Intentionality as Architecture: A Structural Account of Cognitive Organization

Contemporary artificial intelligence systems achieve increasingly sophisticated behavioral performance while remaining limited at the level of cognitive organization. They can generate coherent outputs, adapt to context, and support stable interpretation, yet they lack the architectural conditions required for sustained, self-modifying, and developmentally coherent cognition. This gap reflects a deeper theoretical limitation: dominant accounts of intentionality continue to locate its source in consciousness, interpretation, or evolutionary history, rather than in the present organization of cognitive systems. This paper argues that intentionality should be understood as an architectural phenomenon. It develops the framework of Evolving Cognitive Architectures (ECA), which specifies a set of organizational conditions under which systems can sustain directed, context-sensitive, and developmentally structured activity. These conditions—organizational continuity, dynamic circulatory memory, representational plasticity, non-arbitrary internal semantics, and developmental plasticity—function as structural constraints on systems capable of genuine cognitive agency. The paper makes three contributions. First, it provides a diagnostic critique of influential approaches associated with John Searle, Daniel Dennett, and Ruth Millikan, showing that each identifies correlates, diagnostics, or stabilizing mechanisms rather than constitutive conditions of intentionality. Second, it advances a positive structural account that relocates intentionality at the level of organized cognitive dynamics. Third, it demonstrates how this framework enables a principled reassessment of contemporary AI systems by shifting evaluation from behavioral performance to architectural capacity, thereby clarifying the conditions required for the development of systems with AGI-level cognitive organization. By reframing intentionality as an achievement of system organization rather than a property of substrate, interpretation, or history, the paper positions intentionality as a phenomenon that can be structurally characterized, comparatively analyzed, and, in principle, empirically investigated across both biological and artificial systems.