668 THE AMERICAN JOURNAL OF SOCIOLOGY
but what is precarious and unstable. On the whole, the great majority of social facts which will be recorded in the monograph of the family are ascertained by way of questioning. So the art of questioning, of listening to responses, of inducing sincere and complete outpourings of the heart plays a great role in the success which the observer can look for.
7. Information gained from persons of a superior rank in the locality. One ought to get information only with an extreme reserve from the directing classes of the locality, who know often less than one supposes of the social organization of which they are a part. Their assertions, besides, should always be checked either with the aid of facts observed directly, or with the aid of facts revealed by the declarations of the family.
The family monograph, being, above all, the methodic descrip- tion of that which one has personally observed, could not bor- row anything from a book. An exception can be made only for certain facts authentically ascertained in particular, statistical information offering a character of absolute certainty. In every case, the origin of documents of this class ought always to be indicated in extenso.
8. The double budget and its commentaries. Every monograph has for its essential part a double Budget of Annual Receipts and Expenses, which is preceded by Preliminary Observations and fol- lowed by Notes which have been grouped under the general title : Divers Elements of the Social Constitution.
The preliminary observations will give an opportunity for the observer to gain precisely that confidence of which we have spoken above. They will prepare the workingman little by little to reply to the numerous questions of the budget, and even to understand the necessity of them. They will give to the observer a mass of preliminary notions upon the customs and the life of the family, upon the place which they inhabit, and upon the region of which they are a part.
It is not necessary to seek to complete immediately the pre- liminary observations, and to this end wearisomely to lead back the workingman whom one is questioning to a detail omitted in
