Anomalous Observers in the Subjectively Identical Reference Class

TL;DR

This paper examines the implications of including various types of observers, such as simulations and stored data, in the reference class for anthropic reasoning, raising concerns about the consistency of the approach.

Contribution

It critically analyzes the criteria for defining the reference class in anthropic reasoning, highlighting the problematic inclusion of anomalous observers.

Findings

Including stored or replayed observers complicates anthropic probability assignments.

No clear criteria exist to exclude certain types of observers from the reference class.

The presence of anomalous observers challenges the validity of anthropic reasoning.

Abstract

Anthropic reasoning is a critical tool to understand probabilities, especially in a large universe or multiverse. According to anthropic reasoning, we should consider ourselves typical among members of a reference class that must include all subjectively indistinguishable observers. We discuss here whether such a reference class, which we assume must include computer simulations, must also include computers that replay previous simulations, magnetic tapes that store but do not "run" the simulation, and even abstract mathematical functions. We do not see any clear criterion for excluding these anomalous observers, but their presence is deeply troubling to the idea of anthropic reasoning.