What transforms understanding into recognised knowledge?

Understanding, legitimacy and the politics of recognition.

I never doubted the value of expertise.

Throughout my career I have worked with economists, engineers, statisticians, agricultural scientists, health specialists, planners and policy makers. I have been one of them. I have spent decades conducting research, teaching, writing, advising governments and leading development programmes.

I believe in expertise. I believe in research. I believe in scholarship.

What I no longer believe is that expertise possesses a monopoly on understanding reality.

That distinction took many years to recognise.

There is a difference worth making.

One is a know-it-all.

The know-it-all arrives with answers. Every question already has an answer. Every problem already has a solution. Their knowledge is right. Whatever others know is incomplete, mistaken or irrelevant. Others are there to be taught.

This way of thinking extends far beyond communities. It appears in classrooms. In consultancy. In development practice. It appears whenever people speak of building capacity, as though those expected to learn arrive empty, waiting to be filled.

I have always resisted that language.

I prefer strengthening or enhancing capacity.

Before any training begins, I ask two questions.

What do participants already know?

What do they want to learn?

The first recognises that understanding already exists. The second recognises that learning begins from where people are, not from where experts imagine them to be.

The other is knowledgeable.

The knowledgeable also has expertise. They have studied. They have trained. They understand theory. They understand evidence. But they also recognise the limits of their own understanding. They know what they do not know. They recognise that people whose lives have followed different paths possess forms of understanding they themselves do not.

The know-it-all arrives with answers.

The knowledgeable arrives with questions.

Development has produced too many know-it-alls.

Understanding exists long before institutions recognise it.

Every society understands its world.

Farmers understand soils.

Fishers understand water.

Pastoralists understand grazing systems.

Midwives understand childbirth.

Parents understand children.

Artisans understand materials.

Communities understand seasons.

People observe.

They remember.

They compare.

They adapt.

They transmit what they learn.

Understanding begins there.

Not in universities. Not in ministries. Not in journals. Not in research institutes.

Understanding begins wherever human beings learn how reality works.

Long before societies build institutions, they build understanding.

Knowledge and recognised knowledge are not the same thing.

Understanding does not automatically become recognised knowledge.

Something happens in between.

Understanding becomes recognised knowledge through processes of organisation, testing, documentation, teaching and institutional recognition.

When it becomes part of a recognised body of knowledge.

Disciplines emerge.

Professions emerge.

Standards emerge.

Curricula emerge.

Research methods emerge.

Institutions emerge.

This process gives knowledge legitimacy.

It tells society: "This understanding can now be trusted."

Legitimisation is one of humanity's greatest achievements. Without it there would be no science. No medicine. No engineering. No mathematics. No accumulated learning across generations.

Knowledge does not become legitimate because it is true. It becomes legitimate because institutions recognise it.

False knowledge can become legitimate.

History repeatedly shows that societies have recognised ideas that later proved mistaken.

But legitimacy also creates boundaries. It determines not only what is recognised. It also determines what remains outside recognition.

The quiet hierarchy of knowledge.

This became increasingly visible during my years in development.

An engineer arrived carrying engineering knowledge.

An agricultural extension worker carried agricultural knowledge.

A medical officer carried medical knowledge.

A consultant carried technical knowledge.

A researcher carried academic knowledge.

Communities were expected to provide information. Experts were expected to provide knowledge.

The distinction was rarely questioned.

Degrees. Professional qualifications. Institutional affiliation. Published research. These became signals that knowledge had already been recognised before anyone spoke.

The notebook. The clipboard. The report. The PowerPoint presentation. Even the language people used. All quietly communicated authority.

Not because these people necessarily understood reality better.

Because the institutions behind them had already recognised their understanding as legitimate.

Institutions do not create knowledge. They decide which knowledge will count.

When recognised knowledge returns to learn

One example has stayed with me for years.

A farmer intercrops maize and beans.

The practice is older than the farmer.

Parents did it.

Grandparents did it.

One generation observed.

The next refined.

The next inherited.

The knowledge was never written down.

It was lived.

Tested.

Adapted.

Passed on.

Then an agricultural extension worker arrived.

Degree.

Diploma.

Professional training.

They inspected the field.

They judged intercropping inefficient.

Their advice was unequivocal.

Plant maize.

Or plant beans.

Not both.

The farmer hesitated.

The extension worker dismissed the practice as backward.

Resistant to change.

Another expert advised government to introduce community tractors so farmers could "open up" more land.

The tractors arrived.

They turned over the soil.

They buried the fertile topsoil.

They disturbed the organisms that sustained it.

They left the land exposed.

Rain carried the soil away.

Productivity declined.

The farmer watched.

The farmer understood what the experts did not.

Years later another extension worker arrived.

This time the language had changed.

Climate-smart agriculture.

Conservation farming.

Minimal tillage.

Cover crops.

Organic matter.

Intercropping.

Soil health.

The farmer listened.

Then quietly replied.

Twenty years ago you came here and discouraged me from intercropping. You urged me to adopt modern farming. You told me the old ways were backward. Now you have come back to teach me what you once asked me to abandon.

Nothing had changed on the farm.

The farmer's understanding had remained remarkably consistent.

The understanding never changed. Its legitimacy did.

The farmer had not been waiting for new knowledge.

The institutions had.

How knowledge becomes powerful

Knowledge does not influence the world simply because it exists.

It gains influence when it enters systems that recognise, organise and transmit it.

An understanding becomes a research question.

A research question becomes a study.

A study becomes a publication.

A publication becomes part of a discipline.

A discipline shapes professional training.

Professional training shapes practice.

Practice shapes institutions.

Institutions shape policy.

Policy shapes the realities that societies seek to change.

At each stage, decisions are made about what deserves attention, what deserves investment and what deserves to be carried forward.

Knowledge does not travel alone.

It travels through institutions that determine which understandings become authoritative.

Institutions do not create all knowledge. They create the pathways through which knowledge becomes recognised.

A farmer's knowledge of soils may sustain production for generations.

An agricultural science may transform that knowledge into a recognised body of knowledge.

The difference is not necessarily the existence of understanding.

The difference is the pathway through which that understanding gains legitimacy.

Recognition is therefore cumulative.

Each generation inherits not only knowledge.

It inherits the theories, disciplines, methods and standards through which previous generations decided what knowledge was worth preserving.

The question is not only what we know.

The question is how we came to recognise some knowledge as worth knowing.

Recognition does not merely shape what we know. It shapes the reality that knowledge is able to influence.

The consultant, the community and the notebook.

The same pattern appears throughout development.

The consultant enters the community carrying recognised knowledge. The community receives the consultant as the expert.

The consultant asks questions. Writes notes. Produces a report. The report becomes evidence.

Yet much of what appears in that report did not originate with the consultant. It originated with the people whose realities were being documented.

The consultant did not create the knowledge. The consultant translated it into a form institutions were prepared to recognise.

Sometimes that translation is invaluable. Sometimes it is transformative.

The consultant receives credit for knowledge that existed before the consultant arrived.

We should never mistake recognition for creation.

Artificial intelligence inherits recognised knowledge.

Artificial intelligence enters at the end of this chain.

Large language models do not learn from all human understanding. They learn primarily from understanding that has already been documented, published, archived and institutionalised.

They inherit recognised knowledge. Not all knowledge.

The distinction matters.

A language model cannot easily learn from knowledge that never enters books. Never enters journals. Never enters official reports. Never enters datasets.

The limitation is not technological. It is historical.

AI inherits centuries of decisions about what societies considered worth preserving.

The algorithm does not decide what knowledge matters. It inherits earlier decisions made by institutions.

Yesterday's hierarchy of knowledge becomes tomorrow's automated intelligence.

Legitimacy becomes code. 

The question beneath expertise.

I still believe in expertise. I still believe in disciplines. I still believe in research.

But expertise is strongest when it recognises that it enters conversations where understanding already exists.

The researcher does not arrive where knowledge begins. The researcher arrives where knowledge already lives.

Research contributes. It tests. Challenges. Refines. Connects. Explains. Sometimes it overturns. Sometimes it confirms.

But it does not possess an exclusive right to understanding reality.

The question that remains.

Understanding exists long before institutions recognise it.

Knowledge does not become legitimate because it is true. It becomes legitimate because institutions recognise it.

Institutions do not create knowledge. They decide which knowledge will count.

The question is not who has knowledge.

The question is what transforms understanding into recognised knowledge.

The answer is legitimacy.

Legitimacy is the process through which understanding becomes recognised knowledge. It operates through institutions. Disciplines. Journals. Universities. Governments. Professional credentials. Published research.

Legitimacy is not about truth. It is about recognition.

The question is whether development is willing to recognise what has always been there.

Because what is not recognised cannot inform policy. Cannot shape research. Cannot guide intervention. Cannot be learnt by AI.

Yesterday's hierarchy of knowledge becomes tomorrow's automated intelligence.

Legitimacy becomes code.

Next: The problem with problems including the Roots We Refuse to Uproot — why development invests in symptoms while leaving causes intact.