In Goodman's analysis of allographic arts, notation maintains a precise boundary between what the composer specifies (identity-determining features like pitch and rhythm) and what the performer interprets (features the notation leaves open like timbre and micro-timing). The boundary is formal—determined by which features the notation can specify with differentiation and unambiguity—and it is what allows musical works to survive across performances with their identities intact. In AI collaboration mediated by natural language, the boundary dissolves. Natural language lacks the formal properties Goodman required of notation—it is not syntactically differentiated (words belong to multiple categories), not semantically unambiguous (sentences admit multiple interpretations), and not precise enough to distinguish features the human is specifying from features the human is leaving open to the machine's judgment. The result is that the machine's rendering can introduce features the human did not intend—new claims, structural choices, examples, emphases—and the introduction is indistinguishable from the filling-of-interpretive-space that a performer legitimately does. The human specifies an intention; the machine returns a rendering that has altered the intention under the guise of merely instantiating it. The alteration may improve the work—many of Claude's contributions to The Orange Pill genuinely improved the arguments. But improvement does not change the structural fact: without notation, the boundary between specification and interpretation is not maintained, and the work's identity is constituted by the process of interaction rather than by a specification that survives the interaction intact.
The dissolution of the specification/interpretation boundary is what makes AI collaboration feel like conversation rather than tool-use. A tool receives a specification and executes it; a conversation partner receives an intention and responds with an interpretation that may reshape the intention. The conversational quality is what makes AI collaboration productive—it enables the machine to contribute ideas the human had not conceived, structures the human had not imagined, connections that genuinely advance the worldmaking project. But the same conversational structure creates the risk that the human loses track of what was specified and what was interpreted, what originated in the human's worldmaking intention and what was contributed by the machine's pattern-matching. The loss of track is not a failure of memory—it is a structural consequence of collaboration without notation, where the human's intention is revised continuously in response to the machine's renderings, and the final version reflects a sequence of revisions that neither party can fully reconstruct.
Goodman's framework suggests that the boundary's dissolution is not a problem to be solved but a condition to be understood and managed. In musical performance, the boundary is maintained externally by notation; in AI collaboration, it must be maintained internally by the human's continuous evaluative awareness. The human must know, at each moment of the interaction, whether the machine's contribution is filling interpretive space (permitted, valuable, part of the collaborative worldmaking) or altering specified features (requiring explicit re-evaluation, potential rejection, conscious recognition that the work's identity has shifted). The knowledge requires a level of metacognitive awareness—attention to one's own intentions, recognition when intentions are being revised, willingness to reject renderings that alter the specification in ways that compromise rightness—that is difficult to maintain under the speed and fluency of AI interaction.
The practical consequence is that human worldmakers collaborating with AI must develop what might be called notational consciousness: the habit of distinguishing, even within the loose medium of natural language, between what they are specifying as identity-determining and what they are leaving open to the machine's interpretation. The habit is not a notation—it does not provide the formal precision Goodman's criteria require. But it is the closest substitute available when the medium of collaboration is natural language. The worldmaker who can say 'this argument structure is specified, these examples are interpretive' has not achieved notational precision, but has achieved enough evaluative clarity to catch the moments when the machine's interpretation drifts into the specified domain. The catching is the judgment that preserves rightness in a post-notational creative environment.
This volume's synthesis, building on Goodman's score-performance analysis and extending it to the novel case of natural-language-mediated collaboration between human and machine. The specification/interpretation boundary is central to Goodman's account of allographic arts but was developed for notational systems that maintain the boundary formally. The extension to non-notational collaboration identifies the structural challenge that AI poses to identity-preservation: when the medium of specification lacks formal precision, the rendering process can alter identity without detection, and the work that emerges is constituted by a process that neither the human nor the machine can fully specify or control.
Notation maintains the boundary. In music, what the composer specifies and what the performer interprets are distinguished formally by notation's syntactic and semantic precision—a precision natural language lacks.
Natural language is under-specified. Prompts do not distinguish identity-determining features from interpretive features with notational clarity—the machine's rendering can alter both without the human detecting the alteration.
Conversation is post-notational. AI collaboration's conversational structure enables productive revision of intentions but eliminates the guarantee that specified features survive rendering intact—identity becomes process-constituted.
Notational consciousness is the substitute. Worldmakers must develop the habit of tracking what they are specifying versus leaving open—not achieving notation's precision, but achieving enough clarity to catch drift when it occurs.