The Differential Multiplicity Principle
Imagine that every possible universe is copied a thousand times.
Nothing about their relative frequency changes. If half the observer measure was in one class before the copying, half remains there afterward.
Now imagine that only some universes are copied—specifically, the universes whose laws make it easy for their inhabitants to construct observer-containing worlds.
The relative frequency changes.
Differential Multiplicity Principle: multiplicity changes anthropic probabilities whenever the process producing multiplicity correlates with production of the observer measure relevant to the inference.
Multiplicity is neutral only under an independence condition.
Neutrality is the special case
Let q(theta) be the normalized seed weighting over laws supplied by a particular model, after conditioning on the evidence used at this stage. Let M(theta) be its world-multiplicity factor and F(theta) its capacity to produce descendant observer measure.
Neutrality requires M not to favor laws on the basis of F under that specified seed distribution. In the simplest linear setting, zero seed-weighted covariance is a useful diagnostic:
Cov_q(M(theta), F(theta)) = 0
Zero covariance alone is not a general definition of conditional independence. The substantive requirement is that the multiplicity mechanism supply no information about observer fertility once the relevant model and evidence are fixed.
Uniform multiplication leaves normalized anthropic odds unchanged.
But if world-production favors laws with unusually high or low reproductive capacity under q, the multiplication mechanism is part of the evidence model.
This is why a multiverse cannot simply be declared irrelevant to simulation odds. Its relevance depends on whether the way it populates physical laws correlates with the later production of simulated observer-histories.
Laws participate in reproduction
Physical laws affect whether stars form, complex chemistry persists, intelligence evolves, civilizations survive, computation is affordable, and constructed minds are possible. They also affect what kinds of child worlds those civilizations can implement.
Using the reproduction kernel from the previous essay:
K(theta_i, theta_j)
= observer-history measure with child laws theta_j
produced by worlds with host laws theta_i
The off-diagonal terms matter. A host is not limited to reproducing its own physics.
Selection over the genealogy of realities
Repeated construction can make some law sets common in the observer-weighted descendants even if they were rare among the initial worlds.
This resembles Lee Smolin’s cosmological natural selection, in which black-hole production is proposed as a reproduction mechanism for physical universes with inherited parameters.1 But the mechanisms and the populations are different.
Smolin:
laws -> black holes -> daughter universes -> selection among physical universes
Constructed worlds:
laws -> intelligence -> implemented worlds -> selection in observer measure
The second process need not change the distribution of base universes at all. It changes the distribution sampled by observers across a genealogy containing both upstream and downstream worlds.
No automatic privilege for the root
If the descendant process is productive, most observer measure might lie far from its initial conditions. If reproduction is costly, rare, unconscious, or convergent, it might not.
Nothing in the word base settles the balance.
An upstream world’s ontological priority gives its observers no automatic anthropic priority.
Any preference for base reality has to emerge from the prior over worlds, the limits of implementation and consciousness, reproduction costs, or the behavior of the kernel. It cannot be inserted by naming one layer fundamental.
This brings us to a more disciplined version of the simulation question. We no longer ask only whether civilizations can make simulated people. We ask which worlds produce persistent observer-histories, at what cost, under which laws, and through how many generations.
Only after that foundation should we ask why a civilization would pay for an expensive world at all.
-
Lee Smolin, “The Status of Cosmological Natural Selection” (2006), reviews the proposal and its possible observational tests. The analogy here concerns differential reproduction, not a claim that black-hole offspring are simulations. ↩
Filtered post view · schema v1
Argument map
Loading the claims, evidence, objections, and assumptions behind this post…
- Supports / entails
- Attacks / contradicts
- Structural relation
- Question
- Claim
- Hypothesis
- Evidence
- Objection
- Assumption
- Conclusion
Drag to pan · scroll or pinch to zoom · select a node for details