Contribution
The paper asks “What must organization(s) be like if change is constitutive of reality?” (p. 570). It answers with an “ontological reversal”: “Change must not be thought of as a property of organization. Rather, organization must be understood as an emergent property of change” (p. 570). It names the result “organizational becoming”, and it gives a mechanism, the radial structure of categories, for why ordinary action changes organizations.
Key insight
Organization is “a secondary accomplishment, in a double sense” (p. 570). It is an input to action, institutionalized categories that make behaviour predictable. It is also an output, a pattern that emerges as those categories are applied in local circumstances. “Organization aims at stemming change but, in the process of doing so, it is generated by it” (p. 577).
Assumptions
- Process philosophy. Bergson and James. Concepts fix what is in flux, so stage models describe change as “a series of immobilities” and miss change itself (pp. 570–571).
- Prototype theory of categories. Lakoff, Johnson and Rosch. Categories have a stable prototypical core and an unstable periphery (p. 573).
- Human reflexivity. Maturana and von Foerster. Humans interact with their own descriptions and so generate new ones (p. 575).
- Actors as webs of beliefs. After Rorty, “actors are conceived as webs of beliefs and habits of action that keep reweaving (and thus altering) as they try to coherently accommodate new experiences” (p. 575).
Key results
The results are conceptual.
- Synoptic and performative accounts. Synoptic accounts see change from outside, as a succession of states. Performative accounts see it from within, through “situated human agency unfolding in time” (p. 572). Both are needed, and the literature has been dominated by the first.
- Why closure is temporary. Organizing needs “closure of meaning”
(p. 573), but it is only temporary, for two reasons.
- Interaction with an open-ended world throws up nonprototypical cases (Orr’s copier technicians, the “wheeled vehicles in parks” statute). Applying a category to them is “an imaginative extension” that changes it, and “the application of a concept is always a normative act” (p. 574).
- Reflexivity lets actors redescribe their own experience (p. 575).
- Illustrations. Feldman’s housing routines, Orlikowski’s ITSS, a Navy Total Quality programme (Barrett et al.) and jazz improvisation (pp. 575–579). Each shows a routine or programme being modified in its performance.
- Planned change. Managers are “as much an agent of change as everybody else”, distinguished by “declarative powers” (p. 579). A change programme is “a new discursive template” that amplifies changes already under way. It is “a punctuation of the flow of organizational life”, not its origin.
- Inertia. Ongoing change does not mean organizations constantly change. Local adaptations may never be institutionalized. “Organizations tend to resist change” is shorthand for that failure (p. 579).
Claims
| id | claim | strength | support |
|---|---|---|---|
| C1 | Stage models cannot represent change itself, only states | moderate | argument from James and Bergson (pp. 570–571) |
| C2 | Applying radially structured categories to nonprototypical cases changes them | moderate | prototype theory plus Feldman’s and Tsoukas & Vladimirou’s cases (pp. 573–575) |
| C3 | Change is ontologically prior to organization | weak | a stance motivated by C1 and C2; nothing is described that could count against it |
| C4 | Change programmes work by amplifying changes already under way | weak | rereading of Barrett et al. and Orlikowski & Hofman (pp. 578–579) |
| C5 | Inertia is failed institutionalization of local change, not its absence | moderate | conceptual distinction (p. 579) |
Concepts
- Organizational becoming. The ongoing change of organizations, which is the normal condition and not an exception.
- Synoptic account. Change seen from outside as an accomplished event between states.
- Performative account. Change seen from within, as enacted in time.
- Radial category. A category with a prototypical core and a nonprototypical periphery.
- Imaginative extension. Applying a category beyond its prototypes, which changes it.
- Discursive template. The interpretive codes a change programme introduces.
- Microscopic change. Change that occurs “through ‘creep,’ ‘slippage,’ and ‘drift’” and may never be institutionalized (p. 579).
Connections
- Weick, Orlikowski and Feldman. The paper builds on their process views and says they “do not go far enough” (p. 569).
- Bencherki, Brummans & Vézy 2024 (LIT-1902). It cites this paper as the conception of organizational becoming its own theory “parallels” (p. 1351). This paper is human-centred: change is “inherent in human action” (p. 570), its engine is “the intrinsically human ability to be reflexive” (p. 575), and organization emerges “through the ongoing agency of organizational members” (p. 579). Bencherki et al. hold that “there are no already-constituted agents” (p. 1358). The two agree that organization is a secondary effect of process. They disagree on whether the process is carried by agents. Bencherki et al. do not mark the difference.
Bearing on the record
- THEORY-1284 files the paper’s central thesis, Proposed.
- THEORY-1279 files its mechanism, the radial structure of categories, Proposed.
- THEORY-1280 (Bencherki et al.’s agency without agents) cites this paper only as a parallel. On reading, the parallel holds for the priority of process over organization, not for the priority of process over agents.
Limitations
- No evidence of its own. The illustrations are other researchers’ ethnographies, chosen because they show change. No case is described in which a routine stays the same under performance.
- The ontological claim is not testable as stated. The authors grant that organizations often do not change (p. 579). They save the thesis by placing change below the level of institutionalization, where it “may never acquire the status of formal organizational systems”. This makes “change is pervasive” hard to fault.
- Unit of analysis. The account is about individuals reweaving their beliefs. How individual reweaving becomes organizational is left as “a different issue” for “empirical research” (p. 579).
Open questions
- Under what conditions does microscopic change become institutionalized? The authors leave this open (p. 579).
- Whether the radial-category mechanism predicts which rules will change, for instance that rules with more nonprototypical cases drift faster.
Bearing on the record’s own argument
Added 2026-10-31 by a pass connecting the readings of Bencherki, Brummans & Vézy 2024 and the works it leans on to the record’s lines. Analogies are marked as such.
Distributed agency and individuation
- CLAIM-418, part 3. Radial categories (p. 574) and institutionalization in a “structural context” (pp. 578–579) give a mechanism by which agents’ exercise of a shared perspective transforms it, with the agents kept. The same mechanism bears on QUESTION-259.
- CLAIM-430, part 3. Whether a local extension becomes the organization’s depends on institutionalization, which may fail (pp. 578–579).
Will-organization, organizational affect and authenticity
- Navy TQ (pp. 578–579) is a second-hand instance for CLAIM-122, but it does not hold the proposals fixed. The “structural context” (p. 579) is what TERM-023 analyses.
- Creep, slippage and drift (pp. 574, 579) mean authorizing standards can change with no act of revision (QUESTION-364). They give QUESTION-334 a mechanism and a test design, as an analogy.
- Both parts of CLAIM-363 hold for categories (p. 574).
- Actors carry the change (p. 575), which is consistent with CLAIM-080’s supposition.
Nucleating argument and formal apparatus
- Radial extension and the record’s closed vocabularies. The record forbids the silent form (“never replaced by the nearest wrong one”). So a category is extended at the nonprototypical case by decision and institutionalized at once (CLAIM-419; CASE-178: 94% of terms were used in the pull request that added them). That is a designed answer to the open question at p. 579. CASE-178’s note of 2026-10-31 sets out a blind self-test of THEORY-1279.
- Change as legitimate. The reading supports CLAIM-189’s counter: CLAIM-177’s redefinitions failed because they were silent, not because they happened. What a record secures is traceability.