How do readers actually build meaning from words?

The dominant theoretical inheritance for "what is meaning" runs through compositional semantics: the meaning of a sentence is a function of the meanings of its words and their syntactic combination. The theory has its uses, but it under-describes how meaning is actually made in live language use. The bullseye example — where three of four words activate an archery frame while the fourth sits adjacent but unselected — cannot be derived compositionally. Compositionality treats all four words as equally contributing; the meaning emerges from selecting some and suppressing others.

The alternative formulation: meaning is the live detection of resonance. The mind reading a passage does not aggregate word-meanings into sentence-meaning. It scans for frames that some subset of the available words activates together, and the meaning that emerges is the one carried by the strongest-resonating frame. Other words present in the passage are not aggregated into the meaning — they are passed over. The selection is the meaning-making operation; the meaning is what the selected subset adds up to within the activated frame.

This formulation has three implications. First, meaning is not a property of the linguistic surface alone — it is a relation between the surface and the cognitive operation of frame-activation. Different readers can extract different meanings from the same surface depending on which frames they activate. Second, meaning is non-additive — adding more words to a passage does not necessarily add more meaning, and sometimes adds noise that interferes with frame-activation. Third, meaning is non-monotonic — recognizing one frame can suppress others that the same words could have activated.

This bears directly on AI meaning-making. Why do AI systems miss jokes and wordplay so consistently? specifies the AI failure: parallel-aggregative computation does not perform the selective frame-activation that resonance requires. Does the mind selectively activate frames from only some words? specifies the human operation. The theoretical claim here unifies them: meaning is constituted by the selective-resonance operation, not by the linguistic material on which the operation acts.

The strongest counterargument: this is just frame semantics restated. Frame semantics is a major influence, but the resonance formulation adds the specific claim that meaning-making is the live act of selection within the available frame-space — not the application of a selected frame, but the selection itself. The selection is the cognitive work that produces meaning.

Source: Making Sense - brief for co-authored essay on language