Research and version note This is a version 0.1 research draft in Who Maintains the World? It uses a joint WHO, ILO, World Bank, and WaterAid study of sanitation workers to examine cleaning, risk, and visibility. Labour conditions vary greatly, and the global problems described here should not be attributed to every cleaner or sanitation worker.
Who Maintains the World? · Article 6
Quick read
Cleaning is often understood as restoring a space to the state of “nothing there”: no dirt on the floor, no obstruction in a toilet, and no waste in view. Its outcome is the disappearance of residue, and the people producing that disappearance can leave public attention with it. A clean space looks as though it began that way. Labour becomes visible when cleaning stops.
A sanitation system is not complete when toilets and pipes are installed. Excreta must be collected, moved, treated, and safely disposed of, while sewers and pumping stations remain operational. A joint report by the WHO, ILO, World Bank, and WaterAid documents that sanitation workers in many settings face pathogens, toxic gases, confined spaces, informality, and social stigma. Public-health benefits are broadly distributed while severe hazards can concentrate among people with the least protection.
Automation can reduce direct contact, but it does not automatically remove danger. Equipment must be cleaned, unblocked, and serviced, while inaccessible work remains human. If technology moves exceptional or hazardous tasks to the bottom of a contracting chain, visible modernisation has not changed the distribution of risk.
Respect cannot be limited to describing workers as unsung heroes. Heroic language can praise people for continuing with inadequate protection without changing pay, staffing, personal protective equipment, training, health monitoring, freedom to organise, and the capacity to refuse dangerous work. Recognition should alter conditions and allow cleaners’ observations of leakage, contamination, and poor design to enter facility decisions.
Responsibility also extends upstream. Products, buildings and service schedules determine whether a surface, trap, machine or waste stream can be handled safely. A contract that rewards visible cleanliness at the lowest price can hide rushed work and exposure. Procurement should examine methods, staffing and worker protection rather than assuming that an apparently clean result proves a safe process.
The object being maintained is not visual purity but the safe continuing use of shared space. A society cannot treat its residue as if it disappears by itself, or make people with lower status pay for disappearance through their health. Maintenance should eliminate hazards first, formalise and protect work, provide reporting rights, and redesign facilities and products that cannot be cleaned safely.
Cleaning produces an absence
Many forms of production leave an object that can be counted. Cleaning removes one. When people enter an office, hospital, or station, the floor is already clear, bins have not overflowed, and bathrooms are available. The state in front of them contains little evidence of the work performed hours earlier.
Cleaning is then easily treated as low-skill background. In practice, workers distinguish materials, contamination, chemicals, infection risk, and patterns of use. Hospital cleaning is different from an ordinary office. Human waste and sharps require controls unlike paper. Mixing chemicals can create hazards, while a surface that looks clean is not necessarily microbiologically controlled.
“Clean” is also a social judgment. What must be removed, where it goes, and which standard is adequate depend on a space and public norms. Matter does not vanish; it is transferred, treated, or transformed. A tidy city can mean that waste has left a visible area and entered a landfill, treatment facility, or another community.
Research should therefore go beyond occupational praise. It has to follow the full path of residue and ask how risk is redistributed along it.
A system continues after the toilet
A toilet removes excreta from immediate experience and creates a sense of completion. The sanitation chain includes sewer or on-site containment, emptying, transport, pumping, treatment, and disposal. Failure at any point can return contamination to homes, water, and public places.
The 2019 joint report Health, Safety and Dignity of Sanitation Workers includes toilet cleaners and caretakers, pit and septic-tank emptiers, sewer and manhole workers, and people working at pumps and treatment sites. Its main concern is severe and underdocumented conditions in many low- and middle-income settings. It is not a description of every system, but it shows what a user-end access measure leaves out.
An infrastructure program that counts new toilets may not ensure that waste reaches a safe endpoint. A household gains a facility while workers downstream handle excreta directly under informal conditions. The public-health objective is then achieved partly at the cost of their health. Completeness cannot be established by the number of connections alone.
How stigma and physical hazard combine
Sanitation work can involve pathogens, sharps, toxic gas, oxygen-deficient spaces, collapse, and difficult rescue. Risk control should begin with elimination and engineering before administrative procedure and personal equipment. Giving someone protective clothing is not a complete safety system.
Social stigma amplifies hazard. In some places sanitation labour is entangled with caste, poverty, migration, or informal work, leaving people little bargaining and reporting power. A procedure has limited reality if refusing a task means losing the day’s income. Occupational illnesses may remain outside statistics when workers lack health surveillance.
Stigma also structures public imagination. People want residue gone while avoiding knowledge of its handling. That distance permits unsafe conditions to continue. A task classified as dirty does not make the worker appropriately exposed to dirt and danger. The work handles a collectively produced residue and therefore cannot be reduced to a private contract.
How contracting obscures responsibility
Cleaning and sanitation services are frequently contracted through several levels. Specialist companies may provide real equipment and managerial capacity, so contracting is not inherently harmful. Problems arise when price competition compresses time and training while the client inspects only a visible surface.
A contract paying by area or frequency may leave no time to report a leaking fixture or damaged surface. These design faults make cleaning harder but belong to another department. Workers encounter them most often while their scope instructs them simply to complete the task.
Layers of subcontracting also blur the employment relationship. Following an accident, the site controller, principal contractor, and direct employer can each emphasise a boundary. Work health and safety systems commonly impose coordination where duties are shared. Public procurement should likewise assess training, worker rights, and safety rather than treating the lowest price as efficiency.
Contract performance needs evidence about the process, not only the appearance of the result. Staffing hours, access to equipment, incident and exposure reports, turnover, missed areas and time allowed for hazardous procedures can reveal whether a price is sustainable. Clients should also verify that subcontractors actually receive the protections described in the head contract. Otherwise each tier can present compliant paperwork while risk accumulates at the last tier.
Whom does automation remove from danger?
Mechanical emptying, remote inspection, and cleaning machines can reduce direct human contact. They enter repetitive or hostile areas and increase monitoring frequency. These are important improvements, but technology does not maintain itself.
Machines require loading, cleaning, repair, and attention when they jam. Human intervention may occur at the least stable moment. An automated facility designed on the assumption that a worker can reach inside, without isolation and retrieval, has concentrated exceptional risk.
Access to technology can deepen inequality. A well-funded site mechanises while a low-income community still relies on informal work. A formal organisation buys machinery and contracts the inaccessible remainder. Evaluation should follow final exposure, not count only the hours nominally automated.
Design determines whether automation genuinely removes exposure. Isolation points, retrieval systems, sensors, manuals and spare parts have to match the conditions in which maintenance occurs. Workers who clean and repair the equipment should take part in procurement and trials. If they are consulted only after installation, the organisation discovers maintainability when the machine is already unavailable or unsafe to enter.
The resulting judgment need not oppose mechanisation. A machine can remove a serious hazard and improve dignity. The claim is narrower: risk reduction should be demonstrated across normal operation, jams, cleaning and end-of-life handling. Automation counted at the task boundary can miss the person who later enters the exceptional condition.
Knowledge held by cleaning staff
Continuous contact gives cleaners a longitudinal view of a place. They know when a leak began, which door no longer locks, and when a waste stream changes. Most facilities managers do not observe at the same frequency.
The knowledge is often trapped in informal messages. Contractors lack access to reporting systems, completion forms have no field for defects, and shift or language barriers intervene. What the organisation loses is not general opinion but early warning.
A reporting path need not make cleaners responsible for engineering diagnosis. It can record location, observation, and urgency for review by the relevant profession. Feedback matters: reports that enter a void soon stop. Participation should also occur in procurement and design, because cleanability is a property of the facility.
Why “hero” is insufficient recognition
Cleaning and sanitation are frequently called essential during emergencies. Public acknowledgment corrects some neglect but can translate structure into character. Willingness to take a risk does not make risk acceptable. Need for service does not create an unconditional obligation on particular people.
Material recognition includes fair pay, secure status, rest, suitable equipment, vaccination and health monitoring where appropriate, training, and freedom of association. Workers need routes to report danger and the ability to withdraw from an imminent serious hazard. The institution should arrange continuity rather than make safety subordinate to it.
Language should also respect difference. Workers may not wish to be represented as one group of victims or heroes. Many hold expertise and occupational pride. Dignity lies in participating as labour subjects in the conditions of work, not in receiving an uplifting identity from observers.
Participation has to carry consequences. Cleaning staff need access to defect reporting, feedback and the meetings in which cleanability or sanitation changes are decided. Where language, employment status or fear of retaliation makes ordinary channels unusable, representation and confidential reporting become part of the infrastructure. Praise without a usable voice preserves the same division between those who experience risk and those authorised to define it.
Can cleaning preserve an unequal order?
Cleaning can maintain harmful structures. Public-space cleanliness rules may be used against unhoused people or informal traders. Domestic order can depend on women’s unpaid labour. High consumption creates disposable waste and assigns its handling to low-paid work. Keeping a space usable may preserve an order about who may appear in it.
Effectiveness is not the only question. We need to ask why residue exists, who defines cleanliness, and where removal sends the consequence. Source reduction, repairable products, and safer sanitation design can be preferable to expanding the cleaning workforce.
Workers should not carry the cost of that transformation through unemployment. A transition needs training and fair arrangements. Blaming consumption and dismissing the people who have handled its residue repeats the allocation problem.
Provisional judgment: shared residue creates shared responsibility
Social life produces wear, waste, and excreta. The problem is not whether a world without residue can be created. It is how residue is recognised, reduced, processed, and allocated. Removal from sight is not completion.
Cleaning and sanitation maintain safe usability rather than purity. Systems should eliminate danger where possible, improve design and mechanisation, protect formal and informal workers, and connect field knowledge to facilities management. A contract cannot be the endpoint of responsibility.
Visibility should not become exposure. Society can understand labour conditions without turning workers into a moral spectacle. Procurement standards, incident investigation, aggregate evidence, and worker representation reveal structure while respecting individuals.
Responsibility also reaches the users and designers who generate residue. Segregating hazardous waste, reducing disposable material and providing facilities that can be isolated and cleaned safely may prevent harm before a worker encounters it. These choices distribute maintenance upstream. They are usually more durable than asking the final cleaner to compensate for every unsafe product and space through additional caution.
More local studies will revise the account. The joint global report identifies severe conditions but cannot substitute for local labour law and industry evidence. The general judgment remains: a community cannot enjoy hygiene while transferring the hazards created by its residue to those least able to refuse and keeping that transfer unseen.
Primary sources and further reading
- World Bank, WHO, ILO and WaterAid, Health, Safety and Dignity of Sanitation Workers: An Initial Assessment, 2019.
- World Health Organization, news release on sanitation worker conditions, 2019.
- International Labour Organization, Occupational Safety and Health Convention, 1981 (No. 155).
- Safe Work Australia, Consultation duties, as an Australian reference on worker knowledge and participation.
- Shannon Mattern, “Maintenance and Care”, 2018.
Series navigation: Who Maintains the World? series overview
Discover more from Geoffrey Chen
Subscribe to get the latest posts sent to your email.