Why “may”, “will” and “must” are never used loosely

The smallest words are doing the most work

Reading a formal document, attention naturally goes to the nouns — the amounts, the dates, the named obligations. The words that actually determine what anybody has to do are much smaller and much duller. Whether a sentence says that a party may, will, shall or must do something changes the sentence completely while barely changing its appearance.

Drafters treat these words as a controlled vocabulary rather than as stylistic variation. Many organisations maintain a house rule setting out which verb expresses an obligation, which expresses a discretion and which expresses a simple statement of fact, precisely so that the choice never becomes a matter of taste.

May is a permission, and it belongs to someone

The most misread of them is may, which grants a discretion rather than describing a likelihood. When a document says a party may terminate, it is not estimating what is probable; it is stating that the party is permitted to, at their choice, without needing to justify it.

The question to ask of every may is whose discretion it is. A clause saying that the provider may vary the service and a clause saying that the customer may cancel are structurally identical and point in opposite directions. Reading a document for its discretions — listing who is permitted to do what, unilaterally — is one of the fastest ways to understand where the balance of control sits.

Shall, will and the drift towards must

Shall has traditionally expressed obligation in formal drafting, and it has become contentious because it is also used in ordinary English to indicate the future, and because it has been used inconsistently often enough to generate disputes about which sense was intended.

Plain-language drafting movements have pushed towards must for obligations and will for statements of what is going to happen, and many modern documents follow that convention. Older ones do not, and mixed documents exist where a section was updated and the rest was not. Where a document is clearly using shall for everything, the surrounding words rather than the verb usually carry the meaning, which is a slightly unsatisfying position and an accurate description of the state of things.

The negative forms diverge more than the positive ones

The differences widen under negation. Must not is a prohibition. Need not is an absence of obligation. May not is ambiguous in ordinary speech between prohibition and lack of permission, which is why careful drafting tends to avoid it in favour of an explicit formulation.

This matters when working out what is actually forbidden as opposed to merely not required. A clause saying that a party need not provide something is very different from one saying it must not, though the two can look similar at speed. Where a document has been drafted with care, the distinction is being made deliberately, and it’s worth slowing down for.

Conditions are what most obligations really are

Very few obligations in a document are unconditional. They arrive attached to a trigger: on receipt, within a stated period, if a specified event occurs, provided that something else has been done. The verb states the strength of the duty, and the surrounding condition states when it exists at all.

So reading a single clause in isolation regularly produces the wrong answer. A firm-sounding obligation that only bites once three conditions are satisfied is weaker in practice than a modest one that applies immediately. The trigger, the deadline and the verb together make up the actual commitment, and separating them out is more useful than reading for tone.

A quick pass that gets most of the value

Scanning a document for its obligation words and marking whose obligation each one is produces a surprisingly complete picture in a few minutes. Two columns — what I must do, what they must do — and a third for discretions on either side, will usually reveal the shape of the arrangement faster than reading it through from the beginning.

The gaps in those columns are often the most interesting part. An arrangement where one side holds most of the discretions and the other holds most of the obligations may be perfectly ordinary for that kind of transaction, or it may not be, and that is a judgement about the deal rather than about the drafting. Where the answer matters, it is a question for proper advice, but knowing which words to look at is what makes the question askable in the first place.