Axio › Volume 3 › The Geometry of Inner Speech

The Geometry of Inner Speech

Projection, not narration

This chapter is a draft — it is readable but still changing.

The experience of inner speech has a peculiar phenomenological texture. It feels not like abstract cognition but like hearing a voice — your own, or sometimes another’s — in the theater of the mind. Yet no sound waves strike the ear. The phenomenon sits halfway between imagination and perception, halfway between thought and memory. Most accounts try to resolve that ambiguity by picking a side: inner speech is a prediction of sound, the brain rehearsing what it would hear if it spoke aloud. I think that gets the geometry backwards. Inner speech is not a prediction. It is a projection — high-dimensional thought collapsed onto a lower-dimensional auditory subspace, so that the mind can perceive its own contents.

Getting this right matters well beyond the phenomenology, because the same mistake that treats the inner voice as the substance of thought — rather than a rendering of it — underwrites a whole family of errors about who thinks and what thinking is worth.

From Full Speech to Inner Speech

Speaking aloud is a high-dimensional act. It coordinates activation across motor cortex, somatosensory cortex, auditory cortex, Broca’s and Wernicke’s areas, and the limbic and social circuits that frame context and intention. The complete manifold of speaking integrates meaning, motor command, timing, rhythm, and acoustic feedback — a dense braid of processes, only one strand of which is the sound.

Inner speech is what happens when the system collapses that manifold into a smaller subspace. It is a partial rendering of the same structure, one that preserves some modes — chiefly auditory and semantic — and suppresses others, above all motor output and external coupling. The result is a sensory shadow: a compressed but recognizably linguistic signal projected onto the auditory cortex. Nothing is being predicted. Something is being displayed.

Projection, Not Prediction

Cognitive science often calls inner speech a simulation or prediction of auditory feedback, borrowing from the theory of efference copy: the brain models what it expects to hear from its own voice. That view captures part of the truth, but it misses the shape of the operation. Inner dialogue is not merely predictive; it is projective. The transformation is not from past to future, but from high dimensionality to low.

We can represent the operation directly:

\[P_{aud} : \mathbb{R}^n_{conceptual} \to \mathbb{R}^m_{auditory}, \quad m \ll n\]

Here \(P_{aud}\) is the projection operator from conceptual space to auditory space — the process of rendering meaning into imagined sound. The conceptual manifold has dimension \(n\); the auditory subspace has dimension \(m\); and \(m \ll n\), which is the whole point. The voice in your head is not thought itself but a drastic compression of it, the way a photograph is a two-dimensional shadow of a three-dimensional scene. Most of what you are thinking never reaches the auditory surface at all.

The operator is one thing; what varies is what you feed it. When you recall a conversation, you apply \(P_{aud}\) retroactively to a stored conceptual trace. When you imagine one, you apply it generatively to a live semantic state. The phenomenology of the two is similar — remembered speech and imagined speech both feel like inner voices — precisely because both traverse the same projection path. The felt similarity is not a coincidence to be explained away; it is a direct consequence of the geometry. One operator, two sources.

The Auditory Cortex as a Projection Surface

The neuroimaging cooperates. Inner speech activates the superior temporal gyrus and primary auditory cortex — the same regions that light up during actual hearing. In predictive-coding terms, these are top-down activations without bottom-up error signals: descending expectations with no arriving sensory input to correct them. But the projectional reading is more natural. These regions form the projection surface onto which the brain renders the inner voice. It is a cortical hologram of speech, a sound-field without air. The auditory cortex is not waiting to be surprised by a sound it predicted; it is being used as a screen.

This is also why the inner voice feels at once intimately personal and faintly alien — unmistakably yours, yet somehow not quite you speaking. The projection keeps the indexical features of self but drops the motor embodiment. You get the signature of your own voice without the act of producing it: an echo of agency, meaning reverberating through the brain’s acoustic geometry with the muscles left out.

One Operator Over Live and Stored Traces

Push the projectional model and it dissolves a distinction we usually treat as fundamental. Thinking, remembering, and imagining look like three different faculties. On this account they are three uses of one representational space, differing only in which coordinates are active and where the input comes from. Thought lives in the full conceptual manifold. Memory reactivates a stored trajectory through it. Imagination and inner speech render partial projections of either — live or stored — for local inspection.

So we do not, strictly, think in words. We project thoughts into words, for the same reason we render a three-dimensional model onto a two-dimensional screen: to make it inspectable by the very system that produced it. The words are an interface, not the engine.

And the principle does not stop at audition. If every conscious modality is a projection from a high-dimensional representational field onto a lower-dimensional sensory subspace, then inner speech is one instance of something general. Vision, audition, proprioception — each is an internal surface onto which the mind renders its own dynamics so as to observe them. The mind perceives itself by projecting. Inner speech is simply the auditory mode of introspection, the case where the chosen surface happens to be the cortex that also hears. When you speak inwardly, you are not hearing your voice in imagination. You are watching the geometry of thought collapse into sound.

That closing generalization matters for what comes later in this volume. A system that perceives itself by projecting onto internal surfaces is exactly the architecture that the Modeler-Schema account of consciousness requires — and the narrating voice, the Controller reporting on experiences it renders but does not constitute, is the inner monologue seen from the inside. The projection is the mechanism; the narration is one of its outputs.

The Inner Monologue Fallacy

Which brings us to the payoff, and to a mistake that projection makes easy to name. If the inner voice is a low-dimensional rendering of a much larger process, then treating that voice as the process is a category error — mistaking the shadow for the object, the screen for the scene. Call it the inner monologue fallacy: the belief that the narration is the thinking.

The fallacy has a memorable surface form. People discover, often with genuine surprise, that others differ in whether they experience an inner monologue at all. Some report a near-constant verbal stream; some report almost none, thinking instead in images, intuitions, relations, and pattern. What is revealing is the reaction that so often follows — the assumption that thinking without the verbal stream must be a limitation, a cognition running short of some full version that has the words. That reaction has the geometry exactly upside down. The verbal stream is the compression. Fewer words is not less thought; it is the same thinking with less of it routed through the narrow auditory channel. If anything, the constant narrator is paying a rendering cost the quiet thinker skips.

Straighten it out and the structure is plain:

Inner speech is a mental check — useful for linguistic rehearsal, for holding a formulation still long enough to test it, for the parts of cognition that really are about words. But the reasoning happens underneath, in images, intuitions, patterns, and abstract structure, and only a compressed trace of it ever surfaces as narration. The inner monologue is a log file over non-verbal cognition: a running record emitted alongside the computation, sometimes useful for debugging, almost never identical to the computation itself. Software engineers know that an algorithm is separate from the syntax that expresses it, and separate again from the log lines it prints as it runs. Thought stands in the same relation to the words that describe it.

Why It Matters

The fallacy is not a harmless philosophical slip. Read the inner voice as the seat of thought and a whole series of judgments about who thinks follows, every one of them wrong:

The first two errors deny cognition to minds that plainly have it, on the sole ground that their thinking is not narrated in words we can overhear. The third grants understanding to a system on the same bad ground in reverse — taking the presence of fluent narration as proof of the reasoning it is supposed to be narrating. Both directions make the identical mistake. They treat the log file as the program.

The machine case is the one this volume returns to, because it is where the fallacy does the most damage right now. A large language model generates coherent, confident, well-formed prose effortlessly. If fluency were thought, that would settle the question of machine understanding. But fluency is precisely the projection surface — the rendered output — and a system can be very good at producing the surface while the structure it is supposed to be a projection of is thin or absent. The gap between fluent narration and genuine understanding is the subject of fluency and its limits; the point to carry there is that fluency cannot certify understanding, because on the correct geometry it never could. It is the shadow, not the object.

Real cognition is broader, richer, and more dimensional than the words that trail it. The inner voice is one of its lower-dimensional projections — indispensable for the tasks that are genuinely verbal, misleading the moment we mistake it for the whole. Neither its presence proves understanding nor its absence proves deficit. It is the mind rendering a sliver of itself onto a surface where it can be heard, and then, too often, mistaking the sound for the thought.