Capacity for Irony

Whose capacity, built by whom

“Capacity building” is one of the sector's most-used phrases and one of its most loaded. In Peace Direct's 2021 global consultation Time to Decolonise Aid, participants singled it out as language that “suggests that local communities and organisations lack skills”, a framing that casts the visitor as the source of competence and the host as the deficit to be fixed.¹

The arithmetic underneath makes the irony sharper. Grand Bargain signatories committed in 2016 to channel at least 25% of humanitarian funding “as directly as possible” to local and national actors.² Years on, the share reaching them directly is still measured in low single digits: by independently verifiable data, under 2% in recent years.³ The capacity, the contacts and the funding mostly stay upstream; the local partner supplies the room, the relationships and the legitimacy.

Workshops that do transfer power tend to look different, because they pay local institutions to lead rather than to host. The cartoon's question is the one the budget line usually answers truthfully.

Sources: 1, 2, 3

Scaling the form, not the function

Scale is the sector's reflex ambition: more districts, more states, all states. But Matt Andrews, Lant Pritchett and Michael Woolcock warn that organisations routinely scale the form of success without its function, a pattern they call isomorphic mimicry, which means adopting the charts, manuals and “best practice” of effective programmes to look capable while the underlying capability stays thin.⁴

Stretching that thin layer too fast produces what they term premature load bearing, which means placing weight on a system before it can carry it, and which discredits the very reform and exhausts the people delivering it.⁵ Depth, the slow work of getting one place to actually work, isn't the timid option in this cartoon. It's the radical one, because it resists the incentive to grow the map faster than the competence.

Sources: 4, 5

The pilot that never lands

The development and global-health worlds run on pilots, and they have a name for what happens next: pilotitis, the chronic inability to move beyond the pilot stage. The term crystallised around 2012, when practitioners noticed that the great majority of mobile-health pilots never reached scale.⁶

The pathology is structural, not accidental. Pilots are sized to donor grant cycles, so they are designed to demonstrate rather than to last. Uganda's experience became the cautionary tale. Dozens of uncoordinated mHealth pilots ran at once, none interoperable or scalable, until the government called a moratorium on new ones.⁶ The reply in the cartoon, will it survive the funding cliff, is precisely the variable a pilot is built to ignore.

Sources: 6

Technology amplifies; it does not substitute

Kentaro Toyama spent years building and watching technology projects across the global South before arriving at what he calls the Law of Amplification: technology amplifies existing human and institutional capacity and intent, it does not substitute for them.⁷

Where there are motivated staff, community trust and a way to physically get there, an app can multiply their reach. Where those are missing, the same app amplifies the absence, and the hardware ends up, in Toyama's phrase, gathering dust in a storeroom.⁷ The last mile is rarely a software problem. It is a staffing, trust and logistics problem wearing a software costume.

Sources: 7

The register that measures diligence, not health

The management cliché has an older, sharper cousin, Goodhart's Law: when a measure becomes a target, it ceases to be a good measure, because people optimise the number rather than the thing it was meant to represent.⁸

On the frontline, the cost is counted in hours. A five-country study of primary-care facilities found health workers spend on the order of nine hours a month, sometimes up to fifteen, simply filling reporting forms, often re-entering the same figures across overlapping registers.⁹ India's ASHAs now maintain ten-plus paper registers alongside a growing stack of apps and dashboards, so a single visit can be recorded three times over.¹⁰

“Counted three times” isn't a joke about innumeracy. It describes parallel reporting systems that measure the worker's diligence far more reliably than they measure anyone's health.

Sources: 8, 9, 10

The voices the grid can't hold

Participation is the sector's most democratic-sounding word. In Participation: The New Tyranny?, Bill Cooke and Uma Kothari argued that the rhetoric can mask its opposite, that “the discourse itself, and not just the practice, embodies the potential for unjustified exercises of power,” with funders retaining the real decisions beneath the language of inclusion.¹¹

The logframe is where that filtering happens, out of sight. Objectives, indicators and activities are fixed in advance, in a grid; a voice that doesn't map onto a pre-agreed cell has nowhere to go in the document, however much it matters in the room. The reply asks whether the format can hear what it didn't already plan to ask. It does not oppose participation.

Sources: 11

The patriarchal cooking pot

In 1997 the Zambian feminist Sara Hlupekile Longwe gave the sector its most enduring image of how good intentions disappear: the patriarchal cooking pot. Development agencies, she wrote, take in policies for women's advancement and produce no output, because the commitments “evaporate” inside a bureaucracy whose own culture they threaten.¹²

“Mainstreaming” can be the mechanism of evaporation rather than its cure. If gender is declared everyone's responsibility without budget, authority or accountability attached, it becomes no one's, except the lone officer in the cartoon who is, in practice, the entire gender team. Mainstreamed, in that telling, means dispersed until invisible.

Sources: 12

Putting the first last

Robert Chambers built a career on a reversal he summed up as putting the first last. In Whose Reality Counts? he showed that rural people can analyse their own complex, diverse realities with precision: realities that frequently contradict the top-down version held by visiting professionals.¹³ The scarce resource is rarely local insight; it is outsiders willing to ask first and design second.

India already pays for that insight and then routes around it. More than a million ASHAs, women chosen by and embedded in their own communities, hold exactly the context-sensitive knowledge the flipchart is reaching for.¹⁴ The cartoon's line is not a complaint; it's a sequencing note. The listening was supposed to come before the workshop, not as its findings.

Sources: 13, 14

The humans the report forgets

Impact reports are full of “beneficiaries” and conspicuously short on the people who produce the impact. The WHO estimates that around six million women work in health worldwide for no pay or below-market pay, effectively subsidising health systems with their labour, and roughly a quarter of those unpaid women are in India, mostly in community health roles.¹⁵

Women contribute about 5% of global GDP through health work, some US$3 trillion a year, of which nearly half is unrecognised and unpaid.¹⁶ A human-centred report would have a line for them. The field worker's request is modest and damning at once. Before centring the humans, name the ones already holding the programme up.

Sources: 15, 16

Published, not shared

Dissemination is where knowledge is supposed to return to the people who supplied it. But India's 2011 Census recorded only about 10% of the population as able to speak English at all¹⁷, so a report launched in English, in a district-town hotel, has by construction excluded almost everyone whose lives it documents.

The medium selects the audience. A finding that never travels in the language, format or place of the community it studied has been published, not shared. It is a fitting last panel, because it asks the question the other nine circle around: when the sector talks about impact, who is actually in the room to hear it, and who was only ever in the data?

Sources: 17