The Exercise Is the Message

A military exercise has two audiences and one plan. The plan addresses the first audience: units that need to practise something difficult with people they would have to do it alongside. The second audience is everyone watching, and nothing in the plan is written for them.

This is not an oversight so much as a division of labour. The authorities that govern training are about readiness, and readiness is a measurable thing: proficiencies, certifications, hours. The signalling value is real, unbudgeted, and nobody's assigned responsibility.

The result is a recurring mismatch. An exercise is scheduled years out, for reasons of availability and maintenance cycles, and then arrives in a week when its meaning has changed completely.

A rotation planned around a shipyard schedule will still be read as a decision about a crisis it was scheduled long before.

Cancelling carries its own message, usually a louder one. A planned activity that does not happen is legible to everyone as a retreat, which means the choice is rarely between signalling and not signalling. It is between two signals, neither of which was designed.

Combined exercises complicate this further, because the signal is jointly authored. A partner reads participation as a commitment, an adversary reads it as a coalition, and the administrative arrangements governing it say nothing about either, because they are about liability, funding and command relationships.

The domestic layer matters too and is often the least examined. Exercises are funded through the ordinary budget process, which means their scale is set by appropriations written for readiness metrics rather than by any assessment of what the activity communicates.

So the volume of the signal is a budget artefact. A larger exercise says more, and it is larger because of a line item, not because anyone decided that more needed saying.

The useful reform is procedural and cheap: require that the planning documents for a major exercise contain a short section stating what the activity is expected to communicate, to whom, and what would count as it being misread.

That does not subordinate training to messaging, which would be its own failure. It forces the two into contact, which at present happens only after the fact and usually in response to a question from a journalist.

The alternative is what exists now. A schedule built for maintenance windows produces a sequence of statements about intent, made in a language the planners are not writing in, to audiences they are not required to consider, with consequences that arrive as surprises.

Security cooperation is one of the few instruments that is cheap, repeatable and reversible. It is worth knowing what it is saying.