satisfactory for subjective matters, such as personal traits, in which the motives and capacity of the witness, the atmosphere and spirit of his statement, are all important.

4. Types of Evidence. There are discriminations to be made not only in the types of source but in the types of evidence which sources supply. Real evidence, it will be remembered, is the very fact at issue presented to our senses. Testimonial evidence is the assertions of human beings, to which must always be applied tests of the competence and bias of the witnesses. Circumstantial evidence is any indirect evidence whatsoever which tends to establish the point at issue. The tests applied to it have to vary, because the subject matter is infinitely varied. Any fact in the material universe or in the mind of man may become the basis from which some other fact is inferred.

There is an important distinction in testimonial evidence between the assertions of those who say that they saw or heard the supposed facts themselves and the assertions of those who have the facts only from what others have told them. This latter is hearsay evidence, and in all our interviewing we should discover the extent to which the assertions of the interviewed are founded on observations or on mere rumor.

5. Characteristics of Witnesses. The first, unrehearsed statement of a witness is often the most trustworthy. This first statement can be made less reliable, however, by a careless use of "leading questions," which are a danger not only in the first interview but everywhere. "Every one of us," says Gross, "has made the frightful observation that by the end of an examination the witness has simply taken the point of view of the examiner, and the worst thing about this is that the witness still thinks that he is thinking in his own way."

A witness may be quite sincere also in thinking that he knows more about an event or a person than he really does. His good faith, therefore, is not the only thing to establish. His competence includes both his opportunity to know the facts and the way in which he has used his opportunity. This latter is conditioned by his powers of attention, memory, and suggestibility. What ideas had he in stock, moreover, which would have made him a 1 Criminal Psychology, p. 163.

good or a bad observer? What reason had he for observing carefully?

Apart from his competence as a witness, what risk is there of bias in his testimony? Bias may be racial, national, religious, political, environmental, or some element of self-interest may enter in. Important forms of environmental bias are those of a man's occupation and of the particular habit group to which he belongs.

Obviously it is not enough to add statement to statement, as a phonograph would. The processes of inference, of comparison of material, begin with the first interview and continue through all the steps leading to diagnosis.

II. THE COMPARISON OF MATERIAL

"I am astonished," says Dubois, "to see how many young physicians possessing all the working machinery of diagnosis do not know how to make a diagnosis. It is because the art of diagnosis does not consist merely in gathering together a great many facts, but in co-ordinating those that one has been able to collect, in order to reach a clear conception of the situation." And we are told that the historian first collects his material, then collates it, and only after it has been collated attempts his final interpretation. He weighs his evidence, of course, as we do, item by item when it is gathered, but a reweighing of the total is necessary when all the items are in. "After a student has learned to open his eyes and see," writes Dr. Richard Cabot of clinical teaching, "he must learn to shut them and think." So must we. Nevertheless, this stage of assembling our material, of relating its parts and trying to bring it up into consciousness as a whole, will not be easy to illustrate, since it is the most neglected part of case work technique. Speaking broadly, the social case worker of an earlier day did little visiting of anyone except his client and so observed only within those narrow limits. He was mentally sluggish, moreover, and guilty of much thoughtless prescribing. The case worker of today is more active physically-sometimes doing too much running around, one is tempted to believe-but his advance in useful1 The Psychic Treatment of Nervous Disorders, p. 277. 2 Case Teaching in Medicine, Introduction, p. vii.

ness over earlier workers would be greater if he would oftener "shut his eyes and think," if he would reduce the visible signs of his activity and assemble his forces in order the better to deliberate upon his next move before he makes it. Case records often show a well made investigation and a plan formulated and carried out, but with no discoverable connection between them. Instead, at the right moment, of shutting his eyes and thinking, the worker seems to have shut his eyes and jumped. On the other hand, however carefully the inquiries are recorded and the diagnosis which grew out of them indicated, however carefully a plan of action is decided upon, etc., the processes by which the diagnosis is arrived at-what parts of the evidence have been accepted or rejected and why, what inferences have been drawn from these accepted items and how they have been tested-can none of them be revealed in a record.

Some case workers feel that their conscious assembling of material comes when they present a summary to the case committees of volunteers who assist them in making the diagnosis and the plan of treatment. This is especially true if any of the members of the committee have a social experience that has made them both critical and just in their valuing of testimony. One worker writes, "Repeatedly, flaws in my investigation have not occurred to me in reading over the record, but they have become only too evident at the moment of presenting the case to my committee. The standard in my mind of what the committee ought to know in order to make a fair decision has then suddenly revealed weaknesses to me before they were brought out in the discussion."

The same bracing influence comes from submitting findings at this stage to a case supervisor who is responsible for the work of a group of social case workers. Indeed, the process of comparison, in so far as it can be studied at all at present, is found at its best in the daily work of a few experienced supervisors. Unfortunately they are usually persons who are much overburdened. Although committees, at their worst, can be useless as critics, when well chosen they have an advantage over any one referee in that they bring not only less jaded minds but more varied experiences to bear upon each problem. Either supervisors or committees have the advantage over the worker who makes his analysis unaided,

that they do not know the client or his story, and that consequently they are not already so impressed with any one part of the story as to be unable to grasp the client's history as a whole.1

1. Suggestions for Self-supervision. In the absence of a competent supervisor or of a committee, the case worker will often have to take the place of both by consciously setting aside some time in which he will strive to look at his own work as if he were a critical outsider.2

(a) He can try to review each item of a case with all the others in mind. When each particular piece of evidence came to him, he judged it by what he then knew. How does he judge it now in the light of all the evidence?

Gross suggests another way of testing our material which is psychologically more difficult; namely, to consider a part of it with other material deliberately excluded.3

This is what a probation officer had to do, probably, when a father lodged complaint against his boy for stubbornness and for thieving from his older brothers. The home seemed so satisfactory that she was inclined to seek the cause of the trouble in outside influences that would have led the lad to take first small sums and then much larger ones. When, however, the time came for planning, the explana

1 A case reader of wide experience suggests here that, in fields of work where no committee is possible and no supervisor is at hand, someone with a keen mind be introduced to case record reading and that current problems be "tried out on him." Even where there is a committee it is important that someone on the committee besides the case worker read the record before the case comes up for discussion.

2 Any detailed discussion of the worker's case records must be reserved for a separate book on that subject, though self-supervision might well include not only the case work but its recording. Charles Kingsley warned a young writer that he should never refer to anything as a "tree" if he could call it a "spruce" or a "pine." If that lesson had been impressed upon the present generation of case recorders, the task of writing this book would have been an easier one.

Among the general terms against which collectors of family histories for eugenic study are warned by the Cold Spring Harbor Eugenics Record Office (see Eugenics Record Office Bulletin No. 7, p. 91) are abscess, without cause or location; accident, decline, without naming the disease; cancer, without specifying organ first affected; congestion, without naming organ affected; convulsions, without details and period of life; fever; heart trouble and heart failure; insanity, without details; kidney trouble; lung trouble; marasmus; stomach trouble.

The social case worker's Index Expurgatorius would have to cover a much wider range of subjects; but some of the commonest substitutions are relative for the word expressing the exact degree of relationship; Italian or Austrian or German for the term descriptive of a native of the particular province or other political subdivision; day laborer or salesman or clerk for the particular occupation; and bad, dull, unsanitary, shiftless, incompetent, unsatisfactory, good, bright, industrious, proud, refined, and a host of such adjectives for the specific act or condition.

3 Criminal Psychology, p. 12.

tion had not been found, and, having a mind that demanded specific data instead of falling back upon an unsupported theory, she began her search anew and excluding from her mind for the time being the favorable family appearances, found two court records of the arrest of the father, one for buying junk from minors and the other for peddling without a license. These may seem small offenses, but they were serious enough in the father of a boy who was also developing a tendency to lawlessness.

(b) Sometimes, as Gross suggests also, the grounding of a fact has been so difficult, has taken so much time, that we slur over the task of establishing its logical connection with the whole, or do that part of our work "swiftly-and wrongly." Or sometimes the slurring is due to the desire to make a definite report with promptness, as in the following case:

A charity organization society was asked in August by the state's attorney to interest itself in a non-support case, in which the man of the family had been ar rested for not making weekly payments to his wife on the separate support order of the court. A week later the society submitted a report of its inquiry upholding the wife. In October, however, when the man made application to have his children removed from the home, an exhaustive study of the case revealed bad conditions there. A critic of this case record writes: "Before your first report to the state's attorney was sent, contradictions in the evidence had developed that should have made it clear to you that further investigation was needed. The sources of information were at hand and the winter's rush was not upon you."

(c) As was the case in the foregoing example, a review of our material will often reveal unsuspected contradictions in the evidence. Where these contradictions cannot be reconciled we may safely infer that further evidence is needed; where, though all the evidence points one way, no explanation of the difficulty or guide to its solution has been revealed, we must again look for additional facts.

(d) The rhetorics tell us that the first and last paragraphs of an essay are the two that make the deepest impression upon the reader. It may be well to ask always, therefore, whether the story as told by the first person seen, or the first theory adopted by the worker has received undue consideration in shaping the final conclusion; or whether the last statement made has been allowed this advantage. Anyone who has had occasion to note the eagerness of each of two complainants to tell his grievance first will appreciate that 1 Criminal Psychology, p. 143.