On ordering vs not ordering

A nursing correspondent from Australia writes to ask for a linguistic comment about what he calls 'a medico-legal conundrum'. Evidently nurses are sometimes faced with what are called 'Orders' in patients' notes, one of which is NFR, Not For Resuscitation. He suggests that this is ambiguous, and that 'a more correct use of English if it is to be understood as 'an Order' ought to be, Do Not Resuscitate'. As part of an initiative to get the change implemented, he asks for 'a directive in this use of English'.

I suppose the first thing to say is that linguists don't give directives - at least, I don't. Their role is humbler. I can explore what the nature of the problem is, in the present case, and draw attention to the consequences of taking one path rather than another; but the question of how to proceed is a matter for the professionals in question, and especially so in a sensitive and tricky area such as medico-legal practice. When I was working in clinical linguistics, I would never 'tell' a speech therapist what to do. A linguistic analysis might clearly point to therapy in a particular direction, but it would be up to the therapist to decide how and when and where to implement the recommendations.

In linguistic terms, what we are dealing with in the case of NFR/DNR is an issue in what is called 'speech acts'. It is a routine feature of a language that an instruction to act can be stated directly or indirectly. I can say to someone 'Open the window' or 'It's very hot in here' or 'Isn't it stuffy in here?', with the same intention in my mind each time. The first is a clear-cut direct command to act, and it uses a grammatical construction (the imperative) which is designed to do just that. The others (there are many variants) have evolved to avoid the 'commanding' tone of the imperative, which is always there, even if 'softened' by such words as 'please'. Many people find these 'indirect speech acts', using statements and questions, a more comfortable way of suggesting that someone does something. In conversation, tone of voice, facial expression, and gesture can also help to alter the force of an utterance - but these features are of course not available in written instructions.

Indirect speech acts are by their nature ambiguous. Statements typically assert states of affairs, and do not command. Questions typically ask for information, and do not command. It is therefore always possible for someone who uses an indirect speech act to find the utterance misinterpreted. I remember seeing this happen in a classroom once. The teacher said to a seven-year-old: 'Mary, there's a piece of chalk on the floor by your desk.' And Mary looked at it and said 'Yes I can see it, Miss.' Mary hadn't learned about indirect speech acts, it seems. But the teacher immediately taught her, by switching to a direct command. 'Pick it up then, please, Mary'.

Adults have learned about indirect speech acts, but whether their force is acted upon depends on all kinds of factors. In the window-opening scenario, if I am the listener, I might cooperate straight away, and open the window. Or I might choose not to cooperate. Why? All sorts of reasons. I might not want the window opened. I might not like the speaker. I might not be in a position to do so. There might be consequences which I can see but the speaker cannot (eg I know that it will be noisy). Or I might just be being perverse.

Note that none of these factors are linguistic in character. They are to do with the nature of the world to which the language relates. So any resolution of the NFR/DNR issue requires an analysis of the factors which lie behind the alternatives. Why would anyone not want to issue a clear command, in such circumstances? Given the ethical dilemma presented here, is it to do with questions of ultimate responsibility? DNR is a command, which the person commanded has no alternative but to respect (to fail to do so would be 'disobeying orders'). NFR is a statement, which allows both an action and no action as legitimate responses. Deciding which course of action to follow involves, to a degree, reading the mind of the writer. If readers share the mind-set of the writer, then they will not find the instruction ambiguous, and they will treat it as a command to be obeyed. If they do not, they have a linguistic loophole which would allow them an alternative response, should they be so minded.

Analogies can help. If I saw a notice attached to a vehicle which says 'Do not drive over rough ground', the intention of the owner is clear. But if it says 'Not for driving on rough ground', the intention is less clear. An implication in the latter case is that it isn't impossible to drive the vehicle over rough ground - and I might think the owner was simply telling me 'I wouldn't do this, if I were you'. I might therefore feel justified in ignoring him, on the grounds that if he had definitely wanted to stop me he would have said so clearly. And if I did do it, and the vehicle broke, and he complained by saying 'I told you not to drive the vehicle over rough ground' my defence could be 'Actually, no you didn't!' Whether this is a legitimate defence, in a particular situation, is for lawyers, not linguists, to determine, for it all revolves around the question of whether the intention behind the notice is so clear that anyone disregarding it could be construed as being wilfully perverse.

The DNR text is unambiguous; the NFR one less so. I think my correspondent is right when he suggests that this is a policy matter which requires clarification 'to protect professionals in health care'. Ambiguity by its nature is always a potential danger. But resolving the ambiguity here is a matter for the medical profession to decide, not linguistics.