Definition
Read this first. The position has two parts, and only the first is held. Indexed judgments can be objective once their relata are fixed (CLAIM-363). Realism proper also needs the third condition below, that which frames, perspectives and standards are admissible is itself constrained, and nothing in the record meets it (QUESTION-129). Using the term does not claim that normative realism has been established.
The owner, at U18: “so maybe something like a frame/order/scale/ perspectival-relative normative realism?” A18: “I’d use relational normative realism as the short name, and your fuller formulation—frame-, order-, scale-, and perspective-relative normative realism—as its explicit technical characterization.”
A normative judgment is written, after A18, J(a; F, O, S, P, N): the action, state or policy a, judged in
- a frame F, “the variables, distinctions, and relevant counterfactual comparisons”;
- an order O, which A18 splits in two: “If you mean first-order versus higher-order attitudes, that is different from organizational level. Both matter to our account, so I would distinguish organizational order from reflective order rather than silently combine them.” The owner did not choose. This term keeps both, as separate indices;
- a scale S, “the spatial, temporal, or resource resolution of evaluation”;
- a perspective P, “whose interests, commitments, reasons, or standing are relevant”;
- a standard N.
A18 calls the schema “a schema for analysis, not yet a theory of normative truth”.
The position is realist in A18’s sense when three conditions hold:
- “truth conditions independent of the current judgment.” “Once the relevant relations and standards are specified, an agent can be mistaken about whether an action satisfies them.”
- “counterfactual discipline.” Verdicts answer “to evidence about what would happen under relevant alternatives.”
- “constraints on admissible perspectives.” “We cannot simply let an evaluator choose whichever frame makes a preferred verdict true.”
A18: “That third condition is the difficult one. The first two can establish objective conditional normative judgments without yet establishing objective authority for the conditions themselves.” The term is used with all three. A claim that meets only the first two is indexed judgment, which is CLAIM-363. Whether the third can be met is QUESTION-129.
The distinction the term draws is between a judgment’s dependence on its relata and its dependence on the judge. It admits the first and denies the second.
What it is not
- Not relativism. Relativism, in A17’s table, holds that “No context-transcending standard can adjudicate between standpoints”. The term does not say that. A18 names the challenge from the other side, from Wong’s pluralistic relativism: “why should our position count as realism rather than a species of constrained relativism?” Until the third condition is met the challenge stands, and the position may be constrained relativism under another name.
- Not Frankfurt’s view. NOTE-264 records Frankfurt “Against normative realism”: “I do not believe that anything is inherently important”, and the objectivity of normative standards “consists just in the fact that it is outside the scope of our voluntary control”. That grounds standards in what agents cannot help caring about. The term’s third condition asks for more than that, and Frankfurt denies it can be had.
- Not frame-invariance. A18: “Normative objectivity need not mean invariance under every change of perspective. It may mean invariance under changes that preserve the normatively relevant structure.” And: “what makes a transformation normatively admissible cannot be read off its formal structure alone.”
- Not a derivation from physics. A18’s velocity analogy “is methodological, not a derivation of moral realism from physics”, and the gauge reading of the indices is QUESTION-130, open.
- Not A21’s “multiscale relational normative realism” as a separate term. A21 and A22 use that name for the same position with scale stressed. It is this term.