irst rule with the sick.[13]

For the same reasons, always tell a patient and tell him beforehand when
you are going out and when you will be back, whether it is for a day, an
hour, or ten minutes. You fancy perhaps that it is better for him if he
does not find out your going at all, better for him if you do not make
yourself "of too much importance" to him; or else you cannot bear to
give him the pain or the anxiety of the temporary separation.

No such thing. You _ought_ to go, we will suppose. Health or duty
requires it. Then say so to the patient openly. If you go without his
knowing it, and he finds it out, he never will feel secure again that
the things which depend upon you will be done when you are away, and in
nine cases out of ten he will be right. If you go out without telling
him when you will be back, he can take no measures nor precautions as to
the things which concern you both, or which you do for him.

[Sidenote: What is the cause of half the accidents which happen?]

If you look into the reports of trials or accidents, and especially of
suicides, or into the medical history of fatal cases, it is almost
incredible how often the whole thing turns upon something which has
happened because "he," or still oftener "she," "was not there." But it
is still more incredible how often, how almost always this is accepted
as a sufficient reason, a justification; why, the very fact of the thing
having happened is the proof of its not being a justification. The
person in charge was quite right not to be "_there_," he was called away
for quite sufficient reason, or he was away for a daily recurring and
unavoidable cause: yet no provision was made to supply his absence. The
fault was not in his "being away," but in there being no management to
supplement his "being away." When the sun is under a total eclipse or
during his nightly absence, we light candles. But it would seem as if it
did not occur to us that we must also supplement the person in charge of
sick or of children, whether under an occasional eclipse or during a
regular absence.

In institutions where many lives would be lost and the effect of such
want of management would be terrible and patent, there is less of it
than in the private house.[14]

But in both, let whoever is in charge keep this simple question in her
head (_not_, how can I always do this right thing myself, but) how can I
provide for this right thing to be always done?

Then, when anyt