What a Knowledge Channel is made of

Between 2015 and 2026, allied countries ran a machine for moving military knowledge into Ukraine. Most of the attention went to what was taught. Less went to the machinery around it, which is what decides whether anything arrives.

That machinery had five components. Each had to exist for the others to be worth anything.

A place

Sites, ranges, classrooms, accommodation: funded, staffed, available on a schedule. Knowledge does not transfer in the abstract. It transfers in rooms. Rooms have to be built, heated, reached.

A programme

What would be taught, in what order, over how many days, written down, approved. The question what exactly are we transferring had an answer. The answer was a document.

People whose job this was

Instructors, linguists, coordinators, none of them doing it on top of their real work. Publicly reported figures put more than 63,000 Ukrainian personnel through the British-led programmes over four years, including  11,000 instructors, with more than a dozen partner nations contributing staff. Numbers on that scale are only reachable when the roles are funded and filled.

A shared vocabulary

Terminology databases and glossaries, maintained as working reference material rather than as dictionaries. This is the component I worked on. It turned out to matter for reasons nobody intended.

A check at the end

Assessment and reporting at the close of each cycle. Whatever it measured, somebody at least asked whether the thing had worked.

A place, a programme, roles, a language, a check. All of it expensive. All of it built deliberately over years. Knowledge does not move between institutions on its own.

The glossary is a map of where two systems diverge

A glossary exists so that a rotating population of linguists renders the same concept the same way in week one and in week forty. Most entries are unremarkable. The interesting ones are those that refuse to close: Ukrainian terms with no allied equivalent, where the reason was never linguistic. A missing word meant the underlying thing did not exist on the other side in the same form. Those entries took longest. The honest answer was not a translation but a paragraph. A paragraph does not fit in a glossary.

So the glossaries accumulated something by accident: a record of exactly where the two systems diverge. Not where language is difficult, where institutions genuinely differ. Nobody has treated that record as an object of interest in its own right. Nothing comparable has been compiled in the other direction.

The direction has changed

Europe now states, at senior level, that it intends to learn from Ukraine's experience.

So: where is the place? There are conferences and study visits. A conference is not a site of transfer but of narration. The two behave differently. 

In narration the speaker decides what matters, the audience cannot test what it is told, nobody finds out what they failed to ask. Transfer is slow and repetitive. Its defining feature is that the receiving side gets to be wrong out loud and be corrected. A training cycle provides that. A panel cannot.

What is the programme? No written statement of what the receiving side needs to close appears to exist. Where there is no agenda of one's own, the agenda belongs to whoever is speaking.

Whose job is it? On the Ukrainian side, those with the most relevant experience have the least spare time. On the receiving side, a full-time role converting Ukrainian operational experience into national standards is not something I have encountered. It may exist. It is not visible from the ground.

What is the check? Nobody asks, six months after a visit, what changed as a result.

The objection is fair: the directions are not symmetrical. Basic training is stable. It can be written down. Operational experience from an active war changes monthly, sits with thousands of people rather than centrally, is partly tacit, partly sensitive. All true. None of it means the transfer should be left to chance. It means the reverse channel has to be designed differently from the first one.

The part that already exists

The first channel spent years accumulating people who could operate in both systems at once. Not bilingual people, bicultural ones, who could say why a refusal was heard as contempt or a proposal as insubordination. They held two sets of assumptions simultaneously. The most effective linguists were rarely those with the best English. They were the ones who had spent years inside an allied country and come back.

That population now exists at scale, on both sides. When a rotation ends, the understanding leaves with the individual. The next rotation starts from zero. What it needs is somebody to collect it while it is still assembled, and that is the cheapest item on this list.

What remains open

Who would own such a channel is not obvious: a ministry, an alliance body, a coalition secretariat, something that does not exist yet. Nor is how it would be funded. Nor is how it would handle what cannot be discussed openly.

What is not in doubt is that it can be built, and it has been built once already. The first channel took years, more that a dozen of  partner nations and considerable expense, and it started from scratch. 

The return journey starts with the people who have already worked on both sides. That is a shorter distance than the one covered the first time.

Victoria Zolota worked on United States Department of Defense training exercises in 2018 and 2019. From 2023 to 2025 she was a military linguist within UK Ministry of Defence training programmes for Ukrainian forces. She subsequently worked in humanitarian mine action in Ukraine.