祝融说。

Care is rarely completed by one person alone. Family members know the habits, professionals handle specialized judgment, service organizations provide stable processes, community relations detect anomalies, and public institutions share risks that no individual could bear. Division of labor can raise capability, and it can also spare the person being cared for overdependence on the goodwill of a particular relative. Division of labor simultaneously produces boundaries. The delivery service completes the drop-off, the medical institution is responsible for the prescription, family members remind about follow-up visits, and the community intervenes only when something is abnormal. Each boundary looks reasonable in isolation, yet between the boundaries tasks may remain that no one formally owns.

We continue the fictional Qiaonan neighborhood. After the delivery of Lin Lan's mother's medication resumed, the prescription still had to be renewed, dosages checked, follow-up visits arranged, and any discomfort watched for. Lin Lan assumed the service organization would send reminders; the service organization believed the prescription was the clinic's responsibility; the clinic assumed the family would make contact; the community volunteers could only register a request once the mother herself asked for help. In the end, Lin Lan, who had the fewest options, took temporary leave and stitched the four windows back together.

This thought experiment offers no medical advice and passes no judgment on real institutions. Medication and follow-up visits serve only as one visible task chain. Whatever requires professional judgment must still be handled by qualified personnel. What this chapter concerns is who sees the gap, who has the capacity to act, who bears the waiting, and how responsibility moves through the words "this is not mine."

Where Care Falls: Family, Market, or Public Services

"Taking care of Mother's medication" sounds like a single thing, yet it includes at least obtaining the prescription, delivering the medication, understanding the instructions, observing changes, contacting professionals, arranging transportation, and handling failure. Merging these actions into one moral word makes it easy for family members to be asked to shoulder the burden without limit, and equally easy for institutions to let a completed partial task stand for the whole need having been met. The delivery worker can deliver accurately yet lacks the qualification to modify a prescription; the clinic can make professional judgments yet may not know whether the mother received the notice; Lin Lan knows the daily changes yet cannot substitute for professional assessment. Coordination is not residual labor devoid of technical content. It requires identifying dependencies, remembering times, passing along materials, and deciding the next step when something fails. Because the outcome often appears as "nothing went wrong," this portion easily fails to enter formal accounting. Continuous observation also has boundaries. That family members or neighbors can more easily notice a change in rhythm does not mean they should monitor around the clock. The person being cared for retains the rights to privacy, to refusal, and to reasonably assuming risk herself. Once responsibility is taken apart, professional boundaries become clearer and gaps become more visible. The point of decomposition is not to assign a permanent owner to every step, but to prevent each party from using local completion to declare the whole chain complete.

When the clinic could not reach the mother, it dialed Lin Lan's number first; a delivery anomaly required her to reconfirm; the community form also listed her as emergency contact. No institution formally entrusted her with coordination, yet she became the default outlet shared by all these systems. The family holds long-term knowledge, emotional commitment, and the capacity for rapid action, and as a care relation this is not inherently oppressive. Many people also wish to express love and responsibility through care. An emergency contact designation means that a specific situation permits contact; it is not consent to take over all daily affairs. Lin Lan has work, health, and other relationships, and her available time will change. Distribution inside the family may also form automatically along existing roles. Whoever lives closer, whoever communicates better, whoever once said yes — that person keeps receiving the tasks. In reality different genders, ages, and economic positions may bear different pressures, but this chapter claims no uniform distribution in the absence of data. "Family should be responsible" is a value judgment, and it needs specification of minimum obligations, public support, and the conditions under which bearing becomes impossible. Affection can generate commitment, but it cannot let every institutional gap be privatized without condition. Conversely, the existence of public services does not cancel family relations. The person being cared for may want familiar people involved, and the family may hold experience that professional processes cannot replace. The goal is to make participation a negotiable responsibility rather than a default portal of unlimited entry.

Lin Lan purchases delivery, booking, and transportation services, eliminating a great deal of repetitive labor. Contracts specify time and object, which is more reliable than vague requests to relatives and friends, and it spares both sides from having to build personal dependency. Professional division of labor can raise quality. Writing all commercial services as coldness would push care back onto the family and ignore the skill of service workers. The service provider must know what it has promised, otherwise it cannot configure staff and prices. If a delivery worker had to follow up continuously on every anomaly, the work would lose its boundary and might overstep professional authority. The question is whether the uncovered portion is stated, and whether a handoff path exists. The platform displays "order cancelled"; if it does not route high-consequence anomalies to someone able to handle them, technical notification does not amount to continuity of care. Expanding the contract also has costs. If every delivery were required to include lengthy confirmation, prices might rise and people living independently would lose accessible services. One can layer by risk and by subject rather than convert all orders into comprehensive care. The market can sell execution and partial coordination, but it cannot automatically assume every relational responsibility. Public guarantees and voluntary mutual aid outside the contract still have their place; this is not proof of service failure but a fact about the boundaries of what a contract covers.

Aunt Chen noticed that the mother had not come downstairs, and a volunteer helped contact the service station. Community relations, through daily contact, detect anomalies that formal systems miss, and this has real value. Once that value is seen, institutions may default to "the neighbors will look in" and reduce formal follow-up. Mutual aid shifts from supplement to unpaid infrastructure. Aunt Chen's occasional inquiry is autonomous concern; if she had to check the medications every week, she would be carrying a continuous duty. Without authorization, training, relief shifts, and exit, such a duty is unreliable and may intrude on the mother's privacy. Community organizations count volunteer hours as their achievements yet never record how many night calls a few members absorbed. Average participation conceals concentrated burden. Judgment requires actual records; one cannot infer a universal conclusion from a single story. Mutual aid also brings the power of favor. Helpers may expect compliance, information, or gratitude, and those helped fear losing the relationship if they refuse. The limited boundaries of professional services are sometimes, by comparison, freer. Protecting mutual aid does not mean institutionalizing every action; it means providing resources, training, referral, and a clear exit, so that voluntary relations are not forced to extend without limit because institutions have withdrawn.

The service station says it handles registration and referral and does not directly provide medical or home care. This restriction may come from professional authority and resources, not from simple shirking. Public institutions cannot take over every life choice. Excessive entry into the family harms autonomy and may flatten the differences among relationships with uniform standards. Public institutions can guarantee key entry points, emergency response, information linkage, and minimum support for those with no one to catch them, without personally performing all care. One of their responsibilities is to prevent their several boundaries from jointly producing a state in which no one at all is responsible. Eligibility review can direct resources to those in greater need, but complex verification also hands coordination costs to precisely those who lack capacity. Application procedures should offer assistance, agency, and review in proportion to consequences. Limited budgets require priorities, and one cannot use "care is unlimited" to deny allocation. Decisions should state openly which needs are protected, how people are safeguarded while they wait, and which responsibilities are explicitly left to individuals and families. Public responsibility can also be used to diminish the subject. If any risk triggered compulsory takeover, the person being cared for would lose decision. Support should expand the capacity to act as far as possible, and restrictive measures require relevant evidence, proportion, and a time limit.

How the Transfer Chain Falls Downstream

A responsibility chain should record both the task and the conditions of uptake. The table below is a fictional illustration of care transfers; it defines no legal obligation of any real organization.

The use of the schema is to check whether handoffs are complete. It does not predetermine which party is always responsible, nor treat the person being cared for as an object that merely awaits distribution.

The clinic can say wait for the next appointment; the platform can cite the contract; the community can say resources are short; Lin Lan cannot let her mother go on waiting while the prescription lapses. She holds less formal power, and the real cost of exit is greatest for her. The responsibility was not necessarily assigned to her by anyone in particular. Each party separately chose the boundary it could bear, and together these choices formed concentrated bearing. Lin Lan may be highly capable, and precisely because she has repaired failures many times, the system depends on her more. Capability becomes the reason for continuing to transfer responsibility. If she stopped coordinating, the consequences would fall on her mother first, while the institutions would lose little. Moral pressure and the position of consequences jointly constrain exit. Formally voluntary, actual choice is shaped by the structure of relationships. One starkest institutional corollary is this: different organizations can each keep their own rules reasonable, and simply by repeatedly sending cross-boundary failures to people who cannot let go, they can keep the whole running on unpaid labor. It requires no central command, and it may form from each department optimizing its own metrics. Pointing out this possibility is not a method for offloading blame. The direction of repair is to track cross-boundary failures, to give someone the responsibility to pause and transfer, to compensate coordination labor, and to let the person with the fewest options say "no" without the person being cared for immediately losing basic guarantees.

Lin Lan needs to update the address separately in the clinic, delivery, and community systems, upload the same proofs, and remember different accounts. Digital entry points reduce queuing, yet they add account maintenance and information conversion. People familiar with the systems may find the steps simple. Caregivers must also handle errors in the gaps between work, explain notices, and help the person being cared for confirm — not merely complete clicks. Reminders can lower the cost of memory, yet too many reminders then require judging which of them matter. Autofill reduces input and may also perpetuate wrong records. Convenience and risk arrive together. A unified account reduces duplication, but it creates shared eligibility and a single point of failure. Chapter Nine of another volume has already discussed cross-domain binding; care systems all the more need to distinguish identity facts, professional records, and resource eligibility. Delegated access helps Lin Lan handle affairs, and it may also strip the mother of privacy and control. Authorization should be set by task, term, and revocable scope, with a record the person herself can understand. An offline entrance is not a symbolic window kept in reserve after digital failure. It needs real authority and the capacity to handle anomalies and corrections, otherwise it only retranscribes the online problem.

A road closure once suspended delivery, and the service provider sent out a uniform notice. The clinic suggested arranging ahead; the community organization recruited volunteers on short notice; the family was told to solve it themselves. In crisis every party's capacity declines, and one cannot demand unlimited performance by ordinary standards. Temporary division of labor can save the critical tasks. If Lin Lan and the volunteers took over in sequence, the time, costs, risks, and unfinished tasks should be recorded. After the crisis ends, their success cannot be treated as evidence that formal capacity suffices. Temporary exceptions easily become permanent. The institution discovers that the family can bear it and keeps postponing restoration; the family, having already mastered the process, finds it harder to exit. End conditions and compensation need to be made explicit. Conversely, strict adherence to ordinary boundaries in crisis may cause immediate harm. Critique of responsibility transfer cannot demand that workers in a crisis first complete perfect procedures and then act. Minimum protection comes first, followed by review of authority and bearing. Crisis narratives may also be used to compress choices over the long term. Whether the situation is still urgent, whether the restriction is necessary, whether alternatives have been restored — these need periodic review, not indefinite extension by the same declaration.

The mother may be willing to manage part of her affairs herself, and she may refuse to let Lin Lan see all her records. The existence of risk does not automatically strip her of standing. Care discussions often ask only whether the family or the institution is responsible, omitting the preferences, capacities, and bearing of the person receiving care. If support decides everything in her stead, protection turns into relational control. Demanding that the mother complete complex procedures alone does not necessarily respect autonomy; it may also package institutional burden as training in capability. She can choose delegation, joint decision, or professional assistance. She also bears responsibility for commitments she can understand, such as reporting changes promptly and using services reasonably. Responsibility needs to be proportionate to cognition, resources, and consequences; one cannot cancel all responsibility because help is needed, nor assume that formal notification equals understanding. When the family's opinion conflicts with her own, specific capacity assessment and professional boundaries are required. The thought experiment offers no medical or legal conclusions. The general principle is that the graver the restriction, the stricter the evidence and the review. The person being cared for can also join the design of services, pointing out which reminders help and which delegations overstep. Participation does not obligate her to repair the system; institutions remain responsible for the processes they control.

Refusal, Professionalism, and Explicit Handoff

If Lin Lan cannot refuse any added task, her work and health may erode step by step until the care relation breaks down as a whole. Short-term unlimited responsibility is not long-term reliability. Refusal also brings risk to the mother, so "set boundaries" cannot be offered as a piece of personal advice without consequences. Effective refusal needs alternative entry points, transition time, and task handoff. Lin Lan can say that she cannot manage transportation this month while continuing to visit every week. Partial refusal is not an absence of love, and it should not be interpreted by institutions as the family's total withdrawal. Service personnel likewise need boundaries. Requiring every front-line worker to make up institutional shortfall with goodwill produces hidden labor and inconsistent outcomes. Escalation mechanisms should let them carry the gap to a position that holds authority. The exit of a mutual-aid participant should not meet moral shaming, but for critical tasks explicitly promised, notice should be given in advance. Voluntary does not mean commitment without responsibility, and responsibility does not mean never exiting. Public guarantees provide the minimum condition for refusal: when private relations temporarily cannot carry the load, basic needs do not immediately fall. The scope of the guarantee needs to be decided by concrete institutions rather than extended without limit by this chapter.

The professionals are responsible for judgment on the prescription, a stable service performs the delivery, Lin Lan provides the relational support she is willing to carry, and Aunt Chen asks only when everyday anomalies appear. Each party states its boundaries and its handoffs clearly. Professionalization makes responsibility trainable, recordable, and relievable, so that care does not depend entirely on affection. Relational continuity supplies context, trust, and the detection of non-standard changes. A platform cannot claim a long-term relationship because it sends care notices, and family cannot substitute for professional diagnosis because they know the habits. Each capability acknowledges its own conditions of failure. Joint meetings can sometimes connect tasks, yet they may multiply the person's repeated narration and her exposure of privacy. Information is shared with those who need it, and major decisions are the responsibility of those with authority. A stable contact person can reduce repetition, but a single responsible person again forms dependency. A team needs internal handoff and backup, so that continuity belongs to an arrangement of relations and not merely to some well-meaning individual. Quality cannot be measured by task count alone. Whether anomalies enter handling, who bears the waiting, whether the subject understands, whether the caregiver can rest — these equally affect sustainability. Real indicators need to be designed through professional research; this chapter only proposes dimensions of observation.

One can draw the delivery process as nodes: prescription generated, notice, ordering, delivery, confirmation, observation, correction. Each node marks the primary responsibility, the backup entry, who waits, and who comes next after failure. A responsibility map does not turn ethics into a flowchart. Love, trust, and dignity cannot be exhausted by arrows, but the map can show which necessary actions have been excluded by every formal boundary. If customer service must refer high-risk anomalies, then a usable channel and a response deadline must exist; if the community can only register, it should not be evaluated as having resolved the matter. Responsibility cannot gain a name without the capacity behind it. Cross-agency problems need a temporary coordinator, but the coordination role must not permanently swallow all execution. Whoever controls a link remains responsible for that link's outcomes. Review should look for the common pattern, not merely applaud Lin Lan for saving the day once again. The narrative of personal heroism lets institutions keep depending on those with the fewest options, and it burdens with shame those who failed to save the day. If a deliberate pushing of costs onto the weaker party is discovered, it must be proven through contracts, records, budgets, and consequences; where no concentrated intent exists, structural responsibility remains. Dispersed causation does not mean that no one is responsible.

The clinic states, when renewing a prescription, whom it will notify and what happens if there is no response; the delivery service, when cancelling a high-impact order, provides a reachable handoff; the community, on finding an anomaly, can contact a service entrance that holds authority. Lin Lan chooses which tasks and terms she herself will carry. Every handoff includes at least the object, receipt confirmation, a deadline, and a return path on failure. A high-consequence task cannot be represented as caught by automated sending alone. The temporary coordinator is responsible for confirming that the problem has entered handling, not for completing every institution's work. After resolution, the correction and the subsequent prevention are recorded, so that the same gap does not repeatedly fall back onto the family. The time and cost of care labor enter the discussion of resources. Compensation need not commodify affection; it acknowledges the real investment that institutions depend on, while voluntary relations can retain a meaning not measured by price. The person being cared for can see who is doing what, withdraw unnecessary authorizations, and decide jointly within the scope of her capacity. Privacy limits and emergency exceptions accept professional and legal boundaries. These arrangements will add cost and coordination. Civilization cannot keep prices low by assuming that care has no cost, nor promise unlimited public uptake because costs exist. Open allocation is more honest than letting the gap fall, again and again, on the silent.

Confidentiality, Accounting, and Multi-party Conflict

The clinic cannot tell the neighbors everything about the mother, and the delivery service should not expose medication information to irrelevant parties. Confidentiality protects the dignity of the subject; it is not the name of an obstacle to coordination. But if each party says only "we cannot disclose," the bearer may not know whether action is needed. Task status can be conveyed without publishing the specific content — for example, that a key contact is still outstanding and requires the person herself or an authorized agent to respond. The delegation Lin Lan holds does not mean she may view the entire history, and Aunt Chen's discovery of an anomaly does not thereby grant her the medical records. The scope of information varies with the task and the authorization. Emergency situations may permit faster sharing, with the reasons and notifications recorded afterward. An exception cannot be extended permanently because a risk occurred once. If the person being cared for refuses sharing, the system must confirm that she understands the possible consequences and must retain proportionate support. Autonomy includes risk, but risk does not authorize others to cancel her refusal through vague judgments of capacity. A coordination platform can show who has taken responsibility and when follow-up is due, without centralizing all sensitive material. Visibility of process and disclosure of content are not the same requirement.

Lin Lan takes leave to coordinate windows, pays for transportation, and bears the waiting. If these costs go entirely unrecorded, the service system will appear cheaper than it is. Bringing care labor into accounting does not mean that all affection should be paid by the piece. The gifts within a relationship carry their own meaning, and money cannot express the whole of commitment. One can record coordination time, extra expenditure, interruption of work, and non-substitutable tasks without converting emotion into a uniform price. The material serves to judge who bears the load over the long term, not to grade who loves their family more. Compensation can take the form of expenses, respite, relief shifts, simplified procedures, or stable professional support. Cash is not the only form, and verbal thanks cannot substitute for real recovery. If institutions withdraw support because the family is willing to care, they convert love into a fiscal resource; if they entirely disregard the family's capacity, they may duplicate services or ignore the subject's preferences. Assessment should be carried out together with those concerned and should allow circumstances to change. Real labor and welfare arrangements are governed by concrete law, and this chapter claims no uniform right already in place. It only proposes the evidence needed to judge the allocation of responsibility: who did what, why they could not exit, and how the costs bear on the rest of a life.

The nurse, the customer service agent, the social worker, and the front-line window may all see the problem, yet only a superior or another department can change eligibility. Front-line workers track cases in their private time; professional goodwill absorbs the institutional gap. If this labor rests only on a few capable individuals, the service snaps when they leave. Organizations may also treat successful cases as proof that the process works and keep compressing formal resources. Demanding that the front line "report problems promptly" is not enough. The receiving end needs authority, response deadlines, and reasons for return, so that the same case does not circulate between departments. Professional boundaries protect service users, and they protect staff. Where a judgment exceeds one's qualification, referral is a responsible act; if after referral no party takes over, the gap must be handled by the coordination layer. A worker's refusal of tasks beyond authority should not be moralized as coldness, nor should initiative become the new minimum standard for all colleagues. Only when the organization formalizes necessary work can training, relief, and supervision become stable. Where an individual is negligent or discriminatory, structural analysis does not excuse them. Individual conduct and institutional conditions can both hold at once, and repair must correspond to each.

Lin Lan wants to hand more affairs to professional services; another relative believes the family should handle things themselves; the mother wants less intervention. All three may cite care as their reason, and care cannot automatically decide who is right. If the family meeting is decided only by whoever speaks best or pays most, the labor of others and the person's own preferences are pushed into the background. Equality also does not mean that everyone bears the same hours; distance, capacity, and existing commitments all matter. One can discuss separately who accompanies the clinic visits, who handles the accounts, who takes the night calls, rather than argue over who is more filial. Once the tasks are taken apart, partial consent and partial refusal both become clearer. The one who pays has the right to know how the money is used, but payment does not purchase every decision about a life; the experience of the one who gives time needs to be seen, but it does not confer ownership of the person being cared for. Different contributions correspond to limited grounds of voice. Professionals can help explain risks and options, but they cannot distribute the family's entire moral responsibility. Real conflicts of capacity may require legal and professional procedures, and this chapter renders no verdict on individual cases. If the caregivers cannot cooperate, the independent entrances of public and market services matter especially. Institutions cannot assume that "the family will sort it out among themselves" settles the matter, nor do they need to take over every relational dispute.

The transfer of care responsibility sometimes proceeds through open conflict rather than silent shirking. Recording who refused what, what alternatives exist, and who bears the consequences is more sustainable than covering the dispute with a narrative of family harmony.

The next chapter moves the lens into a classroom with only two advancement slots. Family, school, and public institutions can likewise shift responsibility for cultivation, selection, and risk onto one another, leaving students and caregivers to carry the burden of continuous proof. The fine division of responsibility can also produce the opposite of what it promises: a system can, by endlessly subdividing duties, let every institution meet its own metrics while handing the labor of connecting the world to those who cannot exit. The whole depends on them, yet the official narrative credits success to process and blames failure on deficient family capability. Critique of this structure does not negate family commitment, market services, community mutual aid, or public boundaries. It demands that local completion not substitute for whole-chain continuity, that improvised rescue not become a permanent budget, that profession and relationship each acknowledge their limits, that the right of refusal carry alternative conditions, and that the person being cared for remain a subject. Care responsibility can be handed over, but it must not disappear in the handing over.