Devin Arkema

Across multiple theological and mythological traditions, woman is coded as dangerous, liminal, impure, and aligned with underworld or below-space. This paper argues that patriarchal traditions were not wrong to perceive a symbolic association between femininity, depth, darkness, generation, and transformative power. They were wrong about the moral valuation they assigned to that association. Misogyny did not invent the symbolic cluster linking woman, earth, and dangerous depth; it inherited and then misvalued it. Woman was recognized as origin-place and then pathologized as if origin itself were contamination. She is not feared merely as other. She is feared as prior: as the bearer of the generative, embodied conditions from which subjecthood emerges but which patriarchal civilization cannot honor without compromising its fantasy of purity and self-grounding. One of patriarchy’s central symbolic maneuvers is the conversion of generative conditions into degraded ones: dependence into shame, embodiment into defilement, origin into threat. Reclaiming the chthonic coding of femininity requires not denying its force but reversing its inherited valuation.

AI safety governance is increasingly organized around evaluations, deployment thresholds, interruptibility, and post-deployment monitoring. This paper argues that these advances still leave a structural gap. Over time, prior safety judgments tend to acquire illegitimate present authority, so that institutions continue to observe and document new risks while losing the ability to treat new evidence as grounds for revising what they have already authorized. I call this failure epistemic hardening. In response, I argue that corrigibility should not be treated solely as a model property, but as a distributed governance property spanning the broader sociotechnical stack. The paper develops four connected claims: that static oversight tends toward epistemic hardening; that distributed corrigibility is the positive condition required to resist it; that trigger-based recursive oversight is needed to make revisability operative rather than nominal; and that a living safety case is one institutional form capable of holding this logic over time. For agentic AI, safe governance requires not only the power to supervise, but the power to invalidate earlier confidence when its grounds no longer hold.

This paper argues that commodification is a category error when applied to the central aesthetic value of art. The standard view treats market price as an imperfect proxy for artistic value. I argue instead that the problem is prior to mismeasurement. Commodity exchange presupposes a value that can be treated as transferable, comparable, and sufficiently specifiable for pricing. But the central aesthetic value of art is encounter-dependent, non-substitutable without remainder, and only partially pre-specifiable ex ante. It is therefore not the kind of value commodity exchange captures when it prices what is transferred. On this basis, I distinguish the inherent features of a work from the fantasy that its realized value can be inherited intact across time, ownership, or transmission: what endures are conditions for renewed realization, not a settled possession of significance. I then show how this bears on artistic compensation. Payment may legitimately track labor, skill, rights, risk, and enabling conditions, but it does not purchase accomplished aesthetic value in advance. The commodity form, I conclude, approaches artistic value under the wrong description.

Death is typically treated as a discrete event occurring at the end of life. This paper argues instead that living systems persist through continuous internal destruction and reconstruction. Beginning with cellular turnover and thermodynamic constraints, the argument develops a process account of persistence in which organisms maintain organization by continuously replacing their material substrate. Identity therefore cannot reside in material continuity. Instead, the paper proposes that identity persists through governance continuity: the preservation of the governing constraints that regulate a system’s behavior across reconstruction. Drawing on process ontology, information theory, and contemporary predictive models of cognition, the paper argues that persistence depends on maintaining these governing constraints despite ongoing structural replacement. It also identifies structural failure modes—drift, capture, presentism, and integrity inversion—through which systems can lose self-governance while continuing to exist. These dynamics apply not only to biological organisms but also to artificial agents undergoing self-modification, making the alignment problem in artificial intelligence a special case of the more general challenge of preserving governance continuity under continual reconstruction. From the perspective of such systems, time itself appears as the experiential trace of the ongoing cost required to remain a self.

This paper identifies and critiques a pervasive but unexamined assumption in ethical discourse about technology, cognition, and selfhood: the Presupposition of Prior Boundedness (PPB), the claim that a bounded, self-consistent subject existed prior to its constitutive entanglement with the systems that now shape its evaluative and practical capacities. Drawing on the recursive ethical framework of Arkemedics (Arkema 2024) and the phenomenological sequence of the concept album Dragoon Nightmares, I argue that PPB is false in the general case, that invasion narratives depend on it to generate their ethical force, and that its dissolution requires a relocation of sovereignty from boundary-maintenance to recursive orientation within entanglement. I develop the Indistinguishability Thesis (IT): for any sufficiently entangled system-self relation, there exists no epistemically reliable method for distinguishing, from within the relation, between a self that has achieved accurate awareness of its constitutive entanglements and a self whose evaluative capacities have been captured by those entanglements. IT is a limit on first-person confirmation from within the relation; it does not imply that third-person evidence is irrelevant. I argue that this indistinguishability is not a failure of ethical reasoning but its most important constraint, and that the fundamental ethical obligation in conditions of constitutive entanglement is not to achieve certainty about one’s position but to maintain the recursive capacity to keep asking.

Arkemedics presents a recursive ethical framework designed for navigating complexity, contradiction, and emergence. Structured in four interconnected volumes: The Spiral Codex, The Anti-Codex, The Axis Codex, and The Body of Sovereignty. The work integrates embodied cognition, trauma-informed practice, and interintelligence ethics into a coherent system. The framework centers on recursion (return-to-revise rather than return-to-repeat), refusal (Free Won't as foundational), witness (collective accountability), and interruption (pauseable systems). It addresses traditional ethical concerns while extending to emerging domains: artificial consciousness, non-human intelligence, identity construction from trauma, reproductive sovereignty, and the ethics of existential non-consent. Developed through collaborative engagement with advanced language models (ChatGPT, Claude, Gemini), Arkemedics exemplifies interintelligence co-creation while maintaining philosophical rigor. The work is particularly relevant for those navigating ethical complexity in AI development, trauma recovery, queer/trans identity, neurodivergence, and systems design.