This paper develops a hierarchical account of concept formation that distinguishes the manifestation of originary qualitative feel from the public, norm-governed concepts through which subjects later articulate experience. Its central claim is not merely that concepts differ in abstractness, but that they differ in constraint profile: some remain tightly corrigible by recurrent felt manifestations delivered through relatively stable experiential channels, whereas others depend far more heavily on social reweighting, comparison, and narrative organization. To capture this difference, I distinguish four strata: (1) the manifestation of originary qualitative feel, (2) channel-bound concepts, (3) integrative judgments, and (4) narrative concepts. The pivotal case is smart. In some contexts, smart functions as a third-level integrative judgment anchored by a relatively stable cue-cluster such as rapid learning, flexible problem solving, and quick comprehension. In other contexts, the same lexical item is drawn into fourth-level use, where application depends on broader evaluations of which forms of knowledge, achievement, or tradition count as genuinely worthwhile. This shift shows that semantic drift does not abolish embodiment; it redistributes which embodied, affective, and inferential resources are recruited as operative sense changes. The paper then reorganizes its dialogue with contemporary theory around this problem of functional migration. Barsalou and Dove help explain embodied anchoring but do not sufficiently account for the reweighting of application rules within public evaluative space. Borghi and collaborators illuminate the role of language, sociality, inner speech, and conversational coordination in abstract concepts, yet their frameworks can be sharpened by distinguishing lexical stabilization from discursive reorganization and by connecting internal cue-stabilization with sentence-level and discourse-level renegotiation. Brandom and Jorem and Lohr clarify the public inferential dimension of concept use, but require a stronger account of differential experiential corrigibility. Searle helps explain how high-level concepts approach recognition and status, but not every narrative concept is thereby reduced to a full institutional fact. The conclusion is that the decisive issue is not whether a concept is social, but how lower-level cues, public norms, and narrative frames are combined in use.