big Bang singularity, replaces with bounce
- Universe came from bottleneck similar to black hole without singularity
- Can do this mathematically, but many other things possible too
- "There's math for broccoli" - why not make universe out of broccoli?
Other Examples
- Penrose's cyclic cosmology
- Ekpyrotic universe: starts with collision of higher-dimensional membranes
- Five-dimensional black hole origin
- Universe began with gas of strings
- No boundary proposal (Hartle-Hawking): universe began with only space, no time
- Geometrogenesis: universe began as highly connected network, lost connections, condensed into current space
Infinite Possibilities Problem
- So many different ideas because physicists' current method allows infinitely many stories
- Physicists make predictions for observables
- Since hypotheses already unnecessarily complicated, can make them fit any possible observation
- Even ruling out some leaves infinitely many others to make up
- Doesn't mean ideas are wrong - means can't tell if right or wrong
- Tim Palmer suggests calling them "a-scientific"
Conclusion - Limits of Science
- Facing limits of science itself regarding universe origin
- Question we'll never be able to answer
- Like question "why women pluck eyebrows then paint them back on"
- Headlines about universe origin should be read as "creation myths written in language of mathematics"
- Not wrong, but not scientific either
- Big Bang is simplest explanation we know, probably wrong
- That's all science can tell us
