amounted to almost the total of his insurance. The particular case, then, was different from what it was first supposed to be. This man was not a person with money on hand. The case did not fit the rule. Again we may have the rule "first offenders are hopeful subjects for probation." In a given instance, however, the offender may have been a boy who had been pilfering undetected for years or a girl who had had sexual relations with boys and men since childhood, although only now brought to court. In the sense intended by the general rule neither of these is a first offender.

3. Mistaken Analogy. Inference is often drawn by what appears to be an analogy between some one case under consideration and another which, in those respects that are of importance for the purpose in hand, closely resembles it. For instance, if a worker, knowing of a number of cases of tuberculosis among hatters who had worked in the same factory with Ames,1 conceived the purpose of attempting to render more sanitary the processes of the hat trade, he would look for a probable analogy between the facts of these several cases and of men in other hat factories as to working conditions, incidence of the disease, etc.; whereas if his purpose were to get Ames himself well and self-supporting, he would look for a different set of analogies in making his diagnosis and would emphasize many things in which Ames and his family differed from the other cases of tuberculosis in hat factories.

Again, suppose that Mrs. X's explanation, that the beer seen going into her tenement was left there temporarily for a neighbor, proved to be a falsehood. A worker hearing the same explanation offered under apparently similar circumstances by Mrs. Y might reason by analogy that this was the same old excuse and that Mrs. Y like Mrs. X was a drinking woman. Whereas, if he inquired further he might find that Mrs. Y's neighbor, who was not a teetotaler, had been so kind to Mrs. Y during her children's sickness that Mrs. Y could not refuse her the small favor of taking in her kettle of beer. The analogy, therefore, would prove a mistaken one. The two cases would have had only a superficial similarity as to the points under consideration. This is the danger in reasoning by analogy, and is one into which we all of us fall. We know the frequency with which relief officers reason that since 1 See face card on p. 84.

salt cod and beans are acceptable diet for family A, they must be equally so for family B. The first family may be composed of a sturdy mother with big husky boys, while the second may be a tuberculous woman with tiny children. Both are widows with children, and beyond that the analogy does not hold.

Our tendency to assume more similarity than exists between the circumstances of an old case and of a new one, and hence apply our experience with the one to the other, often checks the acquisition of new knowledge. This tendency may be neutralized in the discussions of any representative case committee-one made up, let us say, of a few professional and business men, of several housewives, a neighborhood tradesman, a nurse, a trade unionist, and of some of the social workers in the various special agencies. Unless an opinion is imposed upon all by someone whom they respect as an authority, each one present will at once draw inferences from the outline of the case presented. A large majority of these inferences will have been shaped by analogy with the different experiences of these people of varying occupations and background. Where discussion is free, the diversities of view tend to offset each other and to bring out facts showing which analogy holds good and which is unwarranted.

4. Mistaken Causal Relation. One fact shown on the face card given on p. 84 is that Mrs. Ames' mother lives with her, although Mr. Ames is ill and the family in need of aid. Suppose one infers as the cause of this situation that the old lady has less pleasant relations with her other daughters and their husbands than with the Ameses. Would this cause operate by itself or would it be likely to act along with other causes to produce the given effect? The common inclination is to seek for one cause. Social workers, however, need to bear in mind that where cause must be sought in human motives, as is apt to be the case in their work, they must expect to find not that it is a single simple cause, but that it is complex and multiple. In the case of Mrs. Ames' mother, should it turn out that the cause after all was simple, the social worker can test first the adequacy of this alleged causal relation. Having accepted the mother's liking for Mrs. Ames as adequate cause, the worker must now ask whether the operation of this cause is thwarted by any circumstance. The mother may prefer

Mrs. Ames' companionship only when she has her to herself. When the husband is at home, jealousy or lack of sympathy with him may seriously mar her pleasure. In this case the preference for Mrs. Ames, although in itself an adequate cause, would be thwarted in its action. Whether this is the case or not the worker must apply the third test of causal relations and inquire whether there may be some other cause for the mother's living with the Ameses. The next alleged cause the worker must examine by the same method as this.

Again, we see on the face card that Ames worked at canvassing for only a month, when he returned to his old employer as doorman. What can we infer as a reason? Either he was not efficient at the canvassing or was not physically equal to it or he preferred a steady job with definite wage in a place he knew. Any one of these causes would have been adequate, none of them is thwarted by any other cause, but all three are equally possible. We can therefore make but a tentative inference.

The facts that the first work Ames sought after leaving the tuberculosis sanatorium was canvassing, and that this was not his usual occupation, suggest that he had received medical advice to get outdoor work. This inference answers all three tests. While this may not be the only conceivable reason for his doing this work, it is certainly the most probable one. In reasoning on people's behavior, one would seldom arrive at more than a strong probability of truth.

Mrs. F, a mother of growing girls, insists on keeping a man lodger and living in a wretched tenement over a saloon, in spite of generous offers from relatives of better living conditions. Shall we infer lax standards of living? This cause is adequate, it is thwarted by no other circumstance, but is it the only possible cause of her apparent stubbornness? She says herself that the objectionable tenement is near the children's school and that the landlord has been good to her. It is also conceivable that she is not eager to move near her relatives. Our inference, therefore, as to her standards is not justified except as a hypothesis to be confirmed by more evidence.

IV. THE RISKS ARISING FROM THE THINKER'S STATE OF MIND

1. Predispositions. Besides the risks involved in thinking as such, there are risks arising from the thinker's state of mind. The social worker, like all others, has certain personal and professional predispositions against which he needs to be on guard. What do we mean by predispositions? For our present purpose we may stretch the application of the word somewhat to include the sum of all those personal and professional habits of thinking and all those feelings and inclinations with which we approach each new problem. Our predispositions are both equipment and handicap. They are an equipment in that they are essential to individuality; they are a handicap in that they limit knowledge in one direction or another. For instance, when social case work agencies first became aware that their records showed a confusion between fact and opinion, they tried to meet the difficulty by instructing their workers to omit impressions, opinions, and inferences of their own, and to enter upon case records "nothing but the facts," reporting each of these colorlessly, "as it happened." Workers who attempted to follow this rule produced records which have been likened to unstrung beads; in the attempt to eliminate all prejudice, they eliminated the judgment and discernment which would have given to the whole investigation unity and significance.1

Predispositions may obscure for us the significance of one set of facts by leading us to exaggerate the importance of another set, while at the same time they are so much a part of us that we may easily be unaware of their existence and consequently of their danger. Such an effort to disregard the fact of our having predispositions as that shown by the case work agencies who wanted nothing but the facts" from their visitors is likely to hamper investigation by deceiving the mind into a belief in its own impartiality. It is the worker's very awareness of his special predisposition on which depends the reliability of his observation and judgment. Once he brings a prejudice into the light of day, he can offset its influence on his thinking.

1 A critic of these pages, who has examined many case records of late in many different parts of the United States, adds: "Not only 'unstrung beads' but all about the same size. Wherever the 'only the facts' rule applies, the tendency is for every fact, big and little, to occupy about the same space in the record. Everything is brought to a dead level."

2. Assumptions. As professional workers become experienced they are apt to gather a set of salted-down rules, or more correctly assumptions, which they thereafter apply as needing no further proof. Some of these assumptions are warranted, others are not. In dealing with cripples, for example, social workers may adopt unhesitatingly the assumption that physical handicap is the cause of unemployment. Closer acquaintance with cripples would show that many do well in employment, and that of those who do not, some fail through temperament, some through the limitations of their earlier industrial experience, and some through one or another physical cause unrelated to the handicap in question. In fact, the common assumption that the cripple's industrial difficulties are insurmountable is his most serious handicap.

Again, there is an assumption among not a few workers that laziness is within a man's control, whereas, while we may not be able to say it is never so, we know that it often is not.

A few years ago it was assumed among social workers in some communities that a girl with a second illegitimate child was hopelessly degraded and that therefore no private agency should attempt to cope with the problem; public authorities should give the necessary care. Indeed this feeling was so strong that private rescue homes receiving these girls were thereby somewhat discredited. Today courageous intensive endeavor has shown that some of the most successful work of reconstruction can be done with the unmarried mother of two children.

The reason assumptions such as these persist is that they are not examined; they are taken on faith. If, however, a worker knows his assumption to be what it is-unproven-he may venture to act upon it. He may know, for instance, that a given man's laziness may conceivably be a manifestation of disease, yet, barring any other indication of illness, he may think best to assume the fellow's culpability and push him to get work.

In any profession one acquires certain predispositions, or habits of thought, from the conditions of one's allotted task and particular occupational environment. A worker in a child-helping or a family rehabilitation agency in a large American city, working among recently arrived foreigners who have become dependent, may regard immigration as a menace, unless his own racial prejudices