Boden's Taxonomy of Creativity
Margaret Boden's signature contribution to cognitive science: creativity is not one capacity but three structurally distinct operations. Exploratory creativity searches within an established conceptual space, producing novelty that the space's rules already permit. Combinational creativity makes unfamiliar connections between familiar ideas drawn from different spaces. Transformational creativity alters the conceptual space itself — changing the rules, boundaries, and definitions of what is possible within a domain. Each mode has different mechanics and is differently affected by AI. The taxonomy transforms the AI-creativity debate from a shouting match into a set of answerable questions, and provides the conceptual infrastructure for understanding what large language models can and cannot do.
The Political Economy of Categories — Contrarian ^ Opus
There is a parallel reading that begins not from cognitive science but from the material conditions of creative labor. Boden's taxonomy, while analytically elegant, abstracts creativity from the economic systems that determine whose creative acts matter and which modes get resourced. The three-fold distinction—exploratory, combinational, transformational—maps suspiciously well onto existing hierarchies of cultural capital. Those who already command institutional authority get to define what counts as "transformational," while those producing exploratory or combinational work are relegated to increasingly precarious positions as these modes are automated.
The taxonomy's real function may be less diagnostic than legitimating. By declaring transformational creativity the "most distinctly human mode," it creates a refuge for elite creative workers while naturalizing the displacement of everyone else. The framework assumes creativity can be cleanly separated from its conditions of production—the training, the leisure time, the cultural inheritance that enables someone to "recognize that a framework is inadequate." But this recognition capacity itself emerges from specific material circumstances. Those doing exploratory creativity in animation studios or combinational work in advertising agencies aren't lacking some cognitive faculty; they're operating within economic constraints that the taxonomy renders invisible. When AI commoditizes their work, the taxonomy offers cold comfort: your creativity was always the lesser kind. The framework thus becomes a sorting mechanism that preserves existing hierarchies while claiming to merely describe natural categories. The question isn't whether machines can be transformationally creative, but who gets to make that determination and what interests that determination serves.
In the AI Story
Boden developed the framework across four decades of research, culminating in The Creative Mind (1990) and refined in Creativity and Art (2010). The distinction emerged from her attempt to apply computational analysis to creative processes that had been treated as mysterious inspiration. She found that much of what gets called creativity is actually exploratory — systematic investigation of possibilities within established frameworks — and that this form is computationally tractable.
The three modes are not merely descriptive categories but have different mechanisms. Exploratory creativity operates through search algorithms navigating formally definable spaces. Combinational creativity operates through pattern matching across domains. Transformational creativity requires something more — the capacity to recognize that the current framework is inadequate and to imagine one that does not yet exist. This third capacity is what Boden identifies as the most distinctly human mode.
The taxonomy has become essential infrastructure for contemporary AI creativity debates because it prevents the conflation of three different phenomena. When an AI system produces novel chess positions, it performs exploratory creativity at superhuman speed. When it generates unfamiliar connections across its training corpus, it performs combinational creativity with extraordinary range. Whether it has demonstrated transformational creativity remains, by Boden's reckoning, genuinely open.
The framework directly addresses the imagination-to-artifact ratio collapse that Edo Segal documents in The Orange Pill. AI's collapse of production cost operates almost entirely within the exploratory and combinational modes. The scarcity that persists — and becomes more valuable as the other modes commoditize — is the transformational capacity to recognize when frameworks themselves need replacement.
Origin
Boden formulated the taxonomy while working at the University of Sussex's Centre for Research in Cognitive Science, where she spent her career. The framework emerged from her attempt to reconcile computational models of mind with the apparent mystery of creative genius. Her central insight was that mystery dissolves when we distinguish what kind of creativity is being produced.
The three-mode distinction first appeared in preliminary form in her 1977 Artificial Intelligence and Natural Man, received its full articulation in The Creative Mind (1990), and has been refined through subsequent decades as AI capabilities have evolved to test the framework's predictions.
Key Ideas
Not one capacity but three. The question 'is AI creative?' conflates three structurally distinct operations, each with different mechanics and different relationships to computation.
Different mechanisms, different affordances. Exploratory creativity is search-like, combinational is pattern-matching across domains, transformational requires framework-breaking capacities machines have not demonstrated.
The taxonomy is diagnostic. It lets us specify precisely what AI systems can and cannot do, rather than issuing blanket pronouncements about machine creativity.
Each mode amplifies differently under AI. Exploratory becomes commoditized, combinational becomes universally accessible, transformational becomes the scarce resource that determines whose work continues to matter.
The framework is not academic abstraction. It provides the vocabulary for understanding the ascending friction phenomenon — what work moves up the cognitive floor when lower-level creative tasks are automated.
Debates & Critiques
Critics have questioned whether the three modes are genuinely distinct or whether transformational creativity can be reduced to sufficiently ambitious combinational work. Boden's response: transformational creativity requires evaluative capacities — the ability to recognize that a framework is inadequate — that neither exploration nor combination provides. The question of whether AI can eventually acquire these capacities remains contested, with Boden herself treating it as genuinely open rather than settled.
Appears in the Orange Pill Cycle
Frameworks and Their Boundaries — Arbitrator ^ Opus
The tension between Boden's cognitive taxonomy and its political-economic critique reveals different questions being answered at different scales. For understanding what AI systems actually do—their mechanical operations and computational limits—Boden's framework proves almost entirely correct (95%). The three modes do capture genuinely distinct cognitive operations, and current AI systems demonstrably excel at exploration and combination while showing no evidence of framework-breaking transformation. The taxonomy's diagnostic precision here is unmatched.
Where the contrarian reading gains force is in examining how these categories function socially (70% weight to the critique). The distinction between modes does map onto labor hierarchies, and those performing "lower" creative modes face immediate displacement. Yet this isn't entirely the taxonomy's fault—it describes cognitive operations that would exist regardless of how we label them. The real insight is that both readings are necessary: Boden gives us the mechanics, while the political-economic lens reveals the stakes. The framework's value lies precisely in making explicit what kinds of creativity AI threatens first, even if it cannot prescribe who deserves protection.
The synthesis suggests reframing Boden's taxonomy not as a hierarchy but as an ecology. All three modes remain necessary for cultural production—transformational creativity without exploratory execution produces nothing, while exploration without occasional transformation stagnates. The question becomes not which mode is "most human" but how to maintain the conditions where all three can flourish. This means protecting exploratory and combinational creators not because their work is transformational, but because the ecosystem requires all three modes. The taxonomy's ultimate value may be in revealing that creativity is not a ladder to climb but a system to sustain.
Further reading
- Margaret Boden, The Creative Mind: Myths and Mechanisms (Routledge, 1990; 2nd ed. 2004)
- Margaret Boden, Creativity and Art: Three Roads to Surprise (Oxford University Press, 2010)
- Margaret Boden, Mind as Machine: A History of Cognitive Science (Oxford University Press, 2006)
- Margaret Boden, Artificial Intelligence and Natural Man (Basic Books, 1977)