Contrary Imaginations

A Psychological Study of the English Schoolboy

Liam Hudson

• Publisher: Routledge
• Year: 2017
• ISBN:
  9781351621052
• Edition: 1
• Topic: Psychology
• Subtopic: Education in Psychology
• Index: Psychology

First published in 1966, this title describes two types of clever schoolboy, the 'converger' and the 'diverger'. The intellectual and personal differences between these two types are examined in detail. This description is used as the foundation for a more general discussion of the motives which lead men and women into the Arts or the Sciences, and of the qualities which enable some to think productively while others do not. Dr Hudson's work is remarkable not only for the fresh light he throws on the relation of intelligence to personality, but also for his method. His research combines the skills of intelligence testing and psychoanalysis in a way which had not previously been attempted.

---

• 1 •

Presuppositions

---

The aim of this book is to delineate two types of clever schoolboy: the converger and the diverger. The earlier chapters offer a fairly detailed description of the intellectual abilities, attitudes and personalities of a few hundred such boys. In the later chapters, this description is then used as the basis for a more speculative discussion – of the nature of intelligence and originality and of the ways in which intellectual and personal qualities interact. Although the first half of the book rests heavily on the results of psychological tests, and the last two chapters involve psychoanalytic theory, I have done my best to be intelligible, and, wherever possible, interesting to everyone interested in clever schoolboys: parents, schoolteachers, dons, psychologists, administrators, clever schoolboys.

There follows a little stage-setting. Psychology is a subject in which the audience should have a knowledge of the experimenter’s presuppositions before interpreting his results. In this first chapter I propose therefore to declare my personal prejudices about what is good psychology and what is not. These are few and simple. The first is the conviction that much of educational psychology is trivial; the second involves a distrust of complex statistics; the third, a rejection of psychological theorizing which is unduly rigorous or precise. I hope that it will be evident from the tone of these remarks that my aim is not to lampoon my fellow psychologists, but to point to certain of the difficulties with which, as psychologists, we are all surrounded.

TRIVIAL MEASUREMENTS

First, the question of triviality. In much of what has passed for educational psychology during the last fifty years we find experimenters absorbed, not in the problems of comprehending human nature, but of measurement. Although measurement is obviously indispensable to any science, it seems that much research becomes trivial through pursuing the problems of measurement for their own sake. These problems are often of absorbing interest, but exert an attraction unrelated to the search for insight into real men and women – and in educational research, this divorce from reality is sometimes extreme. One has visions of (and perhaps one has actually seen) ‘An Investigation into Job Satisfaction Ratings among Physical Training Instructors: Part One, Methodology’ and ‘The Blotsky Art Appreciation Inventory: A Study of Test-Retest Reliability.’

I wish to argue that although psychologists – and mental testers especially – are known for the subtlety and variety of their statistical techniques, these are often inappropriate. At present psychology is an exploratory science, and as a consequence most of our statistical needs are simple. If – in the course of our research – we find ourselves teasing out a result with the statistical scalpel, working out our correlations to three places of decimals, this is surely a sign either of a poorly designed experiment, or of a result too trifling to pursue.

As psychologists we may point out in self-defence that not all our work is so slight. We embrace themes as elevated as any in science: the Nature of Intelligence, and Man’s Relation to his Work, to name but two. But this draws a second salvo: the objection that we select the loftiest themes, yet treat them bathetically. We investigate Man’s Relation to his Work with a paper-and-pencil quiz. Anyone, I sense, not hopelessly biased, must admit that this criticism has something in it. We are responsible, more or less directly, for some of the oddest effects that the urge for rationality has yet produced.¹ But intelligence is no more the ability to do quizzes, than successful marriage is the ability to give right answers to a marital adjustment inventory. There is a gap, a gulf, between tests and the more interesting aspects of human life, and it is the psychologist’s job to span it. Tests are invaluable as written evidence of intelligence (or stupidity), just as examinations, reports, articles, poems or books are. The trouble, at present, is not that psychologists use tests, but that those which we do use are in an exceedingly primitive condition.¹

COMPLEX STATISTICS

My second prejudice is an extension of the first: a distrust of the purely statistical approach to psychological problems. This policy forms a powerful tradition within psychology, and is a defining characteristic of the mental testing movement. Although my own research derives to a large extent from this tradition, it is one which seems in certain respects to have led us astray.

The widespread use of mental tests, especially intelligence tests, dates back to the early years of the present century, and to the efforts of such men as Binet, Terman and Thorndike. One of the most interesting features of the research stemming from their brilliant, pioneering work is its conservatism. After an initial period of innovation and experiment, progress slackened. Instead of developing more subtle tests, psychologists concentrated on the analysis of results culled from the tests they already possessed. The result was a fine flowering of factor analytic studies on the ‘structure’ of the intellect: sophisticated statistics poised on testing techniques of rustic simplicity. One result of this concentration on analysis has been a remarkable stagnation in the technical sphere. The intelligence tests that we use today differ little from those employed during the First World War. By 1920, the intelligence test was already established as a cheap and useful technique for sorting the bright from the dull; yet now, nearly half a century later, the position is much the same. What is more, we are little the wiser about such tests’ predictive value, despite the growth of mental testing during this period from a brave outpost on the fringe of psychology to an empire of staggering dimensions.

One wonders why progress should have been so slow; and it seems that a quality of isolation, or incapsulation, is the crucial one. In a variety of ways, mental testers have sealed themselves off from the human subject-matter which would have ensured them, if not perpetual youth and perplexity, at least a livelier middle age. And in this process of incapsulation, statistics have played an insidious part.

Looking back with the priggish sense of superiority that hind-sight affords, one may see mental testing as an orthodoxy which has inherited both the great strengths and the relative weaknesses of the men who founded it. Many were men of outstanding statistical gifts; but the majority were not equally interested in the observation of individual human beings.¹ Intelligibly, the discipline which formed itself around such formidable men as Spearman and Thurstone was markedly statistical in bias. Indeed, it has been suggested that the contribution of mental testing has primarily been to the theory of statistics itself, rather than to psychology or to education.

From the psychological point of view, the statistical bias of interest amongst mental testers has not been an unalloyed advantage. In the first place, testers have frequently worked with excessive impersonality. They have administered ‘batteries’ of tests to their ‘subjects’ (revealing idioms, perhaps), viewing personal contact as a source less of insight than of ‘bias’. They seem happiest when people are at a safe distance. Many testers seem, too, to have cut themselves off from other aspects of their culture: they display no special knowledge of books, painting, music, science, politics, administration, or any other aspect of intelligent life in the world at large. And the literature generated by such workers has been less stimulating than one might have hoped.² Man’s capacity for intelligent thought is one of Nature’s most astonishing phenomena – and we might reasonably expect that a science devoted to its study should prove commensurately fascinating. Yet we find in mental testing less a thriving area of exploratory science, more a conservative branch of educational technology. And it is so, I would argue, precisely because testers have taken so little account, either of individual people, or of their intellectual accomplishments outside a technical cocoon of tests and examinations.³

Thus, the statistical approach to intelligence not merely makes for dull reading. It also suffers certain characteristic disadvantages of method. In bulk, the mental testing literature is concerned with establishing statistical connections between one variable and another; and this is frequently achieved with technical élan. However, relatively little public attention is paid either to exceptions to the general rule, or to the frightening margins of error that small and medium-sized correlations entail. Many testers are neglectful, too, of the human situations from which their data derive. All but the best seem sometimes to forget that the subtlest analysis in the world cannot make sense of poor primary data: naïve or inappropriate tests, and children who are bored.

On this analysis, the central failing amongst mental testers has been the neglect of inconvenient evidence. Testers have committed a form of scientific solipsism, building a system so tight-knit that all forms of rude human exception are excluded.¹ And as a result of their remarkably long reign, the statisticians have left psychologists a little dazed about the proper nature of their work. Under their suzerainty, the study of intelligence has become the study of tests. This is surely wrong. For once, the plain man’s view of the psychologist’s task seems approximately the right one. Intelligence does consist in thinking efficiently, in doing jobs well. High intelligence means the ability to run an industry, do scientific research, teach, write novels, plan cities, practise politics, law or medicine, or what have you. And, in so far as he is concerned with intelligence, the psychologist’s primary function is to discover what these jobs involve, and to explain why some people are good at them and others are not.

THE LURE OF THEORY

The third of my prejudices concerns theories. In psychology, these come in a bewildering variety of forms. Some are qualitative hunches, others use painstaking statistical analysis. Some concern causes; others merely seek to summarize and describe. We have the factor analytic (various styles), the psychoanalytic (reductive, psychodynamic, and all shades between), the behaviourist, the gestalt, the developmental, the instinctive, the social, the cybernetic, and many more besides.¹

THE STRATEGIC OBJECTION. I now wish to offer three grounds for the belief that, as matters stand at the moment in psychology, elaborate theory-building is out of place. The first is strategic: that precisely formulated theories give rise to long drawn out battles, for and against, which are finally unproductive.

Research designed with respect to theory is also likely to be wasteful. That a theory generates research does not prove its value unless the research is valuable. Much useless experimentation results from theories, and much energy and skill are absorbed by them. Most theories are eventually overthrown, and the greater part of the associated research is discarded.²

Theories (Hull’s theory of learning, to take an example) make claims which other workers immediately try to test. This process of testing seems, on the face of it, relatively straightforward. A rat is claimed to learn when his primary drives are satisfied: satisfy his primary drives, and see whether he learns or not. However, the design of an experiment which will test such predictions is a subtle affair, and the literature concerned with Hull’s theory is a large one.³ What usually happens in the end is that the dispute peters out inconclusively. Experimenters realize either that the theory is not sufficiently precise to test; or that it is too simple to be true. Once this stage is reached, there follows a period of patching up. Then, almost invariably, the theory is allowed to languish, and most of the research aimed to support or refute it is written off as lost.

It seems that this method of organizing psychology around a number of precisely formulated but simple theories must be a poor one:

If undue emphasis is placed on formal theories before successful research strategies have been evolved, there is a clear danger that experimental results may be regarded as evidence for or against rival theories (and nearly all of them are certain to be wrong) instead of letting the data suggest new experiments and new research strategies.⁴

Human beings, it seems, are multifarious; and we achieve little by speculating about them without first finding out what they are like. There is, after all, little merit (and no point) in proposing general ideas about human beings if these are largely or completely mistaken. Nor is there any virtue in claiming that our idea is ‘basically right’ although obscured by the welter of people’s individuality. It is the welter that we must observe and measure; and if we do observe, we are invariably forced to modify our initial hunches out of all recognition. It seems a matter of common prudence, therefore, if theories must be adapted continually to meet new facts, that we should not define any one theory too closely.

Some may say that this criticism is glib; that the work promoted by Hull’s theory is solid, empirical research, and is therefore invaluable, either as a spring-board for the next theorizer, or simply as natural history. Whether this defence is justified in any given case, it ...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Epigraph
• Table of Contents
• Acknowledgements
• 1 PRESUPPOSITIONS
• 2 ARTS AND SCIENCE
• 3 CONVERGERS AND DIVERGERS
• 4 A NETWORK OF CONNECTIONS
• 5 RIVAL SYSTEMS OF DEFENCE
• 6 THE QUESTION OF CREATIVITY
• 7 A TENTATIVE EXPLANATION
• 8 SOME SPECULATIONS ON ORIGINAL THOUGHT
• Appendix A Statistical Tables
• Appendix B The Tests
• References
• Index