e included and what
is to be excluded from the survey. If, for instance, we are making a
survey of the acoustic properties of church buildings in England, it is
not scientific to introduce questions as to the character of the gospel
preached in them. A scientific survey is not necessarily a collection of
all possible information about any people or country; that is an
encyclopaedia; a scientific survey is a survey of those facts only which
throw light on the business in hand. A scientific survey of foreign
missions ought not then necessarily to look at the work carried on from
"every point of view". The point of view must be defined, the end to be
served defined, and then only those factors which throw light upon that
end have any place in a scientific survey. We cannot be too clear about
this, because in survey of a work so vast and so many sided as foreign
missions we might easily include every human activity, unless we defined
beforehand the end to be served and selected carefully only the
appropriate factors. Carefully defined, missionary survey is not the
unwieldy, amorphous thing which people often imagine. There is indeed a
dangerous type of survey which starting with a hypothesis proceeds to
prove it by collecting any facts which seem to support it to the neglect
of all other facts which might disprove it. The procedure advocated here
is the adoption of a definite and acknowledged purpose for which the
survey is to be made and the collection of all the facts which bear upon
the subject in hand. The facts are selected, but they are selected not
by the prejudices or partiality of the surveyor, but by their own innate
and inherent relationship to the subject.

A scientific survey can only be a collection of facts; but inferences
will certainly be drawn from the facts which will direct the policy of
those who administer foreign missionary societies. The drawing of these
inferences from the material collected must be carefully distinguished
from the collection of the material (i.e. the making of the survey). The
latter precedes the former and is independent of it. Inferences hastily
drawn, or prematurely adopted, would only tend to discredit missionary
survey as a means to the attainment of truth. The adoption of a
hypothesis and the making of a survey in order to prove it by a careful
selection and manipulation of facts would not discredit survey as a
means to the attainment of truth; it would only discredit and deba