--- title: "Convergence Point: Wanhong Huang and the Future You Cannot Give Back" slug: "convergence-point-wanhong-huang-and-the-future-you-cannot-give-back" canonical_url: "https://modalpathethics.com/convergence-point-wanhong-huang-and-the-future-you-cannot-give-back/" mirror_url: "https://mirror.modalpathethics.com/articles/convergence-point-wanhong-huang-and-the-future-you-cannot-give-back.md" published_at: "2026-10-23T06:01:00.000-05:00" updated_at: "2026-10-23T06:00:59.000-05:00" tags: - "Convergence Point" - "Modal Path Ethics" authors: - "Aidan Lawson" source: "Ghost Content API — published post" mirror_generated_at: "2026-10-27T18:14:50.285Z" mirror_generator_version: "2.0.0" sha256_plaintext: "f4effad978e143e6d1f31042e4aaa9e35cfacf51fae097dd08673da4f415bc76" --- # Convergence Point: Wanhong Huang and the Future You Cannot Give Back I have discovered someone asking the return-policy version of Modal Path Ethics. Suppose you take something from a person like a job, a research position, a reputation, access to an institution, several years of health, a relationship, or a right. Then, eventually, you admit the mistake. And you give it back. Awesome. > **Are we done here?** Wanhong Huang's answer is: > _What exactly do you think you returned?_ This becomes a problem very quickly. Because the job came back. Except the years did not. The equipment came back. But the experiments that were never conducted did not. The reputation was corrected. And the people who never collaborated with you because of the false reputation did not crawl backward through time and introduce themselves. - The right was restored. - The future it used to open may no longer exist. * * * ## The Dates Come First. The overlap is strong, so once again, chronology first. Modal Path Ethics had committed to its formal reachability account back in early May. - May 7 2026: [_Formal: Contraction Is Harm_](https://modalpathethics.com/contraction-is-harm/) defines harm through degradation of reachable future-space. - May 9: [_Formal: Weighted Reachable Future Space_](https://modalpathethics.com/formal-weighted-reachable-future-space/) and [_Formal: What Makes Something a Locus_](https://modalpathethics.com/formal-what-makes-something-a-locus/) make future-space bearer-relative and ask what kind of extant thing can actually bear contraction. - May 11: [_Formal: Resistance and Harm_](https://modalpathethics.com/formal-resistance-and-harm/) argues that a future may remain formally available while the path toward it is already materially damaged. Huang's public research line then develops through the summer. - By June 3, [_Toward Relational Being in Intimate Life_](https://huang-wanhong.com/2026/06/03/Toward%20Relational%20Being%20in%20Intimate%20Life%20-%20Integrative%20Justice%20and%20an%20Ethics%20of%20Dialectical-Positivist%20Practice/?ref=modalpathethics.com) is treating justice structurally rather than as a numerical optimization problem. - A June 11 paper, later posted to SSRN in July as [_The Sustainability of Intimacy_](https://huang-wanhong.com/2026/06/11/The%20Sustainability%20of%20Intimacy%20-%20Poetic%20Generativity%20and%20Generative%20Justice/?ref=modalpathethics.com), explicitly argues that self-continuation and generativity are morally neutral unless supplemented by a theory of justice. Then the vocabulary gets much closer. - By August 1, Huang is explicitly representing the possibilities available to each party in a relation as a reachable set, [while warning that the size or change of that set does not by itself carry an evaluative sign](https://huang-wanhong.com/2026/08/01/Marriage%20from%20a%20Dialectical%20Generative%20Relational%20Economics%20Perspective%20-%20A%20Preliminary%20Discussion%20of%20the%20Relation%20Between%20Allocative%20and%20Generative%20Accounts/?ref=modalpathethics.com). - By August 16, [_Plasticity, Generativity, and the Acquisition of Constraint_](https://huang-wanhong.com/2026/08/16/Plasticity%2C%20Generativity%2C%20and%20the%20Acquisition%20of%20Constraint%20-%20A%20Preliminary%20Discussion%20of%20Subject%20Formation%20in%20a%20Background-Independent%20Relational%20Ontology%20%E3%80%90%EF%BC%88Preliminary%EF%BC%89Draft%E3%80%91/?ref=modalpathethics.com) defines **plasticity** through horizon-bounded reachable configurations and establishes a foreclosure proposition: realized paths exclude bypassed alternatives where genuine branching existed. That descriptive foreclosure is not automatically a moral loss. - Then, on September 11, Huang drops _Generativity and the Justice of Return: Theory, Practice, and an Atlas of Equivalence_ on Zenodo. [ Generativity and the Justice of Return: Theory, Practice, and an Atlas of Equivalence This paper develops a comprehensive inquiry into the justice of return when harm, deprivation, extraction, or intervention has altered the conditions through which subjects and relational systems generate value, capacities, relations, and future possibilities. The inquiry begins from historically and relationally constituted subjects whose generativity continues to evolve during and after harm. Under such conditions, the return of a resource, status, opportunity, or formal right need not restore its prior generative significance. The paper therefore distinguishes the return of objects and resources from the restoration of relations, capacities, trajectories, conversion structures, and generative processes. It examines these problems across pre-harm prevention, intervention during ongoing harm, and post-harm restitution, compensation, repair, assistance, and future-directed support, while considering the respective roles and limits of law, governance, political economy, and assistance practices. The central analytical contribution is an atlas of restorative comparison. A claim that something has been restored is treated as an indexed comparative judgment requiring specification of a reference system, an object of comparison, a comparative relation, relevant observations or interventions, a temporal horizon, an approximation tolerance, and an epistemic basis. The paper surveys comparison objects including resources, attributes, relations, networks, states, trajectories, reachable sets, transition structures, attractors, basins, stability properties, perturbation responses, causal structures, conversion mechanisms, generative processes, meanings, and historical configurations. It then examines comparative relations including identity, isomorphism, homomorphism, simulation, bisimulation, observational and trace equivalence, topological and dynamical correspondences, statistical indistinguishability, causal and interventional equivalence, approximation, dominance, and refinement. Particular attention is given to dynamical and generative comparison, including the non-equivalence of delayed return, robustness under admissible perturbations, and changes in the processes through which future possibilities are generated. The paper further connects these comparative structures to theories of value and justice, including resource, capability, relational, recognition-based, rights-based, care-oriented, reparative, and generative approaches. Substantive justice in the achieved condition is distinguished from procedural justice in the means through which a proposed equivalence is produced. Idealized technologies and hypothetical powers are used to separate current practical limitations from epistemic, structural, and normative limits. The resulting framework does not identify a universally sufficient form of return. It instead maps the choices concealed within claims of restitution, compensation, repair, and restoration, and provides a common structure for asking what should be compared, in what sense equivalence matters, which values warrant preservation, and which processes of return can themselves be regarded as just. ![](https://storage.ghost.io/c/20/43/2043f11a-6ae3-404c-bb28-01fce8d9ac88/content/images/icon/favicon-88b25a79-3203-4728-889d-1da717372704.png)ZenodoHUANG, Wanhong ![](https://storage.ghost.io/c/20/43/2043f11a-6ae3-404c-bb28-01fce8d9ac88/content/images/thumbnail/orcid-00dc21e5-0690-45f2-861f-3b68c7a5b7ed.png) ](https://zenodo.org/records/22707759?utm_source=chatgpt.com) And that paper talks about: > reachable sets, future possibilities, contraction, trajectories, transition structures, generative processes, historical path-dependence, irreversibility, repair, and the difference between nominal restoration and restoration of what the lost condition could still have generated. Nine days later, Modal Path Ethics publishes v1.2 of _Harm as Contraction of Reachable Continuation_, defining **harm** around degradation of structurally significant continuation or enabling structure genuinely reachable from an extant locus. - Modal Path Ethics' public reachable-future formalism was already online in May. - Huang's generativity project was independently developing by early June and then progressively acquired its own reachable-set machinery through August. - The two projects arrive in September carrying disturbingly compatible equipment. That is what this series is for. * * * ### **Give It Back.** Huang's starting problem sounds almost embarrassingly obvious once stated. Restitution often assumes something like an inverse operation. - You lost **(x)**. - Later, we return **(x)**. Therefore: > **problem approximately reversed.** That works tolerably well for some objects. - I take your hammer. - I return your hammer ten minutes later. Fine. But suppose that hammer was the only hammer in a workshop for six years. Now returning the hammer does not return the six years of things the workshop could have built with it. That difference is Huang's subject. He calls attention to the generative significance of a condition: what further capacities, relations, opportunities, knowledge, resources, meanings, and possibilities it participates in producing over time. A research position does not just sit over there _being a research position_. That produces experiments. Publications. Collaborators. Reputation. New questions. Access. Skills. Later positions. Returning the original position years later does not manufacture that missing causal history. This is stronger than: > Some harms are difficult to compensate. Huang knows that literature already exists. His question is more annoying. > What relation would have to hold between the repaired world and the relevant reference world for the word **“restored”** to be true? And then, because apparently he had free time, he builds an atlas. * * * ## Huang Has Brought an Atlas. A claim like: > **The harm has been repaired.** is radically underspecified. **What was repaired?** - The object? - The person's legal status? - Their income? - Their capability? - Their position in a social network? - The trajectory they were on? - The reachable futures available from that trajectory? - The causal mechanism through which opportunities were generated? - The historical relation itself? **And what kind of similarity counts?** - Identity? - Approximation? - Functional substitution? - Simulation? - Bisimulation? - Reachability preservation? - Dominance? - Some threshold of sufficiency? Huang's framework treats _restoration_ as an indexed comparative judgment. You have to specify the reference system, the thing being compared, the relation under which it is supposed to count as restored, relevant observations or interventions, the time horizon, acceptable approximation, and the epistemic basis for the comparison. Modal Path Ethics has spent months saying things like: > Repair does not erase history. Huang has now arrived with the bill of materials for the word **repair**. * * * ### Then the Reachable Set Walks In. Huang does not require us to pretend that one particular missing future was destined to occur. Suppose an institution wrongly excludes someone from a profession for ten years. We do not know: > They definitely would have written these three papers, met this collaborator, taken this position, and made exactly $417,000. That history simply never happened. Treating one imagined counterfactual biography as a lost object would be epistemically insane. One of Huang's alternatives is the counterfactual reachable set. Instead of inventing one biography that supposedly _would have happened_, ask which families of continuation were available under specified conditions. The harm may have removed regions or routes from that possibility structure without establishing that any particular state would certainly have been realized. That is almost precisely why Modal Path Ethics moved into reachable continuation in the first place. The morally important fact does not require a prophecy. The future does not need to have been guaranteed. It had to have been **genuinely reachable**. Destroy the enabling conditions and something real has happened even though nobody can produce the missing future as Exhibit A. This is the **_convergence_**. * * * ## Size: Not Enough Either. Huang then makes another move Modal Path Ethics should keep. Two reachable sets can have the same size and represent radically different worlds. - One may contain diverse, independent routes. - Another may technically contain the same number of options while every route passes through one controlling institution. - One may be connected. - Another fragmented. - One may place desirable continuations nearby. - Another leaves them behind extreme cost, dependency, or risk. So Huang distinguishes reachable-set volume from reachable-set composition, geometry, topology, inclusion, and viability. That is extremely compatible with this framework's longstanding refusal to turn _weighted reachable future-space_ into **More Branches Good.** Ten worthless or predatory routes do not compensate for one structurally central continuation being destroyed. And fifty nominal options routed through one gatekeeper may constitute less real freedom than five materially independent paths. The tree was always the wrong picture if all the branches secretly share one root permission. * * * ### He Also Finds the Future You Cannot See. Huang distinguishes the **objectively reachable set** from the _recognized reachable set_: the possibilities a subject can actually identify and treat as actionable. This sounds small. It is not. A path can **objectively** reopen while remaining unavailable to the person whose history taught them that the path was closed. Formal access can return. Material access can return. Information, confidence, social support, expectation, or practical imagination may not. The door is unlocked, but the damaged subject does not yet inhabit a world in which the door functions as an option. Modal Path Ethics has mostly approached this through resistance. Huang gives the epistemic side cleaner standing. Restoring a route and restoring the recognized possibility of the route are different operations. A bureaucracy can put the opportunity back on the website and congratulate itself. The field may still remember what the bureaucracy did. * * * ### The Subject You Are “Restoring” Has Also Continued Existing. Then Huang makes the best move in the paper. Suppose restoration becomes technologically extraordinary. Imagine we can reconstruct the lost job. The lost network. The missed opportunity. The original institutional conditions. The **entire counterfactual world**, almost perfectly. > Should we? Maybe not. Because the person standing here now is not just some damaged approximation of the person who would have existed without the harm. [They have continued becoming.](https://modalpathethics.com/multiplayer-catherine-malabou-and-the-transition-problem/) They may now prefer another life. Huang therefore insists on the affected subject's right to reject restoration. A procedure can produce greater similarity to the hypothetical no-harm world while becoming more unjust toward the actual person being repaired. - Erase memories. - Manipulate relationships. - Force treatment. - Compel someone back into an institution. - Restrict their new choices so their life looks more like the counterfactual model. Congratulations. The restoration score has improved. The person would like you to stop restoring them now. Huang's principle is therefore brutal and necessary: > stronger equivalence does not automatically mean greater justice. Outcome fidelity and procedural legitimacy have to remain separate. That one travels directly into Modal Path Ethics. Repair cannot become sovereignty over the repaired. * * * ## **_Convergence Point._** We can now state the overlap without pretending the theories are identical. Huang and Modal Path Ethics independently reject the idea that moral analysis can stop at the present inventory. Both treat history as causally active. Both distinguish possession of a present condition from the future-generating structure attached to that condition. Both use reachable sets to represent real future possibility without pretending one exact counterfactual future was predetermined. Both recognize that a harmful intervention can alter routes, capacities, relations, and future generation before complete destruction occurs. Both reject simple branch count. Both distinguish restoration from erasure of history. And both arrive at a closely related structural intuition: > Giving back what was taken does not necessarily give back what the taken thing was still capable of becoming. That is a serious convergence, and also where the disagreement begins. * * * ### **Huang Refuses to Call Generativity Good.** This is the most important difference. - For Huang, generativity is normatively neutral. A loving relationship can generate. So can an empire. A community can reproduce valuable forms of life. A surveillance institution can become spectacularly effective at reproducing surveillance. A destructive system can preserve and expand its own generative machinery. So Huang refuses the inference: > More generativity = more justice. He requires an independent theory of value to determine which generative structures deserve preservation. Modal Path Ethics agrees with half of this immediately. It does not maximize optionality or say every continuation should survive. Predatory futures can rightly be closed. Protective friction can be justified. The remaining disagreement is subtler. Current Modal Path Ethics makes a stronger primitive claim: > qualifying degradation of a structurally significant reachable continuation is itself a harm, even though the wider transition may still be justified. Huang is more reluctant to let generative loss perform that normative work by itself. His sequence is roughly: - describe generativity → - specify comparison → - introduce value theory → - judge justice. Modal Path Ethics runs more like: - identify locus → - identify actual contraction → - register harm → - judge whether the wider transition was justified. **Extremely** close. Not identical. Good. We have something to argue about. * * * ### Huang Uses the Counterfactual More Than Me. There is another separation worth preserving. Huang's project is about **return**. Return necessarily asks: > Compared with what? So the no-harm trajectory, distribution, reachable set, or historical configuration enters as a reference. He is extremely careful about uncertainty here. He does not pretend the counterfactual is uniquely knowable. But restoration remains comparative. Modal Path Ethics' tightened harm account tries very hard to require less. It only asks whether an actual transition nontrivially degraded a structurally significant continuation or enabling structure genuinely reachable from an actual locus. That means we can sometimes identify damage **without** reconstructing the entire world in which the damaging action never occurred. Huang asks: > What would count as sufficient return? Modal Path Ethics first asks: > What did this actual transition damage? Those are different questions. They belong beside one another. * * * ### A Comparison Object: Not Automatically a Locus Either. Huang's atlas can compare almost anything: - resources, - relations, - networks, - states, - trajectories, - reachable sets, - causal structures, - generative processes, - histories. That does not mean each of those things is the bearer of moral harm. Modal Path Ethics has spent a frankly ridiculous amount of time worrying about precisely that distinction. The object being measured is not automatically the locus whose continuation matters. A damaged employment network may matter because of what happened to the worker. A damaged institution may also sometimes constitute a real locus at another grain. A historical process may be explanatory without becoming a little moral patient hiding inside the graph. Huang gives us a much richer language for comparing what changed. Modal Path Ethics still needs its separate argument about what bears the change. That border should remain. * * * ## Repair. Huang gives us a substantially better repair question. When someone claims a harm has been repaired, ask: > What object was restored? - Relative to which reference? - Under what equivalence relation? - Across what horizon? - Against which interventions? - With what tolerance? - On what epistemic basis? And then ask a completely separate question: > Was the process of producing that restoration itself just? That should become part of the repair machinery. Because “we gave it back” is not an argument. It is the beginning of an audit. * * * ## The Ruling. Wanhong Huang is another **_Convergence Point_** interlocutor. He arrived through generativity, relational justice, restitution, historical development, and the problem of what it means to restore something whose value consisted partly in what it could generate next. Modal Path Ethics began from contraction. It asked how an actual transition can damage the reachable continuation of an extant locus even when no particular future was guaranteed and even before complete closure. The two projects meet in the place between the returned object and the missing future. Huang asks: > What would it actually mean for that future-generating structure to have been restored? Modal Path Ethics asks: > What happened when that structure was damaged in the first place? - One is building a theory of return. - The other is building a theory of harm. - And both have discovered the same unpleasant fact: > The present does not contain everything that was taken from it. You can return the job. You can restore the title. You can reopen the account. You can correct the record. You can hand back the key. None of those acts, by themselves, puts the lost future back where you found it. - The thing came back. - The future did not. That is the **_Convergence Point_**.