Spens (1938)

Notes on the text
Preliminary pages Membership, Analysis, Preface, Introduction
Chapter 1 Development of the secondary curriculum
Chapter 2 The present position
Chapter 3 Physical and mental development of 11-16 year olds
Chapter 4 The curriculum of the grammar school
Chapter 5 Scripture
Chapter 6 English, classics, mathematics, general science
Chapter 7 The School Certificate Examination
Chapter 8 Technical schools
Chapter 9 Administrative problems
Chapter 10 Welsh problems
Chapter 11 Conclusions and recommendations
Appendix I List of witnesses
Appendix II Liberal education (Young)
Appendix III Secondary curriculum (Kandel)
Appendix IV Faculty psychology (Burt)
Appendix V Transfer of training (Hamley)
Appendix VI Curricula in the Dominions (Clarke)
Index

London: HM Stationery Office

Appendix IV Historical note on faculty psychology
by Professor Cyril Burt, Professor of Psychology, University College London
[pages 429 - 438]

'Faculty psychology is the attempt to explain mental phenomena by referring them to certain relatively independent agencies which are in truth only class concepts invested with a fictitious reality.' (1)

It was natural that the efforts of the earliest students of the mind should be towards classifying mental processes and explaining their causes. It was perhaps almost equally natural that each class of process should be attributed to a specific cause of its own and that the causation should be conceived after the analogy of that of bodily processes. Just as we have legs for walking, eyes for seeing, and a tongue for talking - different parts of the body for different bodily functions - so it was supposed that there might be 'parts' in the soul with functions or duties of their own. Of the earlier speculations in this direction, the most famous is the Platonic (2) doctrine of the [Greek] (parts) and [Greek] (powers) of the soul. Similar anticipations are found in pre-Platonic philosophy. Thus several Pythagorean philosophers attempted to locate different mental functions in different parts of the body. (3) Aristotle asserted the ultimate unity of all mental function, and described the Platonic parts or divisions of the soul as activities of this one conscious principle. (4)

The classifications adumbrated in these earlier Greek writings gradually took shape in the distinction between, first, the intellectual or theoretical functions of the mind, and, secondly, the practical or moral functions of the mind, with the emotional sometimes added as a third division. An emphasis on the practical or moral activities appears in St Augustine (5); but otherwise the interest chiefly centres in an attempt to classify what we should nowadays call the cognitive rather than the orectic processes. Thus Avicenna (d. 1037) distinguished the outer senses from the inner, enumerating five of each - the inner senses (commonsense, memory, fancy, imagination, and judgement) being all located in the brain. Similar classifications appear and reappear in the writings of the scholastic philosophers; and it is with them that the doctrine of faculties takes explicit and dogmatic shape.

Of the earlier scholastic writers, Thomas Aquinas (d. 1274) offered an analysis of the faculties which has proved at once the clearest and the most influential (6) His scheme is as follows:

Feeling (fruition) appears not as a separate fundamental faculty but as 'an (incidental) act of the appetitive power'. Duns Scotus (d. 1308), on the other hand, following Augustine, regarded feeling as something separate from and coordinate with desire. (7)

Both Aquinas and Duns Scotus sub-classified the intellectual faculties along much the same lines as their predecessors; and, with numerous minor variations, these subdivisions were preserved almost unchallenged to the 17th century. Thus Giordano Bruno (d. 1600) enumerates as separate faculties those of sensation, memory, imagination, and reasoning - a fourfold classification that is found surviving in books on education even in our own time. Bacon, it may be remembered, analysed the faculties of knowledge, and classified the sciences, under similar headings.

Against this traditional view the first clear protest is sounded by Locke (d. 1704). He retains the distinction between an outer and an inner sense, and admits that the mind has certain 'powers' which are native to it; but he regards these powers as merely abstract descriptions of the mind's way of working. There is only one agent - the person himself - and he has the 'ideas': 'ideas' are not created by the faculties. They are aroused 'by way of' sensation, memory, imagination, and the like; but these names do not denote causal entities. In a famous passage he lays his finger on the fallacy involved: 'We may as properly say that the singing faculty sings, and that the dancing faculty dances, as that will chooses or that the understanding conceives'. (8) Nevertheless, Locke himself did not entirely succeed in freeing psychology from the notion of separate faculties. 'He wished to shift the point of view from agencies to activities; but his rejection of "faculties" is followed by the adoption of "powers", which have no advantage over faculties except in being less real'. (9)

In German philosophy the doctrine still lingered, and reached its most systematic exposition in the writings of Christian Wolff (d. 1754). Wolff sought to define and systematise the views of Leibniz and his followers. He argues that the activity of the mind operates in different directions, and thus displays separate faculties. The faculties of memory, imagination, reasoning, will, and the like, are arranged by him in a kind of hierarchical order, and discussed in somewhat dogmatic terms, but not without considerable insight into actual mental processes. (10)

Kant (d. 1804) was interested more in forms of cognition than in faculties of cognition. He is usually credited with finally laying down a clear distinction between the threefold aspects of mental life - knowing, feeling, and willing, or, as they are now called, the cognitive, the affective, and the conative aspects, but still treats them as separate functions to be discussed in separate treatises. On the cognitive side, his sharp distinctions of sense, intelligence and reason are strongly reminiscent of the traditional faculties, though the epistemological account of them is given a very different turn.

The popular notion of mental faculties is probably due not so much to the recognised authorities on philosophy and psychology as to the writings of the early phrenologists, FJ Gall (1758-1828) and JC Spurzheim (1776-1832). At the beginning of the 19th century the study of the brain had led physiologists to regard it as a single organ homogeneous throughout. The phrenologists advocated a theory of strict localisation. They split up the mind into 37 faculties. These were separated into two main groups - the intellectual and the affective. The intellectual group was further subdivided into a perceptive and a reflective subgroup, while the affective faculties were subdivided into propensities and sentiments. Each faculty was supposed to be lodged in a separate organ of the brain. This doctrine was based on no first-hand study of cerebral anatomy or physiology, and was quickly demolished by experimental work, which showed that, so far as processes can be localised, their distribution corresponds, not to the so-called faculties, but to the organs of the body to which the issuing nerves fibres ultimately go, i.e. the muscles and the several senses.

On the Continent, the doctrine of faculties was successfully attacked by JF Herbart (1776-1841), who, in this respect, has probably influenced subsequent psychologists and educationists more than any other man. His argument is that, once we start postulating a different faculty for every distinguishable process of the mind, there can be no end to our list. Each fresh writer makes a new catalogue, and finality will never be reached. According to Herbart, the essential processes of the mind are due not to the operation of separate faculties, but to the activities of presentations - of what Locke would have called ideas. He distinguishes, not different kinds of faculty, but different degrees of activity; and of these he names four, viz. (i) memory, (ii) imagination, (iii) judgement, and (iv) understanding, the two former being of a lower degree and the two latter of a higher degree. Nor are the activities to be sharply distinguished. Each activity passes gradually into another. Thus 'his picture of the mind is so far new that it may be said to abandon entirely what is usually called the doctrine of faculties.' (11)

In this country the doctrine of faculties has been finally overthrown by the work of the associationists. They have sought to show that all forms of cognition can be described as the effects of association working on or combining sensory elements, and have further indicated that much the same laws of association operate in the domain of feeling and of volition. Thus the mind is no longer regarded as 'made up of a number of separate powers, each of which carries on its operations with supreme indifference to all the rest, and as having no more organic unity than a number of sticks fastened together in a bundle'. All parts of our mental life are regarded as 'having the same fundamental texture'. This standpoint, particularly on the Continent and in America, has been further reinforced by the endeavour to explain the principle of association as essentially a physiological association, i.e. as the formation of nerve paths or nerve connections between cell-groups within the brain. As a result, the lines of division between different mental processes become somewhat blurred. We cannot distinguish an act of will from an act of reasoning, for whenever we reason we are also exercising volition. Nor can reasoning be sharply distinguished from perception; for perception can be shown to have the germs of implicit inference, and when we reason we nearly always perceive.

In rejecting the doctrine of faculties as a basis for the theory of teaching, English educationists have probably been influenced most of all by Sir John Adams' early book on The Herbartian Psychology (1897). He regards the predominance of the faculty doctrine among English educationists as due to the writings of Locke. 'Teachers suck in Locke from the introductions to their earliest school-management books.' 'Locke got rid of innate ideas; but he could not free himself of innate faculties. And Herbart did for innate faculties what Locke had done for the innate ideas. He swept away for ever the whole brood.' (12)

In America the attack starts with James' vigorous polemic against phrenology. 'The faculties are fully equipped persons in a particular attitude. ... The "faculty of language" involves a host of distinct powers (corresponding to memories, ideas, words, judgements, hearing of the ears, utterances of the lips etc). An organ of the brain, to be the seat of such a faculty, would be a complete brain in miniature, just as each faculty is a homunculus doing this or that. A science of the mind must reduce such complex manifestations to their elements.' (13) Thus such arguments bring us ultimately to the point most succinctly expressed by Ward. 'There is only one faculty of the mind - attentive consciousness: instead of a congeries of faculties we must assume a single subjective activity which we may call attention.' (14)

For educational practice, the most important consequence of this change of view has been the progressive discrediting of the old hypothesis which rested on the doctrine of faculties and by which the traditional curriculum has largely been supported - the doctrine of 'mental discipline' or 'formal training'. If the mind consisted essentially of a number of faculties, it would follow that education should consist essentially in the training of those faculties through exercise or discipline. But if the differences between one mental process and another depend solely on the mental contents and not on the mental activities, it follows that education must deal, not with activities, but with contents. For example, if each memory is due to its own specific association, we can no longer speak of memory but only of memories; and the learning of one set of facts will not strengthen any 'organ of memory' which can be used in the learning of another set of facts. Here again Professor Adams was one of the earliest to draw the inevitable deduction from the new views: 'In the war of competing subjects', he writes, 'the main point of popular discussion is: which is best fitted to cultivate the mind - classics, science, mathematics? The Herbartian sweeps aside these claims, and raises the preliminary question: Do any of them train the mind at all? Pupils now learn poetry for the sake of the poetry, not to train the faculty of memory: would it not be well if the same change of view took place with regard to other subjects?' (15)

The first to put the corollary to an experimental test was James himself. Following his lead, Thorndike and his associates in an elaborate series of experiments dealt what is described as the death-blow to the doctrine of mental faculties. These experiments (16) clearly demonstrated that, whether the processes concerned were simple or complex, the effects of special training were transferred to a far more limited extent than previous educationists had supposed. Transference was found to be exceptional rather than inevitable; deleterious as well as beneficial; and apparently the outcome of processes quite different from those which had popularly been assumed. Hence, the problem with which recent investigators have been concerned is not so much whether transfer of training occurs, but rather under what conditions it occurs. The general conclusion to which most of them have been led is that transfer of training may be most effectively ensured when the methods or the ideals learnt during the training period are made clearly conscious and so freed of their context. (17) The practical consequence might be summed up as follows. If the pupil is intelligent enough to understand the logical technique involved in a given subject (mathematics, for example), then we may expect a transfer of this technique to other subjects (e.g. to non-mathematical problems), when - some would say only when - the teaching is so arranged as to make the pupil clearly conscious of the method implied and of the ideals pursued. If, on the other hand, either from his youth or from his dullness he has not sufficient intelligence to grasp this abstract technique or to pursue this abstract ideal, then it may still prove possible to impart the correct method as a habit, but this habit, being mainly unconscious, will now be confined to the particular context in which it has been acquired: for such children, therefore, the training must be incorporated in that specific subject matter to which they will be required to reapply the habits later on.

In making this statement, however, one important reservation must be added, which was too often overlooked by the early anti-formalists. Because we learn a thing best when we attend to it, it does not follow that unless we attend we shall never learn. Some of the most complex of intellectual processes (eg grammatical speech) are picked up, as we say, 'unconsciously': they are acquired out of school and during the very earliest years by what may be termed 'incidental learning'. This, as is now generally admitted, cannot be explained by mere 'association'; and just as faculty psychology sank into discredit under the blows of the associationists, so an exclusively associationist psychology has been gradually discredited by the work of the new 'hormic' and 'Gestalt' schools. The doctrine of Gestalt is too recent to have had much influence on the exposition of the current educational textbooks; and I have therefore been asked to discuss in somewhat greater detail its bearing on the problem of formal training.

The early anti-formalists, as we have seen, built up their revolutionary conclusions on a psychology which regarded mental processes as consisting essentially in the 'association of ideas' or rather of 'the linkage of brain-paths'. They viewed this linkage as purely mechanical and physiological. One early outcome of intelligence testing, however, was to demonstrate that, in the intelligent activities of school children, a purely mechanical or physiological association is by no means the only process at work. The associations themselves may be cognised; and - what is far more important - the cognition may be implicit as well as explicit. Such conscious or cognised associations are termed 'relations'. Thus, one of the best of the early tests of intelligence involved reasoning by what the teacher calls 'rule of 3' and what Aristotle called 'analogy'. The child is given problems based on a scheme such as the following, but generally in words rather than numbers or symbols:

(i) 1/10 X 20 : 2 : : 1/10 X 90 : .......?

(ii) A B C : B C D : C D E : .....?

Professor Spearman (19) has made these two 'neogenetic principles' - the 'eduction of the relation' and the 'eduction of the correlate' as he terms them - the basis of the whole psychology of knowledge. But although the attention of the experimental psychologist has only recently been drawn to this double principle by his search for intelligence tests, under various names it has been frequently mentioned by older writers. (20) Thomas Brown (21) distinguished 'relative suggestion' (i.e. the suggestion of a new idea by 'the perception of a relation' from 'simple suggestion'; Bain called it 'constructive association'; Spencer (22) made his double 'axiom of relations' the basis of all thinking and considered it to characterise the 'highest form of intelligence'. It was, however, Stout who first developed the full implications of what he proposes to rename 'proportional or analogical production'. He points out that the principle involves the final surrender of the old associationist doctrine as a complete account of intellectual process; and he strengthens his argument by emphasising that what is transferred need not be a simple relation but a relational system or (as he terms it) a 'thought-schema'. (23)

Now, whereas simple relations can be clearly perceived, a complex system can, as a rule, only be grasped implicitly. We see the grin without seeing the cat. (24) In the first example given above, the child will often formulate the relation implied by the first pair of terms: 'you knock off the nought'; but in the second example he will more probably continue the series on the basis of a kind of rhythm or 'tune'. Moreover, direct experiment reveals that the explicit education of relations (though simple) involves a high degree of intelligence, whereas the transference of implicit schemata (though complex) may be carried out by young children at a very early age. A child of three who has heard his father hum a song in a tenor key may reproduce the tune in his own treble voice: the tune forms a pattern which can be transferred from the original context to a context wholly new, and that without any special practice in the new field. A child of four, having heard her mother say: 'Tell Jimmy that, if he isn't a better boy tomorrow, I shan't let him go to his auntie's for tea', tells Jimmy the next day: 'Mummy says, if you aren't a better boy today she won't let you go to your auntie's for tea'. (25) How long would it take a French child in the lycée to learn and apply correctly all the rules for changing pronouns, adverbs, and verbs when converting such a sentence from oratio recta to oratio obliqua?

These familiar observations have been confirmed by numerous experiments both on children and on animals. Thus, we seem led back to a view which is as old as Plato (26), namely, that the highest training is often an insensible training; or, to quote the epigram of a famous headmaster: 'Education is what remains when the boy has forgotten all he ever learnt'. It follows that, in rejecting the doctrine of formal training in its traditional shape, the modern educationist must beware of flying to the opposite extreme, and maintaining that all training must be specific and explicitly conscious. If the earlier experimentalists jumped too rapidly to this conclusion, it was doubtless because, in their search for crucial cases, they confined experimental work largely to tasks of a simple and mechanical kind. Always there is a danger that the simplifications of the laboratory may omit what is the most important ingredient in real life, namely, incentive or motive. Too often the problem has been conceived, both by the theoretical psychologist and by the practical teacher, as merely the problem of imparting knowledge or of forming habits; and the emotional and moral aspects of the process, or, in more accurate terms, the affective and the conative aspects, have been overlooked.

Even in the broader implications there is now a disposition to feel that the reaction against faculty psychology may have gone a little too far. Granting that the mind is a unity, and not an aggregate of mental powers, it is not a homogeneous unity, but an organisation. Valuable distinctions, noted and perhaps overemphasised by the earlier classifications, are now in some danger of being lost; and the later notion of the mind as a simple mechanism for linking up elements by association, much as subscribers are linked up by the switchboard at a central telephone exchange, is not only a gross oversimplification of the facts, but fails to explain the peculiar individual differences observable between individual pupils. It is found that certain children may be peculiarly deficient, not in all-round intelligence, but in some special group of cognitive operations - eg in visualisation, in mechanical memorisation, in verbal manipulation, in arithmetical computation, and the like. These peculiarities, which have been studied statistically, have given rise to the description of specific mental 'factors', operating over and above 'general intelligence'. At first sight these new 'factors' are not unlike the old-fashioned 'faculties'. The chief differences are that the factors are statistical abstracts, not causal entities or anatomical organs in the brain, that the lines of distinction are relative rather than absolute, and that the evidence for them rests on an empirical analysis of data collected by means of experimental tests, not upon mere armchair speculation.

The practical corollary is obvious. The problem of the best methods of teaching or training and of the best choice of subjects is to be determined, not merely by a consideration of the general nature of the mind as such, but by a close and first-hand study of the needs and limitations of the particular individuals to be taught. (27)

Footnotes

(1) GF Stout, 'Herbart compared with English Psychologists', Mind, XIV (1889) p. 1.

(2) Tim. 42A, 61C, Polit. 309C, Rep. X. 477C, 580D, 611C.

(3) cf. Philebus, 159.

(4) It has not, I think, been noted that some of the latest and most characteristic doctrines of modern psychology and of modern physics find remarkable anticipations in the writings of Plato and Aristotle. Modern psychology corrects the simple view of the early experimental and physiological psychologists (the followers of Weber and Fechner, for example), and insists that in sense-perception what we perceive are, not isolated stimuli, but patterns or Gestalten - such as shape, movement, size, number, and the like - characteristics which are perceived in common by more than one sense organ: but already in the Aristotelian doctrine of the 'common sense' (a misleading title) we find the same insistence on the 'common or central unifying function of sense by which we perceive [Greek text here] i.e. number, figure, magnitude, movement etc.' (Ross, Aristotle: De Sensu et de Memoria, p. 14; see especially Aristotle, De Anima, 425a, 14-30). Modern physics, quite independently of modern psychology, has reached an analogous conclusion: it insists that 'what physics ultimately finds in the atom or in any other entity' are not isolated individual substances, but only 'structures' or patterns, which thus 'become for us the sole realities', and are perceived as relatively constant or stable beneath a changing interplay of unknown operations between unknown materials (Eddington, New Pathways in Science, p. 262). What is this but a reformulation in more explicit terms of Plato's famous doctrine of [Greek] and [Greek] ('forms' and 'pattern-forms')?
No doubt, the originators of the Gestalt theory would have admitted a continuity of doctrine, though its more recent advocates appear to have forgotten it, for, both in this country and on the continent, the theory is, in its essence, an offshoot of the teaching of Brentano; and Brentano no doubt was greatly influenced by the classical and scholastic tradition surviving in the Catholic University at which he taught. But the wide acceptance today of the Gestalt theory in psychology, as of the analogous theory in quantum physics, is the outcome, not of the plausible advocacy of some intuitive genius, but of experimental research and of rigid logical deduction: for example, in psychology as in physics, the new mathematical instrument is now the comprehensive algebra of 'matrices' (patterns of numbers), which replaces the older algebra of isolated numbers, and will no doubt be in turn replaced by the 'theory of groups'. But what is most striking is that the two bodies of scientists - physicists, on the one hand, and psychologists, on the other - each largely in ignorance of what the other was doing, should have been forced to much the same conclusions - conclusions which to the simple materialists of the 19th century would have seemed revolutionary, but are in fact entirely consistent with the teaching of the older idealists. To discuss the philosophy of education is beyond the province of an educational psychologist; but he may at least be permitted to deprecate the widespread notion that modern psychology (so often wrongly identified in the popular mind with psychoanalysis or behaviourism), or for that matter modern science, must tend to promote a materialistic view of human life and education. It is evident that the trend of present day psychology is in exactly the opposite direction.

(5) Confessiones, Bk. X, Ch. xiii et seq.

(6) Summ. Theol., Pt. I. Q.lxxx, Act. 2 et seq.

(7) Leibniz and Hamilton reserved the term faculty for 'active' experiences and proposed the term capacity for the two 'passive' experiences of feeling and fruition (pleasure and pain). As noted below, the threefold division was reintroduced by Kant; it has been preserved by Ward and Stout, and so lasted to the present day. The work of the correlational and the biological schools in psychology, however, has led many to recombine 'conation' (desire) and 'affection' (feeling) under the single Aristotelian heading of 'orexis'.

(8) Essay on Human Understanding, Book II.

(9) Brett, History of Psychology, Vol. II, page 260.

(10) CF von Wolff (1679-1754), Psychologia Empirica, Leipzig, 1732.

(11) Brett, History of Psychology, Vol. Ill, p. 55.

(12) Loc. cit., pp. 33, 47-8.

(13) Principles of Psychology, Vol. I, pp. 28-9.

(14) Ward, art. 'Psychology', Encyclopedia Britannica, 9th ed., vol. XX, pp. 37 et seq,

(15) Loc. cit., pp. 107, 134.

(16) The first and most famous of these investigations was that carried out by Thorndike and Woodworth in 1901: 'Influence of Improvement in One Mental Function upon the Efficiency in other Functions', Psychological Review, VIII, pp. 250 et seq. The long series of researches which followed this pioneer effort will be found summarised in Whipple's chapter on 'The Transfer of Training' in the Twenty-Seventh Yearbook of the National Society for the Study of Education (Part II, Chapter 13), and in the monograph by Rugg and Blair (prepared for the National Committee of the Mathematical Association of America on Mathematical Requirements) entitled The Present Status of Disciplinary Values in Education.

(17) 'From experiments of an exact nature and from much more numerous rough-and-ready unrecorded experiments which teachers are every day making, it seems very probable that transfer can be greatly aided by methods of teaching: in general, the rule appears to be that any teaching which makes the pupil more conscious of how successful results are obtained is likely to assist transfer.' (Godfrey Thomson, Instinct, Intelligence and Character (1924), pp. 143-5: (author's own italics). cf. Nunn, Education: Its Data and First Principles (1930), pp. 239-242).

(18) Burt, 'Experimental Tests of Higher Mental Processes', Journal of Experimental Pedagogy, I, 1911, p. 101. cf. id. 'Experimental Investigations of Formal Training', Report of LCC Conference of Teachers, 1912.

(19) The nature of Intelligence and the Principles of Cognition, (1923).

(20) Logicians have long recognised that 'all reasoning depends on the properties of relations'. But this view has had little influence on psychology, first, because it has been traditionally assumed that 'a relation cannot exist apart from its terms' and indeed may be reduced to a kind of 'adjective', and secondly because, until recently, almost the sole relation considered by the logician has been the predicative relation. Only since James' emphatic pronouncements on the subject have psychologists been willing to admit that, whatever has been the correct view in logic, in consciousness the awareness of a relation may certainly exist without any awareness of its terms. The first explicit discussion of the 'eduction of correlates' is to be found in Hume - one of the earliest and most thoroughgoing of all associationists. 'Suppose', he says, 'a person to have become perfectly well acquainted with colours of all kinds except one particular shade of blue. Let all the different shades of that colour except that single one be placed before him, descending gradually from the deepest to the lightest ... Now I ask whether 'tis possible for him, from his own imagination, to supply this deficiency. I believe there are few but will be of opinion that he can ... tho' the instance is so particular and singular that 'tis scarce worth our observing.' (Human Nature, Book I, pt. i, sect. 1, p. 315).

(21) Philosophy of the Human Mind, Lectures xxxiii-xxxvii.

(22) Principles of Psychology, Pt. II, chap. vi.

(23) Mind, Vol. XVI (O.S.), p. 50. Analytic Psychology, Vol. II, pp. 78 et seq. In his earlier article he speaks of 'proportional systems' and of 'proportional production' - preferring to borrow his terms from writers on language rather than from writers on logic. In his later book he substitutes Brown's term 'relative or relational suggestion'. I should prefer to speak of a 'mental schema' (rather than of a 'thought schema') to emphasise the fact that it may govern adaptive thoughtless action as well as processes of conscious thought. Here, I fancy, lies the novelty: experiment reveals that 'creative analogy', which the philosopher held to be characteristic of the original thinking of the genius, appears, not only in the thought processes of young children, but even in their adaptive actions, where the re-adaptation is immediately carried out without thought and even without relevant consciousness.

(24) Once again, the fact had been recognised by the early associationists who realised that its occurrence offered great difficulties to the associationist theory. Their favourite instance was that of meaning; and their problem was to explain why 'it is not necessary that significant names should, every time they are used, excite in the understanding the ideas they are made to stand for, being for the most part used as letters are in algebra'. (Berkeley, Works, Vol. I, p. 150. cf. Hume, Treatise on Human Nature, Bk. I, pt. i., sect. 7). Again it should be noted that the patterns may be patterns of action as well as patterns of ideas. A blind man who has sprained his right wrist may sign his name for the first time with his left hand, and the style of his signature will be much the same as before, although he has never used the muscles of the left hand for writing and has never seen what his signature looks like.

(25) It was apparently instances of this sort that led Stout to insist on the importance of the process both in language and in thought. An early discussion of the process and of its unconscious character will be found in MH Paul's discussion of 'word-combination by analogy' (Principles of the History of Language, (1890), chapter V).

(26) Republic, Book III, 401 D. As early as 1631 Comenius argued that children would learn to read more quickly if they were taught from the very outset to read whole sentences instead of associating isolated letters to form isolated words and then associating words to form sentences. In a dictation given to girls in Standard VII I found that 11 out of 37 could spell the word 'phagocyte' correctly although they had certainly never heard or seen it: think of the numerous elaborate rules that would be required were English orthography to be taught by explicitly educed relations. It may be noted that Woodworth, who, as we have seen, was in 1901 a pioneer in destroying the old notion of formal training, writes thirty years later 'the doctrine of Gestalt has made the theory of learning more uncertain than ever'. (Contemporary Schools of Psychology, (1931), p. 132). As I see it, the immediate problem for experimental education is to ascertain for what pupils and for what parts of the curriculum it is better to teach by educing conscious and explicit relations and for what it is better to teach by the organisation of implicit and complex wholes.

(27) In a brief Report on Formal Training (Report of a Committee appointed by the Education Section of the British Association presented at Bristol, Annual Report, 1930, p. 608) I have endeavoured to summarise the educational bearings of current psychological views. A historical survey of the doctrine of faculties is to be found in O Klemm, Geschichte der Psychologie (1912), pp. 44-70. cf. also M Dessoir, Geschichte der neueren deutschen Psychologie (1911), chap. VI, ' German Faculty Psychology following Wolff '; Siebeck, Geschichte der Psychologie; Eisler, Worterbuch d. philos. Begriffe, s.v. 'Seelenvermogen'.