Causal Stance

Abstract: What exactly is the meaning of physical causal closure, a concept frequently discussed in the philosophy of mind? Jaegwon Kim explicitly adopts a conception of causation according to which physical causation is effectively identified with deterministic physical lawfulness, and on this basis equates physical determinism with physical causal closure. While this conception is internally coherent, it differs from the currently dominant theories of causation, which emphasize asymmetry between cause and effect grounded in manipulability and intervention widely employed in contemporary scientific practice. Physics and the theory of causation serve different descriptive purposes, and in this study we refer to them respectively as the Physical Stance and the Causal Stance. Within this framework, physical determinism is a notion that belongs to the Physical Stance, whereas physical causal closure is a notion defined only within the Causal Stance; consequently, the two should not be equated. Since causation is not explicitly defined within the language of physics, physical causal closure is not definable within the Physical Stance alone. By distinguishing between these two stances, this study reconstructs Davidson's Anomalous Monism as a materialist position that consistently acknowledges mental causation without contradicting physical determinism, and examines its relation to the Dual-Laws Model we propose. We further argue that, for the development of scientific theories of mind and consciousness, it is necessary to construct a linguistic framework within which physical causal closure does not hold in the Causal Stance, while physical determinism remains intact in the Physical Stance.

• The paper distinguishes the Physical Stance, which describes deterministic state evolution, from the Causal Stance, which represents asymmetric intervention-based relations, arguing that physical causal closure cannot be defined without causal semantics.
• The paper reconstructs Davidson’s Anomalous Monism by separating strictly deterministic P-Physical Laws from stance-relative C-Physical Laws, resolving the apparent conflict between physical lawfulness and anomalous mental causation.
• The paper uses a Dual-Laws Model to show that mental causes can influence physical events under the Causal Stance while the same system remains strictly deterministic under the Physical Stance, without introducing non-physical forces.

The central thesis

This paper argues that much of the philosophical debate over mental causation rests on a conflation of two distinct descriptive frameworks. Ohmura and Kuniyoshi, following Dennett's notion of stances, distinguish the Physical Stance—the language of physics, which describes state evolution under deterministic laws without distinguishing cause from effect—from the Causal Stance—the language of interventionist causation theory in the tradition of Pearl and Woodward, which is built on manipulability-based asymmetry between cause and effect (2604.05004). Their principal claim is that Jaegwon Kim's exclusion argument depends on identifying physical causal closure with physical determinism, an identification the authors reject as a category error: determinism is a well-defined notion within the Physical Stance, whereas "physical causal closure" presupposes a causal vocabulary that physics itself does not contain.

The paper's strongest and most consequential claim is that the term "physical causal closure" is undefined within the Physical Stance alone. Since physics does not define causation as an asymmetric, intervention-based relation—indeed, reversible dynamics determines past states from future states just as it determines future states from past ones—the closure principle cannot even be stated without importing the Causal Stance. This directly undermines Kim's argumentative strategy, which treats closure as a premise derivable from physics itself.

Physics versus causation theory

The authors ground the distinction in the now-standard observation that physical laws exhibit high temporal symmetry and cannot express the asymmetry that interventionist causation requires. They illustrate this with Pearl's assignment operator: Newton's relation $f = ma$ can be recast causally as $a := f/m$, where $:=$ denotes asymmetric determination by manipulation. The equality sign carries no such asymmetry; reversing the assignment ($f/m := a$) is not generally valid because acceleration cannot practically be manipulated to control force. A thermometer reading tracks temperature but does not cause it—a fact fully consistent with physical law yet unexplained by physics.

Two points follow. First, introducing causal asymmetry does not introduce non-physical forces or energies; it introduces a distinction that physics has no descriptive need for. Second, models indistinguishable under the Physical Stance may be distinguishable under the Causal Stance, since the latter encodes directional information the former discards. Causation theory is therefore not reducible to physics—not because it posits extra ontology, but because it serves a different descriptive purpose.

Reconstructing Anomalous Monism

Davidson's Anomalous Monism comprises three commitments: (1) mental events are causally related to physical events; (2) causal relations conform to strict physical laws; (3) there are no strict laws governing mental events. The authors reinterpret each through the stance distinction. Condition (1) is a claim in the Causal Stance; condition (2) expresses consistency between the two stances; condition (3) concerns mental events as individuated within the Causal Stance.

The reconstruction exposes a latent contradiction in unqualified readings of Davidson. If C-Mental States supervene on C-Physical States and mental laws are anomalous, then C-Physical Laws must also be non-strict, since changes in C-Physical States are influenced by C-Mental Events. Yet condition (2) requires strict physical determination. The contradiction dissolves only if the "physical laws" in (2) and (3) are different objects: P-Physical Laws, defined in the Physical Stance and strictly deterministic, versus C-Physical Laws, defined in the Causal Stance and modulated by mental causes. Without this distinction, physical laws would be simultaneously strict and non-strict. The authors note that Kim's critique of Davidson stems largely from his own conception of causation—effectively identification of causation with deterministic lawfulness—which diverges from the dominant interventionist view; Davidson's separation of causation from laws is, on this reading, more compatible with contemporary causation theory than Kim's.

The Dual-Laws Model

To show that the reconstructed position is coherent, the authors deploy their Dual-Laws Model (DLM), previously introduced elsewhere, explicitly framing it as one possible realization of Anomalous Monism rather than a necessary one. The DLM posits a hierarchical system with a supervenience layer of supervenient functions and a subvenience layer of subvenient entities, each with independent dynamics. Supervenient causes are defined not as the supervenience relations themselves but as changes in equations constructed by selecting and ordering subsets of supervenient functions—a definition deliberately decoupled from the underlying supervenience–subvenience relations so that these causes qualify as genuine interventionist causes. Inter-level causal transmission is implemented via self-referential feedback: error functions computed from the selected equation drive modifications of subvenient states, and causal transmission is unidirectional (supervenience level to subvenience level).

The formal apparatus uses Pearl's assignment operator to encode the asymmetry: the error function satisfies $e := s(c)$, where $c$ is an index sequence selecting and ordering supervenient functions, and index-sequence dynamics $\mathbf{c}_{T+1} = P(\mathbf{c}_T)$ constitute the C-Mental Laws while subvenient dynamics $x^i_{t+1}, err_{t+1} = p(x^i_t, err_t)$ constitute the C-Physical Laws.

The model's dual reading is the crux:

Stance	Interpretation	Consequence
Causal Stance	Index-sequence changes are supervenient (mental) causes influencing C-Physical Events	Mental causation exists; C-Physical Laws are anomalous; C-Causal Closure fails
Physical Stance	Assignment operator replaced by equality; all subvenient states determined solely by subvenient states	Strict physical determinism holds; no non-physical forces required

Under the Causal Stance, the DLM satisfies all three conditions of Anomalous Monism: mental events influence physical events (condition 1), inter-level causation conforms to physical law (condition 2), and both C-Mental and C-Physical laws are anomalous due to inter-individual variation in the supervenience-level dynamics (condition 3). Under the Physical Stance, the same system reduces to $x_{T+1} = f_{c_t}(x_t)$, a closed deterministic evolution. The bold implication is that a linguistic framework exists in which physical causal closure is false while physical determinism remains intact—precisely the framework the authors argue is required for sciences of volition and consciousness that take mental causation seriously.

Limitations and open questions

The paper is candid about several constraints. The DLM is offered only as an illustrative model, not a derivation or unique realization of Anomalous Monism, and the paper's core claim is tied to the stance distinction rather than to any particular model. The formulation in the appendix is abstract—states, bridge functions, index sequences, and feedback dynamics—without empirical instantiation or demonstration on concrete cognitive systems. Davidson himself showed no evidence of distinguishing physics from causation theory in the way proposed, so the reconstruction is an interpretive contribution rather than an exegesis of Davidson's intent. Most significantly, the authors leave genuinely open whether the science of consciousness should adopt the Causal Stance at all: if consciousness is treated as a physical phenomenon like any other, the Physical Stance may suffice, and the choice between stances is deferred to future discussion. The possibility that the language of physics might eventually evolve to explain manipulability-based asymmetry physically is also acknowledged, which would alter the force of the irreducibility claim.

Conclusion

By separating the Physical Stance from the Causal Stance, the paper shows that Kim's equation of physical causal closure with physical determinism conflates vocabularies belonging to different scientific purposes, and that "physical causal closure" is undefined without interventionist causal semantics. Within the Causal Stance, a reconstructed Anomalous Monism—and its concrete realization in the Dual-Laws Model—permits genuine mental causation consistent with strict physical determinism, yielding a framework in which C-Physical Causal Closure fails while P-Physical determinism holds. The practical upshot is methodological: disciplines studying volition and consciousness require explicit stance-relative notation to avoid semantic ambiguity, at least until a common language for the field matures.