A sound score can preserve a relation that the written page changes. Silent binding, homophones and resegmentation give the reader more than one way to encounter the same material. Equal pronunciation does not make two words equivalent in meaning. A declaration of equivalence should say whether it refers to sound, spelling, a local symbolic role or an interpretation.

The source’s ear and eye language offers a literary analogy for this difference. It can be described as a poetic quantum experiment. A familiar word broken into suggestive pieces carries the goal of etymology and phonetic transcription. The score distinguishes sound relation or deliberate mishearing through poetic interpretation.

The example declares knight and night alike in one pronunciation convention. Removing k changes the written figure while leaving the declared sound intact. The resulting scene gives that silence a task: the armor is removed before the corridor is crossed. In common English pronunciation the two words already sound alike, so the visible removal changes the named figure without requiring a corresponding change of sound. The guard’s removal of armor gives that small written operation a bodily counterpart, and the last line grants darkness the capacity to hear. This is an author rendition of the operation.

Night Armor

knight ÷ k = night
The guard removes his armor
before the dark can hear him.

Enter two spellings and their pronunciations using the same convention. Add any segmentation and its intended spoken performance. A segmentation such as an ice / a nice can be proposed for reading aloud, but its acceptability depends on pronunciation and context.

The declarations have matching phoneme strings and different spellings. This can change through pronunciation, dictionary or even dialect. Read aloud and record the convention. Binding can also preserve a suffix: treat an inflection as a separate operation rather than dropping it without explanation.