WHAT RESEARCH EXPECTS OF MANAGEMENT 263 to keep intimately in touch with all aspects, or even one individual field of activity. As a result, there seems to be a tendency to reject the possibility of ever being able sufficiently to understand a portion of a given field or fields to be capable of being imaginative and capable of making intelligent decisions in research matters. This certainly need not be the case, if man- agement develops powers of discrimination. It may and should demand of any professional worker, whether he be scientist, engineer, attorney, or accountant, an interpretative explanation of the application of his work to the problem at hand and it should be possible to obtain this interpretation in language intelligible to the uninitiated. If the executive is not successful in obtaining an answer in lay terms, his education Should stand him in sufficient good stead to resolve the situation to one of intelligent communi- cation. With a proper understanding of the specific work at hand, he must apply an unbiased, analytical viewpoint, in judging the relevance of any field of science to the problem or problems facing his organization. He must discriminate between relevance and irrelevance, since unwarranted and irrelevant research has been the downfall of many a research organiza- tion. Actually, from a practical standpoint, imagination and discrimination usually function simultaneously. As a concrete illustration, let us take the problem of predicting the mechanical behavior of textile yarns from a knowledge of the mechanical properties of textile fibers. For thousands of years, textiles have been made "by guess and by gosh." While tribute must be paid to the remarkable ingenuity and skills of textile craftsmen who have produced a breathtaking variety of useful and beautiful fabrics, craftsmanship has become inadequate in meeting the demands on textile use in our complex society. With the advent of two world wars, the tre- mendous growth in industry, a flood of new fibers, finishes, and processes, the need has become pressing for quantitative methods of predicting and engineering textile products. If textiles are to be engineered even as bridges and highways, then they must be first conceived as engineering materials, and subject to the same kind of mathematical treatment as metal, wood, or concrete. But it is well known that textiles are soft, flexible, and porous, and are made up of mil- lions of individual fibers. Who can imagine such structures behaving as a solid piece of wood or steel? A qualified scientist may state that because of the multitude of fibers in the yarn it is impossible to calculate how the stresses applied to yarns are precisely distributed among the fibers. Yet the most striking advances in textile physics and the prediction of yarn performance in recent years has come by imagining yarns to be ideal engi- neering structures, e.g., solid, round, homogeneous, without any variability along the length or over the entire cross-sectional area. Then by measuring fiber properties and calculating how a yarn should perform, the difference

264 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS between theoretical and actual performance has provided the key to dis- criminating among the many possible causes of deviations, and rationaliz- ing these into mathematical expression. Knowing from empirical experi- ence some of the properties of fibers and yarns, such as crimp and twist, the effect of these variables has been gradually determined. Third--Tolerant Research is the product of human beings, working with and among other human beings. For all the striking advances in science, we still have far, far to go before we can be assured in advance that in setting up any attack upon a problem, we have chosen proper methods for solution. Research progresses by small steps, frequently very slowly. The time of solution and the scope of the work, can rarely, if ever, be guaranteed. Many of the small steps, and much of the time consumed as a result of added wisdom, stem from experimental failure. There is a danger in these times to think of the research method and science as being ahumen and infallible, and thus, when an experiment fails, to question the competence of the scientist. This attitude may be most unfair, and management should not take such a stand. Rather, it should realize that research methods are fruitful as much because they profit t3om failure as because they solve a practical problem. The most intelligent, effective expenditure of research funds depends upon a tolerant under- standing of the limitations of what is known and what can be done. Those who have labored diligently in the field of research, realize the problems that arise when the results of theoretical determinations are first explored at a practical level. Seldom is success achieved on the first trial. More than likely, failure will be apparent on several trials. Such failures should not necessarily be looked upon by management to connote a lack of soundness of work done at the fundamental level, but rather point to the need of additional research at the applied level. Here, the attributes of tolerance are paramount to achieving success. Management must be will- ing to sift the wheat from the chaff. It must determine which lines of attack upon the newly uncovered applied problems will be most fruitful. It must if fi•ed be, stand ready to revise its accepted conclusions in the light of newly discovered facts that become evident in the process of application. Such action must be taken forcefully, but not critically, for the wise man is he who recognizes the limitations of his fellow man, as he should recog- nize his own. Fourth--Humble Science makes no policy--people make policy. Science may tell us how to do something. Science cannot tell us when or whether to do something. The plight of the world today and the shadow of the hydrogen bomb bear grim witness to these facts.