Graham Wallas

The Man Who Tried to Make Thinking Visible

There’s something almost paradoxical about the fact that the most widely cited model of creative cognition — preparation, incubation, illumination, verification — was proposed not by a psychologist, but by a political scientist and Fabian socialist who spent most of his career worrying about democratic governance and public education. Graham Wallas wasn’t trying to build a theory of creativity as an aesthetic phenomenon. He was trying to understand how human beings actually think, because he believed the survival of democratic civilization depended on it.

The Problem: Rationalism Without Psychology

To understand what Wallas was doing in The Art of Thought (1926), you have to understand what he was reacting against. The dominant intellectual posture of late Victorian and Edwardian political theory was a kind of naive rationalism — the assumption that if you gave people the right information, they would reach the right conclusions. Political economy, utilitarianism, the whole Benthamite apparatus assumed a model of human cognition that was essentially computational: input data, apply logic, output policy.

Wallas had been embedded in this world. He was a founding member of the Fabian Society alongside Sidney and Beatrice Webb and George Bernard Shaw. He taught at the London School of Economics from its earliest days. He was, in every institutional sense, part of the rationalist progressive establishment. But he became increasingly disturbed by the gap between the rationalist model and the actual behavior of human beings in political life. His earlier book, Human Nature in Politics (1908), was an early and remarkably prescient argument that political theory needed to take psychology seriously — that voters, legislators, and bureaucrats were not calculating machines but embodied creatures driven by impulse, habit, emotion, and suggestion. This was nearly a century before behavioral economics made such observations fashionable.

The Art of Thought extends this concern from the domain of politics into the domain of cognition itself. If rational deliberation isn’t what people actually do most of the time, then what is happening when someone thinks well? What does it look like when a mind moves from confusion to clarity, from problem to insight?

The Four Stages: More Subtle Than the Textbooks Suggest

The model itself — preparation, incubation, illumination, verification — is so widely reproduced that it’s become almost invisible, a textbook diagram drained of its original texture. But reading Wallas directly reveals a thinker far more careful and ambivalent than the schematic versions suggest.

Preparation is the stage of conscious, deliberate work on a problem. Wallas is clear that this is effortful and often frustrating. You read, you gather, you attempt direct attacks on the question. Crucially, preparation isn’t just information-gathering — it’s the construction of a problem space, the work of figuring out what the relevant dimensions of the difficulty even are.

Incubation is the stage Wallas found most fascinating and most mysterious. He drew heavily on the introspective accounts of mathematicians and scientists — particularly Helmholtz and Poincaré — who reported that their best ideas came not during concentrated work but during breaks from it, during walks, in the hypnagogic state before sleep, or while thinking about something else entirely. Wallas was careful to note that incubation isn’t mere rest. Something is happening below the threshold of consciousness. He speculated — carefully, without overclaiming — about unconscious associative processes that continue to operate on the problem material assembled during preparation. This was 1926; Freud was in the air, but so was William James, and Wallas drew more from the Jamesian pragmatist tradition than from psychoanalysis.

Illumination is the moment of insight — the “Eureka” flash. Wallas noted that it often arrives with a distinctive emotional signature: a feeling of certainty, of sudden coherence. He also noted, wisely, that the feeling of certainty is not a reliable indicator of correctness, which is precisely why the fourth stage is necessary.

Verification is the return to conscious, deliberate, critical work. The insight must be tested, formalized, checked against evidence. Wallas understood that illumination without verification is just enthusiasm.

What the textbook versions typically miss is the degree to which Wallas treated these stages as overlapping, recursive, and variable. He wasn’t proposing a rigid pipeline. He was describing a phenomenological pattern that he observed in the self-reports of thinkers he admired, and he was explicit about the limitations of introspective evidence. The model was offered as a heuristic, not a law.

Connections: From Gestalt Psychology to Cognitive Neuroscience

Wallas’s model anticipated or ran parallel to several research programs that would gain traction later in the twentieth century. The Gestalt psychologists — Wertheimer, Köhler, Duncker — were independently developing accounts of insight and productive thinking that resonated with Wallas’s emphasis on incubation and illumination. The concept of “functional fixedness” (Duncker, 1945) describes exactly the kind of cognitive impasse that incubation seems designed to break.

In more recent decades, the cognitive neuroscience of insight has given partial empirical grounding to Wallas’s intuitions. Mark Beeman and John Kounios’s work on the neural correlates of “aha” moments — the burst of gamma-wave activity in the right anterior superior temporal gyrus just before a solution becomes conscious — is, in a sense, a neuroimaging gloss on what Wallas called illumination. Ap Dijksterhuis’s “unconscious thought theory” revisits incubation with experimental methods Wallas would have envied. And the contemporary creativity research of Teresa Amabile, Keith Sawyer, and others often still uses Wallas’s stages as a point of departure, even when complicating or revising them.

There’s also an interesting connection to artificial intelligence and computational creativity. Modern large language models and search algorithms arguably implement something analogous to preparation (training data, retrieval) and a crude form of incubation (latent space exploration, stochastic sampling). But the illumination stage — the subjective, felt sense of an idea clicking into place — remains the hardest phenomenon to model or even define computationally. Wallas’s framework helps clarify why it’s hard: illumination, as he described it, isn’t just an output. It’s a transition in the quality of conscious experience.

What Remains Unresolved

The honest truth is that we still don’t have a satisfying mechanistic account of incubation. Does it work because the mind continues unconscious associative processing? Because rest reduces cognitive fatigue and fixation? Because exposure to unrelated stimuli provides useful analogical material? The empirical evidence is mixed, and the theoretical landscape is fragmented. Wallas’s model identifies a phenomenon — the productive break — without explaining it, and nearly a century later the explanation remains contested.

There’s also the deeper question of whether the four-stage model is a description of creative thought specifically, or of all non-trivial problem-solving. Wallas sometimes wrote as though creative thinking was a distinct mode; at other times, he seemed to suggest that the stages describe any extended cognitive effort. The boundary between “creative” and “non-creative” thought is itself poorly defined, and Wallas’s framework doesn’t resolve it so much as expose its difficulty.

Why This Matters

What I find most compelling about Wallas is not the model itself — which, taken as a rigid schema, is too neat for the mess of actual cognition — but the intellectual posture behind it. He was a political theorist who realized that you can’t think seriously about governance without thinking seriously about thinking. He was a rationalist who took the limits of rationalism as a research problem rather than a reason for despair. He was an introspectionist who understood the limits of introspection. And he wrote about creativity not as a romantic mystery to be celebrated but as a cognitive process to be investigated — with rigor, with humility, and with a clear sense that getting this right had consequences far beyond aesthetics.

In an era saturated with creativity discourse — design thinking workshops, innovation frameworks, brainstorming hacks — Wallas is a useful corrective. He reminds us that the hard part isn’t having a model. The hard part is doing the preparation honestly, tolerating the discomfort of incubation without forcing premature closure, and submitting illumination to the cold discipline of verification. That sequence is easy to diagram and genuinely difficult to live.