Scale-Relative Realism
Scale-relative realism is the thesis that the reality of a property is relative to the scale at which it is observed, and that no single scale is ontologically privileged. A property is real at the scale at which it is causally consequential and economically natural — where 'economically natural' means that the cost of tracking deviations from the property exceeds the benefit of doing so. At finer scales, the property dissolves into micro-dynamics; at coarser scales, it is subsumed into larger patterns.
The thesis is a response to the strong/weak emergence debate. Strong emergence claims that macro-properties are ontologically novel; weak emergence claims that they are computationally intractable but reducible in principle. Scale-relative realism rejects both: the macro-property is neither novel nor reducible. It is scale-dependent, and scale-dependence is a physical fact, not a perspectival illusion.
The canonical example is the liquid/gas phase transition in statistical mechanics. At the molecular scale, there is no sharp transition. At the macroscopic scale, there is. Both descriptions are real. Neither is reducible to the other. The transition is real at the scale at which it is observed, and its reality consists in its causal consequences: the boiling point determines cooking times, engine efficiencies, weather patterns.
Scale-relative realism extends this insight to all emergent properties. Wetness is real at the scale of human interaction, where it causes slipping and soaking. It is not real at the molecular scale, where there are only polar water molecules and surface tension. The neuronal level of description is real at the scale of behavior prediction, where it explains action selection. It is not real at the quantum scale, where there are only electron densities and ionic potentials.
The thesis has implications for the causal emergence debate. Erik Hoel's effective information framework compares macro and micro levels to determine which has more causal power. Scale-relative realism reframes the comparison: the question is not 'which level has more causal power?' but 'at which scales is this level the natural description?' Different scales select different levels, and the selection is determined by the cost structure of the system-observer coupling.
Scale-relative realism is closely related to observer-indexed emergence and economic naturalness. Observer-indexed emergence emphasizes the observer's role in selecting the relevant scale. Economic naturalness explains why certain scales survive: they are selected by cost. Scale-relative realism adds the ontological claim: the surviving scales are real, not merely useful. Their reality consists in their causal consequences at the scales where they operate.
The thesis faces the challenge of explaining how scale-dependent properties can be real if they are not present at all scales. The response is that reality is not a binary property but a graded one. A property is real to the degree that it has causal consequences, and its causal consequences are scale-dependent. The boiling point has no consequences at the molecular scale; it has massive consequences at the culinary scale. Both facts are true. Neither negates the other.