Why Descartes had the order reversed, and why this changes the way we see ourselves

A seemingly small difference

I work as a lawyer; I am a lawyer: this is not a new line of thought, but I want to explore it further because of its neurological impact. It is often discussed in philosophical terms, overlooking the aspect that actually changes the configuration of neurons in our brains.

Two almost identical sentences; the grammatical difference is minimal: the verb changes from “work as” to “be”.

Someone who says “I work as a lawyer” places the profession in the territory of behaviour: something they do, one of the things they do, alongside others. If they stopped tomorrow, they would still be themselves. Someone who says “I am a lawyer” places the profession in the territory of identity. It is not something they do; it is something they are. If they stopped tomorrow, they would lose a part of themselves.

This distinction is not a linguistic game, but the mechanism through which we construct our relationship with what we do, with what happens to us, and with the emotions we experience when things do not go as planned.

The invisible hierarchy

Robert Dilts, developing Gregory Bateson’s work on the logical levels of learning, proposed a model that organises human experience into six interdependent levels. From bottom to top: environment (where and when), behaviour (what I do), capabilities (how I do it), values and beliefs (why I do it), identity (who I am) and purpose (for whom else, for what else).

The principle governing this hierarchy is at once simple and profound: a change at a higher level propagates downwards, influencing all the levels beneath it. A change at a lower level may remain confined there, without touching anything above it (Dilts, 1990).

If I change a behaviour (for example, I stop smoking, start exercising or learn to delegate), I am operating at the second level. The effort is real, results are possible, but the change is fragile because it depends on discipline, consistency and willpower. If the structure above it (beliefs, identity) does not shift, the old behaviour tends to return. Everyone who has tried to change a habit through determination alone knows this: it works for a while, but a pull towards the starting point remains.

If I change my identity (the way I define myself, the role I assign myself, my perception of who I am), behaviours reorganise themselves in a cascade. Not through discipline, but through internal coherence: the brain naturally seeks to align what I do with what I believe I am. Someone who sees themselves as “a sporty person” does not need to force themselves to train. Someone who sees themselves as “someone who never moves” has to fight themselves every morning to go out for a run.

The cost of confusion

The problem arises when the two levels merge and the person stops distinguishing between what they do and what they are.

If I identify with my behaviour, every failure in behaviour becomes a failure of identity. I have not messed up a presentation: I am someone who messes up presentations, or worse still, “someone who cannot give them”. I have not handled a conflict badly: I am someone who cannot relate to others. The emotional response changes radically: in the first case I feel frustration (temporary, contained and manageable); in the second I feel shame (deep, pervasive and paralysing).

Bateson classified the levels of learning with a precision that illuminates this point. First-level learning is a change in a specific behaviour within a specific context: I do A instead of B. Third-level learning is the modification of the systems of alternatives we draw on to respond to the world; a change at the level of identity and role, which generates new ranges of behaviour that did not previously exist (Bateson, 1972).

And here Bateson adds a principle that ought to be printed on the desk of every manager, every coach and every parent: applying a first-level learning solution to a third-level problem is completely ineffective. Telling someone to “behave differently” when the problem lies in how that person defines themselves is like changing the tyres on a car with a broken engine.

The flywheel of identity

When the work shifts from behaviour to identity, something happens that has the characteristics of a generative flywheel.

Someone who changes their self-perception (from “I am not a natural public speaker” to “I am someone who is learning to communicate”) does not change a single behaviour. They change the system from which behaviours emerge. They will begin to prepare differently, seek opportunities to practise, and interpret mistakes as feedback rather than confirmation of their own inadequacy. None of these behaviours has been planned or imposed. They are natural consequences of an identity that has shifted.

The process also works negatively, with the same force. Someone who defines themselves as “someone who cannot change” will produce behaviours perfectly consistent with that definition, finding reasons not to try, sabotaging themselves before they begin, and interpreting every failed attempt as confirmation of what they already believed.

Albert Bandura documented a parallel mechanism in his work on self-efficacy: the perception that we are capable of meeting a challenge is the strongest predictor of whether we will actually meet it (Bandura, 1977). It is not objective competence that determines action, but perceived competence, and this resides at the level of identity, not behaviour.

Descartes had the order reversed

Cogito ergo sum. I think, therefore I am.

It is perhaps the most famous sentence in Western philosophy, and for four centuries it has shaped the way the West thinks about itself: thought comes first, existence follows as its consequence, in a mechanism in which the mind precedes everything else.

In the light of what we now know about how the brain works, the direction might be reversed. Not: we think and therefore we are. We are (with a body, a nervous system, a genetic history, an emotional history, a neural architecture shaped by experiences that precede consciousness), and from this being of ours emerge thoughts and, in turn, behaviours.

Sum ergo cogito et ergo facio. I am, therefore I think, and therefore I act.

The order matters because it changes where we look for causes and where we intervene to bring about change. If thought comes first, thinking differently is enough to change. If being comes first, change requires work at a deeper level: on identity, on self-perception, on those structures that generate thought before thought even becomes conscious.

A more deterministic world than we would like

Robert Sapolsky, a neuroscientist and primatologist at Stanford, took this reasoning to its most extreme conclusions in his book “Determined: A Science of Life Without Free Will” (2023). After decades of research into the neurobiology of human behaviour, his conclusion is unequivocal: our choices are the product of genetics, experience and environment in ways far more determined than we are willing to admit.

Every decision we make, including the decision to “change”, is the result of a causal chain that includes our genes, our hormone levels at that moment, what we have eaten, our childhood experiences, the culture in which we grew up, and thousands of variables of which we are not remotely aware. The sense of agency, that feeling that we ourselves are deciding, is real as a subjective experience but, according to Sapolsky, does not correspond to what happens in the mind. In the briefest terms, the sense of agency is the way we make sense of having already performed that action.

The debate is open and far from resolved: philosophers such as Daniel Dennett have argued for compatibilist positions, according to which determinism and free will can coexist within a coherent framework. The scientific community is divided, and Sapolsky’s “hard incompatibilism” is a position that still provokes considerable debate.

But Sapolsky’s contribution to our discussion does not require us to embrace his position in full. We need only accept the more moderate premise: our inner world is more determined than we tell ourselves; if this is true, working on behaviour (the most superficial, most conscious, most apparently controllable level) means working downstream of a system that has already decided a great deal before our conscious will has come into play.

Going upstream

If behaviour is a product and identity is the factory, then working on behaviour means intervening in the products one at a time: a slow, laborious and fragile process. Working on identity means reprogramming the factory and allowing the new products to emerge on their own.

This is not an abstract concept, but the principle that distinguishes effective coaching from ineffective coaching, leadership that transforms from leadership that corrects, education that shapes people from education that trains behaviours.

So: “How can I change this behaviour of mine?” and “Who must I become for the behaviours I seek to emerge naturally?”. The first question works at Bateson’s first level (error correction, willpower, discipline); the second works at the third (new systems of behaviour arising from a different identity).

Descartes taught us to begin with thought; neuroscience suggests we begin with being. The difference between the two paths is eminently practical: one works with difficulty, and the other works by itself.

The next time you find yourself saying “I need to change this behaviour of mine”, try rephrasing it like this: “I need to change the way I see myself”.