This paper introduces Emergent Pragmatic Coherentism (EPC), a dynamic extension of W.V.O. Quine’s static “Web of Belief.” EPC reframes inquiry as a form of epistemic engineering: the project of building more resilient, less fragile public knowledge structures. To assess this resilience, we introduce a central diagnostic tool, the Systemic Brittleness Index (SBI)—a measure of a network’s vulnerability to collapse based on the real-world costs generated by its core ideas, or Predicates, when tested against the Pragmatic Pushback of reality.
We argue that networks learn and improve their design through the Functional Transformation, a naturalistic mechanism by which successful, brittleness-reducing propositions are repurposed into the network’s core processing rules. This self-upgrading engine, disciplined by the relentless filter of systemic failure, grounds a novel form of Systemic Externalism.
This evolutionary framework reveals that our fallible maps converge on a real, mind-independent landscape of viable solutions—the Apex Network. The result is a synthesized, three-level theory of truth (Contextual, Justified, and Objective) that resolves the classic isolation objection to coherentism. By providing the missing metabolism for Quine’s web, EPC explains how the practical, engineering project of tracking and reducing systemic costs becomes a self-correcting engine for generating objective knowledge.
W.V.O. Quine’s demolition of the “two dogmas of empiricism” replaced the foundationalist pyramid with the holistic “Web of Belief,” a powerful model of how an individual’s knowledge system maintains its coherence. The image is one of structural integrity: a shock at the periphery prompts conservative revisions to preserve the core. Yet, for all its influence, Quine’s web is a static portrait. It masterfully describes the architecture of justification at a single moment but cannot explain how the private webs of countless individuals give rise to the public, objective structures of science, nor how these structures learn and lock in progress. This transition—from a static web to a dynamic, learning architecture—is the central challenge that EPC addresses for post-Quinean epistemology.
This paper confronts this challenge by reframing inquiry not as a search for ultimate truth, but as a project of epistemic engineering: the ongoing craft of building more resilient, less fragile public knowledge structures. This engineering perspective raises a critical diagnostic question: How can we measure the structural health of our knowledge systems? Our answer is a central conceptual tool: the Systemic Brittleness Index (SBI), a forward-looking measure of a network’s vulnerability to systemic collapse. A network that is brittle—one that wastes immense energy on internal maintenance, suppresses corrective feedback, and accumulates hidden fragilities—is a poorly designed one, regardless of its internal coherence.
This diagnostic framework is not an arbitrary overlay; it is the observable output of a deep pragmatic engine that drives epistemic evolution. To explain this engine, we introduce three core innovations. First, we identify the public ‘Predicate’—a reusable conceptual tool—as the fundamental "gene" of cultural evolution. Second, these predicates are tested against Pragmatic Pushback: the non-negotiable feedback from reality that imposes real-world costs and generates systemic stress. Third, we detail the Functional Transformation, the specific learning mechanism by which validated, cost-reducing predicates are repurposed to upgrade a network's core architecture. This provides the missing metabolism for Quine’s web, explaining how networks learn from success to reduce their brittleness.
Ultimately, this engineering framework provides the foundation for a complete epistemological system. This paper demonstrates that the relentless, cost-based filtering of our ideas grounds a robust form of realist pragmatism. This process explains how our fallible maps converge on a real, mind-independent territory—the landscape of viable solutions we call the Apex Network. The result is a synthesized, three-level theory of objectivity that can distinguish between a claim that is merely coherent within a local context, one that is justified for us now, and one that is objectively true. By focusing on the costs and consequences of our ideas, we can build a naturalistic account of how our knowledge systems become self-correcting engines for generating objective knowledge.
The argument will proceed as a systematic construction of this architecture. Section 2 introduces the core diagnostic toolkit, detailing the Systemic Brittleness Index and its key proxies. Section 3 explains the deeper philosophical engine that drives this diagnostic, grounding it in a forward-looking cost calculus. Section 4 builds the full architecture of objectivity, showing how our negative methodology of studying failure gives rise to a positive, three-level theory of truth. Section 5 details the network’s learning mechanism, the Functional Transformation. Finally, the paper will situate this model, defend it against key challenges, and outline the falsifiable research program it makes possible.
To engineer more resilient knowledge systems, we first need a set of diagnostic instruments. A naturalistic theory requires functional, precise tools for measuring the structural health and integrity of a network, moving beyond mere internal consistency; in this, our approach shares a deep affinity with the diagnostic ethos of complex systems theory (Meadows 2008) and resilience studies (Holling 1973). This section forges that toolkit by defining the basic components of inquiry, identifying the sources of systemic stress they face, and introducing a metric for gauging their vulnerability to collapse.
Our model begins with a deflationary move, shifting the unit of analysis from the inaccessible private ‘Belief’ to the public, testable components of knowledge. Within articulated propositions, we find the functional "gene" of cultural evolution: the Predicate. A predicate is a reusable conceptual technology that ascribes a property or relation (e.g., ...is an infectious disease, ...is a conserved quantity, ...must be falsifiable). It is the core informational tool whose deployment has real-world consequences.
These predicates are tested not in isolation, but within Shared Networks: coherent sets of predicates that emerge from the forced, bottom-up convergence of individual agents tackling shared problems. Science, common law, and bodies of practical craft are all examples of Shared Networks. They are the primary environments in which predicates are tested for their viability, retained, or discarded.
A Shared Network is not a passive library; it is an active, problem-solving system under constant pressure from Pragmatic Pushback—the non-negotiable feedback from reality that occurs when a network's predicates misalign with real-world constraints. This feedback is not an argument but a consequence, generating stresses that can be diagnosed as two distinct tiers of cost.