Machine-readable preservation copy. The PDF is the canonical published edition. # World-Formation Research: Non-Closure, Provoked Empiricism and the Conditions of Responsibility (Studies in World-Formation, Vol. 19) > Canonical source file: `Speed_2026_World-Binding-Vol_19_en.pdf`. If this copy differs from the PDF, the PDF prevails. > Automated text-layer extraction preserves wording but may alter page layout, columns, footnote placement, tables, and image relationships. > Non-textual warning: the PDF contains image objects on 1 page(s); images are not embedded in this Markdown copy. Textual captions remain only where present in the PDF text layer. --- Operatoric Research Corpus - Studies in World-Formation VOL 19 World-Formation Research: Non-Closure, Provoked Empiricism and the Conditions of Responsibility Timothy Speed Operatoric Research Corpus Studies in World-Formation ## Section VIII World-Formation Research and Structural Responsibility ## Volume 19 World-Formation Research: Non-Closure, Provoked Empiricism and the Conditions of Responsibility Timothy Speed 2026 DOI: 10.5281/zenodo.22670280 1 ## Impressum / Publication Information Author: Timothy Speed Independent Researcher Straße zur Eiche 10 04916 Schönewalde Germany Email: info@timothy-speed.com Website: https://timothy-speed.org License: © 2026 — CC BY-NC-ND 4.0 ## GND: 122901991 ## VIAF: 37811735 ## ISNI: 000000001636722X ## Wikidata: Q138504206 ## Scholar ID: hBLHdoAAAAJ https://orcid.org/0009-0002-0143-5949 ## Zenodo Community: Operatoric Research Corpus Archive First published: 2026 This publication is deposited in the German National Library (Deutsche Nationalbibliothek) as a network publication. DOI: https://doi.org/10.5281/zenodo.22670280 2 Abstract Volume 19 of the Studies in World-Formation series marks the transition from the theoretical elaboration of the Operatoric Research Corpus to the explicit formulation of World- Formation Research as a research programme and to its first systematic application to the problem of structural responsibility. Its point of departure is the assumption of a fundamental non-closure of world. Scientific, institutional, and social orders must stabilize differences in order to know, decide, and act. Yet every stabilization also produces exclusions and alters the conditions under which further reality, alternative perspectives, and future knowledge can emerge. World-Formation Research therefore investigates not only what becomes visible and determinable through knowledge, but also how successful closures transform the conditions of further knowing. This second movement is understood not as a subsequent ethics of scientific practice, but as a component of scientific rationality itself. The volume develops an operatoric research approach linking non-closure, Seinsverschiebung, World-Binding, Eigenzeit, Negative Morphology, Counter- Position, and Provoked Empiricism. Provoked Empiricism designates a recursive form of inquiry in which an existing world-order is confronted, through a reasoned positing, with its own normative, epistemic, or institutional claims. Reactions, defensive responses, shifts of jurisdiction, non-intervention, and attempts at closure thereby become empirical material. The purpose is not to manufacture conflict, but to make visible structures that remain difficult to observe precisely because their conditions and exclusions are stabilized as normality. The second part applies this research logic to structural violence. The concept of structural violence made it possible to describe harms that cannot be reduced to a single act or an identifiable original perpetrator. Yet this achievement can itself produce a new epistemic blind spot: as violence becomes increasingly explained through structures, fragmented jurisdictions, institutional routines, and distributed decision-making power, concrete responsibility for its continuation can disappear. The volume describes this as the perpetrator gap and distinguishes between constitutive perpetratorlessness and produced perpetratorlessness, in which responsibility is dissolved again despite increasing knowledge and existing possibilities for intervention. From this analysis emerges the figure of the apparently legitimated inactive witness to structural violence with decision-making power: a person who possesses sufficient knowledge of serious harm, has relevant possibilities to investigate, limit, or interrupt it, and nevertheless allows institutional normality, role-bound conduct, or jurisdictional boundaries to prevail over intervention. Responsibility is therefore treated not as a static property but as something that can become concentrated over time through increasing knowledge, foreseeability, agency, and continued non-intervention. The decisive issue is not institutional inactivity as such, but the relational convergence of serious harm, available knowledge, decision-making power, possible alternatives, and actual response. 3 Provoked Empiricism forms the methodological bridge between both parts of the volume. By documenting indications, opportunities for correction, institutional responses, and repeated attempts at closure, it produces a traceable history of changing epistemic conditions. A system that does not know of a harm is analytically different from one that has repeatedly been confronted with the same harm and nevertheless continues unchanged. The intervening witness does not produce perpetration; rather, the intervention alters the conditions under which responsibility can become visible, attributable, and reconstructable. Volume 19 thus brings methodology and responsibility analysis together within a single research movement. The first part formulates the programme; the second demonstrates how that programme can reopen a successful conceptual stabilization at the point where it begins to obscure its own consequences. The central question is therefore not only which world is stabilized through knowledge, but also what becomes invisible within that stabilization and under what conditions responsibility can re-emerge. Non-closure thereby appears as an epistemic, political, and human-rights condition: the requirement that established categories remain open to revision where they cease to carry the realities they were intended to make intelligible. Keywords: World-Formation Research, Non-Closure, Provoked Empiricism, Structural Responsibility, Structural Violence, Institutional Violence, Epistemic Violence, Seinsverschiebung, World-Binding, Eigenzeit, Negative Morphology, Operatoric Research, Counter-Position, Epistemic Closure, Institutional Closure, Structural Harm, Responsibility Attribution, Perpetrator Gap, Produced Perpetratorlessness, Decision-Making Power, Inactive Witness, Intervening Witness, Accountability, Epistemic Injustice, Human Rights, Institutional Non-Intervention, Recursive Inquiry, Provocation of Accountability, Knowledge and Responsibility, Conditions of Further Knowledge Content Abstract...........................................................................................................................................3 Introduction....................................................................................................................................... 7 World-Formation Research: An Operatoric Research Programme for the Study of Non- Closure, Knowledge, and the Conditions of Further Knowledge DOI: https://doi.org/10.5281/zenodo.22659569..................................................................................... 9 Abstract...........................................................................................................................................9 1. The Missing Second Movement of Science...............................................................................10 2. Non-Closure Is Not a Deficit...................................................................................................... 13 4 3. Seinsverschiebung (shift of being): Knowledge Changes Its Own Ground...........................19 4. Why This Research Programme Is Called World-Formation Research................................21 5. Operatoric Research Corpus and Operatoric Research...........................................................23 6. Operator, Counter-Position, and Provoked Empiricism: How a World Is Brought to Respond............................................................................................................................................ 25 The Meaning of Positing..................................................................................................... 26 Provocation Does Not Mean Escalation............................................................................ 27 Counter-Position..................................................................................................................28 Why Tension Can Produce Knowledge.............................................................................29 Operatoric Reversal..............................................................................................................30 Provoked Empiricism as a Real Response Structure.........................................................31 The Researcher as Part of the Operator.............................................................................31 Self-Use as a Measuring Instrument...................................................................................32 The Operator Does Not Produce Truth.............................................................................33 The Importance of Resistance.............................................................................................34 From Response to Form...................................................................................................... 34 The Operator as an Epistemic Probe..................................................................................35 Counter-Position as De-Monopolization..........................................................................36 Interim Conclusion.............................................................................................................. 36 7. From Counter-Position to Homology: Why Forms Become Comparable Across Fields....37 Analogy as a First Irritation................................................................................................37 From Surface to Relation.....................................................................................................38 Functional Analogy..............................................................................................................39 Structural Homology............................................................................................................40 The Structure Must Survive the Change of Field.............................................................40 Homology Is Not Identity.................................................................................................... 41 Condition Homology............................................................................................................41 Counter-Position as a Test of Homology........................................................................... 43 The Operator as an Invariance Test...................................................................................44 The Failure of Analogy........................................................................................................ 44 Homology Is Gradual...........................................................................................................45 Why This Is Central to World-Formation Research.......................................................45 The Political Significance of Homology............................................................................. 46 From the Individual Case to Structural Knowledge.........................................................47 Interim Conclusion..............................................................................................................47 8. Neurodivergent Epistemology: The Position from Which Non-Closure Becomes Visible 48 Knowledge Can Be Organized Differently........................................................................ 49 Why Closure Is Experienced Differently...........................................................................50 Neurodivergence as an Epistemic Position, Not as Proof of Identity.............................52 The Marginal Position Does Not Automatically See More.............................................52 The Observer Is Part of World-Formation........................................................................53 Autistic Epistemology as a Counter-Position to Epistemic Monoculture......................53 Dissolving Dominant Truth Does Not Mean Arbitrariness............................................54 Counter-Truth as an Epistemic Operator..........................................................................55 The Epistemic Significance of Missing the Target...........................................................55 Neurodivergent Research as Preservation of Alternative Spaces of Possibility............55 5 An Epistemic Asymmetry....................................................................................................55 From Survival to Method.................................................................................................... 56 Autistic Epistemology as a Strengthening of Scientific Non-Closure............................56 Interim Conclusion..............................................................................................................56 9. Negative Morphology: Indirect Reading...................................................................................57 Not the Form, but Its Difference........................................................................................ 58 Negative Form......................................................................................................................58 10. The Two Movements of Science..............................................................................................59 11. Delimitation................................................................................................................................ 59 No New Metaphysics.................................................................................................................. 59 Not Simply Philosophy............................................................................................................... 60 Not Only an Ethics of Science....................................................................................................60 No Replacement for Discipline-Specific Science.....................................................................60 12. Artificial Intelligence: When the Production of Difference Becomes Faster Than World- Binding.............................................................................................................................................. 61 13. Society, Normalization, and Epistemic Annihilation.............................................................62 14. Physical Boundary Cases............................................................................................................63 15. Quality Criteria...........................................................................................................................63 16. Progress as Viability...................................................................................................................64 17. Conclusion: Which World Can Carry Which Knowledge?..................................................65 Provisional Conceptual Architecture........................................................................................ 65 When Structural Violence Has a Perpetrator: The Apparently Legitimated Inactive Witness with Decision-Making Power DOI: https://doi.org/10.5281/zenodo.22206507.....................70 Abstract........................................................................................................................................70 1. Violence Without a Perpetrator.................................................................................................71 1.2. On the Methodology of This Research...................................................................................75 1.3. On the New Concept of the Perpetrator.................................................................................77 2. The Perpetrator Gap in Structural Violence............................................................................. 78 2.1. The Necessary Discovery of Violence Without a Visible Perpetrator............................79 2.2. The Price of Perpetratorlessness.........................................................................................82 2.3. Structure Is Not an Acting Subject.....................................................................................85 2.4. Between the Immediate Perpetrator and Perpetratorlessness.........................................88 2.5. The Epistemic and Ethical Threshold................................................................................92 2.6. The Apparently Legitimated Inactive Witness to Structural Violence with Decision- Making Power............................................................................................................................. 97 2.7. Degrees of Perpetration and the Concentration of Responsibility...............................102 2.8. Perpetration Beyond Criminal Guilt................................................................................106 3. Provoked Empiricism: From Structural Perpetratorlessness to Attributable Responsibility ...........................................................................................................................................................111 6 3.1. The Intervening Witness to Structural Violence.............................................................117 3.2. The Operator of Non-Closure..........................................................................................122 3.3. The Positing: Violence as Recognizable Reality..............................................................123 4. Limits and Risks of the Concept of the Perpetrator...............................................................128 4.1. Minimum Conditions for the Attribution of Perpetration............................................132 5. Consequences: From Institutional Jurisdiction to Responsibility for Violence..................133 6. Conclusion: Structural Violence Needs Perpetrators Again..................................................136 Declaration of AI usage..................................................................................................................139 References...................................................................................................................................139 ## Introduction With Volume 19, Studies in World-Formation enters a new phase. The preceding volumes developed the central conceptual architecture of the Operatoric Research Corpus across different fields: preontology, non-closure, world-binding, Eigenzeit, morphology, simulation, neurodivergent epistemology, work, value, law, structural violence, and the conditions under which stable worlds emerge and persist. What increasingly became visible across these investigations was that the individual fields were not merely separate applications of one theoretical vocabulary. They stood in recursive relations to one another. Later investigations altered the readability of earlier concepts, while empirical and institutional cases in turn forced abstract concepts to become more precise. The Corpus therefore gradually became not only a body of texts, but a research architecture. Volume 19 makes this development explicit. Its first part, World-Formation Research: An Operatoric Research Programme for the Study of Non-Closure, Knowledge, and the Conditions of Further Knowledge, formulates World-Formation Research as the transdisciplinary programme emerging from the Corpus. The distinction proposed there is deliberate: the Operatoric Research Corpus forms the overarching research architecture; World-Formation Research designates the research programme; Studies in World-Formation provides the series of concrete investigations; preontology marks a foundational position of questioning; operatoric research describes the mode of inquiry; and negative morphology and provoked empiricism function as central methodological approaches. This clarification is significant because the central question of the series has gradually shifted. The earlier volumes repeatedly examined how difference becomes bound into viable forms and how such forms acquire stability. World-Formation Research adds a second movement: every successful stabilization also alters the conditions under which further forms, 7 perspectives, and knowledge can emerge. Scientific and institutional knowledge therefore cannot be assessed only by the correctness of its closures. It must also be examined in relation to what those closures make possible, invisible, difficult, or no longer connectable. Non- closure in this sense is not a deficit of knowledge but a condition of further differentiation and future world-formation. The second paper in this volume, When Structural Violence Has a Perpetrator: The Apparently Legitimated Inactive Witness with Decision-Making Power, provides a particularly clear test of this programme. The concept of structural violence represents an important epistemic achievement because it makes forms of harm visible that cannot be reduced to a single immediate act or original perpetrator. Yet this achievement can itself produce a new blind spot: the more strongly violence is explained through structures, distributed responsibilities, and institutional fragmentation, the more easily concrete responsibility for the continuation of known harm can disappear. The paper therefore asks what happens when a successful conceptual opening becomes a new closure. This is precisely where the two studies become mutually illuminating. World-Formation Research argues that knowledge must not simply stabilize a form and treat this stabilization as final. It must remain capable of returning to its own consequences and reopening what its categories have obscured. The perpetrator study performs this movement concretely. It does not reject the concept of structural violence. Instead, it asks what becomes invisible once perpetratorlessness itself becomes stabilized as part of the explanatory structure. The resulting concepts of the perpetrator gap, produced perpetratorlessness, and the apparently legitimated inactive witness with decision-making power emerge from this renewed opening of an already productive category. Provoked empiricism forms the methodological bridge between both parts. Within World- Formation Research, it describes a procedure in which a stabilized order is confronted with a difference it must respond to, allowing previously invisible relations to become observable. The response does not automatically prove an interpretation; it produces forms—such as silence, delay, fragmentation, displacement of responsibility, or institutional self-protection —that can subsequently be examined. In the perpetrator study, this method becomes a recursive analysis of responsibility. Institutional reactions, opportunities for correction, non- intervention, and attempts at closure are themselves treated as empirical material. Knowledge and responsibility thereby acquire a temporal dimension: an actor who initially does not know of a harm occupies a different position from an actor who has repeatedly been informed, possesses relevant decision-making power, and nevertheless allows the process to continue. For this reason, the two texts are not combined here merely because they share terminology. Their relation is structural. The first formulates the research programme; the second demonstrates how that programme can operate on a concrete epistemic and political problem. One describes the necessity of reopening successful closures. The other shows what can become visible when such a reopening is carried out. 8 Volume 19 therefore marks a transition within Studies in World-Formation. The series no longer only develops concepts of world-binding and stabilization across different domains. It increasingly investigates its own mode of inquiry: how knowledge is produced, how categories acquire power, how closures alter the conditions of further knowledge, and how responsibility can re-emerge where institutional and conceptual structures have rendered it invisible. In this sense, the volume does not begin a new research project alongside the earlier volumes. It makes explicit the research programme that has gradually emerged from them. World-Formation Research: An Operatoric Research Programme for the Study of Non-Closure, Knowledge, and the Conditions of Further Knowledge DOI: https://doi.org/10.5281/zenodo.22659569 ## Abstract Science has highly developed procedures for distinguishing, naming, measuring, modelling, and transforming reality into stable forms. Far less developed are procedures for investigating how precisely these successful stabilizations alter the conditions under which further knowledge, other forms of existence, and future world-formation remain possible at all. This paper proposes World-Formation Research as the designation for a transdisciplinary operatoric research programme that begins at this gap. Its point of departure is the assumption of a fundamental non-closure of world. Non-closure refers neither to ignorance nor to provisionally missing data. It is the condition that allows difference, change, perspective, and further formation to remain possible at all. Every act of naming, modelling, or institutional stabilization binds part of this open field into a concrete world-form. In the terminology of the underlying Operatoric Research Corpus, this process is referred to as Seinsverschiebung (shift of being). A Seinsverschiebung does not neutrally represent reality. It changes the space in which further reality can appear. It shifts that space relationally. 9 From this follows a second movement of science. Science must close in order to know and to act. But it must also investigate how its closures alter the conditions of further knowing. World-Formation Research understands this second movement not as a subsequent ethics of science, but as a component of scientific rationality itself. The paper defines the relationship of World-Formation Research to preontology, operatoric research, and negative morphology, and distinguishes the programme from metaphysics, classical epistemology, the ethics of science, and discipline-specific explanation. Using examples from artificial intelligence, social normalization, and physical boundary problems, it shows why the decisive question is not only how knowledge can be increased, but which world remains sustainable with which knowledge. The core demand is therefore: science bears responsibility not only for the knowledge it produces, but also for preserving those conditions under which other and future knowledge can continue to emerge. This has not only a scientific dimension, but also a political and human-rights dimension. Keywords: World-Formation Research, Preontology, Non-Closure, Shift of Being, Operatoric Research, Negative Morphology, World-Binding, Eigenzeit, Provoked Empiricism, Neurodivergent Epistemology, Philosophy of Science, Artificial Intelligence, Epistemic Violence, Future of Knowledge ## 1. The Missing Second Movement of Science Science is good at finding things. This is not a criticism. Without the ability to hold differences in place, to separate bodies from one another, to measure effects, to formulate regularities, and to establish repeatable relations, there would be no science. World must be brought to a halt in some form so that it can be spoken about. Something is named. Something else is excluded. A variable is isolated. A model is built. A diagnosis is made. A boundary is drawn. A state is provisionally taken as given. The problem lies not so much in closure itself, but in the assumption that closure is what science must always be about. That it is the primary goal. For with every successful act of naming, what emerges is not merely more knowledge about a space that has remained unchanged. The space itself has changed. What previously existed as an open field of different possibilities already appears, after stabilization, within a particular order. Relations are weighted differently. Certain differences become relevant, others disappear. Institutions align themselves with the categories. Technologies are built on their basis. People begin to understand themselves through the concepts. Further research takes the objects once produced as its points of departure. Results become habits, and habits become an apparently fixed world. A Seinsverschiebung (shift of being), an important concept in this field of research, is not simply a change within an already existing space. It designates a shift in the conditions under which 10 further forms can arise and appear in that space at all. Once one possibility is stabilized, the remaining possibilities do not remain unaffected beside it. Their relations change. A path has been taken, and because it has been taken, the landscape afterwards is no longer the same landscape as before. This may initially sound abstract. In fact, it is profoundly everyday. A scientific diagnosis does not merely describe a person. It changes insurance entitlements, access to therapy, institutional expectations, self-understanding, and future interpretations of that person’s behaviour. An economic model does not merely describe a market. As soon as actors orient their decisions towards it, the model changes the market. A technology does not merely solve an existing problem. It changes which problems can still become visible afterwards, which remain economically relevant, and which remain politically actionable. Science possesses sophisticated methods for the first half of this process. It investigates whether a measurement is correct, whether a model has predictive capacity, whether a hypothesis is falsifiable, whether data can be reproduced. For the second half, comparably developed procedures are lacking. What has successful knowledge done to the space from which further knowledge is supposed to emerge? Which possibilities were strengthened? Which disappeared? Which disciplines have suddenly become connected through the intervention? Which Eigenzeiten were accelerated? Which other Eigenzeiten became uninhabitable as a result? Which questions can still be asked at all after a particular technological or conceptual stabilization? Yet this thought reaches far deeper than one might initially assume in terms of imprinting, adaptation, fashion, trend, or even culture. It is also an ontological process, fundamentally a preontological moment. For what is altered are not things among things, but spaces of possibility for ontological development. This is where World-Formation Research begins. The thesis of this paper is not that science should determine less, know less, or intervene less. It is that a science which investigates only the quality of its closures remains scientifically incomplete. Science must close in order to be able to know. But it must reopen in order to remain capable of knowing. This may initially sound banal, but actually doing it is extremely demanding and succeeds more rarely than is often assumed. Science must not only recognize space and thereby define and fix it; it also bears a responsibility to allow new spaces, new possibilities, altered conditions, other methods, and other epistemic directions of view. It must not understand itself only as an interpretation of reality. Even if scientific knowledge is not a creative act in the sense that something is simply imagined into being, the exclusion and concealment of other possibilities are far more 11 uncontrolled and unconscious in their consequences, because there is no peer review of what every individual act of knowledge obscures, shifts, or transforms elsewhere. But knowledge cannot be thought of in such isolation. It comes from a space and it acts within a space. Knowledge therefore cannot be thought independently of distortion. One person’s clarification can elsewhere lead to the loss of affordances that might have enabled expanded knowledge. For this reason, research ultimately cannot remain within its own rules, because rules that do not hold everywhere must communicate and carry their own limits. In an enlightened world, people must still be able to exist and live as themselves. In this respect, research must become more participatory. Research must not ignore the space of a society that is constrained by its knowledge. It must remain capable of acting while at the same time preserving future capacity for action. (Speed 2019) Looking at the academic system and the changes it has undergone in recent decades, one might doubt precisely this. In times of maximum globalization combined with an increasing concentration of influence and power relations, knowledge once again requires a local, subjective, incomplete quality. The Enlightenment, too, has a sweet spot. If it goes too far, it loses its capacity for integration, destroys diversity, and reduces the freedom that it claims to preserve and expand through progress. The research developed here is an attempt to research with openness without dissolving it into arbitrariness or treating it merely as a flaw. It seeks fruitful, robust knowledge while at the same time keeping the limits of that knowledge visible. Knowledge should not advance by forgetting its own space of conditions. In this way, precisely those forms of knowledge that are easily excluded by classical objectification are preserved: local knowledge, subjective knowledge, Experiential Knowledge, Situated Knowledge. Not as counter-concepts to scientific knowledge, but as part of the conditions under which knowledge emerges and can prove itself at all. The same applies to boundary regions of physics. Even where something can be completely described as an object, its reality is not necessarily exhausted by its objecthood. An object is always also bound within relations, and these relations can point beyond what is captured by the description of the object itself. The boundary of the object is therefore not automatically the boundary of the relation. This can lead to similar observations even in mathematics. World-Formation Research is interested precisely in this excess: in what a stable form makes visible, but also in what it cuts off, shifts, or makes accessible only indirectly. 12 2. Non-Closure Is Not a Deficit The underlying difficulty begins already in the way scientific thinking usually treats openness. What has not yet been explained appears as a provisional deficit. What cannot yet be measured is regarded as not yet sufficiently accessible technically or methodologically. What remains imprecise seems to be waiting for a better instrument, a more precise theory, a finer resolution. In this understanding, openness possesses no epistemic status of its own. It appears as a transitional state on the way toward greater determinacy. This attitude is necessary and productive for a large part of scientific work. Without it, there would be no reason to improve measurement procedures, correct models, further investigate unknown relations, or empirically explore areas that have so far remained inaccessible. It becomes problematic only where a method of gaining knowledge imperceptibly turns into a statement about the structure of the world itself. The justified assumption that many questions that are open today may be answered in the future then becomes the more far- reaching assumption that openness is fundamentally only a provisional lack of determination. Instead, the thesis developed here is that openness is a fundamental condition of existence itself. Openness is just as real. It is not a “not yet anything.” An epistemic affinity exists first with fallibilist conceptions of science, particularly those of Charles S. Peirce (1877) and Karl Popper (1963). There too, scientific knowledge is fundamentally provisional, corrigible, and dependent on further inquiry. Peirce understands science precisely not as a closed body of knowledge, but as continuing inquiry whose results remain in principle revisable; Popper correspondingly ties scientific progress in knowledge to conjecture, criticism, and refutability. World-Formation Research shifts this insight, however, onto a different level. Non-closure here does not refer only to the epistemic fallibility of present knowledge. The further-reaching thesis is that openness itself is a condition of further difference and world-formation. Fallibilism therefore forms an important epistemological point of connection, but it is not identical with the preontological non-closure claimed here. The All–Nothing Paradox developed in the Operatoric Research Corpus begins at this point (Speed 2025b). It describes neither a simple opposition between being and nothingness nor a dialectical movement in which both poles are ultimately sublated into a higher unity. What is meant instead is a structural impossibility of complete closure. A completely determined All would not be the perfectly known world. It would be a constellation in which every difference had already been completely fixed, every relation conclusively determined, and every possibility entirely transformed into actuality. Such a structure could be perfectly stable, yet precisely for that reason it would no longer be capable of world. Where no real possibility of difference remains, nothing can become otherwise, nothing new can appear, nothing can respond to something else. Conversely, a pure Nothing would likewise not be capable of world. Where no difference, relation, or tension exists, nothing can appear either. 13 World therefore does not arise from either of the two extremes. It exists only where determination and openness remain effective at the same time without dissolving completely into one another. Stabilization is necessary because without it no form, no identity, no relation, and no observability would be possible. At the same time, stabilization must not completely destroy its own condition. A world must be able to close without closing itself finally. This position touches several well-known philosophical traditions, but is not identical with them. Most obvious at first is the proximity to dialectics. The relation between determination and openness, however, is not understood here in Hegelian terms as an opposition that, through its movement, arrives at a higher unity or Aufhebung. Precisely this final reconciliation is excluded. Difference need not remain because a synthesis has not yet been reached, but because complete synthesis itself would eliminate the condition of further world-formation. Non-closure is therefore not a provisional stage of dialectical development, but a structural condition for development to remain possible at all. There is also a proximity to process philosophies, particularly where reality is conceived not as a collection of static substances but as becoming, relation, or event. World-Formation Research, however, does not simply replace the object with “process.” A process too can be positively determined and theoretically closed. The decisive question therefore lies on a different level: Under what conditions can stable objects, processes, and relations arise at all, and what happens to their space of possibility through this stabilization? In this respect, the research programme does not claim that “everything is process.” It investigates why neither object nor process can completely contain its own conditions of emergence and continuation. There are also recognizable points of contact with Bergson’s concept of becoming or with later concepts of virtuality and difference, for example in Deleuze. The Gap as used here, however, is specifically not to be understood as a positive reservoir of possibilities not yet actualized. It is not a second ontological domain of the virtual from which concrete forms emerge. As soon as such a space of possibility is positively described, it has already been stabilized once again as a world-form. The Gap therefore designates less an existing set of alternative states than the structural impossibility of completely transferring the space of possible further differentiation into a present order. This position likewise differs from the Heideggerian use of the preontological. What is meant is neither a pre-theoretical understanding of Being on the part of Dasein nor an existential relation to Being or Nothingness. Preontological here designates a methodological position of questioning toward the conditions of world-formation themselves: What must remain open so that anything at all can become stable as an object, relation, observer, or world? Scientific concepts of open systems, irreversible processes, or non-equilibrium dynamics may also recall these considerations. They must not, however, be equated with them. Thermodynamic openness or irreversibility would be physically describable forms within the 14 world. They do not immediately imply preontological non-closure. Conversely, the All– Nothing Paradox should also not be understood as a generalized theorem in the manner of Gödel, as though formal incompleteness in mathematical systems allowed one to infer directly an incompleteness of the world. Such relations may open research questions, but they do not replace discipline-specific derivation. Finally, non-closure also differs from merely linguistic or discursive undecidability. The thesis is not that reality remains open because language can never represent it completely. Language and scientific models are themselves forms of world-formation. Their limitation is a particular case of a broader problem: Every stabilization produces determinacy, but precisely by doing so it can alter the conditions of further determination. The All–Nothing Paradox therefore stands within a philosophical space of affinities involving dialectics, process thought, philosophies of difference, and theories of openness, but it shifts the emphasis. The primary question is not what the ultimate nature of Being is, but under what conditions world can continue to arise at all. The decisive thesis therefore remains unusually simple: A world must be able to close in order to form. It must not close completely if further world-formation is to remain possible. Non-closure in this context therefore does not designate a remainder of the unknown. It is not some final piece of information that is currently still missing and that future research might one day add. Nor is it a hidden object behind known objects. The concept of the Gap used in the Corpus (Speed, 2025d,e,f) must specifically not be understood as some as yet undiscovered something that sufficiently precise research might eventually determine positively. If the Gap were described as an object, substance, field, or localizable instance, it would already have become part of a particular world-form. Non-closure would at that moment not have been explained, but transferred into a new closure. (Speed, 2026d) It is precisely at this point that it becomes understandable why the concept of the preontological becomes necessary and at the same time can so easily be misunderstood. The preontological does not designate a hidden floor of the universe. There is not first a preontological sphere, above it physics, above that chemistry and biology, and above that finally consciousness, society, or culture. Such a stratification would already treat the preontological as a spatially or hierarchically ordered level and would thereby presuppose precisely those categories whose conditions of emergence are supposed to be investigated in the first place. The image of a deeper ground beneath physics would therefore be misleading as well. Depth, level, ground, inside, and outside are themselves already forms of a stabilized description of world. Preontological, in this research programme, is therefore not a region but a position of questioning. The term designates the point at which investigation no longer begins with things already presupposed, but asks about the conditions under which anything at all can become stable as a thing, relation, state, boundary, or difference. The preontological question 15 does not first ask: What is there? It asks: What must happen or be presupposed for a particular “there,” a particular “something,” or a particular relation to appear at all as a stable world-form? This produces a difficulty that cannot simply be eliminated through more precise terminology. As soon as such a condition is positively named, a new form has already emerged. Naming itself performs a Seinsverschiebung (shift of being). (Speed, 2025g) It establishes a difference, stabilizes a relation, and produces a conceptual space within which the condition is henceforth read. The description is therefore not identical with the open field to which it attempts to respond. It is already the translation of that field into a particular world-form. This non-identity between condition and description is not a methodological accident. It belongs to the matter itself. If every act of naming produces a Seinsverschiebung (shift of being), the preontological cannot be scientifically approached adequately simply by searching long enough for the final positive formulation. Direct representation would be precisely the wrong movement of research. It would transform the condition into an object and thereby blur the difference between the two. Preontological research therefore necessarily remains indirect. One might say that it sees out of the corner of its eye. This image does not designate imprecision or a renunciation of accuracy. What is meant instead is another form of accuracy. The direct gaze fixes. It turns what is to be investigated into an object within an already stabilized grid. The indirect gaze, by contrast, turns toward the changes that arise at the boundaries of this fixation. It does not attempt to see the Gap itself. It investigates the forms, tensions, and shifts that occur where world responds to non- closure. Non-closure itself therefore does not appear positively. Closures become visible. One sees how they form. One sees where they become viable. One sees where they displace other possibilities. One sees where they bend, where they encounter limits, where they generate new openings, or where their own stability alters the conditions of their continuation. In moments of form-decision, however, topological traces become visible in the transitions between forms, and these in turn make conditions of forms recognizable. If a man lifts an overturned chair in order to make the setting “right” again, conditions become visible that reach far beyond the setting. Conditions that may still apply even if one does not know at all that it is a man and a chair moving here. In that moment, something may become visible behind the chair that has nothing to do with either the man or the chair, or the lighting conditions may suddenly change. The scene possesses, so to speak, several negative forms of itself layered over one another. They are topological consequences, but they are often not things. What is decisive is that none of these additional relations need be contained in the object “chair.” They become readable only through the alteration of the setting. The form-decision 16 therefore functions like a probe: By altering a relation, conditions emerge that previously remained invisible in the stable state of the field. There is a recognizable proximity here to Harold Garfinkel’s breaching experiments: There too, a stable social order becomes epistemically readable precisely through its disturbance. The methodological differences from provoked empiricism are discussed in Chapter 6. What at first glance appears to be an analogous observation by an artist can become a research methodology that, without a fixed discipline and without a fixed position, makes new differences recognizable through deliberate positing, irritation, and provocation— differences that initially evade object-centred interpretation in a destabilizing way. Fundamentally, one also finds here a primordial movement of research that is often lost in its institutionalization. It is a movement back toward the level of pattern relations. It is precisely this indirect readability that makes morphology important within the research programme. Morphology here does not mean connecting widely separated objects on the basis of superficial similarities. It is not about comparing a tree with a galaxy, equating a social institution with a black hole, or inferring a shared ontological cause from a similar graphic form. Such analogies may be heuristically interesting, but on their own they possess insufficient explanatory power. A form becomes relevant only where the question of its condition can be posed. What kind of openness is being bound here? What difference must be stabilized for this form to arise? Which possibilities are excluded in the process? What new boundary is produced? How does the form that has emerged alter the space of further formation? This may initially sound abstract, but in the author’s work it appears as research on negative forms, structures, and patterns. (Speed, 2026e) For example, many of the author’s investigations between sociology and state violence showed how, within apparently fixed academic concepts, a new nexus of violence (povertism, welfare violence) could be made recognizable through what those concepts concealed. This was done not only by investigating the problem directly, but through cross-connections and lateral connections between apparently unrelated institutions and actors, through which finer patterns of violence within events could then be triangulated. (Speed, 2026f,g) What initially appeared to be legally regulated and scientifically settled was thus disturbed in a way that made not only unseen forms of violence visible at the margins, but led to an epistemic opening of the entire system. This linking of different levels, perspectives, irritations, and disturbances generates new data. These do not initially aim at a finished piece of knowledge. Their first function is to produce a difference within the closed field—an eruption, a rupture, a disturbance of the existing order. Only from the forms that become visible at this rupture does knowledge begin to condense. 17 Only when such questions can be answered across different domains in a way that preserves concrete differences and produces new testable research questions does superficial similarity within the new connections become a structurally relevant relation. Not everything in chaos is therefore knowledge. Rather, one peels away layers of the onion from the Gap and makes the distortion of every determination usable. Every definition of a new thing distorts the space and thereby creates a new one. By recursively repeating this, a fruitful destabilization remains. The image of onion layers must therefore not be understood as though the Gap itself were ultimately lying beneath the layers. Every layer that is peeled away already produces a new position from which further research proceeds. The Gap never appears at the end of the movement as an uncovered thing. Morphology therefore does not primarily concern finished forms, but sequences of forms. What matters is not only that a particular shape appears, but how it emerges from a previous condition, which possibilities are bound within it, and which altered conditions it produces for the next step. Form thereby becomes the trace of a history of binding. This trace is never the condition itself. But it carries information about it. It is precisely here that the methodological possibility of preontological research lies. If the preontological is not directly accessible as an object, its conditions can nevertheless be reconstructed through recurring patterns of deformation, formations of boundaries, shifts, irreversibilities, and movements of opening. The form does not replace the condition. It points toward it. For this reason, non-closure is epistemically not the same as ignorance. Ignorance can be reduced through additional information. Non-closure, by contrast, remains even as knowledge increases, because every new piece of knowledge itself produces another stabilization and thereby new conditions, new exclusions, and new open relations. More knowledge does not eliminate non-closure. It shifts its form. This is the decisive difference. Non-closure does not designate the point that science has not yet reached. It designates the condition under which science can continue at all. The form is therefore not proof. It is a trace. If non-closure is a fundamental condition of every world-form, form and morphology can be read as stabilized responses under conditions of non-closure. The question then is not how the Gap itself can be made visible. It is precisely not available as such. What becomes researchable instead are those differences, deformations, and alternative possibilities of form that arise when an existing stabilization is varied or brought to its limit. Put pointedly, one 18 might say that this is about working productively with distortion: not optimization toward a final form, but the question of which conditions sustain a form and which other forms become possible under altered conditions. And whether reality remains world precisely where it cannot completely close itself into a thing. If we want to know which response might still be possible, then within a relational field containing a Gap we cannot simply investigate a response in relation to something and derive form and knowledge from it, because the Gap is not recognizable as such. Rather, the space of possibility of positive and negative topology in the context of morphological possibilities becomes the level of research. As does the question of how the Gap can be made recognizable again in form in order to make the space of possibility of alternative form researchable at all. What is required, therefore, is the opposite movement of science. Put pointedly, one might say that this is about playing with distortion. Not optimization toward product maturity, but the question of what the condition of a form is, until one moves ever further toward its boundaries, where form and Gap almost dissolve into one another. In a certain sense, this is where art and research meet. 3. Seinsverschiebung (shift of being): Knowledge Changes Its Own Ground The concept of Seinsverschiebung (shift of being) is decisive because it prevents us from thinking of knowledge as addition. In the additive model, there is a world, and science adds more and more correct descriptions to it. In principle, the world remains the same. Only our map becomes more precise. Seinsverschiebung (shift of being) contradicts this image. Every stabilization consumes possibility. This does not mean that all possibilities not chosen must physically disappear. What is meant, initially, is something more structural: Once a world-form has become effective, later possibilities exist under altered conditions. Space has history. And history, in this sense, is not an archive. It is deformation. The world curve (Speed 2025a), which was derived from Seinsverschiebung (shift of being) in earlier works, follows precisely from this. If every stabilization alters the conditions of its own continuation, world-formation cannot proceed linearly. Repetition is never pure repetition, because the conditions at the second occurrence are no longer the same. Even when a form returns, it returns into a space that has already been altered. This is also where the connection lies to the concept of Eigenzeit described by the author in other works. (Speed, 2026h,i) Within this research programme, Eigenzeit does not merely designate a subjective perception of physical time. It marks the history of a local world-binding. Different systems, 19 organisms, institutions, and forms of life carry within themselves different histories of binding and shifting. They therefore cannot be synchronized arbitrarily without altering their conditions. Modern technology, however, creates the impression that synchronization is always merely a technical question. What can be calculated faster should be decided faster. What can be transported faster should become available faster. What can be digitized should be reconstructable at any time. What can be simulated appears reversible. But computational speed and world-binding speed are not the same thing. A society may be able to generate millions of new options within seconds and still require years to meaningfully integrate a single profound institutional shift. An ecosystem can be altered within a few hours by a technological intervention and require centuries to form a new stability. For this reason, effect is not a later addition to knowledge. If an effect alters the conditions under which further knowledge emerges, then effect itself becomes part of the epistemic problem. This is one of the more radical consequences of the research position proposed here. An ecosystem is then not only an object of biology. It is at the same time a condition of future knowledge. A society is not only an object of investigation for the social sciences. It is a space of conditions determining which researchers can exist, which questions can be asked, funded, understood, and thought at all. In this way, lifeworld and knowledge become more inseparable than a subsequent ethics of science usually assumes. One could also propose the thesis that the world has not become more complex in the sense of becoming less understandable, but that our methods of investigating it have increasingly standardized, objectified, and thereby simplified the grid, the filter through which we view the world. (Speed, 2026j,k) For research, objective proof counts as Everything, while the concrete experience of people counts as almost nothing. This massive imbalance must be stabilized anew. But if knowledge always means a distortion of space, then living space cannot be considered separately from knowledge; and if knowledge can be controlled while its consequences cannot, then more is required than an ethics of science. What is required is an opening toward shared responsibility—not only for the correctness of knowledge, but also for the altered conditions of further responses that every act of knowledge helps to bring forth. Knowledge distorts the space of possibility. Not because it produces the Gap, but because every stabilization alters the relational form in which non-closure subsequently remains effective and readable. 20 For this reason, the negative form of knowledge must also be taken into account. It is entirely conceivable that research might thereby succeed not only in exploring knowledge within a world, but in understanding the alteration of a world in turn as an alteration of knowledge, thereby making the question of which world enables which knowledge more central than the assumption that progress consists solely in exploring the knowledge of a world until that world is, as it were, fully mapped. This also has many political dimensions. 4. Why This Research Programme Is Called World-Formation Research In the previous works of the Operatoric Research Corpus, the concept of the preontological has repeatedly been used. It remains necessary for the theoretical ground of this research, but it is not sufficient to designate its entire field of inquiry. The distinction can initially be formulated quite simply: Preontology designates the categorical position of the question. World-Formation Research designates the problem space in which this question is pursued. Operatoric Research designates the way in which this research works. Preontology asks under what conditions anything at all can appear stably as an object, relation, boundary, observer, or world-form. It does not designate a temporally earlier or spatially deeper level of reality. This distinction was already developed in the preceding chapter. What matters here is only this: preontology is not a separate sphere that could be investigated like an additional object of physics. It is a position of questioning toward conditions that are not fully exhausted in the forms arising from them. A similar shift in the direction of ontological questioning can recently be found in Axelkrans (2026). In his outline of a Generative Relational Ontology, he argues that stable objects should not be presupposed as ontological primitives, but that their stabilization itself should be treated as requiring explanation. The question thereby shifts from already constituted entities to the conditions under which stable objects, relations, and levels of organization can arise at all. World-Formation Research clearly touches this perspective, but begins differently: it does not postulate a positive generative ground, but keeps the preontological condition open precisely as a position of questioning that cannot be fully positivized. World-Formation Research begins at this point, but does not remain there. Its research interest is directed toward world-formation. Here, “world” does not necessarily mean the totality of the universe, nor merely a subjective worldview, nor an exclusively socially constructed reality. World here designates a relatively stable field of distinctions, relations, possibilities, and boundaries within which something can appear, be known, be responded to, or become effective. A diagnostic order can in this sense form a local world-form. A legal system as well. A scientific model. A technical infrastructure. 21 These world-forms are not identical and must not be reduced to the same substance. They possess different materials, scales, conditions of validity, and disciplinary languages. What they share is only the question of whether and how a stable order arises, which differences it binds, and how this stability alters the conditions of further formation. “Formation” does not necessarily designate a temporal transition from a worldless state into a finished world. The concept is specifically not intended to claim a cosmogonic history of emergence. What is meant is the operative stabilization of a field and the alteration of its further conditions that is inseparably connected with this stabilization. World-formation therefore does not mean only: Something becomes form. But at the same time: With this form, what subsequently remains possible, visible, connectable, or thinkable changes. This is precisely where the connection to Seinsverschiebung (shift of being) lies. A model does not merely describe a market. As soon as actors act according to it, it changes the market. A technical infrastructure does not merely solve a problem. It produces new dependencies and new possibilities. A scientific category does not merely sort differences that already exist. It influences which differences can still be recorded and investigated at all in the future. World-Formation Research therefore investigates neither exclusively things nor exclusively processes. It investigates the relations between stabilization, form, and altered conditions of continuation. This also explains why its field of inquiry lies across established disciplines. A problem can arise in one discipline and unfold its decisive consequences in another. A medical category has clinical, legal, economic, and biographical effects. A technical innovation can simultaneously alter computer science, work, language, education, and political institutions. An economic decision can alter ecological conditions that later, in turn, influence the space of possibility for biological and social research. Transdisciplinarity here therefore does not mean bringing together as many disciplines as possible as broadly as possible. It arises where the effect of a world-form crosses its original disciplinary boundary. World-Formation Research consequently lies neither above nor below the specialized sciences. It lies across their forms of world-building. It does not decide what is disciplinarily true within physics, biology, medicine, or sociology. Rather, it asks what happens to the wider space of conditions when a discipline-specific stabilization becomes effective. This is also why World-Formation Research is not a universal theory. It does not search for a single form behind all forms. It does not claim that a diagnosis, an event horizon, a market, and a public authority are the same thing. 22 It investigates whether comparable problems of conditions can arise in very different fields. The methodological prerequisites for such a comparison—analogy, homology, operatoric displacement, and the possible failure of transfer—are treated separately later. The name World-Formation Research therefore designates less a new discipline than a research programme. Its object is not “the world” as a finished totality. Its object is world-formation under conditions of non-closure. This can also be clearly distinguished institutionally within the Corpus: The Operatoric Research Corpus is the overarching research architecture. World-Formation Research designates the research programme emerging within it. Studies in World-Formation designates the series of concrete investigations. Preontology forms a theoretical ground. Operatoric Research designates the mode of research. And Negative Morphology or provoked empiricism are central methods. This also explains why such different fields of research can appear within the same Corpus. Their shared interest lies in the question of how world-forms become viable, which boundaries they produce, and how their stabilization alters the conditions of further world- formation. World-Formation Research therefore does not investigate a world behind worlds. It investigates the formation of world. And the decisive question is not whether this formation can ever be completely concluded. Precisely its non-concludability is its point of departure. ## 5. Operatoric Research Corpus and Operatoric Research The overarching body of work within which this research direction emerged has for some time carried the designation Operatoric Research Corpus. The series Studies in World-Formation is a concrete publication series within it. These designations were not originally developed in order to institutionally establish a new scientific discipline. They emerged from the work itself. In retrospect, however, it becomes clearer what is meant by “operatoric.” Operatoric research does not begin with the finished object. It asks about operations. Opening. Binding. Separating. 23 Folding. Condensing. Stabilizing. Returning. Excluding. Shifting. From this perspective, a thing does not appear as a self-evident point of departure, but as a provisionally stable response. This does not mean that things are unreal. It means that their reality has a history. Here again, this may initially appear abstract. But when applied concretely to social systems, it produces, for example, movement, irritation, internal perspectives, and reversals. In fields of the natural sciences, this can lead to the recognition of blind spots or ontological contradictions. The Operatoric Research Corpus is therefore not to be understood as a collection of mutually independent applications of a basic theory. Its texts stand in feedback relations with one another. A later investigation changes the readability of an earlier one. A concept that first emerged within consciousness research can become readable differently in an analysis of artificial intelligence. A welfare-state case study can in turn make visible the political consequences of an abstract concept of closure. A physical boundary case can make more precise a preontological condition that had previously been formulated too vaguely. The Corpus itself is therefore not fully archivable, although its texts can of course be archived. Its significance lies not only in the individual texts. It lies in their shifting relations. In this sense, the Operatoric Research Corpus is the research architecture. The Corpus pursues the aim of repeatedly making research possible anew, of opening it, rather than primarily producing individual products. It is an investigation into the requirements of knowledge and the preservation of a form of knowledge that expands living space instead of narrowing it so severely that this knowledge in turn becomes obstructed. This, however, is not a classical “freedom of research,” because such freedom could produce precisely the opposite, but rather an ongoing recalibration of the conditions of form- formation within a world. It is not without reason that Artistic Research is another valuable approach within this research. In practice, activist moments are also often added, as are altered forms of experiencing world shaped by neurodivergence. The more human beings intervene in world, the clearer becomes the necessity of developing knowledge about what world-binding, world-capability, or emergence can mean. For we have long ceased to be merely observers of stones, plants and animals, atoms, particles, or waves; knowledge itself also changes world. 24 World-Formation Research therefore designates the research programme within which this can be observed and worked on. Studies in World-Formation designates a series of individual investigations within this programme. 6. Operator, Counter-Position, and Provoked Empiricism: How a World Is Brought to Respond Up to this point, World-Formation Research has been described primarily as research on conditions, closures, and shifts in conditions. What remains open, however, is how such research proceeds in practice. The decisive methodological question is: How can a structure be made visible that, in its normal state, stabilizes itself precisely by keeping its own conditions invisible? Part of the answer lies in the operator. In this research programme, the operator is not a mathematical operator in the strict sense, nor is it a hidden object transferred between different fields of research. It is a deliberately posited relational movement. A question, an action, a contradiction, a demand, or a consciously produced difference is introduced into a field in order to make visible how that field responds to it. The operator is not arbitrary, but is based on a relational displacement in relation to an already stabilized order. The author experiences such a displacement particularly from his neurodivergent epistemic position. It is, however, not tied to neurodivergence. Comparable positions of difference can arise from marginalization, institutional marginality, disciplinary outsider status, or from a deliberately introduced methodological irritation. What matters is not their biographical cause, but that a position emerges that cannot be fully stabilized within the existing order and can thereby make its conditions visible. The operator therefore does not primarily attempt to represent an object. It sets something in motion. This happens because the introduced difference does not fit seamlessly into the existing order. The field’s response to this difference thereby becomes epistemically relevant. In this respect, operatoric research differs from a classical model of observation in which the researcher is supposed to intervene as little as possible in the object. The classical ideal is understandable: the less the observer changes, the more the observed seems to be recognizable independently of them. In many areas, this procedure is indispensable. It reaches a limit, however, where the structure under investigation is itself relational or where precisely the alteration of a relation becomes the object of investigation. This is particularly evident in social and institutional fields: minorities or marginalized persons can experience orders from a position in which not only their normality, but also the conditions of that normality become visible. In boundary areas of the natural sciences, the situation is different and must not be equated with this. Yet there too, observational position, accessibility, boundary conditions, or measurement relations can be decisive for which 25 structures become visible at all. The possible commonality therefore lies not in an identity between social marginality and physical observation, but in the more general research question of whether relational displacement can make forms readable as patterns of conditions that remain invisible within an already stabilized description. A marginalized position can, instead of the supposed self-evidence of a normality, make visible precisely those power structures through which this normality is stabilized. Conversely, a public authority, for example, does not possess a reality entirely independent of the citizen. A court does not exist only as a building, legal text, or organizational plan, but in the way it responds to applications, objections, deadlines, evidence, and concrete people. A clinic does not reveal its diagnostic structure fully in its guidelines or publications, but also in which evidence it accepts, which it rejects, how it deals with contradiction, and which forms of patient knowledge it can process at all. Relational distortion can produce moments in which normally invisible rules emerge or encounter their limits. Distance as a methodological stance, by contrast, can reinforce the impression that an order is more stable and independent of the observer than it actually is under real strain. In such fields, response is part of the object. It is not a lack of distance, but the proximity and involvement in a phenomenon required for observation. If the researcher removes themselves completely, precisely what is supposed to be investigated may disappear. A positing can therefore become the point of departure for an investigation. ## The Meaning of Positing Operatoric research often begins with a positing. A positing here is not a claim that is already established as true. It is a difference introduced into a space. Such a positing does not already have to be maximally differentiated in order to be scientifically precise. Its precision may only become apparent in the relations it produces. Sharpness does not necessarily lie in limiting a statement as narrowly as possible. It can equally lie in the strength of a positing that produces enough difference for a form to emerge in the reactions of the field and later be determined more precisely. (Speed, 2026c) Precision can therefore emerge retrospectively: not solely before the investigation through limitation, but within the investigation through the stability of the relations produced by a positing. For example: A legally guaranteed right is now actually to be claimed. An institution that claims transparency is asked to justify a concrete decision. A diagnostic institution that processes scientific evidence is confronted with extensive contradictory evidence. A public authority that has duties of protection is informed of a concrete danger and asked to respond. 26 The positing consists in taking an abstract or institutionally asserted order seriously in practice. A difference is thereby produced. For as long as a right exists only in law, it can remain fully compatible with the self-image of a state governed by the rule of law. Only when someone actually attempts to enforce this right against resistance does it become visible whether the formal and practical world-forms correspond to one another. Precisely this logic underlies the provoked empiricism developed in earlier works. In 1667:25 (Speed 2026b), applications, complaints, and danger notifications are explicitly described as empirical measuring instruments. The researcher is there not only an observer, but part of the experimental arrangement. Institutional responses—delay, shifts of responsibility, silence, formal narrowing—become data themselves. The positing is therefore, for example: This right exists. I am now attempting to make it practically effective. The operator is the actual execution of this experiment. The data are the deformations that arise between positing and institutional response. ## Provocation Does Not Mean Escalation The term “provoked empiricism” can easily be misunderstood. Provocation here does not necessarily mean aggression, scandal, or deliberate escalation. It initially designates the eliciting of a response. The Latin provocare contains precisely this movement: to call forth, to call out. A structure is not simply observed. It is placed in a situation in which it must act. This can happen through a lawsuit. Through a concrete request for justification. Through a contradictory body of evidence. Through a legally compelling question. Through the demand that a generally asserted norm be applied to a concrete exceptional case. The strength of this method lies in the fact that many systems appear largely free of contradiction in their normal state. A legal system has courts. A healthcare system has diagnostics. A university has quality standards. A public authority has procedural rules. The structure only becomes visible where its normative claims come under strain. The 1667:25 paper makes this point particularly clear. Formally, social courts, disability rights, and legal remedies exist. Only the actual attempt to enforce these rights under conditions of real dependency makes visible whether formal existence and material effectiveness are the same thing. This is a different form of legal research because it does not target the norm itself, but the real deviation and distortion produced by its existence. 27 The paper therefore argues that certain institutional structures emerge only where someone actually presses against their boundary. Provocation is, in this sense, a stress test of world-forms. It can, for example, serve to make perpetrators within structural violence recognizable in the first place. Because only through this does the friction required for structures to emerge arise. (Speed, 2026g) ## Counter-Position A second central procedure is the counter-position. It too is not primarily rhetorical. A counter-position brings two statements, models, or realities into a common field that can both appear stable within separate contexts. For example: The right exists. and The right is not practically enforceable. As long as both statements remain separate from one another, no necessary contradiction arises. The first statement belongs to the legal order. The second can be treated as individual failure, misunderstanding, or exception. Operatoric research brings both statements into the same relation. The question then is no longer: Does the right exist? But: What happens when the formally existing right is to be practically actualized? The difference between the two world-forms becomes the research space. The same procedure appears in the Humboldt paper (Speed 2026a). There, the statement is not simply: The clinic has decided that autism is not present. A second, independently viable architecture of evidence is set against it: There exists a decades-long biographical, functional, psychometric, and scientifically reflected structure that may not be sufficiently explained by this result. The central question is then not immediately which of the two sides is right. The more interesting question is: 28 How does one of two competing models of reality become institutionally dominant? What happens to the other evidence? Which parts are integrated? Which are fragmented? Which are reinterpreted? Which lose their epistemic status over the course of the procedure? This is precisely what the Humboldt study reconstructs by examining how extensive long- term evidence is selected, translated, weighted, or excluded within a time-limited diagnostic procedure. The follow-up study on the consequences of violence then shows how one form of knowledge affects the other. (Speed, 2026f) The counter-position is therefore not a simple pro-and-con. It is a method for producing epistemic tension. ## Why Tension Can Produce Knowledge In the classical research process, contradiction is often treated as a problem that is to be resolved. A hypothesis is compatible with the data or it is not. Two models compete until one offers the better explanation. This logic remains important for World-Formation Research as well. Operatoric research, however, is additionally interested in the time before resolution. What becomes visible while two not fully compatible orders are maintained simultaneously? This intermediate position is epistemically productive. For a stable world-form can conceal many of its presuppositions as long as it does not come under pressure. Only when a second form does not fit into it do its boundaries become visible. The counter-position therefore does not produce truth. It produces friction. And friction, as already stated, makes structure visible. In the diagnostic example, the patient’s self-report can be regarded as subjective. The clinical interpretation, by contrast, appears institutionally as professionally objectified. The counter-position now reads: If the subjectivity of the patient is taken into account as a possible source of error, why is the subjectivity of the clinical interpretation not examined with the same systematic attention? The demand for symmetrical error sensitivity in the Humboldt paper arises precisely from this reversal. Self-deception on the part of the person being assessed is diagnostically relevant. Institutional self-confirmation is equally so. The operator here consists in reversing the direction of observation. Not only: Can A be wrong? But: 29 What happens when the same assumption of error is applied to B? This operation appears simple. But it can shift an entire epistemic order. ## Operatoric Reversal Many of the counter-positions used in the Operatoric Research Corpus have this structure. The dominant statement is not simply denied. It is reversed. Not: Why can the patient not assess their own situation objectively? But: Why should the institution be able to assess its own situation completely objectively? Not: Why does the person not fit into the system? But: Which persons can this system carry at all? Not: Why is a right not successfully enforced? But: What kind of right is it if its actual enforcement structurally fails under the given conditions? Not: Is a person atypical? But: Which population was previously produced for this person to appear atypical at all? Not: What deviation does the individual display? But: What deviation does the grid produce? This reversal is not a trick. It changes the point of observation. The original truth initially remains in place. It is not immediately replaced. But it loses its monopoly position. Precisely this is the operatoric function of the counter-position. It opens a stabilized space. 30 Provoked Empiricism as a Real Response Structure In the next step, the counter-position is not only formulated theoretically. It is worked through in practice. This is the point at which provoked empiricism begins. If a public authority claims that a right is available, an attempt is made to actually use that right. If a clinic claims to comprehensively assess evidence, extensive evidence is introduced. If a procedure promises participation, corrective objection is made. If an institution claims transparency, justification is demanded. The response of the system is documented. This produces an important shift. The research no longer investigates only what an institution says about itself. It investigates what it does when its own claims are to be realized in practice and it is taken at its word—in other words, when one attempts actually to live the ideals of society. In the Humboldt paper, this procedure is explicitly described: applications, inquiries, objections, and demands for justification place institutional structures in situations in which actual decision-making logics emerge. The resulting reactions are not regarded merely as side effects, but as empirical material in their own right. This changes the status of conflict. Conflict is no longer merely a disturbance of research. It can be a research situation. The Researcher as Part of the Operator This leads to a difficult consequence. In provoked empiricism, the person of the researcher cannot be completely separated from the instrument. This is methodologically risky. At the same time, it is the condition for certain data. In the 1667:25 paper, this problem becomes particularly clear. The response of an institution to a genuinely dependent, vulnerable person is not necessarily the same as its response to an external scientist. Precisely for this reason, an external observer cannot fully reproduce certain situations. The institutional response there is itself a function of the relation of power and dependency. This does not mean that the affected person is automatically right. Nor does it mean that subjective experience replaces proof. It means only: The position of the researcher can be an empirically relevant variable. In certain research fields, it therefore cannot simply be calculated out. The classical demand for maximum distance can paradoxically destroy information here. If a system responds to differences in power, the power difference must be present in the experiment. If an institution responds differently to neurodivergent communication than to standardized academic communication, precisely this difference belongs to the object of investigation. If a diagnostic system responds to contradiction, the contradiction cannot be removed without making the corresponding structure invisible. 31 A methodological affinity exists here with Harold Garfinkel’s ethnomethodological breaching experiments. Garfinkel made social orders visible by deliberately disturbing taken-for-granted expectations and then examining the responses through which participants attempted to restore the violated normality. The epistemic gain lay not primarily in the disturbance itself, but in the repair movements, justifications, and irritations it elicited. Provoked empiricism follows a related logic, but shifts it in a different direction: the researcher is not merely someone who introduces an artificial violation of rules into a social setting, but can themselves be the bearer of the real vulnerability, dependency, or difference to which the system under investigation responds. Precisely for this reason, their position cannot simply be removed from the experimental arrangement. The response to the person is then not merely an interfering effect, but part of the structure to be investigated. In this sense, the researcher becomes not only an observer of a boundary reaction, but an element of the operator through which the hidden presuppositions of the order become visible. The proximity to Garfinkel therefore has clear limits. Provoked empiricism does not necessarily depend on violating a social norm. Conversely, it can consist precisely in consistently actualizing an institutionally asserted norm and thereby eliciting the difference between formal order and real functioning. Whereas breaching experiments primarily make implicit background rules and the repair of social normality visible, provoked empiricism can investigate longitudinal institutional form-sequences, power asymmetries, and shifts in conditions. The person of the researcher is not arbitrarily interchangeable: their real dependency, vulnerability, or neurodivergent position can itself form a variable of the institutional response. The epistemic interest also does not end with the reconstruction of a social rule. The question is which world-form stabilizes itself in the response, which possibilities it excludes, and how this stabilization alters the conditions of later responses. In this sense, the breaching experiment forms an important methodological relative, while provoked empiricism extends it toward longitudinal, power-related, and world-forming processes. Self-Use as a Measuring Instrument Within this research tradition, the researcher’s own person is therefore sometimes deliberately used as a measuring instrument. This is described particularly clearly in the Humboldt paper. There, the researcher’s own position is used as a catalyst for institutional processes; conflicts are not only experienced personally, but documented as provoked empiricism. This step is methodologically delicate. A measuring instrument can influence the measurement. But precisely here lies the difference from naive subjectivism. Operatoric research does not deny this influence. It attempts to make it explicit. In the Humboldt paper, documented event, subjective experience, scientific interpretation, and political intensification are therefore separated from one another. At the same time, it is 32 explicitly acknowledged that personal involvement can influence perception, selection, and intensification, and that the internal motives of institutional actors cannot be reconstructed with certainty. The researcher is therefore not neutral. But they must be transparent. This is a different epistemic contract. And here too it holds: many of the author’s research contributions would not have been possible, or would not have produced these data, had he not proceeded in this way. Classical methods can also scarcely reach such data and insights. Situated knowledge can be central and can help dissolve blind spots of a purely objective approach. Nor should it be forgotten that data also emerge where the author makes mistakes—that is, also where he fails with this approach. Certain data, however, can only truly be evaluated retrospectively and externally. What the author is doing here is initially documentation and partial analysis, but not, or to a much lesser extent, an asserted claim to final determination. ## The Operator Does Not Produce Truth At this point, a boundary is necessary. A provoked response does not automatically prove the interpretation by which it was elicited. If an institution remains silent, no particular motivation necessarily follows from this. If a court delays, intention cannot be inferred from this alone. If a clinic responds defensively to criticism, this does not yet prove diagnostic bias. Operatoric research must not confuse response with cause. What it initially produces is a form. For example: Silence. Delay. Fragmentation. Role displacement. Pathologization. Diffusion of responsibility. Self-protection. These forms must then be interpreted. And the interpretation remains fallible. The operator therefore does not serve to produce a previously desired result. It serves to place a space under sufficient strain that previously invisible relations can emerge. 33 ## The Importance of Resistance A particularly important data point is resistance. If a positing is taken up without difficulty, this may mean that it is connectable within the existing world-form. If, by contrast, a positing produces strong deformations, precisely this response can become epistemically interesting. For example: A patient statement is accepted without difficulty as long as it fits an expected diagnostic narrative. As soon as it contradicts the institutional interpretation, its epistemic status changes. A public authority responds routinely as long as an application remains within familiar categories. As soon as a person simultaneously raises several structurally connected problems, the concern is fragmented. A scientific theory processes data without difficulty as long as they fit within the expected space of variables. A boundary case, by contrast, produces additional assumptions, exclusions, or auxiliary constructions. The question then becomes: What does the resistance reveal about the boundary of the world-form? It is precisely here that provoked empiricism already touches negative morphology. Resistance is the deformation. The deformation later becomes a trace. ## From Response to Form The operator therefore produces a sequence: Positing. Response. Resistance or connection. Shift. New form. This new form is the actual object of research. In the 1667:25 paper, for example, not only a single court proceeding is examined. Several proceedings are placed alongside one another. Shifts of responsibility, legal fragmentation, delays, and formal narrowing condense into a recurring pattern in which law remains formally open while its practical effectiveness is blocked. Precisely here, events become morphology. A single event may be accidental. A single delay may be banal. A single shift of responsibility may be legally necessary. But when similar 34 deformations under different conditions repeatedly produce the same functional direction, a form emerges. This form can then be compared with other fields. But only then. The Operator as an Epistemic Probe The operator can therefore be understood as an epistemic probe. A probe introduces a defined movement into a field and registers the response. The operator need not always take the same form. It can be a demand. A reversal question. An experimental contradiction. A legal action. An artistic intervention. A theoretical positing. What matters is not its external form. What matters is that it produces a relational difference to which the field under investigation must respond. For example: Operator: Establish symmetry. Patient and clinic are subjected to the same possibilities of error. Operator: Practically actualize the norm. A formal right is actually claimed. Operator: Maximize evidence. A system is confronted with substantially more material than its routine normally processes. Operator: Reverse perspective. Not the deviation of the person, but the system’s capacity for selection is investigated. Operator: Place the boundary under strain. A system is used within its own rules until it becomes visible where its capacity for binding ends. The result is not predetermined. Precisely for this reason, the operator can be scientifically productive. 35 ## Counter-Position as De-Monopolization In this way, the counter-position also acquires political significance. A dominant truth does not have to be false in order to become dangerous. It can become problematic when it is no longer recognizable as a particular stabilization under particular conditions. The counter-position does not destroy this truth. It relativizes its ontological claim. This is particularly important for marginalized groups. A medical category can be statistically and clinically highly useful. It must nevertheless not become identical with the entire reality of the people it includes or excludes. (Speed, 2026f) An economic category can be useful. It must nevertheless not determine that everything outside its measurability is not work. A legal order can be formally correct. It must nevertheless not infer from its formal existence that real legal protection is necessarily given. The counter-position keeps a second reality within the space. Difference thereby emerges again. And difference is the condition under which a closed world-form can be investigated anew. It is also the condition in the relation between living space and knowledge. If knowledge can no longer take place within life, life also takes place less and less within knowledge. We see the danger of such pure simulations today on many levels. What is reality and what is fake has become a very different challenge for us today than in earlier epochs. This paper disputes that reality is solely a question of fixing objects and rules. Without permanent negotiation with experienced reality, it may under certain conditions become a threat or a one-way street of knowledge. ## Interim Conclusion Operatoric research therefore does not simply mean asking unusual questions. It produces controlled differences. It places world-forms under tension. It forces abstract norms into practical actualization. It reverses established directions of observation. It uses resistance as a source of data. And it follows the form that emerges from this movement. It is research that co-forms world, but in doing so it does not seek the one knowledge; it seeks to preserve the capacity for knowledge. What matters to it primarily is not proof, but the next question. It begins from a world that can never arrive at a finished answer. This is a different worldview from that of classical science, but it is not a religion either; its aim is the investigation of experienced reality and existence. Provoked empiricism is a particularly radical variant of this approach because the positing does not occur only theoretically, but is introduced into real institutional fields. The person of the researcher can thereby become part of the operator. This increases the risk of 36 distortion. At the same time, it opens spaces of data that are not necessarily accessible through distanced observation. This, however, still does not explain why a form obtained in this way may be compared with forms from other fields. Precisely here begins the next methodological question. How does a local deformation become a structural relation? When is a comparison merely analogy? When does analogy become homology? And under what conditions may a form be transferred from one field to another? This question leads to the next chapter. 7. From Counter-Position to Homology: Why Forms Become Comparable Across Fields The operatoric research described so far initially produces local knowledge. An institution is placed under tension. A counter-position is introduced. A system responds. A boundary becomes visible. A contradiction condenses. A form emerges. This, however, does not yet explain why such a form should possess significance beyond the individual field. It is precisely at this point that the problem of analogy begins. Transdisciplinary research easily enters a dangerous zone here. As soon as similar patterns are observed in very different areas, the temptation arises to generalize these similarities immediately. Social exclusion then appears like a physical boundary. A diagnostic logic of selection resembles a filter. An institutional stabilization resembles a condensation. An economic crisis can be read like a phase transition. Such comparisons can be heuristically productive. But they can just as quickly become meaningless. For similarity alone proves nothing. A tree resembles a river system. A neural network resembles a transport network. A distribution of galaxies can visually resemble biological structures. No shared condition follows from such similarities. World-Formation Research therefore requires a more precise distinction. Not every analogy is a homology. And not every homology already points to a shared condition of world-formation. Analogy as a First Irritation An analogy initially arises where two phenomena appear similar in a particular respect. This is by no means scientifically worthless. Many research movements begin with analogies. They can shift perception. They can make a previously separated relation visible. They can produce a question that would not have arisen within a single discipline. 37 The problem begins only where the analogy itself is already treated as an explanation. If, for example, one says: An institution functions like an organism. Or: A financial market behaves like a physical system. Or: A diagnosis functions like a filter. Then only a direction of comparison has initially been opened. One does not yet know which properties may actually be transferred. Nor does one know which properties specifically must not be transferred. World-Formation Research therefore does not treat analogy as a result. It is an entry point. An irritation. An indication that a structural comparison might be worthwhile. The point, after all, is to make the trace of the Gap visible, thus the response to... This in turn cannot be read from the object itself, but from relations of conditions. ## From Surface to Relation The first step is to detach the similarity from its visible form. Not: Both things look similar. But: Which relation is preserved in both cases? This is a crucial difference. The paper Thinking in Formation (Speed 2026c) describes a form of cognition in which knowledge does not primarily adhere to isolated statements, but to structures that remain under variation. A configuration is regarded there as recognized when it organizes itself in a comparable way under altered conditions. What is decisive is not identity, but transformability. Precisely this idea can be transferred methodologically to World-Formation Research. A form is then not relevant because it remains externally identical. It is relevant when its relational organization remains preserved under change. Let us take a simple example. 38 A diagnostic institution filters certain profiles out of a heterogeneous population. An algorithm filters certain patterns out of a data space. A court filters out from a complex life situation those aspects that are legally relevant within the respective proceeding. These three processes are not identical. The materiality is different. The aims are different. The norms are different. The consequences are different. But a relational structure may occur: A complex field is reduced. Certain differences are privileged. Other differences lose relevance. The result is subsequently treated as a sufficient representation of the original field. When the same relational structure recurs under very different conditions, the comparison begins to become scientifically more interesting. It is not yet a homology. But the analogy has become deeper. ## Functional Analogy The next stage is functional analogy. Here the question is no longer only whether forms appear similar, but whether they fulfil a comparable function. A diagnostic grid can transform heterogeneity into clinically manageable categories. An administrative grid can transform complex life situations into legally decidable questions. A technical model can transform complex environmental states into calculable representations. The functions differ. But in all three cases there could be a common functional movement: Reduction of openness in favour of manageability. This is already more informative. But functional analogy is still not sufficient. For many systems reduce complexity. That is almost a triviality. World-Formation Research becomes particularly interested only where it can additionally be shown that the reduction alters the conditions of further development. The diagnostic filter produces a selected population. The selected population influences later research. The research in turn stabilizes the diagnostic model. A sequence of forms emerges here. The Humboldt paper (Speed 2026a) describes precisely this possibility: if certain profiles are preferentially recognized as autistic, the diagnostic and scientific populations that emerge from this may later confirm precisely those theoretical assumptions according to which they were previously selected. The institution then does not merely observe an already existing object of research, but participates in its production. At this point, the question of homology becomes important. 39 ## Structural Homology Homology here does not designate biological homology in the narrow disciplinary sense. The term is used operatorically. What is meant is a structural correspondence in which not the external form, but a relational architecture remains preserved under transformation. A structural homology is present, for example, when the following sequence repeatedly becomes visible in different fields: An open field possesses more possibilities than a concrete order can hold simultaneously. An operation reduces this field. A stable form arises from the reduction. The stable form becomes capable of action. Through its capacity for action, it alters the original space. Later forms therefore arise under altered conditions. This sequence can look different in a clinical field than in a legal one. It can look different in a technical system than in an ecological one. The question is whether the relational architecture remains preserved. The cognition paper (Speed 2026c) describes a very similar mode of knowledge: contents can change while the operative structure remains stable. Knowledge there is recognized precisely by the recurrence of a structure under variation. This is probably the most important epistemological bridge between the cognition described there and World-Formation Research. The Structure Must Survive the Change of Field A genuine structural homology must survive a change of field. This means: If a relation remains visible only because its concepts are carried along unchanged, the transfer is weak. If, by contrast, the concepts can change while the operative structure remains preserved, the homology becomes stronger. Example: In the diagnostic field, the operation may be called classification. In law, allocation of jurisdiction. In an AI model, feature selection. In an economic model, risk weighting. The concepts differ. The concrete operations differ. But perhaps a shared structure remains: A field is selectively stabilized, through which certain relations become connectable and others invisible. It must then be examined whether this commonality has merely been produced linguistically or whether it actually carries analytically. Precisely here the actual research work begins. For these considerations can become essential when it comes to a deeper understanding of World-Formation, that is, of how and why forms arise at all. Why are democracies the way they are? Why have the oceans arrived at 40 this or that response? Where do reactions to other fields become visible in climate change, or where have behaviours within capitalism produced forms that marginalize particular forms of knowledge? ## Homology Is Not Identity The concept of homology must not tempt us to flatten differences. This would be particularly dangerous in transdisciplinary research. A physical boundary case has different conditions of validity from a social institution. A clinical diagnosis has different normative functions from a mathematical model. A social relation of violence cannot be explained from physical equations. A physical theory cannot be proven from autobiographical experiences. World-Formation Research must not dissolve these differences. On the contrary. A homology is interesting precisely because different fields remain different. It does not show identity. It shows a possible structural correspondence under difference. One might say: The form is similar enough for a question to be transferred. But different enough for the answer to remain open. All of this is crucial because the effects of knowledge show themselves precisely in the fact that they exert broader influence on formats of thought, categories, and entire cultures, and produce transfers. The overall topology of a society, of a reality, changes within that distortion which was described as Seinsverschiebung (shift of being). If one wants to understand the effect of the distortion, it is fruitful to follow the sequences of forms of the morphologies within the topology. That is, the responses to the responses to the responses… ## Condition Homology World-Formation Research, however, goes one step further than merely considering homologies. A structural homology initially describes that a relational architecture recurs. The deeper question is: Why? At this point, the idea of condition homology emerges. A condition homology is present when different systems not only develop similar forms, but when these forms may represent responses to a comparable type of non-closure. This is the strongest and most risky stage of transfer. An example: 41 In a diagnosis, a complete biographical field cannot be processed simultaneously in all its depth. In a court, an entire life situation cannot be decided as such, but must be broken down into legally relevant subquestions. In a technical model, the entire environment cannot be contained in its full reality, but must be transformed into modelable states. The systems differ fundamentally. But in all three cases, a similar problem of conditions exists: The field to be processed is larger than the form that is supposed to stabilize it. The systems respond to this with reduction. If this reduction is subsequently treated as though it were identical with the original field, a second problem arises. The form exceeds the reach of its own knowledge. Precisely here a condition homology could be present. Not: Court, diagnosis, and AI are the same. But: All three must deal with a difference between field and stabilizable representation. And all three can fail when this difference is made invisible. That would be a genuine World- Formation question. Knowledge within one discipline can therefore already not be considered in isolation because it is subject to distortions from other fields that were themselves earlier responses to non-closure. History does not develop as progress, as the best of all possibilities, but as a response to non- closure following from other responses that occurred before or at the same time. A war forces scientists to emigrate. In another culture, this produces a different view of realities that later shapes physics. But why should we simply accept that knowledge will henceforth develop only as a consequence of responses that arose from a war? This example fundamentally deprives progress of the authority to represent the best or most meaningful solution. Science can also remain subjected to a narrowed path for decades or longer. But how does one work one’s way out of it again? How does opening become possible when everything has been fixed? By researching not only things, but examining responses to the Gap as responses that stand in relations. The relation of proximity to the Gap determines the options when knowledge not only forms world, but world becomes the condition of knowledge. One can, for example, pose as a thought experiment the question of what would happen if universities were dissolved and every citizen were declared a co-researcher. Certainly, many important achievements of academic practice would also be lost, but one can sense which forms of liberation might be connected with this as well. Particularly for the evolution of knowledge in the sense of research. The opposite of this would, in a sense, be the pressure of 42 third-party funding. The question is simply at what point, and in what form, a re-set is sometimes needed. Such a re-set would not mean rendering institutional knowledge, disciplinary standards, or scientific expertise worthless. What would be more interesting is which questions might suddenly arise if research no longer first had to pass through career paths, disciplinary boundaries, funding logics, institutional reputation, and defined responsibilities. Perhaps many of these questions would fail. Perhaps they would be methodologically weak. But perhaps relations, objects, and problem descriptions would emerge at the same time that never become researchable at all within the existing order. Precisely herein lies the epistemic significance of the thought experiment. It does not ask which institution is better, but which forms of knowledge a particular institutional stabilization enables and which it excludes before they can even emerge. The pressure of third-party funding forms an interesting counter-movement to this. Where research must increasingly justify why a project is already relevant, connectable, fundable, and promising before it is carried out, the new can only appear in forms that become readable early enough in relation to the existing order. Research then does not simply become less free. It can also become preselected. The space of possibility of research is narrowed to those questions that can already make their later viability plausible in the language of the present order. A re-set in this sense would not be a romantic return to institutionless knowledge. It would be a deliberate destabilization of the conditions under which knowledge is currently permitted to emerge. The decisive question then is not whether institutions are necessary. They probably are. Rather: When does a research order stabilize knowledge so strongly that its own form begins to obstruct the evolution of knowledge? Counter-Position as a Test of Homology The counter-position described earlier acquires a second function here. It serves not only to open a dominant truth within a field. It can also serve as a test instrument for structural transfer. Take the counter-position: The model is successful. against The model alters the conditions of its own success. This counter-position can be introduced into different fields. In diagnostics: The diagnostic model successfully identifies certain profiles. But does its practice of selection produce populations that later confirm the model itself? In AI: 43 The language model produces linguistically successful outputs. But does its social use alter the data space from which later models learn? In the financial system: The risk model optimizes individual decisions. But does its widespread use produce correlations that alter systemic risk? If the same counter-position becomes productive in different fields, the probability increases that more than mere metaphor is involved. The operator then also serves as an instrument of comparison. Homologies point toward similar fundamental forms of reality. It is somewhat as though one were resetting the operating system of a device to its starting point. The updates are deleted and one returns to the standard configuration. This is the level of homologies. By producing a difference or deviation through counter-positions, one forces a system to give a new response to the Gap thereby provoked. This response shows new traces of the Gap, and thus new variants of possible morphologies. One quasi-restarts reality. The homology is the hardware; the software has to be rewritten. Classical progress works with updates. World-Formation Research works with resets, reloads under altered conditions, and investigates what happens—what appears as a new response. The Operator as an Invariance Test This allows the operator to be specified more precisely. It is not only a probe within a field. It can also function as an invariance test between fields. One takes a relational question and shifts it. For example: What happens when a form alters its own environment? This question is then introduced into different domains. Not with the expectation of the same answer. But with the question of whether a comparable relational structure emerges. If it does, the matter is examined more closely. If not, the transfer ends. This is important. World-Formation Research must not attempt to save every positing. An operator can be productive in one field and explain nothing in another. This is not a failure of the research programme. It is a necessary negative result. ## The Failure of Analogy An analogy has failed when only the language remains. 44 For example: If an institutional boundary appears comparable to an event horizon only because both are called “boundaries,” no relevant homology has been shown. If “condensation” in a social system merely uses the same poetic word as in a physical description, this is not enough. If “energy” is shifted into one another psychologically, socially, and physically without clarification, this produces not transdisciplinarity, but category confusion. World-Formation Research must therefore develop a discipline of translation. With every transfer it must be clarified: Which concepts are used literally? Which metaphorically? Which heuristically? Which structurally? Which possess mathematical or empirical definitions in their field of origin that do not apply in the target field? This clarity is not a restriction of the method. It protects its actual strength. ## Homology Is Gradual Structural homology too should not be treated as binary. Two fields are not simply homologous or non-homologous. The strength of the homology depends on how much structure survives the transformation. A weak homology may concern a single similar function. A stronger homology may encompass an entire sequence of operations. A very strong homology would preserve several conditions at once: Binding. Exclusion. Feedback. Eigenzeit. Irreversibility. Alteration of the conditions of continuation. The more of these relations remain stable across a change of field, the more relevant the question of a shared architecture of conditions becomes. But even a very strong homology proves no universal cause. It generates a research task. ## Why This Is Central to World-Formation Research Without this clarification, World-Formation Research could easily be misunderstood as a collection of creative comparisons. With it, the method becomes more precise. Research begins locally. An operator produces difference. The response produces form. The form is described in its relational structure. 45 It is then examined whether this structure remains preserved under transformation. Only afterwards is it asked whether different forms respond to a related type of condition. The movement therefore is: Analogy. Relation. Function. Homology. Condition homology. New disciplinary question. This is not an automatism. Every stage can break off. Precisely therein lies the scientific control of the procedure. ## The Political Significance of Homology The question of homology additionally possesses a political dimension. Dominant systems often explain exceptions locally. A person does not fit the diagnosis. A person fails within the administrative system. A court overlooks a relation. A model makes an error. Each case appears isolated. Morphological comparison can show that apparently local errors may be expressions of a recurring structure of conditions. The question then changes. Not: Why was this person unlucky? But: What form of system repeatedly produces precisely this kind of invisibility? This is particularly important for marginalized groups. Their experiences are often depoliticized precisely by being treated as individual cases. But when a recurring structure becomes visible under different conditions, individual failure can become a systemic research problem. This does not mean generalizing every individual case. It means treating it as a possible trace. In this way, responses that have led to sequences of forms can be developed backward again through homologies. A particular habit of systems to exclude minorities in a particular way may be historically concealed and may have become so deeply interwoven with other structural patterns that structural violence emerges which is practically invisible or difficult to grasp because it lies hidden within the sequences of forms themselves. It would then be a form of knowledge that conceals knowledge. 46 By irritating the logic of the structure through counter-position and related operations, difference can arise that makes visible what form was “actually meant” to be. An opening toward alternative patterns of order occurs. New order is therefore created not through linear or dialectical progress, but through dismantling clusters of sequences of forms that arose from earlier knowledge. A new world- binding can thereby emerge. A world that enables other knowledge or can carry other realities. It is entirely conceivable that, beyond a certain point, it will become increasingly decisive for human civilization not simply to continue stacking new knowledge on old knowledge, which can lead to a narrowing of society, but to understand the relational management of knowledge anew. Namely as two movements. Producing knowledge, and reopening sequences of forms. Thus expanding conditions. World-shaping and knowledge-finding must not be considered separately. This will also become necessary in order to distinguish pure simulations from lived realities. From the Individual Case to Structural Knowledge This also makes understandable why case studies play a special role in this research programme. A case does not have to be representative in order to be structurally informative. An exceptional case can become important precisely where it places under strain a boundary that remains invisible in the normal case. The movement is therefore not one that leads from distance to secure knowledge, but one that reopens the conditions of knowledge through destabilization. For this reason, qualitative research is often more fruitful here. Someone intervenes in the world and observes what new knowledge emerges as a result. That this happens precisely subjectively is a gain, a requirement of this research. For the subject already brings the difference with them. The Humboldt paper therefore does not claim to derive the frequency of false-negative diagnoses from one case. Its value lies in making otherwise hidden processing steps visible and generating new questions that must subsequently be tested in further cases and with independent data. This is a pattern for World-Formation Research as a whole. The boundary case is not a statistical average. It is a probe. It can make a form visible. The form can be compared. The comparison can generate a hypothesis. The hypothesis must subsequently be tested further beyond the individual case. ## Interim Conclusion World-Formation Research therefore does not work with analogy in the simple sense. Analogy opens the view. Counter-position produces tension. Operators bring structures to 47 respond. Morphology describes the forms that emerge. Homology tests whether relational invariances remain preserved under change. Condition homology finally asks whether different systems may respond to related forms of non-closure. Because they may historically be based on similar grids, that is, narrow from a similar or identical movement. This final stage is not a finding. It is a research question. Precisely for this reason, transfer must never eliminate the differences between fields. The stronger the commonality is claimed to be, the more precisely the differences must be documented. The scientific achievement does not lie in reducing everything to one form. It lies in recognizing when a form retains something of its structure under change—and from this developing a question that was previously not visible. The origin is confronted with the consequences, and through the difference a new response is forced, which in turn produces differences that become the basis for the expansion of morphological diversification. Knowledge can thereby become more complex. Research then moves not only forward along a timeline, but expands the basis of what already exists. A further question finally leads to a core of this entire research and its necessity. For it has emerged predominantly from divergent experiences of reality and world. The author is neurodivergent (AuDHD). Why, then, does this kind of comparison become so urgent for certain researchers, while within classical scientific procedures it can easily appear too relational, too subjective, or insufficiently stabilized? Why does knowledge arise in some cognitive positions precisely through displacement, counter-position, relation, and form? And why can keeping competing world-forms open be not only methodologically interesting, but existentially necessary for people whose existence is only inadequately captured by dominant categories? This begins a question of neurodivergent epistemology that reveals the origin of this entire research. 8. Neurodivergent Epistemology: The Position from Which Non-Closure Becomes Visible Up to this point, World-Formation Research has been described as a method that investigates closures, counter-positions, sequences of forms, and structural homologies. This explains how the research works. What has not yet been explained, however, is why precisely this kind of research emerges at all. This question is not incidental. For methods do not emerge from a neutral space. They emerge from particular forms of perception, particular experiences of world, and particular conditions of knowledge. What appears self-evidently relevant as a difference to 48 one person may be barely visible to another. What an institution regards as a meaningful stabilization can be experienced immediately as loss by someone whose existence is distorted by that stabilization. World-Formation Research therefore cannot be understood independently of neurodivergent epistemology. This does not mean that it is an “autistic science.” Nor does it mean that autistic or other neurodivergent people would fundamentally know more truthfully, more deeply, or more objectively. Such a claim would be just as problematic as the reverse assumption that classical scientific knowledge is cognitively neutral. The relevant point is different. Different cognitive organizations can possess different epistemic properties. And these properties can lead to certain forms of closure, loss, distortion, or relation becoming visible earlier, more strongly, or at all. The paper Thinking in Formation describes precisely such a possible organization of knowledge. Knowledge there does not appear primarily as a finished propositional statement, but as a structure that forms in enactment. Positing, displacement, condensation, recurrence under variation, and Eigenzeit are not merely stylistic characteristics, but components of knowledge production itself. The cognition described there does not stabilize knowledge primarily through distance. It is embodied and acts enactively. It stabilizes through relation. Not primarily through reduction. But through condensation. Not first through explicit definition. But through repeated transformation of a structure until it remains viable under variation. This produces an epistemic point of departure that is central to World-Formation Research. ## Knowledge Can Be Organized Differently In many of its methodological ideal forms, classical science proceeds from a particular movement of knowledge. The object is isolated. Confounding variables are reduced. The observer is expected to control their influence. Statements are formulated so that they can be tested independently of the concrete situation in which they emerged. Knowledge gains stability by detaching itself from the individual case. This movement is extraordinarily powerful. But it is not cognitively neutral. It privileges a form of knowledge in which distance, explicitness, separability, and repeatability are central values. The cognition paper, by contrast, shows another possible mode. There, knowledge arises precisely within situation, relation, and Eigenzeit. The statement is not always the place in which knowledge is fully contained. Meaning emerges across sequences. Concepts gain stability through repeated use. Context is not merely interference. It is part of the form. 49 Knowledge does not become stable by being separated from process, but because a structure repeatedly maintains itself under changing conditions. The difference can be expressed, in a strongly simplified form, as follows: One form of knowledge seeks stability through reduction of relation. The other seeks stability through an increase in relational density. Both can be rational. But they do not make the same things visible. Neurodivergence means a different access to world. This has direct consequences for relevance, perception, and the intensity of pain, recognition, or experience. If science is to be a study of reality, then neurodivergence necessarily means that the epistemic and cognitive foundation of classical science is not singular. This does not mean that neurodivergent people live in another reality, but it does mean that they are subject to different distortions of it, and this is not without consequence for the relations of form, content, sensory impression, pain, or intensity of reality and truth. This also affects research methodology. Many autistic people therefore conduct research very differently. Embodied, emotionally, directly, and in intense monotropism. Behind this lies a deep logic and rationality that can produce highly fruitful results in research. Here too, we see how standards can prevent knowledge. ## Why Closure Is Experienced Differently Precisely here begins the connection to non-closure. For a form of knowledge that gains clarity through reduction, closure may initially appear productive. An object becomes more precise. A category becomes cleaner. A statement becomes more unambiguous. For a form of knowledge that depends on relation, however, the same operation can produce loss of information. If a structure can only be understood as long as several levels, contradictions, and relations remain simultaneously within the field, then forced reduction to a single statement can destroy precisely what was supposed to be recognized. This is not an abstract difference. It appears in practice. The cognition paper describes how, under open, self-directed, and not fully predetermined conditions, complex cross-domain structures can emerge, whereas strongly standardized and externally fixed conditions can lead to marked functional breakdowns. It interprets this not as a global deficit, but as condition-dependent functionality. This makes it understandable why closure in this epistemic position does not appear only as a methodological gain. It can at the same time be experienced as a real loss of accessibility. A concept becomes more precise. But the field becomes poorer. A procedure becomes clearer. But certain relations disappear. A category becomes more stable. But precisely 50 through this, those persons can become invisible whose structure is not organized along that stabilization. World-Formation Research emerges from sensitivity to precisely this process. It recognizes epistemological and ontological boundary cases in which shifts occur that can lead to encountering world epistemologically in a different way. Such boundary cases are not found only in neurodivergence. They also appear, for example, where attempts are made to take animal languages or non-human communication seriously as an epistemic problem. The question is then no longer only whether an animal fulfils our categories of language, but whether our category of language is capable at all of capturing those relations within which communication is organized there. The same question would arise even more radically if one day an extraterrestrial intelligence wished to speak with humans about research. It would by no means be self-evident that its concepts of object, proof, time, causality, observation, or meaning would be compatible with ours. The first scientific task might then consist precisely not in translating its knowledge into our categories as quickly as possible, but first in finding out what kind of world has produced what kind of knowledge at all. This also becomes central in understanding artificial intelligence. Knowledge that emerges from embodiment is not necessarily structurally identical with knowledge that emerges from disembodiment, simulation, or only indirect world-binding. Likewise, knowledge that emerges from strong Eigenzeit-bound relation to a limited field can make different relations visible from knowledge designed for maximum scaling, interchangeability, and speed. What matters here is not to rank one of these forms of knowledge as fundamentally superior. What matters is that their respective conditions are not without consequences for the form of knowledge. Neurodivergence thereby becomes a particularly accessible case of a more general epistemic question. Within the same species, it already shows that different cognitive organizations can produce different forms of relevance, connection, precision, temporality, and world- binding. What appears in neurodivergence as a difference within human cognition could become much more radical in non-human or artificial intelligence. The classical academic paradigm of the last two hundred years could therefore be reaching a boundary. Not because objectivity, discipline, specialization, or institutional science would become obsolete. But because a research model that implicitly proceeds from a relatively uniform form of human knowledge may no longer be sufficient once science encounters heterogeneous subjects of knowledge. The coming question would then no longer be only: Who is allowed to conduct research? But: What can count as research at all when the bearers of knowledge possess different forms of world-binding? 51 Perhaps precisely there begins a science that no longer only produces knowledge about different worlds, but must investigate how different forms of world produce different forms of knowledge at all. Neurodivergence as an Epistemic Position, Not as Proof of Identity It would be wrong to derive an essentialist theory from this. Not every autistic person thinks relationally. Not every neurodivergent person works operatorically. Not every neurotypical person thinks linearly. Neurodivergence must not be used here as a biological guarantee of a particular mode of knowledge. More interesting is the concept of epistemic position. An epistemic position emerges from the interplay of cognitive organization, social position, institutional experience, and concrete relation to world. Neurodivergence can structure such a position. Marginalization can intensify it. Poverty can alter it. Institutional exclusion can make certain contradictions visible that appear less relevant to privileged persons. The point is therefore not to say: Autistic people see the truth. But: Certain neurodivergent and marginalized positions can produce forms of knowledge that are less accessible within dominant epistemic protocols. The cognition paper formulates this consequence clearly: If different protocols of cognition exist within the same scientific space and cannot be fully translated into one another, then science is more plural in its cognitive foundation than it usually appears. The decisive question then is not only whether a divergent form of thinking conforms to established criteria, but which forms of cognition remain hidden from us as long as only criteria corresponding to a particular cognitive mode are valid. ## The Marginal Position Does Not Automatically See More Standpoint Epistemology (Harding, 1991, 2004) has long pointed out that marginalized positions can perceive certain power structures more clearly. World-Formation Research does not derive from this any automatism of epistemic privilege. Marginality does not produce infallibility. It can distort perception just as well. It can place emotional strain on research. It can overweight certain topics. But at the same time, it can produce data that do not emerge from a privileged position. The 1667:25 paper demonstrates this with particular clarity. It argues that certain institutional responses cannot be reproduced by an external observer because the institution responds differently to genuinely dependent and vulnerable people 52 than to researchers standing outside the system. The response is itself a function of the power position of the counterpart. This means: The marginalized position does not merely see something different. Under certain circumstances, it produces a different empirical world. The system responds differently. Different data therefore emerge. This insight is fundamental to World-Formation Research. For it shows that observer positions do not merely provide different perspectives on the same finished world. They can elicit different world-responses. ## The Observer Is Part of World-Formation This shifts the classical question of subjectivity. Usually, it is assumed: The more strongly the researcher is involved, the greater the danger of distortion. This remains correct. But it is incomplete. For distance too produces a specific form of world. Anyone who can observe only what appears under neutralized conditions does not automatically see reality as such. They see reality under conditions of distance. A court responds differently to an abstract legal-scientific text than to a person attempting to enforce a right in practice. A clinic responds differently to a standardized case vignette than to a complex patient who introduces extensive counter-evidence and contradicts the institutional interpretation. A public authority responds differently to a scientific inquiry than to someone whose livelihood depends on its decision. The observer is therefore not outside world-formation. The observer is a condition of its concrete form. This is particularly relevant for neurodivergent research. If an institution responds differently to unusual communication, high levels of detail, persistence, unusual self-representation, or atypical forms of relation, then this response is not merely a problem of the person. It is data about the institution. The interaction itself becomes the field of research. Autistic Epistemology as a Counter-Position to Epistemic Monoculture In this context, Autistic Epistemology does not merely designate “knowledge about autism by autistic people.” The concept can be understood more broadly. 53 It designates the question of which forms of knowledge can emerge from autistic or neurodivergent cognition and which scientific consequences arise when these forms are no longer treated merely as deficit, communicative peculiarity, or subjective perspective. This question becomes particularly sharp in autism research. For here, a science frequently structured neurotypically produces knowledge about people whose perception, communication, and cognition can diverge from precisely these norms. This produces an epistemic asymmetry. The institution decides which form of expression counts as credible. Which self-report counts as relevant. Which communication counts as coherent. Which deviation is treated as symptom and which as knowledge. The Humboldt paper makes this asymmetry particularly visible. ## Dissolving Dominant Truth Does Not Mean Arbitrariness Neurodivergent epistemology thereby directly touches the political significance of non- closure. For privileged epistemic positions, a stable truth can primarily mean security. A scientific achievement can be measured by how clearly a finding is fixed, cited, confirmed, and historically stabilized. For people whose existence is not carried by this truth, the same stabilization looks different. It can become a boundary. A labour-market logic says: This is not productive work. An institution says: Your form of communication is inappropriate. A science says: This form of knowledge is not methodological enough. Then the possibility of placing a dominant truth back into a relative space of conditions becomes existential. Not because truth would be unimportant. But because the alternative is otherwise that the person themselves becomes unreal. The central distinction is therefore: A statement can be empirically well founded. It does not therefore have to exhaust the whole of reality. This distinction is particularly important for marginalized groups. It keeps open the possibility that a scientifically successful model can at the same time be selective. 54 Counter-Truth as an Epistemic Operator The counter-position described earlier thereby acquires another meaning. Marginalized research often produces counter-truths. Not necessarily in the sense of an alternative ideology. But as statements that, within the dominant order, are initially not permitted to be true at the same time. The Epistemic Significance of Missing the Target The cognition paper describes another point that is important for World-Formation Research: knowledge does not always arise because a positing immediately hits its target. A positing that misses produces movement. Movement produces relation. Relation produces form. Form stabilizes retrospectively as knowledge. Classical epistemic logic initially evaluates missing the target as error. In the cognition described here, precisely the difference between positing and object can become productive. The relational precision of positing already described in Chapter 6 becomes cognitively understandable here. Neurodivergent Research as Preservation of Alternative Spaces of Possibility Here finally lies the direct connection to the political demand of World-Formation Research. If different cognitive forms produce different questions, data, and relations, then epistemic diversity is not merely a question of fairness. It is a condition of scientific future viability. A science that homogenizes its own cognitive modes of access too strongly can thereby reduce the space of possible knowledge. If only certain forms of writing are valid. Certain career paths. Certain forms of communication. Certain forms of objectivity. Certain speeds. Certain forms of social cooperation. Then not only a group is excluded. Possible epistemic operations are excluded. If science is to preserve the conditions of future knowledge, it must also preserve the conditions of different forms of knowledge. This does not mean recognizing every form of knowledge as equally valid. It must still be tested. It can fail. But it must not be excluded before testing simply because its form does not correspond to the dominant cognitive protocol. ## An Epistemic Asymmetry This allows the opening question of this chapter to be answered more precisely. Why might the necessity of non-closure become visible earlier from a marginalized or neurodivergent position? Because the price of closure is distributed unequally. Someone who is well represented in a dominant category does not necessarily experience its limitation. Someone who is falsely represented by it lives at its boundary. Someone who functions successfully within an institution can experience its grid as neutral. Someone who fails at this grid 55 experiences directly that grid and reality are not identical. Someone who easily masters a scientific language experiences its form as transparency. Someone who loses part of their cognitive structure in this form experiences it as translation with loss. Marginality thereby becomes epistemically interesting. The marginalized body becomes the place where the difference between model and world no longer remains abstract. It is lived. ## From Survival to Method In this sense, World-Formation Research can also be understood as the transformation of a survival strategy into a scientific method. Anyone who is incorrectly described by a dominant world-form must learn to distinguish between description and reality. The system describes me this way. But that is not the complete reality. This distance can be existentially necessary. It prevents institutional truth from fully coinciding with self-truth. From this elementary movement, a scientific possibility emerges. If a dominant description can be read relative to its conditions, then one can also ask: Which conditions produced this description? Which possibilities did it exclude? Which world does it produce? Which counter-description would produce another world? Autistic Epistemology as a Strengthening of Scientific Non-Closure World-Formation Research therefore also contributes to a stronger Autistic Epistemology. Not by claiming that autistic knowledge is fundamentally superior. But by showing that science can be cognitively plural in its organization and that this plurality is methodologically relevant. Autistic and other neurodivergent forms of knowledge can enrich science not only through additional perspectives. They can produce other operations of knowledge. ## Interim Conclusion World-Formation Research did not emerge from a neutral epistemological space. Its sensitivity to non-closure, counter-position, relation, and form is closely connected to a neurodivergent and marginalized epistemic position. This position provides no privileged access to truth. But it makes certain epistemic costs visible. It shows that closure does not only produce clarity. It can destroy affordances. It can render people invisible. It can exclude alternative forms of knowledge. Through its own stability, it can shape later reality. Precisely for this reason, the question of non-closure is not merely theoretical for marginalized groups. It can be a condition of survival. Dominant truth must be capable of being placed back into a relative space of conditions so that an existence not fully captured by it remains thinkable. 56 How, then, can knowledge be stabilized without unnecessarily destroying the conditions of other and future knowledge? With this, the methodological point of departure for negative morphology is fully prepared. If direct fixation always also produces loss, if different cognitive positions make different world-responses visible, and if structural homologies are recognized not through identity but through relation under transformation, then research must learn to read more indirectly. It must read the deformations. The omissions. The resistances. The sequences of forms. The places where a world shows its own boundary. Precisely this is what becomes visible in negative morphology. 9. Negative Morphology: Indirect Reading Within World-Formation Research, Negative Morphology designates the attempt to make conditions of world-formation readable through the forms of their responses. The necessity of this approach was already developed in Chapter 2. If non-closure does not exist as a positive object, and if every direct determination itself produces a new world-form, research cannot hope eventually to move behind all forms and uncover the Gap there as the final object. It can only follow what happens when something is determined, bound, irritated, shifted, or reopened. Negative Morphology is therefore not a search for the invisible behind the visible. It is research on the changes of the visible. It does not proceed from a world that can ultimately be known completely, but from the assumption that stabilized reality becomes readable in forms that can be understood as responses to non-closure. This fundamentally changes the concept of knowledge. Knowledge is likewise not a process moving toward finished knowledge, but cannot exist without counter-responses, without gaps, without distortion. A modern science of the next century might therefore seek not only to represent knowledge directly, but to work with more flexible systems of response. That is, with knowledge that is bounded and at the same time opened within multidimensional relation and within a context of life. The object of its attention here is the form-decision. What changes when something is stabilized as this and not as that? Which relations become stronger as a result? Which disappear? Which previously possible movement becomes less likely? Which new boundary arises? Which tension is absorbed, and which is merely displaced? And which further form is prepared by this decision? The form is not the condition itself. It is a trace of how a concrete world under concrete conditions responds to something that it cannot completely close within itself. 57 Not the Form, but Its Difference Negative Morphology thereby differs from a classical theory of forms. It does not catalogue shapes. It does not primarily ask why an object possesses precisely this form. It is more interested in the difference between possible forms. It seeks to expand the basis rather than merely determine it. A chair lying on the floor and the same chair placed upright again differ, in object-related terms, only in position and orientation. Within a setting, however, this small change can make an entire order visible. Someone apparently regards one position as “correct.” Paths of movement change. Lines of sight open or close. Light falls differently. A social expectation emerges. Perhaps it is only through the correction that it becomes visible that a norm existed at all. The relevant information is then not fully contained in the chair. It lies in the difference between the configurations. Precisely this difference is productive for Negative Morphology. At this point, intervention through Artistic Research can be particularly fruitful. A stable order can conceal its conditions. As long as all elements stand where they are expected to stand, the order appears self-evident. Only displacement makes visible that an order was present. For this reason, disturbance, irritation, exception, and boundary case possess a particular epistemic significance within this research programme. Not because deviation automatically produces truth. But because the response to deviation provides information about the form that was not visible in the stable state. ## Negative Form The term negative here means neither non-being nor mere absence. A negative form becomes readable through what a positive form does not contain, does not permit, or alters in the course of its stabilization. A category, for example, possesses a positive definition. Its negative form, by contrast, becomes visible in the cases, relations, and possibilities that become less visible precisely through this definition. The same applies to institutional procedures: an individual section can be processed correctly while, through the division of the object, precisely its overall relation disappears. The negative form is therefore not a secret second reality behind the positive form. It appears in the difference between what an order can carry and what, through this carrying capacity, no longer has a place. Absence thereby also becomes readable relationally. 58 Negative form must not be confused with arbitrarily interpretable absence. What a form does not contain does not yet tell us why it is missing. Negative Morphology therefore requires documentable differences, recurring profiles of deformation, or controlled variations. The form is not proof of its condition. It is a trace whose viability is determined only by whether a more precise question emerges from it that can be tested outside its own conceptual vocabulary. ## 10. The Two Movements of Science Science therefore requires two movements. The first movement is familiar. It produces closure. It distinguishes one object from another. It isolates variables. It develops taxonomies. It produces models. It stabilizes knowledge to the point at which it becomes communicable, testable, and technically usable. Without this movement, there is no science. The second movement begins after success. It does not simply ask whether the result is correct. It looks again at the space. What is different now? Which presuppositions have been altered by the new knowledge itself? Which previously independent domains have been coupled? Which possibilities have been rendered invisible by the successful stabilization? Which actors can respond to the new world-form and which cannot? Which Eigenzeiten collide? Which previously open questions suddenly appear meaningless because the new model has already established its own presuppositions as self-evident? This second movement must not be confused with criticism. Its aim is not to tear down every closure again. A world that binds nothing would be no more capable of world than a world of complete closure. The problem is not stabilization. The problem is stabilization without renewed examination of the altered conditions. Science must therefore close in order to know. It must reopen in order to continue knowing. If knowledge alters its own space of conditions, then the investigation of this alteration is part of scientific rationality. Non-closure thereby becomes an epistemic resource. It is the garden in which future responses can arise. The garden may be ordered. But an order that eliminates everything unplanned eventually no longer has a garden. ## 11. Delimitation ## No New Metaphysics World-Formation Research does not introduce an additional substance. The Gap is not an ether. Preontology is not an invisible field beneath known physics. MNO (Speed, 2025e) designates, within the underlying Corpus, structural operations of openness, differentiation, and stabilization, not the assertion of a new particle or a hidden force. The research does not ask: What lies behind the world? 59 It asks: Which conditions become visible when world-formation reaches its limits? That is a different claim. ## Not Simply Philosophy World-Formation Research inevitably has philosophical dimensions. Its fundamental questions touch ontology, epistemology, and philosophy of science. It is not exhausted by these, however. The test of an operatoric analysis lies precisely in whether it can establish new relations to concrete empirical, technical, social, or formal problems. A diagnostic practice, an AI system, or a welfare state are not metaphors. They produce real consequences. If their forms are investigated using concepts such as closure, grid, Eigenzeit, or Seinsverschiebung (shift of being), the analysis must become connectable again to those consequences. Not Only an Ethics of Science The ethics of science usually asks what consequences research may have, which risks are acceptable, and how knowledge should be applied responsibly. World-Formation Research begins earlier. If every act of knowledge already produces a Seinsverschiebung (shift of being), then effect does not begin only with later application. The choice of object, the construction of a model, and the definition of a category already alter which reality can become institutionally or scientifically visible. The question therefore is not only: What should we do with this knowledge? But: What has this knowledge already done to the space of what can be known? ## No Replacement for Discipline-Specific Science World-Formation Research cannot decide which physical theory is correct. It cannot replace a clinical study with morphology. It cannot make a mathematical claim true without proof. That would be a category error. Its task is to identify boundary questions, make structures of conditions visible, and open research directions that might not arise within a single discipline. 60 In this sense, its strength lies precisely in limiting its own claim. 12. Artificial Intelligence: When the Production of Difference Becomes Faster Than World-Binding The current development of artificial intelligence makes the necessity of this second movement of research particularly visible. A large part of the public and scientific debate treats the performance of artificial systems as a question of scaling. More data. More parameters. More computing power. More context. More agents. More speed. Even where limits are discussed, they are often formulated as technical limits. Energy consumption, hardware, data availability, error rates, or algorithmic complexity. World-Formation Research asks a different question. What happens when the speed of difference production grows faster than the capacity of a world to bind this difference? (Speed, 2026l,m) A system may be able to generate millions of solutions without a society being able to meaningfully integrate millions of solutions. Computational capacity is not binding capacity. This remains true even if future quantum computers should radically accelerate certain calculations. Physical acceleration does not automatically abolish institutional Eigenzeit, human relation, ecological regeneration, or the formation of social responsibility. This could produce a paradoxical situation. The more intelligent a technical system becomes in the narrower sense, the more clearly it may become visible that intelligence itself is not the decisive limit. (Speed, 2026k) The limit could be bindability. A society requires grid-relations within which new differences can find connection. Law, public discourse, education, language, social relations, individual Eigenzeiten, and political institutions, however, do not possess an arbitrarily scalable speed of integration. If these grids lose their capacity for binding, additional intelligence does not necessarily produce additional progress. It can accelerate disintegration. This would also explain why the central question of artificial intelligence may be posed incorrectly. Not: How intelligent can a system become? But: 61 How much produced difference can a world carry without destroying its own conditions of response? AI research thereby shifts from the question of performance to the question of World- Binding. This would be an example of a research approach from the field described here. ## 13. Society, Normalization, and Epistemic Annihilation The same structural type becomes visible in social institutions, although in a completely different material and historical form. Administration must classify. It cannot begin from zero with every person. Law requires categories. Medicine requires diagnoses. Social policy requires criteria. Here too, closure is not the error. It becomes problematic where the category produced is confused with reality itself. An institutional grid has only limited resolution. What is not connectable within this grid can appear as deviation, unreliability, illness, refusal to work, or statistical noise. Neurodivergent forms of life and knowledge in particular make visible that this need not be a merely incidental communication problem. The institution does not simply see too little. Through its way of seeing, it can actively produce another world. The person must then conform to the category in order to become institutionally real. Epistemic injustice thereby becomes a problem of World-Formation. The decisive question is not only whether an institution has correctly recognized a person. It must also be asked which forms of existence the institution can carry at all through its conditions of recognition. This has immediate consequences for science. If research systems recognize only certain forms of productivity, education, language, publication, or methodological connectability, they do not merely lose individual persons. They lose possible forms of knowledge. Normalization can therefore produce epistemic annihilation. Poverty can produce epistemic annihilation. A research culture that permits only accelerated life trajectories that are easily connectable institutionally can reduce its own space of possibility. Social justice does not thereby automatically become science. But it acquires an additional dimension in the philosophy of science. Anyone who wants to enable future knowledge must also ask which forms of life can carry future knowledge production at all. 62 ## 14. Physical Boundary Cases The strongest temptation of a transdisciplinary approach lies in reading physical phenomena too quickly as confirmation of a general theory. World-Formation Research should resist this temptation. Singularities do not prove the Gap. Cosmic censorship does not prove preontology. Quantum-mechanical uncertainty is not simply another term for non-closure. Nevertheless, physical boundary cases are particularly important for the research. Why? Because a theory of world-formation must be able to do something where formally stabilized descriptions of world reach their limits. Singularities, event horizons, measurement problems, decoherence, Quantum Speed Limits, or fundamental limits of information can therefore serve as test fields. The question is less whether this or that phenomenon would be a form of MNO, and more: what kind of boundary of conditions appears here? Which quantities can be stabilized simultaneously? Which form of closure remains possible? Which information becomes inaccessible? Which relation changes through observation or measurement? Which physical limits prevent complete simultaneity, complete control, or unlimited differentiation? If different boundary problems reveal recurring morphological structures, a research question can emerge from this. Physics must then decide whether this question is physically viable. That would be an example of the actual usefulness of World-Formation Research. The point is therefore not to identify the next thing, but to expand the space of possible questions. At the same time, homologies also involve a return to more fundamental formations, which can then, in a re-start, be read partly anew and productively through the artificial positing of difference. ## 15. Quality Criteria A research programme that works across disciplines requires strict boundaries precisely because of its openness. World-Formation Research must not become immune to criticism by interpreting every criticism simply as another example of closure. That would itself be complete closure. An operatoric investigation should therefore meet at least the following requirements. It must specify which concrete condition is being investigated. It must distinguish between morphological similarity and structural homology. It must preserve the differences between the disciplines involved. It must make clear when a statement is preontological, when morphological, when empirical, and when normative. It must allow counterexamples. And, wherever possible, it should generate new questions that can be tested outside its own theoretical vocabulary. 63 One further condition is particularly important. World-Formation Research does not stand outside its own theory. This paper too produces a Seinsverschiebung (shift of being). By defining World-Formation Research, certain readings of the Operatoric Research Corpus are strengthened and others weakened. An open research context is brought into a new form. This form is necessary in order to make the programme communicable. But it must not be confused with its final state. The definition is therefore not a conclusion. It is a new working hypothesis within the same process that it describes. ## 16. Progress as Viability This allows the question of scientific progress to be posed anew. The classical conception is cumulative. More knowledge is initially better than less knowledge. Greater technical possibility initially appears as progress. World-Formation Research does not reject this in general. It does, however, add a second axis. Knowledge must be capable of being carried. A society can know more and still lose its ability to live meaningfully with that knowledge. A technology can become more powerful while at the same time reducing the diversity of possible future responses. A scientific theory can enable increasingly precise predictions while making its own presuppositions increasingly invisible. Progress would therefore not be measured solely by an increase in information or control. It would also have to be measured by the viability of the world-relations that arise from it. A possible working definition would be: Progress expands the set of viable world-relations without unnecessarily destroying the conditions of future differentiation. The word “unnecessarily” is decisive. Every world-formation closes possibilities. There is no innocent form. The task therefore cannot consist in keeping every possibility open forever. That would itself be worldless. The question is rather which closures are necessary and which closures destroy the space of possibility without a corresponding gain. The cultivation of non-closure thereby becomes a real scientific task. Not as a romantic demand for openness. But as the preservation of future capacity to respond. 64 17. Conclusion: Which World Can Carry Which Knowledge? Perhaps the fundamental scientific question of modernity has for too long been posed in one direction. How much can we know? How precisely can we measure? How far can we predict? How strongly can we control? How completely can we simulate? These questions have produced an extraordinarily productive world. But at the same time, they have created conditions under which their own productivity produces new forms of blindness. And they have partly forgotten the world within which they themselves must be bound. It has also partly been overlooked how strongly world- shaping affects the conditions of knowledge. The development of artificial intelligence makes this visible because the speed of modelling is now beginning to overtake the Eigenzeiten of many areas of society. The ecological crisis makes it visible because technical effectiveness can be faster than the regenerative capacity of the systems on which it acts. Neurodivergent and other non-normative forms of knowledge make it visible because standardized grids can obscure precisely what produces new information outside their own form. How do we exist, conduct research, and act meaningfully under the condition that complete closure is neither achievable nor desirable? World-Formation Research does not understand non-closure as a defeat of science. It is the condition of its future. Science bears responsibility for the space from which the next knowledge must emerge. This means not automatically assuming that the next thing is always the most meaningful response. ## Provisional Conceptual Architecture ## Operatoric Research Corpus Overarching architecture of work and research. ## World-Formation Research Transdisciplinary research programme for investigating the conditions, forms, and consequences of world-formation. ## Studies in World-Formation Series of individual investigations within the Operatoric Research Corpus. ## Preontology Foundational theoretical research position toward conditions that cannot themselves be fully stabilized as positive objects within the world they enable. 65 ## Gap / Non-Closure Not missing information, but structural openness as a condition of further differentiation and world-formation. Seinsverschiebung (shift of being) Irreversible alteration of the conditions of further world-formation through stabilization, naming, or intervention. ## Negative Morphology Indirect reconstruction of conditions that cannot be represented directly through boundary cases, sequences of forms, distortions, and recurring response profiles. ## World-Binding / Weltbindung Process through which difference is transformed into a viable world-form bound temporally and relationally. ## Eigenzeit History of local world-binding and thus a condition of further form-formation that cannot be substituted arbitrarily. ## Cultivation of Non-Closure Scientific and social task of preserving the conditions of future capacity for response, knowledge, and world-formation despite necessary closures. 66 Declaration of AI usage Generative AI tools, including ChatGPT, were used during manuscript preparation to support language editing, structural refinement, translation and the critical review and clarification of formulations. The research, empirical material, theoretical concepts, interpretations, and scholarly conclusions are those of the author. All AI-assisted material was critically reviewed, revised where necessary, and verified by the author, who assumes full responsibility for the final manuscript. References: Link to operatoric corpus: https://zenodo.org/communities/operatoric-research-corpus- archive Axelkrans, Erik. (2026). “Beyond Substance: Contemporary Science and the Case for a Generative Relational Ontology.” Philosophies, 11(4), 147. https://doi.org/10.3390/philosophies11040147 Bergson, H. (1907). L’Évolution créatrice. Paris: Félix Alcan. Deleuze, G. (1968). Différence et répétition. Paris: Presses Universitaires de France. Garfinkel, H. (1967). Studies in Ethnomethodology. Englewood Cliffs, NJ: Prentice-Hall. Gödel, K. (1931). Über formal unentscheidbare Sätze der Principia Mathematica und verwandter Systeme I. Monatshefte für Mathematik und Physik, 38, 173–198. https://doi.org/10.1007/BF01700692 Hegel, G. W. F. (1812–1816). Wissenschaft der Logik. Nürnberg: Johann Leonhard Schrag. Heidegger, M. (1927). Sein und Zeit. Halle an der Saale: Max Niemeyer. Harding, S. (1991). Whose Science? Whose Knowledge? Thinking from Women’s Lives. Ithaca, NY: Cornell University Press. Harding, S. (Ed.). (2004). The Feminist Standpoint Theory Reader: Intellectual and Political Controversies. New York: Routledge. Peirce, C. S. (1877). The Fixation of Belief. Popular Science Monthly, 12, 1–15. Popper, K. R. (1963). Conjectures and Refutations: The Growth of Scientific Knowledge. London: Routledge & Kegan Paul. Speed, T. (2025a). The Curve of the World - Why World-Binding Cannot Be Linear — Shift of Being, Time, and the Impossibility of the Archive (Version 1). Zenodo. https://doi.org/10.5281/zenodo.18097931 Speed, T. (2025b). The All–Nothing Paradox - Ontological Openness as a Condition of World-Formation - Why Closure – Not Complexity – Marks the Limit of Artificial Systems (Version 2 English). Zenodo. https://doi.org/10.5281/zenodo.18000820 67 Speed, T. (2019). Unfolding Gap - On the Structural Gap Between Work and Value (Version 2 English). Zenodo. https://doi.org/10.5281/zenodo.17951579 Speed, T. (2026a). The Epistemic Crisis of Autism Diagnosis: False-Negative Autism Assessment, Epistemic Injustice, and the Case of the Humboldt University Autism Clinic in Berlin (Versions 2.0 Corrections Page 18 / Abstract Correction). Zenodo. https://doi.org/10.5281/zenodo.22065932 Speed, T. (2026b). 1667:25 Violence, Health Harm and the Denial of the Rights of Autistic People in Germany - Proceedings before the Social Court Cottbus and the Higher Social Court Berlin-Brandenburg - On the Practical Impossibility of Enforcing Effective Legal Protection for Neurodivergent People (Version 1 English). Zenodo. https://doi.org/10.5281/zenodo.18887765 Speed, T. (2026c). Thinking in Formation: On a Form of Cognition That Produces Knowledge in Process (Version 1). Zenodo. https://doi.org/10.5281/zenodo.20049745 Speed, T. (2025d). The Physics of the Poor - A Neurodivergent Meta-Theory of Consciousness (Version AAM DNB Version English). Zenodo. https://doi.org/10.5281/zenodo.19051837 Speed, T. (2025e). MNO and Ontological Recurrence: A Non-Representational Account of Quantum Measurement and Conscious Experience (Version 1). Zenodo. https://doi.org/10.5281/zenodo.17913823 Speed, T. (2025f). Orch-OR with Recurrence: A Minimal Dynamical Condition for When Objective Reductions Yield Conscious Experience (Version 1 English). Zenodo. https://doi.org/10.5281/zenodo.17942531 Speed, T. (2026d). From Gap to World - Pre-Ontology, Indimergence and the End of Representational Physics (Studies in World-Formation, Vol. 1) (Version 1). Zenodo. https://doi.org/10.5281/zenodo.18983094 Speed, T. (2025g). Seinsverschiebung (Shift of Being) as a Pre-Ontological Category - On the Incompatibility of Existence and Understanding in Modern Regimes of Stabilization (Version 2 English). Zenodo. https://doi.org/10.5281/zenodo.18007628 Speed, T. (2026e). The Negative Movement of Eigenzeit: Non-Closure, Operator Formation, and the Conditions of Morphogenesis (Version 1). Zenodo. https://doi.org/10.5281/zenodo.20590372 Speed, T. (2026f). "You Just Want the Label": Embodied Identity, Diagnostic Power and the Violence of False-Negative Autism Assessment (Version 1). Zenodo. https://doi.org/10.5281/zenodo.22641619 68 Speed, T. (2026g). When Structural Violence Has a Perpetrator: The Apparently Legitimated Inactive Witness with Decision-Making Power (Version 1). Zenodo. https://doi.org/10.5281/zenodo.22206507 Speed, T. (2026h). Morphology Without Memory – Eigenzeit, Folding and the Limits of Simulation (Studies in World-Formation, Vol. 5) (Version 1). Zenodo. https://doi.org/10.5281/zenodo.18996583 Speed, T. (2026i). Eigenzeit. On the Limits of Block Time and the Ontological Costs of Decision, Responsibility, and Simulation - Implications for Science, Politics, and Societal Reality Formation (Version 1). Zenodo. https://doi.org/10.5281/zenodo.18481704 Speed, T. (2026j). A General Theory of Eigenzeit Relations: Difference, Overlap, Glitch and Emergence (Version 1). Zenodo. https://doi.org/10.5281/zenodo.20522842 Speed, T. (2026k). The Sweet Spot of Intelligence: World-Binding, Complexity, and the Limits of AGI (Version 1). Zenodo. https://doi.org/10.5281/zenodo.20828692 Speed, T. (2026l). The Limits of Predictive Intelligence: World Models, World-Binding, and Responsibility (Version 1). Zenodo. https://doi.org/10.5281/zenodo.20920844 Speed, T. (2026m). Simulation, Eigenzeit and Civilizational Crisis (Studies in World- Formation, Vol. 17) (Version 1). Zenodo. https://doi.org/10.5281/zenodo.20177396 69 When Structural Violence Has a Perpetrator: The Apparently Legitimated Inactive Witness with Decision-Making Power DOI: https://doi.org/10.5281/zenodo.22206507 Abstract The concept of structural violence makes it possible to identify harms that cannot be traced back to a single immediate act of violence or to a clearly identifiable original perpetrator. Yet precisely this theoretical achievement creates a possible perpetrator gap: the more strongly violence is described as the result of structures, institutional routines, fragmented responsibilities, and distributed decision-making power, the more easily concrete personal responsibility for its continuation can disappear. This paper develops, for this gap, the concept of the apparently legitimated inactive witness to structural violence with decision-making power. This refers to a person who has knowledge of a sufficiently concretely described serious situation of violence or harm, possesses real possibilities to examine, limit, or interrupt it, and nevertheless places institutional normality, role-bound conduct, or jurisdictional boundaries above the prevention of harm. Perpetration does not necessarily arise from the original production of the violence, but can become consolidated over time through increasing knowledge, continued non-intervention, and existing decision-making power. The paper therefore distinguishes between constitutive perpetratorlessness in structural violence and produced perpetratorlessness, in which concrete responsibility is once again dissolved into institutional fragmentation despite existing knowledge and decision-making power. In addition, provoked empiricism is described as a recursive procedure through which institutional reactions, opportunities for correction, non-intervention, and attempts at closure themselves become empirical material. The intervening witness to structural violence thereby creates an epistemic counter-position that confronts institutional self- descriptions with documented consequences and can make personal responsibility visible. The proposed concept of the perpetrator is explicitly not identical with perpetration in criminal law. It is understood as a scientific, ethical, and human-rights-based category of responsibility attribution. Its application requires serious harm, sufficient knowledge, real agency, an opportunity for intervention, continued non-intervention, relevant continued 70 effects, and verifiability. The aim is to remain able to describe structural violence as structural violence without allowing the people who enable its continuation under known conditions to become invisible. Keywords: structural violence, institutional violence, administrative violence, administrative harm, administrative evil, state violence, state harm, state crime, social harm, zemiology, bureaucratic violence, epistemic violence, epistemic injustice, testimonial injustice,, structural harm, institutional harm, systemic harm, coercive violence, structural coercion, institutional coercion, violence by omission, violence through inaction, perpetration by omission, bystander responsibility, complicity, crimes of obedience, responsibility diffusion, diffused responsibility, institutional responsibility, personal responsibility, perpetratorhood, perpetrator attribution, perpetrator responsibility, structural perpetration, decision-making power, institutional agency, distributed agency, responsibility concentration, responsibility attribution, accountability, public accountability, positive obligations, due diligence, duty to protect, human rights violations, state responsibility, institutional non-intervention, willful blindness, strategic ignorance, culpable ignorance, institutional ignorance, produced ignorance, epistemic refusal, institutional closure, epistemic closure, provoked empiricism, intervening witness, structural violence witness, testimonial agency, epistemic counter-authority, counter-knowledge, recursive inquiry, non- closure, structural violence accountability, institutional complicity, legitimized inaction, normalized non-intervention, street-level bureaucracy, administrative burden, rule-bound violence, bureaucratic fragmentation, institutional fragmentation, re-fragmentation, produced perpetratorlessness, constitutive perpetratorlessness, perpetrator gap, decision nodes, violence responsibility, structural violence perpetrator, inactive witness with decision-making power ## 1. Violence Without a Perpetrator Over the past 13 years, the author has investigated state violence. Central works in this regard were and are: Speed, T. (2025). SPEED'S WORK - An Autistic Intervention in the Concept of Work In the Age of AI and Robotics (Version AAM DNB Version English). Zenodo. https://doi.org/10.5281/zenodo.19035228 Speed, T. (2026). 1667:25 Violence, Health Harm and the Denial of the Rights of Autistic People in Germany - Proceedings before the Social Court Cottbus and the Higher Social Court Berlin- Brandenburg - On the Practical Impossibility of Enforcing Effective Legal Protection for Neurodivergent People (Version 1 English). Zenodo. https://doi.org/10.5281/zenodo.18887765 71 Speed, T. (2026). The Epistemic Crisis of Autism Diagnosis: False-Negative Autism Assessment, Epistemic Injustice, and the Case of the Humboldt University Autism Clinic in Berlin (Versions 2.0 Corrections Page 18 / Abstract Correction). Zenodo. https://doi.org/10.5281/zenodo.22065932 Speed, T. (2026). The Game No One Can Win: The Bedarfsgemeinschaft as a State Experiment in Violence Involving Two Disabled, Neurodivergent Artists (Version 1). Zenodo. https://doi.org/10.5281/zenodo.21978519 In addition, the author developed several works on the legal architecture affecting neurodivergent people, in an attempt to conceptualize violence against neurodivergent and poor people more clearly: Speed, T. (2026). Welfare Regimes as Structural Harm Neurodivergence, Welfare Conditionality, and Institutional Violence (Studies in World-Formation, Vol. 15) (Version 1). Zenodo. https://doi.org/10.5281/zenodo.19002085 Speed, T. (2026). Ontological Coexistence and Structural Violence in Law (Studies in World- Formation, Vol. 14) (Version 1). Zenodo. https://doi.org/10.5281/zenodo.19001795 Speed, T. (2026). Povertism, Exploitation, and the Political Economy of Manufactured Vulnerability (Studies in World-Formation, Vol. 16) (Version 1). Zenodo. https://doi.org/10.5281/zenodo.19002300 Over the years, forms of institutional violence in job centres, courts and public prosecutors’ offices, as well as in ministries, were investigated—forms of violence that were repeatedly institutionally erased, displaced, concealed, or not understood, as well as erased within administrative procedures, welfare-state decisions, diagnostic practices, and judicial responses. It was repeatedly possible to show that serious harm occurred. It was repeatedly possible to reconstruct which decisions, omissions, and institutional arrangements this harm resulted from. And repeatedly, at a certain point, the same peculiar gap appeared. The violence could be described. The perpetrator disappeared. As a result, the violence continued. The concept of structural violence is initially extraordinarily powerful for such processes. It makes it possible to identify violence where no immediate physical assault takes place and no single actor intends or controls the entire harm. No one has to explicitly order that a person’s health be damaged for an institutional arrangement to produce precisely this effect. Violence can materialize through rules, procedures, institutional routines, material dependencies, and the combination of apparently independent decisions. Yet precisely here a second problem arises. The more precisely the structure becomes visible, the more easily the people who concretely sustain it can disappear. 72 The job centre refers to the applicable law or to its calculation. The court examines a specific matter in dispute. Another body declares that it has no jurisdiction over another part of what is happening. Health consequences are assigned to a medical sphere and individualized. Criminal-law questions supposedly belong to a different institutional context. Political bodies can refer to the autonomy of administration and the judiciary. Each institution considers only one fragment. The context in which the individual decisions jointly affect a concrete human being disintegrates. This problem is certainly known in adjacent fields of research, but is usually described from a different direction. Research on street-level bureaucracy, since Lipsky, has shown that state action cannot simply be understood as the linear implementation of centrally established policy. Public services instead emerge through a multitude of decentralized decision-making situations in which individual actors possess considerable discretionary and interpretive scope. Hupe and Hill accordingly describe modern governance as a multidimensional system of nested decisions in which questions of responsibility become more difficult precisely because decision-making power is distributed across different levels and actors (Hupe & Hill, 2007). More recent research on administrative harm similarly describes this fragmentation increasingly as a problem in its own right. In the context of activation-oriented social policies, McGann speaks of administrative harm and explicitly connects this concept to work on administrative evil. The decisive observation here is that modern bureaucracies can diffuse responsibility: division of labour, technically rational procedures, and institutional roles create a separation between action, intention, responsibility, and the consequences actually caused. Harm can thereby be produced within ordinary organizational processes without those involved in producing it having to perceive their own actions as violent or even as causing harm (McGann, 2025; Adams & Balfour, 1998). Similarly, research on accountability points out that the actual practice of public administration is shaped by local discretionary decisions, organizational contexts, and differing frames of reference. The formal existence of accountability mechanisms therefore does not mean that responsibility for a concrete overall outcome can be clearly attributed. Particularly at the level of public services, responsibility can be distributed among professional autonomy, organizational hierarchy, legal requirements, and different institutional control systems (Brodkin, 2008; Sager, Thomann & Hupe, 2019). For the problem of violence examined here, however, this description is still insufficient. Fragmentation does not merely make accountability more difficult. It can destroy the perception of the overall context of violence itself. Each body may arrive at a decision that appears at least internally plausible within its own fragment, while the accumulation of these decisions produces harm that none of the institutions involved any longer recognizes as its own action. The institutional apparatus thus fragments not only jurisdiction, but also causality: the administrative decision is located here, the financial deprivation there, the psychological or physical consequence in a third place, and the question of criminal responsibility somewhere else again. 73 It is precisely here that a limit of the established concepts becomes apparent. Street-level bureaucracy, administrative burden, administrative harm, or diffused responsibility explain important mechanisms of what is happening. They primarily describe, however, why harmful administrative practices arise and why responsibility within complex organizations is difficult to attribute. The perpetrator problem developed here begins one step later: it asks what happens when this fragmentation has already been made visible, when a concrete decision-maker is explicitly presented with the connection between the individual institutional fragments and the harm arising from them, and nevertheless continues to adhere to the fragmented institutional perception. At that point, fragmentation is no longer exclusively a property of the system. It can itself become a decision. The violence exists as a whole. Responsibility appears only in fragments. The following analysis begins precisely with the question of when this fragmentation no longer suffices as a description of institutional complexity, but must instead be understood as a mechanism through which concrete actors continue to evade responsibility for the continuation of a violent process despite increasing knowledge. A paradoxical situation can thereby arise: the harm can be documented over months or years, while responsibility becomes increasingly dispersed with every additional institution involved. The violence exists as a whole. Responsibility appears only in fragments. At this point, a mechanism becomes visible that is also significant for the investigation of other forms of institutional violence. Conventional complaint and oversight procedures are often insufficient to make complex relations of violence visible. A complaint is already institutionally formatted. It has a designated recipient, a specific subject matter, and usually also a prescribed mode of resolution. The affected person describes a grievance. The institution examines this grievance within its existing categories, not infrequently reformats it, and then decides what is relevant, to which procedure it belongs, and when the matter can be considered resolved. The actual harm may develop over long periods of time. Several institutions may be involved in its emergence. Administrative-law, medical, social, economic, and psychological consequences intersect. Institutional processing, however, repeatedly fragments this context into separate jurisdictions. What takes place in a person’s life as a coherent process of violence appears in the respective files as a sequence of separate factual issues. Concrete individuals thus continue to contribute to the maintenance of structural violence even when they have been informed of the consequences. 74 1.2. On the Methodology of This Research In the author’s work, a procedure therefore developed that was later referred to as provoked empiricism. The perpetrator theory developed here did not initially arise as an abstract model that was subsequently applied to institutional cases. Its concepts emerged from decades of practice in which institutional violence, responses to warnings, repeated opportunities for correction, shifts in jurisdiction, non-intervention, and continued harm were documented and compared over a very long period of time. Only through the recurrence of certain patterns did the need arise to expand existing concepts. The theoretical movement of this paper therefore proceeds essentially from practice to theory: recurring structures were reconstructed from observed institutional processes, concepts were developed from these structures, and these concepts were subsequently tested again against further institutional reactions. Provoked empiricism thus refers not only to a method for investigating the perpetrator position described here. It also forms an essential part of its empirical genesis. It did not emerge from a desire for provocation, but from the experience that institutional systems possess a strong capacity to close complex contexts of harm again. Provoked empiricism, as it were, turns the system back onto itself, folds structural violence back into the decision-maker, confronts that person with it, and gathers from this data points that make the overall picture of the violence recognizable. At the beginning of this procedure, there is usually no accusation of violence. There is an indication. An institution is made aware of an error, a contradiction, or possible harm. It is given the opportunity to examine and correct the matter. If such a correction does not occur, the problem is described more precisely. Further evidence is added. Consequences are documented. Decisions are placed in relation to one another. The institutional response in turn becomes part of the material. In this way, the situation changes step by step. The research position here often already arises from being directly affected, because this is frequently what makes intervention possible in the first place, as well as making visible how the institution reacts to the affected person and what it does differently than in dealing with external actors, or with mere observers or commentators on what is happening, such as the press. What initially appears to an authority as a possible error can prove to be a recurring mechanism. What was treated as an isolated decision can become part of a longer chain of harm. And what could initially be explained by lack of knowledge takes place, after repeated information, under altered epistemic conditions. At some point, the language of error is therefore no longer sufficient. 75 The concept of violence emerges where the individual decisions can no longer meaningfully be separated from one another because their cumulative effect on physical, psychological, social, or material integrity has become visible. And it is precisely here that the shifting between the position of the affected person, research, journalism, and activism becomes important as a source of data. Institutions can in turn respond structurally to the naming of structural violence. They can deny that the concept is recognized within their respective legal or professional domain. They can refer to jurisdictional boundaries, reject individual causal relationships, or recognize only those fragments that can be represented within their own categories. In this way, even violence that has been explained in detail can continue to exist without personally attributable responsibility. This is precisely where the perpetrator problem begins. For without the affected person, without that person’s intervention, the perpetrator remains within an abstract relation of structural violence acting upon anonymous victims. When harm first occurs, much may remain unclear. Its causes may be unknown. The institutional consequences of a decision may not have been foreseeable. Jurisdictional boundaries may genuinely be unclear. A person within an apparatus may have access to only a small part of the overall process. This situation changes, however, with increasing knowledge. And this knowledge should also be conveyed by someone who personally experiences pain. For this knowledge is different from the knowledge of external observers. A purely objective position can, by virtue of its function alone, already relativize violence. What is decisive is not whether something is subjective or objective, but whether it is correctly named and explained in each case. A decision-maker can be informed that a particular practice produces serious harm. The causality can be explained. Consequences that have already occurred can be documented. Further harm can be predicted. Possibilities for intervention can be identified or at least examined. The person concerned can reject the description of violence, develop an alternative explanation, initiate an investigation, or plausibly explain the absence of sufficient agency. If none of this occurs and the harmful process nevertheless continues, at some point it is no longer sufficient to speak exclusively of an abstract structure. At this point, the question of a specific perpetrator position arises. Precisely in cases of structural violence, we often see the phenomenon that affected persons become increasingly emotional. This can culminate in an emotional attribution of perpetratorhood based on the simple fact that there is no response to violence, or that maintaining factual restraint leads to further violence. 76 This emotionality is another data point. Without it, violence would often remain unrecognized. The task is therefore to prevent the point at which violence becomes unspeakable, at which epistemic violence marginalizes the perspective of those affected. The emotional cry is important. But the recognition of violence is consolidated through the act of working out its mechanisms. This is precisely why an open process is needed that prevents structural violence from anonymizing itself. Provoked empiricism can fulfil this task. 1.3. On the New Concept of the Perpetrator But to whom is provoked empiricism directed? What kind of perpetrator are we speaking about here in the context of structural violence? For this perpetrator position, this paper proposes the concept of the apparently legitimated inactive witness to structural violence with decision-making power. The concept refers to a person who has knowledge of a sufficiently concretely described process of structural violence or harm, who, by virtue of their institutional position, has relevant possibilities to examine, interrupt, or limit the process, and who nevertheless allows a situation to continue whose serious consequences have become known or sufficiently foreseeable. The focus is therefore on the perpetrator who allows structural violence to continue despite knowledge and substantial decision-making power, even though the consequences are known. The apparent legitimation of this passive stance is not necessarily expressed in legal terms. Rather, it arises from the institutional form of inaction. Jurisdiction, procedures, judicial or administrative independence, professional roles, hierarchies, and the absence of sanctions can create a situation in which non-intervention appears externally as neutrality. It is precisely this neutrality that is questionable. A person who knows nothing about an ongoing serious harm is in a different position of responsibility from a person to whom this harm has repeatedly and comprehensibly been presented. A person without relevant possibilities for action is in a different position from someone whose decision is significant for the continuation or interruption of the process. The analysis of structural violence must be capable of capturing these differences. Otherwise, a peculiar situation arises in which research can explain with ever greater precision how violence is produced, while at the same time being able to say less and less about who is responsible for its continuation. The structure would thereby unintentionally become a protective space for the perpetrator. 77 The concept of the perpetrator proposed here therefore does not seek to reverse the insights of research on structural violence. It takes the analysis one step further. Structures remain decisive for the emergence and reproduction of institutional violence. Their concrete continuation, however, takes place within situations of action in which people possess different degrees of knowledge and power. Where knowledge, decision-making power, foreseeable serious harm, and continued non- intervention become sufficiently concentrated, a question therefore arises that reference to the structure can no longer answer: Who decides now that the violence may continue? This question forms the starting point of the following concept of the perpetrator. How, then, are we to proceed with structural violence if we have no means, no tool, to stop the perpetrators who newly emerge from it? Enlightenment becomes meaningless if it has no effect. ## 2. The Perpetrator Gap in Structural Violence The theoretical achievement of the concept of structural violence consisted precisely in detaching violence from the idea of an immediately acting perpetrator. This shift was necessary. As long as violence is recognized only where one person directly inflicts an injury on another, those social conditions remain invisible in which people systematically experience worse life chances, avoidable illness, dependency, or exclusion without the harm being traceable to a single act. Johan Galtung’s distinction between personal and structural violence therefore made a decisive expansion of the concept of violence possible. Violence could be understood as the difference between the condition of a person or society that is actually realized and the condition that would be possible under given circumstances. Where social structures prevent people from attaining their possible physical, social, or material living conditions, violence could be present even though no immediately visible attacker existed. Yet precisely this expansion creates a problem that can easily disappear from view in the further use of the concept. The theoretical discovery of violence without an immediately visible perpetrator can develop into a social conception of violence without responsible actors. The two are not the same. The fact that a violent process cannot be reduced to a single original perpetrator does not mean that concrete actors do not assume responsibility within its continuation. A social structure can distribute possibilities for action, make certain decisions more likely, fragment jurisdictions, and make the perception of particular harms more difficult. It cannot, however, 78 itself reject an application, ignore a warning, prevent an investigation, fix a medical interpretation, or uphold a decision despite known consequences. At these points, the person re-enters the structure. The question of the perpetrator therefore does not move back behind the concept of structural violence. It arises from it. This also has legal consequences. For if structural violence never gives rise to a perpetrator who can be identified and held accountable, not only will the violence never be stopped by the state, but there will also be no further development of legislation. If, for example, state action then becomes increasingly automated, including through the use of AI, society may lose its grasp on the question of what can still be named as violence. If structural violence knows no perpetrator, then dealing with structural violence resulting from automation becomes even more difficult. This should speak in favour of limiting automation. 2.1. The Necessary Discovery of Violence Without a Visible Perpetrator The classical conception of violence is closely tied to an act. One person does something to another. There is an immediately recognizable relationship between perpetrator, act, and injury. This form possesses a high degree of social legibility. A blow has someone who strikes it. A shooting has a shooter. An imprisonment presupposes someone who orders or carries it out. Even where justification, guilt, or causality are disputed, the basic architecture of what is happening initially remains visible. A considerable part of human harm, however, does not possess this form. People die earlier because certain resources are structurally denied to them. Others lose their physical or psychological health over years because their living conditions remain permanently below what would be socially avoidable. Educational pathways are closed, medical care remains inaccessible, social groups are systematically disadvantaged, and poverty reproduces itself across generations. In such cases, there is often a long chain of social conditions between a single decision and the resulting harm. Many people contribute to it without controlling the overall effect. Some of those involved do not even know it. The concept of structural violence responds precisely to this problem. Johan Galtung’s foundational 1969 text Violence, Peace, and Peace Research detached violence from the necessity of an immediately acting personal perpetrator. What became decisive was no longer solely whether someone carried out a recognizable act of violence, but whether social structures systematically prevented people from living a less harmed life within the social possibilities available at the time. The difference between what was possible and what was actually realized could thereby itself become the object of an analysis of violence. This expansion was epistemically necessary. 79 Without it, a large part of socially produced harm would either have to remain outside the concept of violence or be artificially reduced to individual persons at fault. Poverty would then be a condition, illness an individual fate, institutional exclusion a consequence of abstract rules. Precisely where no one personally declares an intention to harm another person, the harm could disappear from the analysis of violence. The concept of structural violence interrupts this logic. It shifts the question from the visible act to the conditions under which possibilities for action, resources, security, and life chances are distributed. This, however, also changes the position of the perpetrator. In personal violence, the perpetrator can often be directly conceived alongside the violence. In structural violence, the search for a single causative person can precisely miss the object of analysis. An order of poverty has no single author. A discriminatory system of care cannot be traced back to a single physician. A welfare regime is produced through laws, administrative practice, political decisions, jurisprudence, budgetary constraints, organizational cultures, and social conceptions of normality. The violence lies in their interconnection. The fact that the perpetrator initially recedes is therefore not a deficiency of the concept. It is one of its preconditions. This theoretical step becomes problematic only when the necessary insight that structural violence does not require a single original perpetrator gives rise to the further assumption that it fundamentally possesses no individually attributable perpetrators. There is a decisive difference between these two statements. A structure may have developed over a long period of time. No one acting within this structure today need have created it. Nor must any single person be able to control all of its consequences. None of this, however, tells us what responsibility arises once concrete actors within this structure possess knowledge, power, and possibilities for action. This distinction is particularly important in the case of state institutions. The individual caseworker did not create social welfare law. A judge did not design the institutional architecture of the judiciary. A physician did not single-handedly produce diagnostic standards. A minister is not responsible for all the historical conditions of their department. These observations, however, answer a different question from the one that arises within a concrete process of violence. At some point, what becomes decisive is no longer who originally created the structure, but who is responsible for its concrete continuation under present conditions. This introduces a temporal dimension of structural violence that is central to the question of the perpetrator. Responsibility can change. At time X, a person may know little. A harm may not yet be recognizable. A decision appears plausible on the basis of the information available. 80 At time Y, consequences emerge. At time Z, these consequences are documented and communicated to the institution concerned. At a later point, the connection between institutional practice and harm is described in a comprehensible way. Eventually, a decision must be made as to whether the same practice is continued, examined, changed, or interrupted. The structure may have remained largely the same between these points in time. The responsibility of the acting person has not. This is precisely why the study of structural violence requires a second analytical movement. The first movement led from the visible perpetrator to the structure. It made visible that social violence can exist without the classical perpetrator-act relationship. The second movement must once again ask about concrete agency within the structure that has become visible. In doing so, it must not return to the old individualism. The task is not to artificially reduce complex social violence to a single person. Rather, what is decisive is to investigate at which points within an order of structural violence concrete decision-making power arises and how this power is exercised under conditions of increasing knowledge. This also changes the question of the perpetrator. It means that structural violence can initially indeed be violence without an identifiable perpetrator and that, in its further course, a concrete perpetrator position can nevertheless emerge. This possibility is central to the present paper. Without it, we lose any analytical grasp on what is happening. Particularly in the practice documented by the author, this proved to be a considerable problem. It makes it possible to retain Galtung’s expansion of the concept of violence without deriving from it a permanent depersonalization of responsibility. The structure remains necessary in order to understand the overall context of the violence. The acting person becomes relevant again where concrete decisions are made within this context. The theoretical movement therefore is not: back from the structural to the personal concept of violence. It is: from personal violence to the structure – and within the structure once again to the responsible person. 81 Only this second movement makes it possible to describe a perpetrator position that remains difficult to see within the classical opposition between personal and structural violence. For public prosecutors in particular are today often neither able nor willing to prosecute structural violence that kills people before their very eyes, including those who know this but do not act, because this bridge is not established as a causal relation of violence. The perpetrator who acts out of structural violence carries, so to speak, structural violence on their back; but because public prosecutors do not regard themselves as responsible for structural violence, they do not derive a criminal offence from the violence that follows from it, because structural violence is for them neither a motive nor a starting point of perpetration. Instead, through structural violence the violence itself is erased, and it then appears as an incidental consequence of legitimate state action whose causal connection remains legally unproven even when, scientifically, it is unequivocally structural violence. Access to concrete perpetrators is often prevented precisely because personal responsibility is once again forced into the classical individualized concept of the perpetrator. The search then focuses on a person who directly causes the violence, intends it, controls it, or carries it out in a clearly attributable individual act. If such a figure cannot be found, the analysis easily falls back again on the abstract structure. A large area of concrete responsibility thereby remains unoccupied between the immediate perpetrator and the apparently perpetratorless structure. Modern institutional violence, however, often unfolds precisely in this intermediate space. The individual actor need neither have created nor fully controlled the violence. What may instead be decisive is that, at a particular point in the process, the actor possesses sufficient knowledge, real agency, and the possibility to continue, limit, or interrupt an already existing harm. Anyone who searches for personal responsibility exclusively according to the model of immediate perpetration therefore renders invisible precisely those perpetrator positions that may be particularly significant for the continuation of structural violence. 2.2. The Price of Perpetratorlessness The perpetratorlessness of structural violence is initially an analytical achievement. It prevents complex social harms from being reduced to individual persons. At the same time, however, this necessary expansion can give rise to a second form of invisibility. The structure begins to absorb responsibility. What was initially intended as an explanation can in this way become an exculpatory figure. No one has caused the violence in its entirety. No one controls the entire system. No one possesses all the information. No one is responsible for all the consequences. Each of these statements may be true in itself. Taken together, however, they can still create a situation in which a serious process of violence persists over time without any concrete person having to declare themselves responsible for its continuation. 82 Modern administrations are particularly susceptible to such diffusion of responsibility. Decisions emerge through division of labour. Information is located in different files and institutional spaces. One body knows the financial situation, another the health consequences, a third decides legal questions, a fourth exercises political or administrative oversight. The violence emerges from the connection between these levels, while each individual institution recognizes only its own fragment. This creates a peculiar asymmetry. For the affected person, the effect exists as a whole. For the apparatus, it exists in fragments. The body upon which different decisions act simultaneously cannot perform this institutional separation. Loss of income, existential insecurity, coerced relationships, diagnostic invalidation, social dependency, judicial delay, or psychological overload can reinforce one another. The affected person experiences their accumulation as a single reality. Institutionally, however, the same processes are dealt with in separate files, jurisdictions, and areas of law. This difference is decisive for the analysis of violence. It concerns not only the organization of administration, but the question of which reality can become visible at all. An institution that considers only its own fragment can arrive, within that fragment, at an apparently plausible decision. Several such plausible individual decisions can nevertheless together produce a condition that is destructive for the affected person. A form of institutional violence emerges that can stabilize itself precisely because it never appears in its entirety at any single location. The literature on administrative evil, diffusion of responsibility, and bureaucratic fragmentation describes essential parts of this problem. The concept of administrative evil in particular points to the fact that serious harm can arise within ordinary institutional routines without the actors involved perceiving themselves as ethically responsible or as causing the overall harm. Technical rationality, professional roles, and organizational division of labour can weaken the connection between one’s own actions and their consequences. For the question of perpetration developed here, however, a further step is necessary. Diffusion of responsibility explains why actors may initially fail to recognize their own involvement. It does not adequately explain what happens after this involvement has been made visible to them. A temporal boundary thus emerges. Before confrontation, institutional fragmentation can genuinely prevent perception. After confrontation, the same fragmentation can become a justification for its continuation. This process can be described as re-fragmentation. What is meant is the renewed decomposition of an already visible context of violence into separate institutional 83 jurisdictions after its interconnection has previously been established in a comprehensible manner. Whereas original fragmentation can produce genuine non-perception, re- fragmentation under conditions of sufficient knowledge can become a personally attributable technique of produced perpetratorlessness. This distinction is significant. If a person with decision-making power is shown in a comprehensible manner that their decision forms part of a larger chain of harm, their epistemic situation changes. They now at least know of the claim that such a connection exists. If they are additionally presented with documents, consequences, temporal developments, and possible causalities, the need to justify their subsequent decision increases. From this point onward, mere reference to institutional jurisdiction is no longer sufficient as a complete description of what is happening. The question is no longer merely why a complex system disperses responsibility. The question is what a concrete person does after this dispersal has been reassembled before their eyes. Here, perpetratorlessness itself can become a technique for reproducing structural violence. The structure provides the acting person with a language in which their own decision can be detached from the overall context of violence. They merely decide on an application. They merely assess a legal matter in dispute. They merely produce a diagnostic finding. They merely examine formal jurisdiction. What occurs outside this fragment appears as an effect of other systems or as a personal problem of the affected person. The institutional role thus functions as a reduction of perception. It does not merely limit jurisdiction. It can also limit responsibility. This limitation becomes particularly problematic once the consequences are explicitly carried back into the institutional space. For what had previously been excluded is then no longer entirely outside perception. It has been named. It has been described. It has been documented. The connection has been established. A continued return to the fragmented perspective is, under such conditions, qualitatively different from original lack of knowledge. It can become the active maintenance of institutional non-perception. At this point, re-fragmentation intersects with research on strategic ignorance, culpable ignorance, and institutionally produced non-knowledge (McGoey, 2012; Rosen, 2003). For the present approach, however, what matters less is whether ignorance is psychologically intended than its observable function: an already accessible connection is not integrated into the subsequent decision. Ignorance can thereby itself shift from an initial condition to an institutional practice. The question of responsibility thus shifts again. An actor is not a perpetrator simply because they are part of a problematic institution. Nor do they become a perpetrator solely because 84 their decision has negative consequences. The question of perpetration arises where a person, despite increasing knowledge, continues to reproduce the institutional separation through which the overall context of violence is kept invisible. At this point, perpetratorlessness itself acquires a political function. It protects the continuation of what is happening. The statement “the system causes the harm” can then be both true and insufficient. It describes the conditions under which the violence emerged, but it no longer answers the question of who, under present conditions, enables its continuation. A theory of structural violence must therefore distinguish between two forms of perpetratorlessness. The first is constitutive perpetratorlessness. It refers to relations of violence whose emergence genuinely cannot be traced back to a single person. The second is produced perpetratorlessness. It arises where concrete responsibility continues to be dissolved into institutional fragmentation despite existing knowledge and agency. This second form is central to the present paper. For once perpetratorlessness has to be produced, it is no longer merely a property of the structure. It itself becomes a mechanism of the structure’s stabilization. It becomes part of the violence. The question is then no longer merely why no one appears responsible. It becomes: Who contributes to making no one appear responsible? ## 2.3. Structure Is Not an Acting Subject The concept of structure possesses a peculiar power. It explains relationships that elude a focus on individual acts. At the same time, it can begin to appear as an actor in its own right. The structure then “produces” poverty, the system “generates” illness, the administration “causes” dependency, and the institution “decides” over possibilities of life. This language is analytically useful as long as it remains clear that it is an abstraction. A structure does not sign an administrative decision. A system does not read a warning of danger. An institution does not decide in the same sense in which a concrete human being decides. Structures organize possibilities for action, distribute resources, create roles, define jurisdictions, stabilize expectations, and limit perception. They can make certain decisions 85 more likely and make others appear almost impossible. They can reproduce violence over long periods of time without any single person having planned the entire context. Their effects are real. Their agency, however, is mediated. This distinction is decisive for the question of perpetration. If it is not made, a peculiar effect emerges: the more strongly violence is explained structurally, the less it appears to be carried out by human beings. The perpetrator is replaced by a grammatical shift. “Person X decided under the conditions of a particular institution” becomes “the institution decided.” A chain of concrete decisions becomes “the system.” The analysis thereby loses precision precisely at the point where it would need to become politically relevant. This does not mean that every institutional decision can be fully attributed to an individual. A caseworker may be bound by mandatory law. A judge may only decide a specific matter in dispute. A physician works within diagnostic standards that they did not create themselves. Political decision-makers operate within departmental jurisdictions, budgetary constraints, and hierarchical structures. But these limitations are not an absence of action either. They describe the space within which action takes place. Effects cannot be ignored within that space. The question must therefore be: What possibilities actually existed within this space? What information was available? What alternatives existed? What risks were known? What decision was made? What decision was omitted? What consequences were already visible at that point in time? Only at this level does structural violence become attributable again without being reduced to a simplistic individualism. The problem of modern institutional violence lies precisely in the fact that agency and responsibility do not disappear, but are distributed. Many actors possess limited but real agency. No single person controls the overall context. Nevertheless, what happens may depend on a multitude of concrete decisions made at particular points. For the analysis of perpetration, this means that the search must not be for an omnipotent originator. Instead, the search must focus on those decision nodes at which an order of structural violence becomes concrete and at which the acting person knew the consequences, or at least had knowledge of a risk. Such a node may be an administrative decision. It may lie in a judicial decision. It may arise where a diagnostic assessment is maintained despite extensive counter-information. It may lie in the decision of an oversight body not to investigate a matter further. It may arise where political responsibility is deflected by reference to administrative autonomy. At each of these points, however, the question of agency arises. 86 The analysis of perpetration thereby shifts from the search for an origin toward the examination of chains of decisions. An institutional process of violence may develop over a long period without central direction. Nevertheless, its continuation may at a particular point become concentrated in a small number of concrete decisions. The stronger this concentration becomes, the less sufficient it is to speak of an abstract structure. In this context, the concept of responsibility must also be defined more precisely. Responsibility here does not refer only to ethical guilt in the sense of an internal attitude. It refers to the relational connection between knowledge, agency, decision, and consequences. A person may bear little responsibility within a system if their possibilities for action are minimal. Another person may bear considerably more responsibility within the same structure because they possess greater decision-making power. Responsibility is therefore initially distributed in degrees. At certain points, however, it can become concentrated. This concentration is not a psychological process. It does not depend on whether the person concerned feels responsible. It arises from their objective position within what is happening. A person who possesses relevant information, can influence a substantial part of the context of violence, and makes a decision about its continuation occupies a different position from someone who merely carries out a small part of the process. Precisely for this reason, the institution’s self-description is insufficient. A person may understand themselves merely as a functionary. They may say that they merely applied the law in force. They may regard themselves as a neutral decision-maker, expert, or administrative employee. This self- description does not necessarily alter their actual position within the process of violence. What is decisive is the function that their action or non-intervention has within the overall process. Here, an important difference emerges between subjective and structural responsibility. Subjectively, a person may feel innocent. Structurally, their decision may nevertheless be significant for the continuation of harm. That alone is not yet sufficient to call them a perpetrator. It is, however, sufficient to keep the question open. This openness is essential because institutional systems tend to confuse subjective normality with objective harmlessness. Someone acting within a routine often experiences their conduct as ordinary. Yet precisely this normality can contribute to the stabilization of violence. At this point, the question of perpetration intersects with research on administrative evil. This research has shown that serious harm can arise within normal organizational processes without the persons involved experiencing themselves as evil or violent. The ethical danger of bureaucratic systems therefore does not necessarily lie in extraordinary wickedness, but 87 precisely in the possibility that people understand harmful processes as the normal performance of their role. For the approach developed here, however, this perspective also remains incomplete. It primarily explains how actors can operate within institutional normality without fully perceiving the violent character of their actions. The decisive question begins again at a later point: What happens when this normality is interrupted? What happens when the context of violence is explicitly named? What happens when a person is confronted with the consequences of their actions? What happens when they are given the opportunity to review their decision? At that moment, the institutional routine changes. It becomes available for reflection. The person may continue to adhere to it. But this adherence now takes place under different epistemic conditions. It is precisely there that the later perpetrator figure developed in this paper begins to become visible. Structure and person therefore do not stand in competition with one another. The structure explains why certain situations of action arise. The person decides, within those situations, about concrete continuations. A theory of structural violence must hold both together at the same time. It must be able to understand how violence arises without a central perpetrator. And it must be able to recognize when concrete perpetrators become documentable within that violence. 2.4. Between the Immediate Perpetrator and Perpetratorlessness The difficulty of grasping personal responsibility within structural violence also lies in the fact that the existing concepts of the perpetrator predominantly originate from other constellations. The immediate perpetrator is the clearest figure. There is a recognizable relationship between act and harm. Someone strikes, shoots, imprisons, deprives, commands, or destroys. The act can be attributed to a person even where motive, intent, justification, or guilt are disputed. For complex institutional violence, however, this figure is often too narrow. There, harm does not arise in a single act, but in a sequence of interdependent decisions, omissions, and institutional arrangements. The concept of the follower likewise solves this problem only partially. The follower is defined above all by their relationship to an already existing social or political order. They participate, adapt, do not object, or benefit from a system without being its central originator. Their responsibility lies essentially in supporting and normalizing an existing process. 88 The perpetrator examined here can likewise stabilize a system. Their position, however, differs in one decisive respect: the focus is not merely on adaptation, but on concrete decision-making power under conditions of increasing knowledge. A person may not be a follower in any broader social sense and nevertheless possess, at a particular institutional node, a decision upon which the continuation of a violent process depends. The figure of the desk perpetrator also suggests itself. The desk perpetrator does not necessarily operate through immediate physical violence. They organize, plan, command, or administer acts carried out by others. The concept makes visible that spatial and organizational distance does not eliminate perpetration. It is precisely in this that its historical significance lies. For the constellation examined here, however, an important difference remains. The desk perpetrator is typically involved in the production or organization of the act. The perpetrator of structural violence at issue here may appear much later. They may encounter a violent process that already exists. Their decisive act may consist not in actively organizing a harmful process, but in failing to interrupt its continuation after acquiring sufficient knowledge. The figure thereby approaches the perpetrator by omission. In criminal law, however, this term does not designate a general ethical or political category. Criminal liability by omission presupposes specific legal conditions. In particular, there must be a duty to act whose violation can, under the relevant conditions, be attributed in a manner equivalent to active conduct. The criminal-law concept is therefore deliberately narrow. For the analysis of structural violence, this creates a considerable gap. A person may possess substantial institutional power, know the serious consequences of what is happening, and nevertheless lack a legally sufficiently defined guarantor position under criminal law. They may be politically, administratively, or professionally able to initiate an intervention, order a review, or alter a decision without this necessarily giving rise to a criminal-law duty of the same scope. The guarantor position is also often deliberately limited politically, and thus legally, as can frequently be observed in welfare systems, where broader responsibility would much more quickly lead to the crossing of thresholds of criminal liability. If responsibility were tied exclusively to the criminal-law offence of omission, precisely this zone would remain analytically empty. The concept of the bystander captures another part of the problem. Research on witnessing, bystander behaviour, and the diffusion of responsibility has long examined why people fail to intervene in the face of emergency, violence, or injustice. 89 Perception, social uncertainty, diffusion of responsibility, group dynamics, and the assessment of one’s own possibilities for action all play a role. The bystander, however, is initially a spectator. Their position need not include institutional decision-making power. This is precisely where an important boundary lies. A person who observes a violent situation but possesses no relevant possibility of influencing it occupies a fundamentally different position from a person whose institutional decision helps determine whether the harmful practice continues. The concept of the inactive witness is therefore likewise insufficient. The central addition is: with decision-making power. This power need not be absolute. Nor need it be sufficient to end the entire relation of violence alone. What matters is whether the person concerned possesses a real possibility to examine, interrupt, alter, or effectively refer a relevant part of what is happening to another body for decision. Mere witnessing thereby becomes a different form of institutional participation. Research on administrative evil comes particularly close to this constellation. It describes how serious harm can arise within ordinary administrative processes while the actors involved understand themselves as rule-compliant, professional, and ethically neutral. Division of labour, technical rationality, and role-bound conduct can contribute to no one perceiving the overall result as their own responsibility. This perspective explains an essential part of modern perpetratorlessness. It explains how people can reproduce violence without experiencing themselves as perpetrators of violence. For the concept developed here, however, the point at which this lack of knowledge or self- exculpation is disrupted is once again decisive. The concern here is with a concept of the perpetrator that acquires particular relevance in the context of structural violence. An actor within an administrative apparatus may initially genuinely fail to recognize that their decision forms part of a larger structure of violence. If, however, this connection is repeatedly, concretely, and comprehensibly presented to them, the situation changes. Whereas society remains passive toward the classical perpetrator until police investigation or conviction, the concept of perpetration described here operates in a space in which civil society intervenes beforehand. A dialogue with the perpetrator takes place while that person is carrying out, allowing, or participating in violence. It is therefore perpetration under direct witnessing, in which the witness speaks to the perpetrator repeatedly and makes 90 possible a negotiation within the ongoing process in which the perpetrator can desist from the violence. This is a different instance from adjudication. It is an ethical discourse from which the perpetrator withdraws by acting as though the violence does not exist and as though they need not become part of the discourse about it. This concept of the perpetrator precisely does not presuppose that the violence has already been established under an existing legal norm. It may only have been recognized yesterday. The characterization of the violence may even still be contested. What is decisive is that, pending clarification, violence continues despite knowledge of documented harms and victims, while the participation of civil society is consciously excluded in order to continue that violence. The original explanation in terms of institutional blindness should then lose explanatory reach. Yet it is continued. The person now at least knows that their previous interpretation is contested. They know of the alleged harms. They may already know of consequences that have occurred. They have an opportunity to examine the claim. It is precisely from this point onward that the distinction between the mere reproduction of a system and the assumption of personal responsibility for its continuation becomes relevant. The type of perpetrator proposed here therefore lies between several established figures. They are not a classical immediate perpetrator because the violence may already exist structurally. They are not merely a follower because their specific decision-making power is decisive. They are not a classical desk perpetrator because they need not have organized or ordered the violence. They are not merely a bystander because they possess institutional possibilities for intervention. And they are not necessarily a perpetrator by omission in the criminal-law sense because their responsibility for violence may extend beyond the boundaries of positive criminal liability. What characterizes this figure is a specific combination: They do not merely see the violence. They can take an institutional position toward it. And more than that: Their decision regarding that position itself becomes part of the violent process. At this point, the question of participation becomes a question of perpetration. The decisive transition does not lie in an internal ethical transformation. No one has to intend to become a perpetrator. Nor is it sufficient to psychologically attribute malice or indifference. What is relevant, rather, is a change in the objective situation of action. The person possesses knowledge. They possess power. They know of the possibility of further harm. They have an opportunity for intervention. And they make a decision about 91 how to position themselves in relation to this context. This is precisely why the analysis of perpetration requires a threshold. Not every participation is perpetration. Not every omission is perpetration. Not every bad decision is perpetration. But there is no empty space between complete innocence and immediate exercise of violence. Different degrees of concrete responsibility are located there. The following question is therefore under what conditions this responsibility becomes so concentrated that an institutional actor becomes the perpetrator of an ongoing process of structural violence. ## 2.5. The Epistemic and Ethical Threshold The argument developed so far presupposes that responsibility can change over the course of a violent process. A person who initially knows only a small fragment is in a different position from the same person after repeated confrontation with harms that have occurred, possible causal relationships, and the danger of their continuation. Knowledge concentrates responsibility. This, however, captures only one side of the question of perpetration. The second concerns agency. At first glance, it appears comparatively easy to determine. A person can either intervene or they cannot. In institutional contexts, however, this distinction is considerably more difficult. Possibilities for action are limited there by jurisdictions, procedural rules, hierarchies, competences, and professional roles. State actors in particular can therefore fall back on a sentence that initially appears extraordinarily plausible: I am not permitted to act differently. For the analysis of structural violence, this statement is of central importance. It can describe a genuine limit. But it can equally mark the point at which institutional rule-bound conduct begins to collide with a more fundamental obligation to prevent serious harm. A court, for example, genuinely cannot remedy every social harm within a single proceeding. A judge does not possess unlimited jurisdiction. They are bound by law, procedure, and institutional competences. This constraint is itself one of the conditions of the rule of law. What cannot follow from this, however, is unlimited ethical exculpation through rule- bound conduct. For the claim that a person is institutionally not authorized to perform a particular act does not yet answer the question of what that person must do once they recognize that the rule- compliant continuation of their action or omission leads to the gravest, potentially irreversible harm. Here, a second threshold emerges. 92 It no longer concerns only the question: When does a person know enough? But: How severe must an impending harm become before attachment to institutional normality loses its exculpatory force? This problem is known in political ethics in a different form. The debate on dirty hands examines situations in which the demands of a political or institutional role collide with fundamental ethical obligations. Michael Walzer’s classical figure primarily addresses the opposite constellation: a political actor may find themselves compelled to violate an ethical boundary that normally applies in order to avert a catastrophe. The later debate developed from this a broader question concerning the relationship between role obligations, general ethics, and situations of extreme harm. Particularly in situations of extreme danger, it cannot simply be assumed that ordinary adherence to rules possesses normative priority under all circumstances. The dirty hands debate sometimes speaks here of a “supreme emergency,” in which the prevention of catastrophic harm collides with otherwise applicable ethical constraints (Walzer, 1973; see Coady, 2023). For the perpetrator figure examined here, this problem must, as it were, be reversed. It is not initially the willingness to act beyond the rules that becomes suspect. Under certain conditions, what can become suspect is precisely the refusal to move beyond the institutional role. A judge, administrative employee, physician, or political decision-maker cannot indefinitely exculpate themselves by referring to the limits of their normal role when it simultaneously becomes apparent that the continuation of this normality produces serious harm. As the severity of the harm increases, therefore, not only does the obligation grow to search for possibilities for action within the existing rules. The obligation also grows to examine the rules themselves ethically as part of the situation. This is not intended to assert a general duty to break the law. Such a thesis would neither be compatible with the rule of law nor necessary for the concept of perpetration developed here. Institutional rules exist for good reasons. Judges in particular must not arbitrarily deviate from applicable law according to personal ethics. A system in which every office-holder decided for themselves when the law should yield to their own values could itself produce considerable violence. The opposite extreme, however, is equally problematic. Adherence to an institutional role cannot be treated as absolute either. A relationship must therefore be determined between these two poles: The more severe, immediate, and irreversible the impending harm, the greater the justificatory burden on a person who retreats behind institutional limits on action. 93 In the case of a minor administrative burden, the statement that a particular procedural route allows no other decision may constitute a sufficient justification. In a situation that threatens a person’s existence or seriously damages health, the situation changes. In the case of a concretely impending irreversible health harm, it changes again. And where human life is immediately endangered, reliance on institutional normality reaches a boundary at which the role itself becomes an object of ethical responsibility. The pressure to act therefore does not increase independently of the severity of the violence. It stands in relation to it. This can be described as proportionality between impending harm and the required willingness to depart from institutional normality. The smaller the possible harm, the more strongly ordinary adherence to rules may determine conduct. The greater the harm, the more must be expected of an actor with decision-making power to exhaust all possibilities for intervention available within the system. Initially, the protective, objection, review, and escalation mechanisms available within the respective institution take priority. Where remonstration, internal review, additional professional assessment, referral to higher-level bodies, or other regular protective mechanisms exist, they must be seriously exhausted. Only where such routes are absent, fail, or remain manifestly inadequate in view of the harm does the question of more extensive institutional deviation become more pressing. As the harm increases further, actions beyond these may become necessary: explicit documentation of the conflict, involvement of other bodies, a public or internal warning, refusal personally to carry out a particular action, the creation of institutionally visible dissent, or acceptance of personal professional disadvantages. Up to and including loss of employment. In the most extreme case, the question finally arises whether the obligation to prevent the gravest violence also includes a willingness to violate an institutional rule or relinquish a role. This point is decisive. Otherwise, every institutional system of violence would acquire an almost perfect ethical defence. It would merely have to translate its violence sufficiently strongly into rules. The more precisely it were regulated who is not permitted to intervene, the lower, paradoxically, the responsibility of all those involved would become. A system could produce the gravest harms while simultaneously allowing every individual actor to explain that their hands were tied. Such a construction is incompatible with a serious concept of personal responsibility. An institutional rule can explain why an action is difficult, risky, or associated with personal consequences. It cannot under all circumstances explain why nothing at all had to be done. 94 Precisely for this reason, agency must be understood more broadly than the formal competence to make a particular decision directly. An actor possesses possibilities for action even where they cannot themselves end the overall harm. They can object. They can document. They can demand a review. They can involve other competent bodies. They can explicitly make the harm visible within their decision. They can use an available interpretation in favour of preventing harm. They can refuse a decision if they cannot reconcile it with fundamental ethical obligations. They can accept institutional or personal consequences. And at an outer limit, an ethical obligation may arise not to continue performing a role if its proper performance requires one’s own participation in the gravest violence. This is not an additional heroic requirement. It follows from the concept of responsibility itself. A person with institutional power may not shift the costs of preventing violence entirely onto the person who is already bearing the harm merely because their own intervention would create professional, institutional, or legal risks. It is precisely here that the question of personal risk enters the analysis of perpetration. A judge, for example, may reach the conclusion that their formal possibilities have been exhausted. For the question of their responsibility for violence, however, this would not yet settle everything. It would also have to be examined whether they document the recognized harm, whether they use existing room for manoeuvre, whether they enable further legal review, whether they alert other competent bodies to the danger, and whether their conduct makes it recognizable at all that they do not treat the continuation of the harm as institutionally irrelevant. Under conditions of the gravest violence, the minimum ethical requirement can therefore be: The actor must at least recognizably act on the side of interrupting the violence. This may be institutionally uncomfortable. It may damage careers. It may lead to conflicts with superiors or other courts. In the most extreme case, it may require no longer performing a particular function. Precisely for this reason, one’s own institutional security cannot simply be weighed against the physical or psychological integrity of another human being. This idea is close to the political theory of dirty hands, but shifts its perspective. There, the question is often when office-holders may or must violate ethical rules in order to avert extraordinary dangers. Here, the question is when office-holders may no longer use institutional rules, role expectations, or at least their routinized interpretation as ethical exculpation, because the consequences of their rule-compliant non-intervention themselves cross a threshold into the gravest violence. Research on dirty hands recognizes precisely the idea that political responsibility in extreme situations may require action that 95 conflicts with ordinary normative constraints; at the same time, such action remains ethically accountable and does not become innocent merely because of its asserted necessity. For the concept of the perpetrator developed here, an important asymmetry follows. Breaking a rule requires justification. But following a rule also requires justification once its foreseeable consequence is serious violence. Institutionally, the second side is often perceived far more weakly than the first. A state governed by the rule of law understandably monitors with great attention whether an office-holder exceeds their competences. Much more difficult is the question of when not exceeding a competence, in the face of concretely known harm, itself becomes an ethical problem. The approach developed here therefore does not demand a general replacement of law by individual ethics. It demands an accounting of responsibility in which the severity of the impending harm, the certainty of the available knowledge, the proximity of one’s own decision to the harm, the legally available possibilities for intervention, and the costs of a necessary institutional deviation are considered together. Such a threshold cannot be determined by a single number. It has several dimensions: the severity of the harm; its immediacy; its irreversibility; the quality and certainty of the knowledge; the causal proximity of one’s own decision; the actual power to intervene; the availability of less intrusive alternatives; and finally the costs that breaking a rule, refusing an action, or relinquishing a role would impose on the decision-maker themselves. These factors stand in a relative relationship to one another. A minor and reversible harm does not justify arbitrary institutional self-authorization. A foreseeable severe and irreversible harm, by contrast, may require an office-holder to go far beyond the routine processing of a matter. A concrete danger to life or fundamental physical integrity alters the relationship once again. At such a boundary, the statement “I was not permitted to do anything” can lose its exculpatory force if forms of objection, warning, escalation, refusal, or ultimately personal withdrawal in fact remained available. 96 This also allows the epistemic threshold to be determined more precisely. Perpetration does not arise merely because an actor receives some complaint. It becomes concentrated where knowledge, severity of harm, and agency increase to such an extent that continued neutrality itself requires explanation. The decisive question is then no longer: Was this person permitted to act within their ordinary role? It becomes: What could this person still reasonably be expected to refrain from leaving undone in view of the known harm? This reverses the perspective. The institutional rule no longer automatically forms the outermost boundary of responsibility. It itself becomes a factor whose ethical weight must be weighed against the impending violence. Particularly where a person possesses substantial decision-making power, responsibility may therefore require that, in a conflict, they do not merely protect their institutional role. At some point, they must decide what risk they themselves are willing to bear in order not to impose the entire risk of their obedience on another human being. At this threshold, non-intervention finally begins to lose its apparent neutrality. 2.6. The Apparently Legitimated Inactive Witness to Structural Violence with Decision-Making Power From the preceding considerations, a perpetrator figure can be identified that occupies a particular position within structural violence. It does not necessarily arise at the origin of the violent process. As stated, it need neither have planned nor created it. Nor is it characterized by directly carrying out the violence. Its perpetration develops within an already existing context when knowledge, decision-making power, severity of harm, and continued non- intervention converge. For this figure, the concept of the apparently legitimated inactive witness to structural violence with decision-making power is proposed here. The concept is deliberately elaborate. Its individual components designate different conditions that cannot be collapsed into one another. The actor is first a witness. Witnessing here means more than the incidental perception of an event. The person concerned possesses sufficient knowledge that an institutional practice is producing serious 97 harm or will with high probability produce such harm. This knowledge may arise from direct observation. More often, however, it is conveyed through complaints, warnings of danger, medical records, scientific analyses, court files, reports from other institutions, or repeated direct confrontation with those affected. What is decisive is not that every individual detail has been established beyond doubt. Institutional responsibility cannot begin only once a situation has been completely clarified. Particularly where serious harm is impending, a sufficiently well-founded suspicion may already create an obligation to investigate. The more serious the possible harm, the less complete epistemic certainty can be made a prerequisite for a response. Second, the witness is inactive. Inactivity here does not refer exclusively to doing nothing at all. Institutional inactivity may itself appear extraordinarily active. Letters are answered, files are forwarded, meetings are held, decisions are drafted, jurisdictions are examined, or formal procedures are opened. What is decisive, therefore, is not the amount of institutional activity, but its relationship to the violence. An institution can do a great deal and nevertheless remain inactive with regard to the specifically described harm. The concept therefore refers to non-intervention in the violent process. A response that does not examine the harm, a referral that creates no responsibility, or a decision that merely reproduces the existing institutional order does not necessarily count as an intervention merely because something has happened administratively. For the analysis of perpetration, what matters is whether the action is capable of investigating, reducing, interrupting, or at least visibly problematizing the continuation of the known harm. Third, this witness possesses decision-making power. This component distinguishes them from the mere observer. Decision-making power does not necessarily mean complete control. Modern institutions, in particular, distribute power. A judge cannot change the entirety of social welfare legislation. A physician cannot remedy the care situation alone. A minister cannot directly replace every individual decision. A head of an authority is subject to legal constraints. The relevant question is therefore not whether a person could end the entire violent process alone. It is whether they could have made a relevant difference. Could they investigate? Could they change a decision? Could they use an existing margin of discretion differently? Could they enable a provisional protective measure? 98 Could they effectively involve another body? Could they escalate a warning within the institution? Could they refuse their own participation? Could they at least document the conflict in such a way that the violence could no longer be treated as institutionally unproblematic? Decision-making power therefore possesses different intensities. It may be direct, indirect, administrative, judicial, political, professional, or epistemic. For the question of perpetration, what matters is not the person’s title, but the actual capacity for effect inherent in their position. Fourth, this inactivity is apparently legitimated. It is precisely this feature that distinguishes the figure from many classical images of the perpetrator. The actor need not hide. They need not necessarily conceal their decision. They may sit in their office, process a file properly, and perform their role in full compliance with the rules. The violence may continue precisely under conditions of institutional normality. The apparent legitimation arises through a space of reasons that appear socially and institutionally acceptable: jurisdiction, procedural law, limits of discretion, professional standards, hierarchy, judicial independence, internal directives, data protection, the absence of a legal basis, or the claim that another body must act. Many of these reasons may in fact be legitimate. Their existence, however, does not conclusively determine whether a person bears responsibility for a known harm. It is precisely for this reason that the term apparent legitimation is used. The institutional order can make an action or omission appear formally legitimate while an ethical responsibility remains that this order does not fully capture. Here, a fundamental difference between legality and responsibility for violence becomes visible. An act may be legal and nevertheless reproduce violence. An omission may remain unpunished and nevertheless continue serious harm. A person may act correctly within institutional terms and still enter an ethical perpetrator position. This possibility is constitutive of the concept proposed here. The perpetrator therefore does not arise only where their institutional legitimation completely collapses. They may arise precisely where this legitimation remains externally intact. This explains part of the stability of modern structural violence. The classical perpetrator may have to conceal their conduct. The apparently legitimated inactive perpetrator may document their conduct openly. A letter may explicitly state that no further examination will take place. A decision may continue an existing practice. An authority may declare that it 99 recognizes no jurisdiction. A political body may state that it cannot intervene in the actions of another institution. All of this may be transparent and formally comprehensible. The question of responsibility for violence arises only when these actions are connected with the known process of harm. The perpetrator figure can therefore initially be defined through several necessary dimensions: First: There is a serious process of structural violence or harm. Second: The person concerned possesses sufficient knowledge of this process or at least of a well- founded risk of serious harm. Third: They possess relevant possibilities to influence its course. Fourth: They are given time or an opportunity for examination, intervention, or a reasoned response, insofar as the urgency of the harm permits this. Fifth: A relevant intervention does not occur. Sixth: The violent or harmful process is thereby continued, facilitated, legitimated, or insufficiently limited. Seventh: The non-intervention is substantially justified through institutional normality, role-bound conduct, or a formal limitation of responsibility. None of these characteristics should be considered in isolation. A person may possess substantial decision-making power and know nothing about a harm. Another may know the harm precisely and possess no possibility whatsoever to influence it. A third may possess knowledge and power but actually intervene, even if that intervention remains unsuccessful. The third constellation is particularly important for the concept of the perpetrator. Responsibility must not be measured by success alone. A person who seriously attempts to prevent violence does not become a perpetrator merely because their institutional power ultimately proves insufficient. This brings another criterion into view: the recognizable orientation of the action. If an actor takes a harm seriously, examines possibilities for protection, makes internal limits visible, uses routes of escalation, and, where necessary, personally bears institutional costs, 100 their failure may still be tragic. It is, however, something different from a non-intervention that merely reproduces the existing condition. The concept of the perpetrator is therefore not directed against institutional powerlessness. It is directed against institutionally protected indifference toward one’s own agency. This indifference, in turn, need not be psychological indifference. The person concerned may feel personally affected. They may express regret. They may even be subjectively convinced that they could do nothing else. For the analysis of perpetration, subjective attitude is insufficient. What matters is what actually happens within the violent process. If a person declares that a situation is regrettable while their decision simultaneously secures its continuation, the analysis of violence must examine the decision and not merely the expressed regret. The perpetrator figure is therefore determined in an important sense functionally. It designates a function within the continuation of structural violence. This functional determination, however, must not cause the person once again to disappear behind their role. That would be precisely the error the concept is intended to avoid. A function is exercised by a human being. The person can reflect on their role. They can act. They can object. They can assume responsibility. They can weigh the institutional risk of their intervention against the risk that their inaction imposes on another human being. And they can decide against intervention. Where the conditions described above become sufficiently concentrated, it is therefore legitimate to speak of a perpetrator of structural violence. This designation is not an assertion of criminal guilt. Nor is it merely a metaphor. It designates the scientifically, ethically, and human-rights-based attribution of a concrete person to a violent process whose continuation is substantially shaped by that person’s action or omission. The concept thereby acquires a function that goes beyond description. It interrupts the semantic exculpation provided by the structure. The statement “structural violence” continues to designate the overarching relation of violence. The statement “perpetrator of structural violence,” by contrast, designates a concrete responsibility within that relation. Both levels must be able to exist simultaneously. For the alternative leads back to the problem from which this investigation began: the violence is described with ever greater precision while the person who decides on its continuation disappears conceptually. The concept of the perpetrator prevents this disappearance. 101 It places a personal address where institutional complexity no longer adequately explains responsibility, but increasingly obscures it. 2.7. Degrees of Perpetration and the Concentration of Responsibility The concept of the perpetrator of structural violence must not be understood as though perpetration arose in an identical way in every case. Knowledge, agency, proximity to harm, and possibilities for intervention are distributed differently. Institutional actors therefore do not all occupy the same position of responsibility. These differences, however, do not speak against the concept of the perpetrator. They are what make it precise. The relevant question is not whether responsibility is absolute or complete. It is how far responsibility has become concentrated within a concrete process of violence. One person may possess only limited institutional power. Another may directly determine a decision upon which the continuation of the harm depends. A third possesses no direct decision-making authority but can initiate oversight, a review, or political intervention. Yet another knows only a small fragment of what is happening, while another actor has the entire chain of harm documented before them in detail. These differences are significant. They alter the degree of responsibility. They do not eliminate it. For the purposes of analysis, several dimensions can therefore be distinguished. The first is epistemic proximity. It designates the degree to which a person knows, or is capable of knowing, the context of violence. An incidental indication carries a different weight from detailed documentation. An abstract warning of possible consequences is different from proof of harm that has already occurred. Repeated warnings also alter the situation. The more frequently and concretely an actor is confronted with the same context of violence, the more difficult it becomes to explain continued non-perception as mere lack of knowledge. The second dimension is causal proximity. It designates the significance of one’s own action or omission for the continuation of what is happening. Some decisions have only an indirect effect. Others directly determine whether a benefit is withdrawn, a diagnostic determination is maintained, a protective measure is denied, or a harmful practice is continued. The closer a decision lies to the concrete harm, the greater the responsibility for its consequences. 102 The third dimension is decision-making power. This concerns not only formal jurisdiction, but the actual scope for action. A person may possess discretion, choose alternative interpretations, initiate a review, involve a higher authority, or suspend a decision. Political or administrative authority may also be relevant even where it does not directly replace an individual decision. The fourth dimension is the severity of the harm. It carries particular weight. An institutional error that causes a minor and easily reversible disadvantage creates different demands for intervention from a situation in which serious deterioration of health, lasting social destruction, or a danger to life becomes recognizable. As the severity of the harm increases, therefore, so does the obligation to move beyond institutional routines and examine extraordinary possibilities for intervention. The fifth dimension is duration over time. Structural violence can develop over long periods. Precisely because of this, responsibility changes. A person who once makes a possibly incorrect decision occupies a different position from a person who, over months or years, repeatedly receives opportunities to recognize and limit the harm. Continuation creates additional knowledge. Continuation creates additional decision points. And continuation can therefore create additional responsibility. The sixth dimension concerns the availability of alternatives. The more real possibilities exist to limit or at least examine a harm, the less convincing the claim of complete institutional powerlessness becomes. These alternatives need not be simple or free of risk. Particularly in cases of serious violence, responsibility may require the acceptance of costs. An actor can risk institutional conflict. They can oppose superiors. They can problematize a decision publicly or internally. They can refuse participation. In the most extreme case, they can place their own position at stake. The fact that such actions create personal costs does not deprive them of their ethical relevance. Rather, precisely in situations of the gravest violence, the question arises as to why the affected person alone should bear the costs of the institutional order. These dimensions do not produce a mathematical formula. They produce a landscape of responsibility. Within this landscape, personal responsibility can become concentrated or diminished. Two persons who formally belong to the same institution can therefore occupy very different perpetrator positions. 103 A caseworker with minimal freedom of decision and incomplete knowledge may bear only a small share of a violent process. A senior person who has the entire harm documented before them and possesses far- reaching possibilities for intervention occupies a different position. A judge whose concrete decision determines whether existential harm continues, in turn, has a different proximity to what is happening than a political body that has merely been informed in general terms about the problem. These differences must remain visible. They must not, however, once again be used to dissolve every attribution of perpetration. This is precisely where a frequent problem in academic discourses on responsibility lies. Because responsibility is complex, it becomes increasingly relativized. Because many factors are involved, no single person appears sufficiently responsible any longer. Because agency is distributed, shared responsibility ultimately becomes irresponsibility. The approach developed here rejects this conclusion. Gradual conditions do not necessarily lead to a merely gradual existence of responsibility. A person can bear different degrees of responsibility and nevertheless be a perpetrator. This idea is familiar from other areas of human attribution of responsibility as well. Participation can carry different weight. Knowledge can extend to different degrees. Power can be distributed differently. Nevertheless, the category of responsibility does not disappear merely because several persons are involved. The same must apply to structural violence. The perpetrator position is relational. The concrete action or omission of the person is not. A person may not have freely chosen their institutional situation. They may have entered a problematic structure before becoming aware of its violent character. It does not follow, however, that later decisions can be understood merely as passive “drifting into” the situation. With increasing knowledge, new decisions arise. Does the person remain? Do they intervene? Do they object? Do they investigate? Do they document? Do they change their practice? Do they accept institutional disadvantages? Or do they adhere to the existing order? The answer to these questions is not structurally predetermined. It is precisely here that personal responsibility appears. The concept of the perpetrator must therefore distinguish between the emergence of the perpetrator position and the assumption of perpetrator responsibility. 104 The perpetrator position arises from the objective constellation of knowledge, power, harm, and non-intervention. Responsibility arises from how the concrete person acts within this constellation. No one can therefore be completely exculpated by explaining that they merely found themselves in a perpetrator position. An institutional context can place a person in a difficult or contradictory situation. It can limit their possibilities for action. It can make certain decisions risky. It can even impose considerable personal costs. But once the violence has become sufficiently visible, continued participation itself becomes a decision. This is the point at which personal responsibility emerges from structural involvement. The concept of the perpetrator is intended to capture this responsibility. It is not intended to claim that all perpetrators act in the same way. It is not intended to claim that all bear the same share of the violence. Nor is it intended to claim that every perpetrator position would be assessed in the same way under criminal law. It makes a more fundamental claim: There is a point at which the differentiation of responsibility must no longer serve to prevent its attribution altogether. That point is reached when the existing differences in knowledge, agency, and proximity to harm continue to determine the degree of responsibility but can no longer plausibly explain why a person should remain outside the concept of the perpetrator. Precisely for this reason, the concrete naming of a perpetrator is not a theoretical imprecision for the approach developed here. It is its consequence. The analysis can establish that one person bears less perpetrator responsibility than another. It can distinguish between direct and indirect participation. It can take limited room for action into account. It can recognize that institutional resistance creates personal costs. What it should not do is continue these differentiations until the concrete person once again disappears behind the structure. For then the perpetratorlessness previously identified as a problem would merely have been restored in more differentiated language. The decisive question therefore remains simple: Under the conditions actually available to them, did this person contribute sufficiently to the continuation of known structural violence? If, after careful reconstruction, this question must be answered in the affirmative, describing the person as a perpetrator is not an exaggeration. It is a form of attribution. 105 ## 2.8. Perpetration Beyond Criminal Guilt The concept of the perpetrator developed so far inevitably leads to a boundary question: May a person be described as a perpetrator of structural violence if no court has established a criminal offence and possibly not even an existing criminal provision fully captures the conduct in question? For the theory developed here, this question must be answered in the affirmative. Otherwise, positive law would not merely decide on criminal liability. It would at the same time decide which forms of human harm may be described as violence at all and to which persons this violence may be attributed. Such a primacy of law would be epistemically difficult to justify. Criminal law fulfils a specific social function. Under particularly demanding requirements, it determines when state sanctions against a person are permissible. It therefore operates with offences, rules of evidence, subjective requirements, standards of causality, grounds of justification, and institutional procedures. Its high threshold is not a weakness. Precisely because a criminal conviction has serious consequences, the attribution of criminal guilt must be limited and reviewable. It does not follow from this limitation, however, that everything outside criminal law is free of violence. Research on social harm and zemiology emerged precisely from this problem. It criticizes the fixation of classical criminology on state-defined crime and shifts the starting point toward harm actually experienced. Hillyard, Tombs, and other representatives of this approach point out that a considerable part of socially relevant harm is not criminalized and is in part produced precisely by state, economic, or institutional orders. The question of whether an event has been defined as a crime must therefore be analytically distinguished from the question of whether people are seriously harmed by it. This shift is central to the concept of violence used here. A person can lose their home, suffer serious harm to their health, lose the material basis of their existence, or be subjected to an institutional practice that permanently impairs their physical or psychological integrity. Whether positive criminal law already provides an appropriate offence for this context does not alter the harm that has occurred. Nor does it alter the concept of violence. The harm possesses a reality prior to its legal classification. A similar movement can be found in research on state crime. Here, the problem arose early on that state action can hardly be assessed solely by reference to those criminal norms that the state itself defines and enforces. Green and Ward therefore sought, among other things, to define state crime through organizational deviance and human-rights violations. Here too, the state’s self-definition of what counts as criminal is not accepted as a sufficient standard. 106 Particularly in the case of state violence, it would be circular to allow the state alone to decide which of its own actions may even appear as violence or criminality worthy of fundamental investigation. This does not mean rendering adjudication meaningless. It means determining its scope more precisely. A court can authoritatively determine whether, under the conditions of a particular legal system, a criminal offence can be attributed to a concrete person. It cannot, in the same sense, authoritatively determine whether a social process may scientifically be described as violence, ethically as serious wrongdoing, or in human-rights terms as a violation of human integrity. This distinction concerns different orders of knowledge. The criminal-law level asks whether a particular statutory offence has been fulfilled and whether there is individual criminal guilt. The scientific level asks about mechanisms, causalities, effects, repetitions, and structures of responsibility. The ethical level asks what conduct toward other human beings is responsible in view of possible and actual consequences. The human-rights level relates fundamental protected interests such as dignity, life, physical and psychological integrity, autonomy, and equality to the actions of state and other powerful actors. These levels influence one another. But they are not identical. Judith Shklar’s critique of legalism is particularly relevant in this context. Shklar opposed the idea that law could be understood as a system largely isolated from ethics and morality, politics, and social reality. Legalism becomes problematic when legal categories begin to determine the boundaries of what can socially be perceived as wrongdoing at all. For the analysis of state violence, this critique is particularly important because legal institutions do not stand outside the society whose violence they assess. The rule of law is therefore not the final observer of human reality. It is itself part of this reality. Its decisions produce effects. Robert Cover emphasized this relationship particularly radically in Violence and the Word. For him, legal interpretation does not remain on the level of text and meaning. It is embedded in institutional practices through which people are restricted, imprisoned, coerced, or subjected to other material consequences. Legal language is therefore not disembodied. Behind a decision there may stand a real chain of institutional action that affects people and their bodies. For the concept of violence developed here, a simple but consequential conclusion follows: 107 The legal description of an act is not identical with its entire social reality. An administrative act does not remain merely an administrative act if it foreseeably produces existential harm. A diagnostic decision does not remain exclusively a professional assessment if it influences access to protection, care, and social recognition. Judicial non-intervention does not remain exclusively a procedural event if the continuation of a known serious harm depends upon it. The legal form of these actions remains relevant. But it does not exhaust their meaning. The same applies to concepts such as bodily harm. In positive criminal law, bodily harm has a specific doctrinal meaning. Whether concrete conduct fulfils the corresponding criminal offence is a legal question. Yet the concept also describes a reality that precedes criminal law: a human body or human health is injured. Science and civil society therefore cannot be required to reduce every use of the concept to its respective criminal-law classification. If an institutional practice demonstrably leads to significant physical or psychological harm, it may be scientifically meaningful to speak of an injury to physical or health integrity and, depending on the conceptual definition used, of bodily harm. It must remain clear, however, on which level the statement is being made. The scientific-ethical statement “Within the concept of violence used here, this act constitutes bodily harm” is not identical with the criminal-law statement “This person has fulfilled the criminal offence of bodily harm.” The second claim contains specific legal requirements that must be examined in an appropriate proceeding. The first describes a harm and its attribution within an explicitly stated scientific and ethical model. This distinction must not be confused with a prohibition on speaking. Social concepts such as violence, killing, torture, abuse, or bodily harm do not arise exclusively in legal codes. They possess historical, political, medical, cultural, and human- rights meanings. Frequently, it was precisely civil-society movements, affected persons, and scholars who named a particular form of harm long before the law developed an adequate category for it. Law responds to social reality. It does not fully create it. 108 This point is particularly important where the violence under investigation is produced by the state or by institutions. If only the existing legal order were permitted to determine whether state action could be described as violence, an epistemic circle would arise: The state defines the permissible forms of state action. The state defines the criminal offences. State institutions decide on their application. And from the absence of a criminal conviction, the conclusion would then be drawn that no violence had occurred. Something that has frequently been documented in the author’s work. Such a conclusion would be scientifically untenable. Research on social harm has therefore insisted on taking harm seriously as an independent starting point. Research on state crime supplements this with the insight that human-rights norms and social responses may be necessary in order to analytically grasp state-inflicted harm at all. For the concept of the perpetrator in this paper, the following therefore applies: Perpetration in structural violence is not a subcategory of criminal perpetration. It has a different starting point. The first question is not: Which criminal offence has been fulfilled? The questions are: What violence is taking place? What harm is arising? How is it being produced or continued? Who knows about it? Who possesses relevant agency? Who intervenes? Who decides, under these conditions, that what is happening will continue? Only after this does the additional question arise whether the reconstructed conduct is also criminally relevant. The order is decisive. If the analysis of violence began with the question of existing criminal offences, precisely those forms of institutional violence for which positive law does not yet possess adequate language could disappear once again. The concept of the perpetrator developed here therefore begins with the reality of the harm and reconstructs the relationships of responsibility from there. This also means that describing a concrete person as a perpetrator places high demands on the research. Independence from criminal law is not independence from evidence. 109 On the contrary. Precisely because the attribution of perpetration takes place outside a judicial judgment, its criteria must be made explicit. Knowledge, agency, temporal development, possible alternatives, severity of harm, institutional limits, and actual responses must be documented in a comprehensible manner. The concept of the perpetrator does not derive its legitimacy from state authorization. It derives it from the quality of its justification. This distinguishes it both from a mere insult and from a criminal judgment. It is a form of public, scientific, and ethical attribution. In cases of structural violence, this attribution fulfils a particular function. It prevents the absence of criminal sanctionability from being confused with the absence of responsibility. A person can be responsible for the continuation of violence even where existing criminal law recognizes no fitting perpetrator figure. A person can be the perpetrator of a scientifically described process of violence even where a criminal court does not classify their conduct as a criminal offence. And a society must retain the possibility of articulating such responsibility. Otherwise, a peculiar asymmetry would arise: through its conceptual gaps, the law could not only produce impunity, but at the same time determine that what it does not capture may socially have no perpetrator. It must not be permitted to have precisely this second effect. Positive law possesses decision-making authority over the legal consequences of its categories. It possesses no monopoly on truth regarding human harm. For the concept of the perpetrator developed here, this boundary is constitutive. For structural violence becomes politically dangerous precisely where its institutional normality results in serious harm taking place in reality while neither an act nor a perpetrator relationship can be articulated. The possibility of a scientifically and ethically justified attribution of perpetration keeps this space open. It allows civil society to name violence before the law has fully understood it. And it preserves the possibility that the law must later respond to a reality that has already become visible. Only in this way does violence ultimately become something that demands a political response, which can then be developed further in the form of new laws. Nor can an unjust state be confronted in any other way. At this point, provoked empiricism becomes both a human-rights and a scientific tool. 110 3. Provoked Empiricism: From Structural Perpetratorlessness to Attributable Responsibility Structural violence possesses a strong capacity for self-stabilization. This stabilization does not occur only through power, coercion, or legal protection. It also occurs epistemically. Institutions determine which information is regarded as relevant, which jurisdictions exist, which relationships are examined, and when a matter is regarded as closed. In this way, a violent process can be fragmented, linguistically neutralized, and ultimately transformed into a form in which no one is any longer compelled to position themselves in relation to the overall context. Provoked empiricism begins precisely at this point. It is not a procedure that artificially produces violence in order subsequently to claim its existence. Nor does it create conflict for the sake of conflict. Its function is to prevent institutional closure as long as central questions of harm, causality, and responsibility remain unanswered. The concept of provocation therefore initially designates an epistemic operation. An existing institutional description is confronted with material that it cannot readily integrate. The contradiction is not removed from the process, but fed back into it. The institutional response is documented and again made an object of investigation. A response generates new material. A non-response does as well. Referrals of jurisdiction, rejection, justification, silence, or termination of communication generate further data about the way an institutional system deals with the description of its own possible violence. The method therefore operates recursively. It feeds the reaction of the system under investigation back into the investigation. An institution may, for example, declare that a particular harm does not fall within its jurisdiction. This statement does not end the investigation. It generates a new question: What institutional architecture makes it possible to exclude a known harm from one’s own sphere of responsibility? A body declares that it possesses no possibility for action. This statement, too, is not automatically accepted as an endpoint. What must now be examined is which possibilities for action actually exist, which are merely not institutionally provided for, and what costs would be generated by exceeding the ordinary role. The author, for example, as described in the book Speed’s Work (Speed, 2025), used criminal complaints against public prosecutors to have violence repeatedly re-examined and to integrate each respective deviation as a data point. In this way, a comprehensive picture of state concealment of violence could be uncovered. In this way, provoked empiricism creates epistemic need for explanation. This is its central function. 111 An institutional system can continue to object. It can continue to reject the thesis of violence. It can continue to adhere to its original decision. But it can no longer do so under the same epistemic conditions as before. The first decision may have been made in ignorance. The second is made after knowledge of an objection. The third perhaps after presentation of concrete harms. The fourth after a detailed reconstruction of the causal relationship. The fifth after renewed warning of further consequences. The structure does not necessarily change as a result. But responsibility within that structure changes. Provoked empiricism is therefore also a method of concentrating responsibility. It gradually reduces the spaces in which ignorance, uncertainty, or diffuse jurisdiction can continue to function as complete explanations. For this reason, in some investigations the author linked different authorities together in loops of justification. This created new data points within the investigation through which violence could be triangulated. This process must not be confused with mere escalation. Its initial form is often distinctly low-threshold. It begins with an indication. A person draws an institution’s attention to a possible error, contradiction, or harm. At this point, precisely no perpetration is yet being asserted. The institution is given the opportunity to review the matter and, where appropriate, correct it. If correction occurs, the process can end. Provoked empiricism has no methodological obligation to escalate. Rather, the possibility of de-escalation forms part of its evidentiary value. An institution that responds to comprehensibly presented evidence, acknowledges an error, and limits a harm produces a different empirical finding from an institution that maintains its original position despite increasing counter-information. Precisely for this reason, repeated opportunities for correction are important. They serve not only fairness toward the actors under investigation. They increase the evidentiary strength of the later attribution of responsibility. The more real possibilities for examination and correction existed, the more difficult it becomes to describe the continuation of what is happening as mere error. A typical movement thus emerges: First, a problem is named. Then its effects are described. Subsequently, the relationship between institutional decision and harm is reconstructed. The institution is given the opportunity to examine this reconstruction. Further consequences are documented. The description of violence is made more precise. Concrete decision-makers are identified. Their possibilities for action are examined. Finally, it is observed how they behave under these conditions. 112 Provoked empiricism thereby changes the epistemic position of the affected person. In many institutional constellations, this person initially begins as the object of a decision. The applicant is examined. The patient is diagnosed. The claimant is assessed. The benefit recipient is administered. The institution observes. The affected person is observed. In the course of provoked empiricism, this direction can be reversed. The affected person begins to observe the institution. They collect its decisions. They compare its statements over time. They document contradictions. They reconstruct jurisdictions. They connect institutionally separated processes with one another. They investigate the consequences. They observe what happens after these consequences are communicated. The person originally defined by the institution thereby becomes an evidence-bearing actor. This shift is essential to the method. It creates a form of counter-authority that does not rest on office. The affected person acquires no judicial, ministerial, or professional decision-making power. Their counter-authority arises from another source: from the capacity to document and conceptually order what is happening in such a way that the institutional self-description no longer remains the only available description. This changes the power relation. Not necessarily permanently. Not necessarily completely. But at certain points, significantly. Institutional power also rests on the ability of an institution to define what a case is. It determines which documents belong to it. It decides which questions are examined. It determines which language is used. It can close a matter. If, however, a second, sufficiently documented description is produced outside this order, a competition between realities emerges. The institutional description may be: a properly conducted procedure. The counter-description is: a chain of harm reconstructable over time. The institution speaks of jurisdictions. The counter-description reconstructs their connection. The institution speaks of individual decisions. The counter-description shows their cumulative consequences. The institution declares the case closed. The counter- description takes precisely this closure as another event in the chronology. Here, provoked empiricism begins to operate as an operator of non-closure. Non-closure does not mean that a procedure may never be ended. The concept designates another level. An institutional system may administratively close a matter. It cannot, however, simply prevent this very closure from subsequently becoming an object of investigation. Closure thereby loses its absolute character. It becomes an event. 113 Why was the matter closed here? Which questions remained unanswered at that point? Which harms were known? Which questions of responsibility were not answered? What institutional function does the closure fulfil? What happens after it? The recurrence of the method prevents institutional self-description and observed reality from being permanently separated from one another without that separation itself becoming visible. The response is fed back. The rejection is fed back. The silence is fed back. The termination of contact is fed back. Every closure can thereby generate renewed opening. This produces a particular form of epistemic instability. Rigid institutional systems depend on certain categories remaining sufficiently stable. An administrative problem remains an administrative problem. A medical finding remains a medical finding. A judicial decision remains a judicial decision. Health consequences are attributed to another system. Provoked empiricism reconnects these separated levels. It thereby makes the structure within structural violence comprehensively visible in the first place. It works against fragmentation. For a certain period of time, it can therefore become unclear which description is the stronger one. The institutional category remains in place. But it is no longer fully closed. Alongside the “administrative procedure,” there can now stand “institutional violence.” Alongside the “diagnostic result,” there can stand “epistemic harm.” Alongside “lack of jurisdiction,” there can stand “non-intervention despite knowledge.” Alongside the “closed case,” there can stand a continuing chronology of harm. Within this intermediate space, an epistemic power gap emerges. The institution retains its formal power. But it loses part of its capacity to stabilize reality exclusively through its own category. It is precisely there that change becomes possible. New concepts can be established. New causalities can become visible. New forms of witnessing can emerge. And finally, new positions of responsibility can be identified. Provoked empiricism therefore does not expand reality by inventing arbitrary new facts. It expands the domain of what is socially recognizable. What previously had no perpetrator can become accessible to an analysis of responsibility. What was previously treated merely as the subjective experience of the affected person can, through documentation, repetition, and institutional reaction, become an intersubjectively verifiable object. It is precisely here that provoked empiricism differs from an ordinary complaint. A complaint initially seeks an institutional decision concerning the matter presented. Provoked empiricism simultaneously investigates how the institution arrives at this decision and what its reaction reveals about the context of violence under investigation. 114 The method also differs from activism. Activism can create publicity, formulate demands, generate political pressure, and bring about social change. Provoked empiricism may touch upon all these elements, but its primary epistemic interest lies elsewhere. The confrontation itself becomes a research instrument. The question is not only whether an institution yields under pressure. What is investigated is how it reacts when its own reality is systematically confronted with a competing, evidence-based description. In this respect, the method approaches a form of institutional stress test. This concept, too, however, requires explanation. The stress test is not intended to create artificial overload. It does not arbitrarily increase pressure until an institution makes some mistake, even though this can not infrequently also generate data. What is tested, rather, is whether a system can fulfil its own promises of correction under conditions of increasing evidence. Can it acknowledge an error? Can it integrate contradictory information? Can it perceive serious consequences as part of its own decision-making space? Can it redistribute responsibility? Can it reopen a case? Can it change its own procedure? Can it tolerate an institutionally uncomfortable finding? Can it continue to treat the person who produces that finding as a legitimate epistemic actor? The last question in particular is important for institutional violence. For systems can react to the loss of epistemic control by attacking the credibility of the person producing the counter-description. The affected person may be described as difficult, emotional, vexatious, aggressive, unreasonable, or professionally incompetent. An epistemic question is thereby personalized once again. The focus is no longer on the evidence presented, but on the person presenting it. For provoked empiricism, this reaction too is not merely an obstacle. It is a possible finding. If an institution responds to an increasingly documented description of violence primarily by delegitimizing the witness, this reveals the function that control over witnessing has for its stability. The intervening witness therefore stands in a particular relationship to the institution. Initially, they possess no formal authority. Precisely for this reason, their existence is problematic for rigid systems. An officially appointed commission of inquiry possesses a status that is institutionally legible. An oversight authority possesses jurisdiction. A court possesses decision-making authority. Their intervention is, in principle, provided for by the system. The intervening witness does not possess such a position. They create a counter-authority without the institution having granted it to them. Their authority must arise from evidence, coherence, documentation, and public verifiability. It is precisely here that its destabilizing potential lies. 115 The institution can ignore them. It can delegitimize them. It can terminate contact. It can reject the evidence produced. But each of these decisions can in turn be documented and fed back into the context of the investigation. This also changes the meaning of termination of contact. A termination of contact is not automatically proof of violence or perpetration. It may have good reasons. Within an already strongly concentrated relationship of responsibility, however, it can acquire particular significance. If an institution has repeatedly been confronted with documented harms, knows the description of violence presented, possesses relevant possibilities for action, and finally terminates communication without addressing the central relationship, the termination can become part of the institutional closure that, in turn, reveals violence. It may then end the dialogue. It does not end the evidence. It is precisely here that the function of the operator of non-closure becomes particularly clear. The system can remove the witness from its communicative space. It cannot retrospectively erase what has been documented up to that point. The attempt at closure itself becomes part of the history. In this way, provoked empiricism can produce something essential for the later analysis of perpetration: a documented history of increasing knowledge. The perpetrator of structural violence then does not appear as a person to whom responsibility is retrospectively attributed in the abstract. It can be reconstructed when they knew what. Which indications they received. Which harms had already occurred. Which interventions were demanded. Which possibilities for action existed. Which responses were given. And what happened afterward. This chronology changes the quality of the attribution of perpetration. It shifts it from assertion to reconstruction. Provoked empiricism thereby connects two levels that often remain separated in structural violence. It makes the violence visible as a whole. And it makes visible how concrete people react after this whole has been made visible to them. Only the connection of both levels allows personal responsibility to emerge from structural perpetratorlessness. Provoked empiricism is therefore not a provocation of violence. It is a provocation of accountability. The author’s work contains countless examples in which this could be carried out in practice. Sometimes successfully, sometimes not. 116 ## 3.1. The Intervening Witness to Structural Violence Provoked empiricism does not only change the view of the institution. It also changes the position of the affected person. In classical institutional procedures, the affected person initially appears in a passive or at least asymmetrical role. They apply, report, explain, submit documents, answer questions, and wait for a decision. The institution possesses the categories, the procedural power, and the authority to produce from the material presented a binding or at least institutionally effective description. This order is not limited to administration. It can be found in medicine, social welfare law, courts, psychiatric and diagnostic procedures, assessments, complaint systems, and many forms of professional case processing. The affected person provides material about themselves. Others decide what epistemic status this material receives. Foucault already described in Discipline and Punish how people are reformatted as cases. It is precisely here that a familiar form of epistemic injustice lies. With the concept of testimonial injustice, Miranda Fricker described how speakers can suffer an unjustified loss of credibility because of social prejudice. Later research on epistemic injustice expanded this perspective and showed that certain groups are not only believed less, but in some cases also possess fewer recognized conceptual resources through which to make their experiences intelligible at all. For structural violence, this second level is particularly important. A person may experience harm and at the same time possess no institutionally recognized language in which that harm appears as a coherent process. They can describe economic losses. They can document deterioration in health. They can challenge individual decisions. They can point to fear, exhaustion, dependency, or loss of possibilities for action. As long as these elements remain within separate institutional categories, however, no shared description of violence emerges. The affected person possesses experience. The institution possesses the order of its meaning. The intervening witness to structural violence breaks through this distribution. They remain a witness because their position arises from the experience and observation of the violent process. They become an intervening witness, however, as soon as they no longer limit themselves to communicating their experience within the categories offered to them. The author enacted this in his work when, as an affected person, he began to publish scientifically on violence he had himself experienced and to confront authorities and courts 117 with these publications, which existed in a different epistemic space with an authority of their own. The intervening witness begins to investigate the categories themselves. They connect processes that were institutionally separated. They reconstruct temporal sequences. They relate decisions to their consequences. They compare statements from different bodies. They document changes. They identify contradictions. And when the existing language is insufficient, they develop new concepts. The intensity, repetition, or institutional inconvenience of such an intervention does not determine its epistemic status. Precisely where affected persons repeatedly disrupt existing closures, there is a danger that their intervention will be devalued through attributions such as inappropriateness, irrationality, or vexatiousness. The decisive line of demarcation therefore lies in the method: If claims are documentable, temporal sequences reconstructable, evidence and interpretation distinguishable, counter-information incorporated, and one’s own attributions correctable, the mere persistence of the intervention cannot disprove its epistemic quality. An epistemic shift thereby occurs. The witness is no longer exclusively a source of information. They become the producer of an order of the material of their own. This change is significant. A classical witness reports on an event. The intervening witness additionally investigates the conditions under which their report is believed, rejected, fragmented, or made institutionally invisible. They therefore observe not only the original violence. They observe the processing of the description of violence. It is precisely this second level that is decisive for institutional violence and is far more than a subjective perspective. For violence can remain invisible not only through an original decision. It can also become invisible because its later description is not taken up. The institution can acknowledge harm but dispute its cause. It can acknowledge a cause but reject its own jurisdiction. It can acknowledge a connection but name no concrete responsibility. It can acknowledge responsibility but see no possibility for action. Each of these movements changes the social existence of the violence. The intervening witness records these movements. In this way, a form of counter-authority emerges. This counter-authority is neither identical with institutional authority nor merely its opposite. It possesses no sovereign power. It cannot revoke an administrative decision. It cannot replace a judgment. It cannot legally alter a diagnosis. Its strength lies elsewhere. It can problematize an institutional description to such an extent that it no longer remains self-evidently the only reality. The intervening witness thereby creates epistemic competition. A finding stands alongside a counter-analysis. An administrative act stands alongside a documented chronology of harm. A judicial decision stands alongside a reconstruction of its 118 material consequences. An institutional explanation stands alongside a second description that makes its omissions visible. Formal power initially remains unequally distributed. The epistemic situation no longer does so to the same degree. It is precisely there that a shift in power can begin. Institutions possess considerable power over which realities count as official. This power, however, depends on their categories remaining sufficiently stable and competing descriptions not reaching a comparable degree of coherence, evidence, and public verifiability. The intervening witness intervenes precisely at this point. They do not create counter- authority through mere assertion. It arises from the concentration of evidence. Documents are brought together. Temporal sequences are reconstructed. Statements are compared. Consequences are observed. Institutional reactions are archived. Scientific literature is consulted. One’s own position is disclosed. In this way, an initially individualized experience can become a publicly verifiable object of research. This is an important difference from the mere position of the victim. The concept of the victim initially describes a relationship to harm suffered. The concept of the intervening witness describes an epistemic act. A person may be a victim without assuming this role. They may likewise be an intervening witness without themselves being the person primarily harmed. Relatives, researchers, journalists, activists, whistleblowers, or other observers can perform a similar function if they are able to reconstruct a structurally fragmented context of violence and carry it back into the institutional space. The concept is therefore not tied to autobiographical research. In autoethnographic or case-based investigations, however, the position can acquire particular density because the witness simultaneously possesses continuous access to the effects of institutional decisions. They can observe changes directly. They can determine when information was communicated. They can trace how the institution responded to it. They can record over long periods the difference between institutional description and lived consequence. This proximity creates possibilities for knowledge. At the same time, it creates risks. The intervening witness is not neutral in the classical sense. They have an interest in the violence ending. Precisely for this reason, they may be accused of having their research contaminated by personal involvement. Such a criticism, however, presupposes a specific understanding of scientific distance. Proximity to the object does not automatically eliminate the capacity for knowledge. It changes the conditions under which knowledge is produced. The relevant question is 119 therefore not whether the intervening witness is disinterested. It is whether their claims are verifiable. Are the underlying documents available? Can temporal sequences be reconstructed? Are alternative explanations considered? Are institutional responses documented with sufficient completeness? Can observation, interpretation, and normative evaluation be distinguished? Is it comprehensible when and why the description of a process changed? These requirements are higher than the mere demand for subjective neutrality. They shift scientific quality from asserted distance to the transparency of the evidence. For research on institutional violence in particular, this may be necessary. For institutional proximity is often, paradoxically, regarded as less problematic than proximity arising from being affected. The professional actor appears neutral even though they work within the institution under investigation, use its categories, and may depend on its stability. The affected person, by contrast, appears biased because they directly experience the consequences of that institution. This asymmetry itself requires epistemic explanation. The intervening witness therefore rejects the idea that being affected means epistemic inferiority. Nor do they claim that being affected guarantees truth. Their position rests on a different claim: Proximity can produce evidence when it is methodologically reflected upon and made verifiable. And institutions react differently to affected persons than to external actors. Different data therefore become visible. The intervening witness thereby also moves beyond the classical logic of the complaint. The complainant asks the institution to review a matter. The intervening witness may continue to demand such a review. They do not, however, wait entirely for the institution to decide which reality is valid. They begin the reconstruction themselves. This is where the actual shift lies. The institution is no longer the only place in which a case can be ordered. A second epistemic space emerges. This space can arise in scientific publications. It can arise in archives. It can arise through investigative documentation. It can arise in civil-society reports, artistic research, or public interventions. What is decisive is that institutional reality is reproducibly confronted with another ordering of the material. This is precisely why the intervening witness also differs from the activist in the narrower sense. Activism is often directed toward change. It mobilizes, protests, formulates political demands, creates publicity, or organizes collective counter- power. The intervening witness can do all of this as well. Their specific function, however, lies in the production of an evidence-capable counter-reality. 120 They do not merely seek to make an institution act differently. They seek to make reconstructable why its previous conduct forms part of a violent process and what happens after this has been communicated to it. The intervention therefore possesses both a political and an epistemological character. It is intended to interrupt violence. And it produces knowledge about why violence is interrupted or why this does not occur. It is precisely this combination that distinguishes provoked empiricism from purely observational research. The intervening witness changes the object by supplying it with information. This change is methodologically intended. An institution that does not know of a harm is a different object of research from an institution that knows the same harm in detail and nevertheless continues to act unchanged. The intervention therefore produces a new empirical situation. It removes one possible explanation: ignorance. Further interventions can weaken other possible explanations: uncertainty. Lack of documentation. Misunderstanding. A one-time error. An unforeseeable consequence. The institutional response thereby becomes increasingly informative. In this way, the intervening witness does not produce perpetration. They change the conditions under which perpetration can be recognized. This is a decisive difference. If they produced perpetrators only by provoking people, the method would be circular. In fact, they make an already existing decision-making power epistemically connectable to an already existing or impending harm. It cannot, however, be excluded that in some cases this may lead to an act occurring only because a perpetrator feels pressured by provoked empiricism. That, too, is part of the investigation. The investigator, however, ensures that the relationship cannot remain permanently abstract for the actors. This increases the possibility of personal attribution. By making a process personal, it does not necessarily become less objective; rather, this can represent a concentration that is necessary in order to make action appear necessary. The intervening witness is therefore the counterpart to the apparently legitimated inactive witness with decision-making power. Both possess knowledge. Their relationship to the violence, however, differs fundamentally. The inactive witness possesses institutional power and does not use it sufficiently to interrupt the known process. The intervening witness may be almost entirely powerless in formal terms and nevertheless acts in order to make the violent process visible, attributable, and interruptible. This contrast reveals a remarkable reversal. The institutionally powerful person can invoke their limitations. The institutionally powerless person exceeds the passivity assigned to them. One must react to the other. 121 One possesses formal authority and may produce non-intervention. The other possesses no formal authority and produces epistemic intervention. Both figures therefore become fully intelligible only in relation to one another. The intervening witness creates those epistemic conditions under which the inactive witness with decision-making power can no longer fully conceal their position behind ignorance, fragmentation, or institutional normality. Their function therefore consists not merely in bearing witness. They make accountability possible. And where institutional accountability is systematically avoided, they prevent this avoidance from being confused with the disappearance of the violence. ## 3.2. The Operator of Non-Closure Within provoked empiricism, the intervening witness assumes a specific function: they become the operator of non-closure. Non-closure does not mean that institutional procedures may never be ended. A court must decide, an authority must process matters, a diagnostic institution must reach conclusions. Problematic closure begins where the institutional end of a matter is taken to mean the end of its reality. The administrative decision is concluded, but the harm is not. The proceeding is over, but the health consequence continues. The diagnosis has been made, but the contradiction between the finding and lived reality remains. Precisely in cases of structural violence, closure can therefore itself become part of the mechanism of violence. It separates what can be institutionally processed from the consequences that continue to have effects and enables the apparatus to treat its own operation as completed while the affected person continues to bear its effects. The operator of non-closure refuses this equation. They return the unresolved remainder. Not necessarily into the same procedure, but into the epistemic context. A decision is connected with its consequences. A termination of communication becomes part of the chronology. A referral of jurisdiction is examined in terms of which responsibility has been displaced through it. A concluded diagnosis can continue to be the object of scientific criticism. Closure itself thereby loses its status as the final point and becomes another event within the process under investigation. In this sense, provoked empiricism operates recursively. 122 Its recurrence prevents institutional self-description and observed reality from being permanently separated from one another without precisely this separation becoming visible. This is where the particular function of the operator lies. It does not produce truth. It prevents a closure that determines truth prematurely. A temporary epistemic intermediate space thereby emerges. The institutional description remains in place but loses its exclusivity. Alongside the administrative procedure, there can stand a description of violence. Alongside the diagnosis, a counter-analysis. Alongside the lack of jurisdiction, a reconstructed responsibility. Alongside the formally closed case, a continuing history of harm. This non-closure is therefore more than persistence. It is an epistemic intervention into the capacity of institutional systems to stabilize their own reality conclusively. It is precisely here that its significance for the concept of the perpetrator lies. As long as institutional closure functions, personal responsibility can disappear together with the case. If the context is kept open, the question also remains of who, under the conditions that have meanwhile become known, decides on its continuation. The operator of non-closure therefore does not merely keep a conflict open. They keep responsibility visible. ## 3.3. The Positing: Violence as Recognizable Reality Non-closure therefore prevents an institutional process from disappearing definitively into an existing category. It keeps contradictions open, connects decisions with consequences, and prevents the administrative termination of a procedure from being confused with the end of its social reality. The next step is then to name the open condition again. It is precisely here that the positing begins. By positing, no arbitrary assertion is meant. The concept designates the process in which a previously fragmented reality that has not been adequately described institutionally is brought into a new conceptual form. Individual health consequences become a context of harm. A context of harm can become structural violence. Repeated non-intervention despite knowledge can become a perpetrator position. The positing thus connects elements that were already present but could not appear as a common reality within the institutional order. Its function is simultaneously epistemic and 123 political. Epistemically, it produces an object. Politically, it changes the conditions under which that object can be spoken about. The word “violence,” for example, has a different effect from the formulation “adverse consequence of a decision.” Both may refer to the same process. They organize it differently. The “adverse consequence” can remain isolated. It can appear as a regrettable side effect. It can be processed statistically, medically, or administratively. The concept of violence, by contrast, establishes a connection between harm, power, and responsibility. It asks not only what has happened, but also through which relationship between people and institutions this harm became possible and why it continues. Such a positing can destabilize the institutional order. The positing adds a second description to this space: institutional violence; serious harm to health; epistemic oppression; produced perpetratorlessness; non-intervention despite knowledge; personal responsibility for violence. These concepts do not automatically replace the institutional description. They do, however, prevent it from deciding alone what the process is. A struggle over the description of reality thereby emerges. This struggle is not merely semantic. Concepts determine which questions can be asked. If a process is regarded as an administrative error, the question is how it can be corrected legally or administratively. If the same process is investigated as structural violence, other questions are added, such as who has been harmed, who knew about it, and so on. As long as a violent process is described exclusively in the categories of the institution involved in producing it, that institution possesses a considerable epistemic advantage. The positing by the intervening witness interrupts this asymmetry. It declares: There is another description of what is happening. This description must justify itself. It must be measured against evidence. It must withstand contradiction. It may prove to be wrong. But it requires no prior permission from the institution in order to exist as a scientific or social hypothesis. Precisely this is essential in relations of power. Institutional authority rests not only on the ability to enforce decisions. It also rests on the ability to make certain descriptions appear reasonable, professional, or legally relevant and to reject others as inappropriate, emotional, outside the relevant field, or exaggerated. The 124 concept of violence can disturb this order. The concept of the perpetrator disturbs it even more strongly. For with the naming of the perpetrator, the direction of observation changes. The question of the perpetrator asks: Which concrete person decides, under these conditions, about the continuation? A personal address thereby emerges. This address is particularly important for structural violence because precisely its dissolution into impersonal relationships can constitute a substantial part of its stability. The positing of a perpetrator therefore does not mean that the structure has ceased to exist. It means that the structure is no longer sufficient to absorb all responsibility. The perpetrator remains part of the structure. But they become visible within that structure. Here, the positing reaches its sharpest form. A decision-maker is no longer merely informed that an abstract system produces harmful consequences. They are confronted with the thesis that their own action or omission has by now become part of this violent process. The situation thereby changes again. The person must now respond not only to the harm. They must respond to their own position within that harm. They can reject the attribution. They can set out their possibilities for action. They can describe an alternative causality. They can intervene. They can dispute the concept of violence. They can explain why their conduct remains ethically justified despite the known consequences. The defence against the attribution of perpetration therefore also generates new information. It is precisely here that the connection between positing and provoked empiricism lies. The concept is not simply attached as a label to a completed case after the research is finished. Under certain conditions, its introduction can itself become part of the investigation. This can be seen particularly in the author’s court records. The attribution of perpetration creates a new epistemic burden. Previously, the person may have been able to say: “The system works this way.” Now the question is: “What responsibility do you bear for the fact that, under your decision, it continues to work this way?” 125 This shift cannot be answered through a simple reference to the structure. The personal address remains. The naming of the perpetrator thereby possesses a confrontational function. This confrontation is intentional. It is not, however, equivalent to an insult. The naming of the perpetrator is not intended to produce maximum emotional escalation. It is intended to produce maximum capacity for attribution. Its legitimacy therefore depends on the criteria developed above. The weaker the evidence, the more problematic the positing becomes. The more uncertain the causality, the more cautiously it must be made. The smaller the person’s agency, the less the concept of the perpetrator can bear. Conversely, the more serious the documented harm, the clearer the knowledge, the greater the decision-making power, and the longer the non- intervention, the more strongly the positing can be justified. The severity of the concept must follow the strength of the evidence. Precisely for this reason, its public use can form part of a serious analysis of violence. The social naming of a perpetrator fulfils a function that a purely structural concept of violence cannot fulfil. It generates accountability pressure. A structure suffers no reputational loss. An institution can distribute responsibility. A concrete person, by contrast, possesses a biography, a professional role, a name, and a responsibility on which they can publicly take a position. This is part of the political effectiveness of the concept. The attribution of perpetration prevents responsibility from disappearing completely into institutional abstraction. This process is not without danger. Personal attributions can be wrong. They can simplify. They can distort complex causalities. They can themselves become an instrument of political persecution. Precisely for this reason, the concept of the perpetrator developed here requires the thresholds described above. But the risk of falsely naming a perpetrator cannot lead to the general exclusion of every naming of a perpetrator within structural violence. For non-naming also has consequences. It can contribute to the continuation of violence. It can protect those responsible. It can create the impression that no one could have acted differently. And it can result in the affected person having to prove the harm while no one has to explain why this harm continued to be permitted despite existing knowledge. The decision not to name a perpetrator is therefore not a neutral position either. It can become part of produced perpetratorlessness. This is where the political significance of the positing lies. It claims for science and civil society the right to articulate a relationship of responsibility before, or independently of whether, positive law already provides a fully fitting category for it. 126 A counter-authority to institutional self-description thereby emerges. This counter- authority possesses no ultimate truth. It does, however, possess the right to contest institutional truth. Here too, it is not primarily focused on definitive fixation, but on a process of engagement with injustice. The method is intended above all to uncover what would otherwise remain concealed, to place those responsible within a space of possible alternative actions, and to create awareness of experiences of violence and causality beyond institutional logic. Precisely within this conflict, a new social reality can emerge. A practice that yesterday was still regarded as normal implementation may tomorrow be recognized as violence. A harm long treated as an unavoidable side effect may become visible as avoidable institutional harm. Such shifts are not exceptions in social development. Many concepts of violence that are taken for granted today first had to be asserted against established institutional descriptions. The positing therefore often precedes general recognition in time. Someone must first say: This is violence. Someone must show why. Someone must document the consequences. And if the violence continues despite this knowledge, someone may have to say: This person bears perpetrator responsibility for its continuation. Without such positings, social violence could change only within already existing categories. Precisely those forms of violence for which these categories are inadequate would therefore remain permanently difficult to see. The positing is therefore not the conclusion of knowledge. It is an intervention into its conditions. It creates a form around which contradiction, evidence, and responsibility can reorganize themselves. In interaction with provoked empiricism, this creates a recursive process: The violence is named. The institution reacts. Its reaction changes the evidence. The position of responsibility is reassessed. Where necessary, it is named more precisely. The system reacts again. The positing itself thereby becomes part of the empirical movement. It can fail. It can be rejected. It can generate institutional resistance. It can trigger a correction. It can create publicity. It can draw new actors into the process. What matters is that, after it, the previous epistemic situation cannot be fully restored. There is then no longer an appearance of institutional innocence. Someone who is merely doing their job becomes recognizable as someone participating in violence. Someone who functions, obeys, or merely carries out orders can no longer evade responsibility so easily. A culture of accountability toward witnesses emerges, through which witnessing itself becomes an epistemic position with a function of participation. 127 It is precisely here that the positing possesses its political force. It expands the space of what must socially be treated as possible reality. The positing of a perpetrator nevertheless remains a verifiable attribution of responsibility. The person concerned must have the possibility to substantively refute the underlying reconstruction: by demonstrating lack of knowledge, lack of agency, a different causality, or intervention that has already taken place. Provoked empiricism must not immunize the perpetrator figure against refutation. Its strength lies precisely in the fact that increasing evidence can both strengthen an attribution and correct an earlier one. The severity of the concept must therefore follow the strength of the evidence. 4. Limits and Risks of the Concept of the Perpetrator A concept intended to make structural violence personally attributable again itself possesses power. Precisely for this reason, it must be limited. The argument developed so far is directed against a form of institutional exculpation in which concrete responsibility disappears behind structures, jurisdictions, and roles. This must not, however, produce the opposite danger of prematurely reducing complex relations of violence to individual persons. The concept of the perpetrator can only be analytically viable if it reflects on its own possibilities for misuse. The first danger lies in over-personalization. Structural violence exists as an object in its own right precisely because it cannot be traced back to a single act or a single originator. If every harmful decision is immediately personalized, the concept of structural violence loses its analytical function. An actor may operate within a harmful structure without fully knowing its consequences. They may possess only minimal possibilities for action. They may be bound by mandatory requirements. They may cause harm that was not foreseeable at the time of their decision. In such constellations, naming a perpetrator would be premature. The concept developed here therefore does not begin with mere involvement. It begins where knowledge, agency, severity of harm, and continued non-intervention have become sufficiently concentrated. The structure thereby remains analytically intact. The person emerges only where, within this structure, they assume an independent responsibility for its continuation. The second danger concerns epistemic overconfidence. 128 Whoever names violence claims interpretive authority. Whoever names perpetrators claims even more. An intervening witness can therefore themselves be tempted to equate their counter-description with truth. Provoked empiricism must not be more critical toward institutional authority only to place a new uncontrolled authority in its place. Its own analysis must remain refutable. It is more an offer, an invitation to civil society to engage with the violence that has been identified, than a final determination in the sense of criminal law. Documentation does not replace infallibility. Personal experience does not replace causal examination. Scientific language does not replace evidence. Even an analysis of violence developed over a long period can be wrong, incomplete, or distorted. The concept of the perpetrator must therefore remain open to counter-information. An institution or person may convincingly demonstrate that certain information was not known. They may provide evidence of actual attempts at intervention. They may establish an alternative causality. They may show that the agency attributed to them did not in fact exist. Such a refutation must not be reinterpreted as merely another act of resistance simply because it contradicts the original thesis. Otherwise, provoked empiricism would become a closed system in which every reaction confirms the initial thesis. That would be methodologically fatal. A good theory of perpetration must possess the possibility of removing someone from the perpetrator position again. The third danger lies in circularity. The intervening witness confronts a person with a description of violence. That person rejects the description. If this rejection alone were then treated as evidence of their perpetration, the concept would not be testable. The fact that an actor rejects an attribution has, in itself, only limited evidentiary value. What matters is how they deal with the underlying facts. Do they dispute the harm? Do they dispute the causality? Do they dispute their agency? Do they present other data? Do they intervene? Or do they merely retreat into institutional formulas without addressing the described relationship at all? Only this level is analytically relevant. The fourth danger is the endless loop of non-closure. The method of provoked empiricism rests on not automatically accepting institutional closure as the end of a reality. This does not mean extending the principle indefinitely for its own sake. It must remain directly related to the violence it captures. Sometimes the methodology also exhausts itself, or cannot be continued for other reasons, because it may, for example, tip into stalking or itself become a form of violence no longer connected to self- defence or the protection of others from the consequences of violence. 129 Non-closure is therefore not an end in itself. It is justified only where a relevant unresolved contradiction remains between institutional description and documented harm. The fifth danger concerns the proportionality of naming a perpetrator. The term “perpetrator” has considerable social effects. It can damage reputation. It can intensify institutional conflict. It can create publicity. These effects are not necessarily reasons against its use. But they fall within the responsibility of the person using the term. The stronger the public impact of an attribution of perpetration, the higher the requirements for its justification must be. Several factors are relevant here: the quality of the evidence, the severity of the harm, the person’s position of power, the centrality of their decision, the possibility of responding, and the question of whether naming the person is actually necessary for understanding the violent process. Anonymization may be appropriate in some cases. In others, it may render the structure of responsibility invisible once again. Both must be weighed against each other. The sixth danger lies in the confusion of perpetrator analysis with psychological character diagnosis. The concept of the perpetrator developed here says nothing about whether a person is evil, sadistic, callous, or narcissistic. It requires no such assumption. The attribution concerns a concrete position within a violent process. A person can be a perpetrator of structural violence while at the same time subjectively regarding themselves as caring, professional, or conscientious. They may even genuinely act in caring and responsible ways in other areas. The analysis of perpetration does not describe the whole person. It describes a relationship of responsibility. Precisely for this reason, it should not turn into character destruction. The seventh danger lies in political instrumentalization. A concept of the perpetrator can be used against opponents. State, political, or social actors could use it to delegitimize dissenting persons. Civil-society groups could also use it inflationarily. The more socially effective a concept becomes, the less its original author controls its use. Its definition must therefore be narrow enough not to transform every form of political responsibility automatically into perpetration. Not every wrong decision is violence. Not every act of violence immediately produces personally attributable perpetration. Not every non-intervention is ethically reprehensible. Not every ethically problematic non-intervention crosses the threshold into perpetration. The threshold remains essential. 130 The eighth danger concerns the return to an individualized concept of violence. Precisely because the paper develops a personal perpetrator figure, the impression could arise that structural violence must ultimately once again be understood as the sum of individual wrong decisions. That would be a misunderstanding. The new concept of the perpetrator does not replace Galtung’s concept of structural violence. It supplements it at a particular point. The structure continues to explain why certain harms arise systematically, why they recur, why individual actors behave similarly, and why responsibility is dispersed. The concept of the perpetrator, by contrast, explains when a concrete person within this structure assumes a responsibility that can no longer be fully explained by their structural embeddedness. Structure and perpetrator therefore remain analytically distinct. A violent process can remain structural even when perpetrators are identified. Conversely, the structural cause does not disappear merely because individual persons become responsible. The ninth danger is subtler. It lies in a possible exculpation of the system through the naming of perpetrators. Once an individual actor is identified, an institution may attempt to shift the entire harm onto that person. The perpetrator then becomes the new scapegoat. The system explains the case as individual misconduct and thereby protects its own structure. Precisely for this reason, the concept of the perpetrator must remain directed in both directions. It must no longer allow responsibility to disappear behind the structure. But neither may it allow the structure to disappear behind the perpetrator. An individual person can be a perpetrator and at the same time an expression of an institutional pattern. Removing them may be necessary and still change nothing about the conditions that produce the next similar decision. A complete analysis of violence must therefore always ask both questions: Who bears personal responsibility? And what structure produces situations in which this responsibility repeatedly arises? The tenth danger finally concerns the asymmetry of evidentiary power. Institutional actors often possess greater access to files, legal expertise, personnel, time, and public authority than affected persons. A concept of the perpetrator that required perfect evidence exclusively could therefore once again privilege those who already possess better possibilities for documentation. Conversely, 131 the lesser institutional power of the intervening witness cannot mean that every subjective account counts as sufficient justification for attributing perpetration. The solution can only lie in transparent weighting. Evidence must be assessed according to its actual accessibility. Missing documents may have to be evaluated differently depending on who controls them. Institutional non-documentation can itself also become relevant. At the same time, a distinction must be made between plausible reconstruction, strong suspicion, and sufficiently established attribution of perpetration. The concept therefore requires an internal gradation of epistemic certainty. This protects not only the accused person. It also protects the analysis of violence. A concept of the perpetrator used too quickly loses its social sharpness. A concept that may be used only under conditions of practically unattainable certainty, by contrast, loses its function. The actual task lies between these two extremes. The limits of the concept are therefore not an afterthought. They belong to its construction. A concept of the perpetrator of structural violence must be hard enough to name responsibility. It must be open enough to be refuted. It must be personal enough to break through perpetratorlessness. And it must remain structural enough not to fall back into naïve individualism. It is precisely this tension that makes it analytically useful and fruitful in practice. 4.1. Minimum Conditions for the Attribution of Perpetration For a robust attribution of perpetration within structural violence, at least the following conditions should be examined together: 1. Serious harm: There is a relevant process of violence or harm. 2. Knowledge: The person concerned knew the context or, on the basis of sufficiently concrete indications, was required to engage seriously with it. 3. Agency: They possessed real possibilities to examine, limit, interrupt, or refer the matter to effective decision-makers. 4. Opportunity for intervention: There was an actual opportunity to respond. 5. Non-intervention: An adequate response did not occur, or the existing practice was continued despite knowledge. 6. Continued effect: The decision or omission contributed in a relevant way to the continuation, stabilization, or legitimation of the harm. 7. Proportionality: The intervention that could reasonably be expected stood in an appropriate relationship to the severity, immediacy, and irreversibility of the harm. 132 8. Verifiability: The attribution remains open to refutation through new evidence. The more strongly these conditions converge, the more strongly personal perpetrator responsibility becomes concentrated within structural violence. 5. Consequences: From Institutional Jurisdiction to Responsibility for Violence The concept of the perpetrator developed here creates no new courts, no formal jurisdiction, and no criminal offence. Its effect lies elsewhere. It changes the question that can be addressed to institutional action. So far, an institution can determine its responsibility largely within the boundaries of its own jurisdiction. A court examines a matter in dispute. An authority decides within a legal framework. A medical institution assesses a diagnostic matter. A political body refers to the autonomy of subordinate or other institutions. This order is insufficient, however, where harm arises from the connection between several such orders. Existential harm does not distinguish between social welfare law, medicine, administration, politics, and the judiciary. Institutional fragmentation may therefore be functionally correct and at the same time insufficient for the analysis of violence. It is precisely here that the concept of the perpetrator changes the perspective. It supplements the question “What am I responsible for within my jurisdiction?” with a second: “What responsibility arises from what I now know about the consequences of my actions?” This second question cannot be answered completely through institutional jurisdiction. An actor may not be responsible for the overall solution to a problem and may nevertheless bear responsibility for whether they intensify, limit, make visible, or ignore a known harm. The standard therefore shifts from mere jurisdiction to responsibility for violence. Responsibility for violence designates the responsibility that arises where a concrete human being within an order of structural violence possesses knowledge and relevant possibilities for action. This shift has several consequences. For research, it first means that structural violence does not have to end analytically at the point where no immediate perpetrator can be found. The structure remains the starting point. 133 It explains the emergence of the relation of violence, its repetition, its institutional stability, and the distribution of possibilities for action. Within this structure, it can then be investigated where responsibility becomes concentrated. This makes possible a connection that has often fallen apart: the analysis of the system and the analysis of the human being who decides within that system. Particularly in violence research, this connection is significant. An exclusively personal concept of the perpetrator underestimates the power of the structure. An exclusively structural concept of violence, by contrast, can underestimate the fact that violence is continued in concrete situations through concrete decisions. The concept proposed here attempts to hold both levels together at the same time. For institutions, this results in a changed conception of accountability. Accountability cannot consist entirely in demonstrating that a procedure was formally conducted correctly. The formal correctness of a decision does not necessarily answer the question of its known consequences. The more serious these consequences become, the more strongly institutional accountability must also include the material reality of the affected person. This applies particularly where an institution is repeatedly informed of a connection between its decision and serious harm. From this point onward, it is no longer sufficient merely to reproduce one’s own procedural logic. An additional justificatory burden arises. For example, the question of why no intervention takes place. These questions alter institutional action even where the concept of the perpetrator is never explicitly invoked. They generate a form of preventive responsibility. An actor knows that their role does not automatically guarantee ethical neutrality. Institutional normality thereby loses part of its exculpatory force. This also concerns the training of professional roles. Judges, physicians, administrative employees, or political decision-makers usually learn their function within clearly defined boundaries of competence. This would need to be supplemented by the ability to recognize situations in which adherence to one’s own role itself becomes part of serious harm. The decisive professional competence would then not consist only in knowledge of rules. It would also consist in the ability to recognize the boundary of rule-bound conduct. The actor would have to be able to understand when harm has become so serious that additional intervention is required. Where serious injuries become known, the question therefore always also arises as to who possesses the possibility of preventing or at least limiting their continuation. The concept of the perpetrator thereby connects structural violence more closely with the logic of positive protective obligations. 134 Non-intervention no longer appears automatically as a neutral state. It can itself become part of the violent process under investigation. This is particularly relevant where institutional architecture systematically results in no one assuming overall responsibility. A state may possess extensive protective rights and nevertheless create situations in which every individual organ can declare that it has no jurisdiction over a substantial part of the harm. The concept of the perpetrator makes visible that protective obligations need not be examined only at the level of abstract state responsibility. They materialize at decision nodes. Human beings act there. This simultaneously creates a new function for civil society. If responsibility for violence is not defined entirely by positive law, science, affected persons, journalists, human-rights organizations, and other social actors can reconstruct and name responsibility. This possibility supplements social accountability and does not replace courts. A state governed by the rule of law does not require less civil-society counter-power because it is organized according to the rule of law. It requires it precisely because institutions governed by the rule of law can also possess their own blind spots. The intervening witness occupies a particular position in this context. They do not create state authority. But they possess the possibility of stabilizing a description of violence outside the institutional order that has so far been unable to incorporate it. In this way, a new form of social accountability can emerge. A person may have to explain why they continued a particular decision despite known consequences. An institution may have to engage with a description of violence that does not originate from its own conceptual apparatus. A political public may assess a process differently from the institution that initially declared it closed. This shift may later also feed back into law. Law does not develop exclusively from within itself. New social concepts of violence, new forms of witnessing, and new conceptions of responsibility can place legal categories under pressure. What initially appears as a scientific or ethical attribution of responsibility can, in the long term, influence which protective obligations, official duties, or criminal-law categories a society considers necessary. The concept of the perpetrator developed here does not claim to anticipate this legal development. It first creates a language for a zone of responsibility that is currently often lost between structure and classical perpetration. It is precisely here that its practical significance lies. Structural violence no longer has to end with the statement: No one alone is responsible. The more precise question is: Who, under the given conditions, could have done something, by then knew enough, and nevertheless decided that the existing condition should continue? 135 At this point, institutional jurisdiction becomes responsibility for violence. And responsibility for violence can become perpetration. ## 6. Conclusion: Structural Violence Needs Perpetrators Again The starting point of this paper was a recurring observation within the author’s work, in which violence by public authorities was investigated over many years. The violence could be described. But the perpetrator disappeared within the logic of the apparatus. As a result, the violence continued because the mandate and the direct or indirect consequences that lay outside the asserted purpose of an order or an action became invisible within the institutional process. Often, even the concept of violence itself was absent, let alone the concept of the perpetrator. The explanation of structural violence was and remains necessary in order to make visible those forms of social harm that possess no immediately recognizable perpetrator. Its theoretical strength lies precisely in detaching violence from the idea of a single person acting intentionally. This expansion must not be reversed. A society produces violence through laws, institutions, economic conditions, diagnostic orders, political decisions, administrative practice, and cultural normalities. No one need have fully planned this violence. No one need control all of its consequences. Frequently, there is in fact no single original perpetrator. It does not follow from this, however, that structural violence must remain permanently perpetratorless. It is precisely at this point that the present investigation began. A person may initially be part of a harmful structure without being able to recognize its violent character. They may make a mistake. They may follow a procedure whose consequences are unknown to them. This situation changes once the harm becomes visible. It changes further when causal relationships are documented. It changes again when a person is explicitly made aware of their own role within the violent process. From this point onward, non-intervention increasingly requires explanation. Where knowledge, decision-making power, serious harm, and continued non-intervention converge, a perpetrator position emerges that is only inadequately captured by the existing classical concepts. For this constellation, the concept of the apparently legitimated inactive witness to structural violence with decision-making power has been proposed here. Its particular feature lies in the fact that this person need not have created the violence. 136 They need not intend it. They need not carry it out directly. They may even act within an institutionally recognized role. Their perpetration arises where they allow a sufficiently visible violent process to continue despite possessing relevant agency. Their apparent legitimation does not fully protect them. An institutional rule can explain why intervention is difficult. It can explain why personal or professional risks arise. It can explain why particular actions are not formally provided for. But it cannot under all conditions explain why serious violence was allowed to continue. As the severity of the harm increases, therefore, so does the obligation to be willing to depart from institutional normality. Following rules is then no longer automatically neutral. It too requires justification. The concept of the perpetrator thereby acquires an ethical level that is not identical with criminal guilt. Positive law decides on criminal liability. It does not finally decide whether violence has occurred. Science, human rights, and civil society possess their own possibilities for describing harm and responsibility. This distinction is particularly necessary where state or institutional violence is under investigation. Otherwise, through its own limitations, law could simultaneously determine which forms of state violence may socially be named as violence at all. The concept of the perpetrator developed here opposes this epistemic monopoly. It nevertheless remains bound to evidence. Its independence from criminal law is not independence from justification. Perpetration must not arise from mere outrage. It must be reconstructable. Provoked empiricism has a particular function within this reconstruction. It does not merely make the harm visible. A theory of structural violence that, under sufficiently documented conditions, fundamentally hesitates to name concrete perpetrators can thereby reproduce part of the very diffusion of responsibility that it seeks to criticize. The concept of the perpetrator proposed here attempts to close this gap. It does not return to a simple world of personal guilt. It leads from the structure back to the human being. The structure explains why violence could arise. The concept of the perpetrator asks who allowed it to continue after it had become visible. Both are needed. Although structural violence has long been internationally recognized as a fact, it continues to persist in its many forms. It is not infrequently protected by a gap between individual perpetration and structural violence. 137 Structural violence is therefore often assigned primarily to the sphere of politics and, to a lesser extent, to NGOs. Criminal law withdraws along the lines of a narrowed concept of perpetration. The author’s work has shown that, in practice, structural violence without perpetrators is extremely difficult for affected persons to prevent. The distance between the victim’s experience and political reality is too great, and the protective function of public prosecutors regularly fails here because of the narrowed criminal-law concept of the perpetrator. Civil society therefore needs concepts of perpetration that make it possible to stop the individual perpetrator of violence who enables structural violence in the first place. Not because they created it, but because they actively continue it: because they do not intervene, do not refuse participation in it, but allow it to happen. 138 Declaration of AI usage Generative AI tools, including ChatGPT, were used during manuscript preparation to support language editing, structural refinement, translation and the critical review and clarification of formulations. The research, empirical material, theoretical concepts, interpretations, and scholarly conclusions are those of the author. All AI-assisted material was critically reviewed, revised where necessary, and verified by the author, who assumes full responsibility for the final manuscript. References Brodkin, E. Z. (2008). Accountability in street-level organizations. International Journal of Public Administration, 31(3), 317–336. https://doi.org/10.1080/01900690701590587 Hupe, P., & Hill, M. (2007). Street-level bureaucracy and public accountability. Public Administration, 85(2), 279–299. https://doi.org/10.1111/j.1467-9299.2007.00650.x Lipsky, M. (1980). Street-level bureaucracy: Dilemmas of the individual in public services. Russell Sage Foundation. McGann, M. (2025). Beyond administrative burden: Activation and administrative harm. Australian Journal of Social Issues. https://doi.org/10.1002/ajs4.371 Adams, G. B., & Balfour, D. L. (1998). Unmasking administrative evil. SAGE Publications. Coady, C. A. J. (2023). The problem of dirty hands. In E. N. Zalta & U. Nodelman (Eds.), The Stanford Encyclopedia of Philosophy. Stanford University. Cover, R. M. (1986). Violence and the word. Yale Law Journal, 95, 1601–1629. Foucault, M. (1977). Überwachen und Strafen: Die Geburt des Gefängnisses (W. Seitter, Trans.). Suhrkamp. Fricker, M. (2007). Epistemic injustice: Power and the ethics of knowing. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780198237907.001.0001 Galtung, J. (1969). Violence, peace, and peace research. Journal of Peace Research, 6(3), 167– 191. https://doi.org/10.1177/002234336900600301 Green, P., & Ward, T. (2004). State crime: Governments, violence and corruption. Pluto Press. Hillyard, P., Pantazis, C., Tombs, S., & Gordon, D. (Eds.). (2004). Beyond criminology: Taking harm seriously. Pluto Press. Sager, F., Thomann, E., & Hupe, P. (2019). Accountability of public servants at the street level. In H. Sullivan, H. Dickinson, & H. Henderson (Eds.), The Palgrave handbook of the 139 public servant. Palgrave Macmillan. https://doi.org/10.1007/978-3-030-03008-7_5-1 Shklar, J. N. (1964). Legalism: Law, morals, and political trials. Harvard University Press. Walzer, M. (1973). Political action: The problem of dirty hands. Philosophy & Public Affairs, 2(2), 160–180. McGoey, L. (2012). The logic of strategic ignorance. The British Journal of Sociology, 63(3), 533–576. https://doi.org/10.1111/j.1468-4446.2012.01424.x Rosen, G. (2003). Culpability and ignorance. Proceedings of the Aristotelian Society, 103, 61– 84. https://doi.org/10.1111/j.0066-7372.2003.00064.x Speed, T. (2025). SPEED'S WORK - An Autistic Intervention in the Concept of Work In the Age of AI and Robotics (Version AAM DNB Version English). Zenodo. https://doi.org/10.5281/zenodo.19035228 Speed, T. (2026). 1667:25 Violence, Health Harm and the Denial of the Rights of Autistic People in Germany - Proceedings before the Social Court Cottbus and the Higher Social Court Berlin-Brandenburg - On the Practical Impossibility of Enforcing Effective Legal Protection for Neurodivergent People (Version 1 English). Zenodo. https://doi.org/10.5281/zenodo.18887765 Speed, T. (2026). The Epistemic Crisis of Autism Diagnosis: False-Negative Autism Assessment, Epistemic Injustice, and the Case of the Humboldt University Autism Clinic in Berlin (Versions 2.0 Corrections Page 18 / Abstract Correction). Zenodo. https://doi.org/10.5281/zenodo.22065932 Speed, T. (2026). The Game No One Can Win: The Bedarfsgemeinschaft as a State Experiment in Violence Involving Two Disabled, Neurodivergent Artists (Version 1). Zenodo. https://doi.org/10.5281/zenodo.21978519 Speed, T. (2026). Welfare Regimes as Structural Harm Neurodivergence, Welfare Conditionality, and Institutional Violence (Studies in World-Formation, Vol. 15) (Version 1). Zenodo. https://doi.org/10.5281/zenodo.19002085 Speed, T. (2026). Ontological Coexistence and Structural Violence in Law (Studies in World- Formation, Vol. 14) (Version 1). Zenodo. https://doi.org/10.5281/zenodo.19001795 Speed, T. (2026). Povertism, Exploitation, and the Political Economy of Manufactured Vulnerability (Studies in World-Formation, Vol. 16) (Version 1). Zenodo. https://doi.org/10.5281/zenodo.19002300 140