1.1. SOURCES
The content of Part I is based on work at the Institute since the seminar at the University of Sussex in January 1967, reported in our March issue. The comments and criticisms of the papers presented at the seminar led to a re-examination of the logical and psychological basis of the method and this in turn to a far wider conception of its significance and potentialities.

The main sources on which we have drawn are:

1. Psychological investigations into the process of understanding complex situations made at the Institute.
2. A co-operative study of the logic of communication made with Mr. R. S. Arbon of the Hirst Research Centre of the G.E.C., Wembley. This study, though of a general character, was directed to the use of structural communication in the Systemaster teaching machines and in Computer Guided Instruction.
3. Co-operative projects with school and college teachers interested in the use of structural communication as a means of increasing the effectiveness of lectures and discussion groups.
4. A short course for experienced secondary school teachers arranged on behalf of the Curriculum Laboratory. Goldsmith's College, University of London, April 26th-28th (see Appendix I).
5. An Interdisciplinary Workshop introducing the techniques of structuraI communication to administrators, business executives, engineers, medical practitioners and others. May 28th-29th (see Appendix II ).
6. Co-operative tests with colleges and schools (listed in Appendix III), to develop the technique and determine its suitability for various disciplines.
7. Discussion groups of the Institute staff and associated writers in which the basic theory has been considerably developed.

1.2. HISTORY
1.2.1. THE NOTION
In earlier papers, the concept of structure was applied to the presentation of information to be communicated. From practical experienced gained in the use of the technique, it became apparent that the act of communication itself must be structured to include the response of the recipient and the development of his grasp of the message These form an indivisible whole, and we propose in future to use the term structural communication to comprise all forms of interaction in which the integral character of the theme is preserved throughout, and a shared understanding between the participants develops by the action taken by the recipient of the message.

We believe that structural communication can make an important contribution not only in the fields of education and training, but in all cases where fragmentation of a message risks losing its meaning, and where total understanding is more important than transmission of separate items of information.

We have come during 1967 to realize that structural communication is not an extension of existing forms of communication by means books, programmed texts, pictorial devices, or even personal interviews. It has even been suggested that a special term should be coined for structural communication in depth, but there is still much work to be done before the scope and potentialities of the method-as well as its limitations and defects-can be assessed with sufficient clarity to justify its claim to be a new discipline.

1.2.2. PROGRAMME OF THE INSTITUTE
The Institute has, since its inception, been mainly interested in man's potential for developing the higher powers and qualities latent in human nature and has, as an integral part of its programme, approached the problem of understanding in terms of creativity and consciousness rather than of conditioning and automatism. The consequence has been that an approach to the theory and practice has been followed that is radically different from that of the behaviourist schools and is at the same time pragmatic and operational.

1.2.3. EDUCATIONAL INTEREST
Our direct concern with education began in 1960 with the formation of the Integral Science Education Research Group (I.S.E.R.G.) under the leadership of Mr. A. M. Hodgson. This group of young graduate scientists worked for several years on the task of producing an interdisciplinary curriculum for science teaching which would serve also for encouraging creativity and independence of judgment in the students. Soon after ISERG began its work, the Institute was invited by the Education Department of Middlesex County Council to look into the problem of helping students at colleges of technology to make better use in their professional work of what they had learned at college. It soon became evident that this is basically a problem of understanding, requiring the same conditions as those which make shared experience possible in adult groups. With the co-operation of two colleges of technology, experiments were made in group tutorials using the method of challenge and response. These brought out the immense value of the free enquiry tutorial groups, but also emphasized the practical impossibility of achieving this on any large scale within the resources of education manpower available now or likely to become so.

The next step was to see if the process of communication could be standardized in such a way as to reduce the expenditure of time and energy on the part of the group leaders. The creation of an open-ended learning situation in which understanding can evolve requires far more preparation and also far more inspired handling than simple exposition and discussion. Unless the preparation is carefully done, the tutorial either degenerates into an aimless discussion that may be pleasant enough but seldom generates understanding or else, it reverts to a teacher [dominated session in which the leader seeks to condition the members of the group to his own way of thinking. The preparation requires either a very gifted teacher or good team work on the part of teachers with much experience of the process. There is the further difficulty that an open-ended tutorial group works well with from seven to twelve members. Any larger number results in the majority failing to participate, and this sets another limitation upon the practicability of tutorial groups in understaffed schools and colleges.

We are familiar with the benefits or centrally prepared audio-visual aids; but these do not allow for interaction between the creative mind of the author and the receptive minds of the students. At first sight, it would seem impossible to combine the advantages of central preparation of very high quality teaching material with genuine dialogue between the author and the recipient. The attempt to do so by 'programmed texts' has given results very limited in scope.

1.2.4. EVOLUTION OF SYSTEM
A review of the situation made in 1965, led to the suggestion that the kind of dialogue required in open-ended group discussion could be achieved by using a pre-designed information retrieval system which could also serve as a means of communication. This appeared to be an appropriate problem for a digital computer and an approach was made to the Ministry of Technology for advice. In the outcome, the General Electric Company was approached and agreed to look into the problem which was referred to the Hirst Research Centre and a fruitful co-operation was launched that has resulted in the project described in this paper.

It was soon recognized that if the adaptive dialogue aimed at was to be conducted solely through the data-processing and storage resources of a digital computer, the programme writing required for each tutorial session would be totally uneconomic. This led to the concept of hierarchy of teaching machines that would enable teacher-student participation to carry the greater part of the processing load and also dispense with most of the data storage that would over-burden a computer. It was recognized that the teaching machines could provide all the data processing required, while the memory could be furnished by a visual display system. The machines collectively named Systemaster, were designed at the Hirst Research Centre and patented jointly with the Institute in 1966. An early version of the Systemaster was described in the March issue of this journal. Since then a decisive improvement in the flexibility of the logic has been introduced and this is outlined in section 2.4. below. A fuller account will be published elsewhere by Mr. R. S. Arbon of Hirst Research Centre.

1.3. THE COMMUNICATION OF UNDERSTANDING
The art of communicating understanding consists in devising a means whereby the recipient of a message can reproduce, or at least simulate, the action by which the author of the message reached his own understanding of its content. This can be achieved in verbal communication providing the author has a clear idea of the steps that led to his own intuitive grasp and providing also that he has the skill to present the challenge to the recipient in the right way. This action takes the form of a dialogue in which the author and recipient are equally active Not all verbal communication is successful. Unless the recipient makes a response which demonstrates his grasp-or lack of it-of the author's intention, the latter has no means of telling whether his intention has been put over. If the response can be put in a form easy to evaluate, without impairing the recipient's freedom of expression, the act of communication can make rapid progress. A few exchanges of challenge and response will clarify both the author's intention and the recipient's understanding and enable them both to verify the degree of agreement they have reached In principle, the exchange need not be verbal and the two parties need not meet in person, providing the two conditions of freedom of expression and a response susceptible of rapid and exact evaluation are satisfied.

Whether or not the communication is verbal and whether it is made in writing or by a teaching aid or computer, the basic requirement is that it should take the form of a dialogue conducted within a framework that ensures unambiguous expression of the author's intention and the recipients' response. This we hope to achieve by the method of structural communication.

The requirement of unambiguous evaluation is quite different from that of verbal or factual accuracy. The latter can be verified by putting the message in unit terms and processing the response by means familiar in telecommunication practice and linear programmed instruction to give a feedback loop. This can be made precise, but it cannot be made flexible.

The limitations of the procedure can be illustrated in the computerized translation of technical literature which, while not yet economic, is certainly feasible. Though lacking in literary merit and idiom, such translations can be very useful; but solely on condition that the reader is already familiar with the subject matter and with the intentions of the author. He can get at the meaning of a literal translation by an intuitive leap that enables him to bridge the inadequacies of the text. This leap is required, not because the language may be uncouth, but because the computer programme cannot be written in such a way as to pick up the mutual relevance of different parts of the message, and present it as an integral whole. Clarity of expression plays little part here. Even when a text is carefully written by an experienced author and the meaning is unambiguously stated, the reader may not notice some essential point and grossly misinterpret the message. However clearly the meanings may be expressed, their understanding depends upon the interplay of a variety of factors, so that it is only the totality of the message that contains the meaning. If this meaning is broken down into units small enough to be verified on a digital basis, the mutual relevance of the elements is lost, and this very relevance is usually what is most important.

1.3.1. FOUR KINDS OF COMMUNICATION
We can distinguish, for the purpose of the present study, four kinds of communication:

Telling.
The author A and the recipient R have a one-way contact. A tells his message to R who grasps it as well as he can, but has no means of ascertaining whether he has grasped A's intention correctly. R cannot even be sure that he has assimilated the factual content of the message except by hearing it or reading it through again. This is the situation if A gives a lecture not followed by discussion or puts his message in the form of a written statement or a book.

Discussion.
A gives his message. R responds with questions which A answers. Or A may question R to find out what he has grasped. This is the situation of a lecturer who encourages questions after he has spoken. It applies, though in a very restricted sense to programmed instruction both of linear and branching types. There is a feedback, but not assurance of understanding by R of A's line of thought.

Dialogue.
A does not give his message in one piece but builds it up by an exchange of ideas in which R is an active partner. The limitation of this method is that A and R must be in personal contact either verbal or by correspondence. The demand upon A's time is such that he cannot transmit his message to many recipients and be must be very gifted in the art of dialogue to bring his hearers4 to the point where they not only grasp his intention but demonstrate that they do so. The Socratic dialogue as described by Plato -- not Plato's own written works-exemplify the method.

Dramatic Presentation.
The dramatic performance is, on the face of it, an example of 'telling'. The audience receives the message through interpreters and has no direct contact with the author. This may seem to imply a passive condition, but the reality is far different. Not only is the audience response an essential part of the drama-without it the play would be dead. Even more significant is the four-part interaction in which the playwright, producer, actors and audience are equally involved in every stage of rehearsal and production. The interaction enriches and integrates the production and ensures, as far as possible, that the audience will enter into the author's intention. This intention is not a static fact to be precisely 'taught' to the audience. It takes shape and develops through four major stages. First, it is conceived and written by the author Second, it is worked out for the stage presentation by the producer with an eye upon his cast and his audience. Third, it is given body and mind in the performers. Lastly, it is transformed into a shared experience by the presentation and response of the audience. The communication is not complete until the last curtain falls.

1.3.2. DRAMATIC CHARACTER OF TRUE COMMUNICATION
The intense personal interactions of a stage performance are not an obligatory part of the dramatic communication. A deep rapport with the author may be experienced by one who reads the play and never sees it acted. This is because great drama can evoke the experience it portrays-but it does not follow that the reader has correctly grasped the author's intention. The unresolved controversies regarding Shakespeare's intentions are sufficient evidence that the evocation of experience in itself does not guarantee understanding. For this, we must have the give and take of the Socratic dialogue. But even Plato, using the form of the dialectic, not infrequently failed to convey his own intention and the reason for this is very relevant to our theme. He failed when be was not himself clear what he wanted to say and yet was sure that he was touching upon a theme of great significance to himself and others: for example in his doctrine of ideas. Plato had a wonderful sense of drama and when he constructed a plot as in the Symposium or Crito and Phaedo, his intention is clear and is communicated.

Our claim for originality in the method of structural communication derives from the active participation of the listener or recipient of the message in the actual act of transmission and not as an onlooker who reads a narrative or a Platonic dialogue.

The communication is open-ended to the extent that the recipient of the message has a far wider degree of freedom in making his response than he can fully utilize. At the same time, the author retains his hold of the situation by the way he constructs his message and arranges for the recipient's response to be made in such a form that he can comment upon it without being present. The action is rehearsed, but it is not predetermined. The plot is definite, but its working out is left open, as indeed it is with a dramatic presentation in which a good actor will never play the part twice in the same way. The organization of the subject-matter is carried out in such a way that the student is not told the author's intention, but discovers it for himself. The response is so organized that the student not only verifies his degree of success in grasping the message, but builds up his grasp in the act of testing it.

1.4. METHOD OF STRUCTURAL COMMUNICATION
In structural communication the aim is to evoke understanding, not to convey facts except as a by-product. Facts are always necessary, but they are seldom of much value unless they are placed in a framework of understanding. This is true not only of advanced studies, but even at the elementary level. One of the most encouraging trends of modern education is the determination to cease from teaching bare facts and simple skills and to aim rather at educating pupils to think and judge for themselves and to look for the wider relevance of what they are taught. These are also the central objectives of structural communication. In claiming originality and the ability to perform a task at present reserved for the gifted teacher with one or a small number of pupils, we do not suggest that structural communication solves all teaching problems. We believe that it has its place in the educational resources of the future alongside of conventional methods brought up to date and other techniques, aiming at other objectives.

1.4.1. THE STRUCTURAL PRINCIPLE
We have now to describe the principle upon which stuctural communication operates and explain why it should have a place in the education of young people. References to the Socratic dialetic and Shakespeare's drama may have suggested a depth of significance far beyond the comprehension of a child. It can be argued that simplicity is more important than profundity and that complexity tends to confusion.

To this we would reply that simplicity can be deceptive. The modern society is a very complex organization and life grows more complex from decade to decade. Education in the future will be far more than teaching knowledge and skills; it will have to prepare men and women for a life in which the higher abilities of creativity, synthetic judgment and grasp of complex situations will be at a premium over any kind of routine skill. The explosion in science and technology, and the size of organizations calls for a similar explosion in education. The revolution in education is a breakaway from mere teaching towards co-operative learning, and this will require also new techniques. This alone justifies serious consideration of a method which is centred upon the dialogue and aims at understanding rather than knowledge. We believe this to be true for all ages from primary school to adult training and mid-career retraining.

The principal we have adopted is that nature and all situations with which we have to deal are organized structures. They are neither atomic and random nor are they homogeneous and continuous. We recognize and understand situations to the extent that they have a well-defined structure. The same applies to our own human behaviour. We do not act through a string of elementary gestures held together by instinctive behaviour patterns - like animals - but by a very adaptive combination of foresight and hindsight. Human behaviour is successful to the extent that it corresponds to the patterns of the situation, both internal and environmental. The operations we perform must be coherent, and any set of instructions intended to set a behavioural sequence in motion must have a structure that will be an effective link between the structure of the situation and the pattern of behaviour. The instructions serve a purpose analogous to a jig or template in transforming a design from an engineer's drawing to a component in production.

The structure of a communication is dynamic, in other words it has a time component as well as space components. The time component is more than simple sequence, because it allows repetition of operations and return to operations previously made. This gives a significance to the structure that it is hard to appreciate except by carrying out the process. It can be described as unconstrained feed-back to distinguish it from cybernetic systems which are constrained to work towards optimum values of a prescribed set of parameters. This constraint is replaced by the free selection of the recipient of the message who is allowed to choose his own strategy for reaching an optimum. He can vary at his discretion both strategy and aim.

Another way of putting it is that the message itself comes into existence by the very process of its transmission. It is therefore different for each recipient, just as it would be in a face-to-face colloquium.*1

Every significant theme-even the most elementary-has a potential for development A twelve-year-old confronted with the transition from arithmetic to algebra or from descriptive to apodeictic proofs in geometry has to assimilate notions that were built up over the many centuries by adult thinkers of great genius. The notions and the operations that they lead on to are complex and cannot be grasped by a series of small steps. They can be presented as a dynamic transition from the particular to the general which once grasped makes the subsequent study of all branches of mathematics a natural and inevitable development. The transition was prepared during the descriptive and experimental stages; but the intuitive leap required must be made once and for all. A good teacher working with a homogeneous group of pupils can help them to make it, but if some of the group are quick to see and others slow, the latter may never achieve it. In this way, many a potential mathematician has missed his or her vocation. Structural communication mitigates the disparity of fast and slow, but does so without losing the dynamic character of the process of understanding.

The use of the word 'dynamic' in this context is intended to distinguish structural communication from programming in which the message is conveyed in steps each of fixed content and calling for a ed response from the recipient. Although there is authentic feedback, the process is static inasmuch as the recipient's response is for the purpose of verification only and not a part of the build-up of the message.

1.4.2. CONTENT OF A MESSAGE
We must now examine the content of a structural communication. It is fair to assume that in every good communication each act transmission concerns a single theme, which is the kernel of the structure. If there are two themes there are two communications. An act of communication must not mislead by introducing elements alien to the theme and it must not be lacking in any elements that are essential for a complete grasp of the author's intention. These requirements need not mean that everything in the message must be spelled out in detail. On the contrary, the recipient must face a challenge that will 'make him think', and this may be done by requiring him to develop for himself elements presented in embryonic or incomplete form. In verbal communication, it is frequently advantageous to present the theme in outline, sometimes with simplifications and omissions to make the recipients feel at ease. The necessary corrections and amplifications are inserted in the course of discussion. This dynamic approach is inherent also in structural communication by the written word or through visual presentation mechanically controlled (computer-aided instruction).

The aim is to convey a total and integral meaning. This can be achieved only if the integrity of the theme is preserved. The meaning may be approached by successive approximation, but it must not be broken down into a sequence of steps. Very few situations are strictly 4 sequential. There is nearly always a lateral or cross-action between lines of activity. There is a feedback and self-corrective effect that requires, repetition and recycling. These characteristics are present even in theoretical studies and they are, of course, of major importance in any practical undertaking. If any subject is taught as a sequence of steps, the student will tend to assume that the same sequence characterizes the subject-matter and this can sometimes prove a serious block to his understanding. The same applies to operating instructions where attention to cross-sections may often be as important as the linear sequence of operations to be performed.

Every author of technical literature is aware of the impossibility of describing complex mechanisms, situations and operations without frequent use of cross-references. These have the effect of adding a dimension to the presentation but excessive recourse to the device is tedious for the author and irritating for the reader. The same result be achieved more economically and with greater ease of reading, if the meaning is built up by the method of prepared dialogue. The subsidiary points that would be made by cross-reference then become part of the act of communication.

1.4.3. STRUCTURING THE PRIMARY MATERIAL
The structuring of the primary material can now be considered. We shall distinguish later between subject-matter, which is information, and theme, which is the carrier of intention. (See Section 2.4.1.). The first requirement is to identify the theme in such a way that there is no ambiguity in the author's mind, but without fixing precise limits. The theme must be rich enough in content to allow sub-themes significant in their own right to be specified. The complexity and extent of the theme defined in 2.4.3.1. must correspond to the capacity of the recipients. The message must present a real challenge to their understanding, but not make an impossible or excessive demand. By inserting the structure into the act of communication, the author makes it possible for the recipient to reconstruct the mental process behind the message. The latter does not have to do this consciously-though it is useful for him to try-but is provided with a means of testing his understanding which has the same effect. To illustrate this point, let us suppose that in preparing his message, the author has made a set of notes of the points he wants to get over and has grouped these into sub-themes in the form of a chart. When he has completed his message, he will go over it with his notes and tick the items off one by one to make sure that every relevant item has been included. He can go further and ask himself if he has presented his material in a balanced way so that each idea has been given prominence corresponding to its importance for the theme. He will also satisfy himself that the sub-themes,*2 are shown in their correct relationship. This second requirement can be met by taking his notes and putting them in groups each labelled with one of the sub-themes of the message. He can now verify not only that he has included all the points severally, but also that he has brought out their mutual relevance.

1.4.4. COMMUNICATING THE INTENTION
The recipient of the message wishes to do more than read and memorize verbatim what the author has written: he wants to understand the intention behind the message so that he can go beyond the limitations of its verbal form. If he can grasp the intention, he will be able to deal with unforeseen - and perhaps unforeseeable - contingencies that may arise in the course of putting the message into effect. He can do this by extracting the points he considers significant and putting them in his turn into groups or sub-themes. In other words, the recipient can work backwards from the message to the point at which the author made his own preliminary notes.

If now the two sets of notes could be compared, the parties to the communication would see at a glance if the intention had been successfully transmitted. If not, they will see how and where the breakdown occurred and by means of an appropriate supplementary explanation, the author could bridge the gaps. The recipient of the message now stands in the shoes of the author and-if he is of equal capacity-will be able to deal equally well with any situation that may arise.

Since, in general, the required consultation between author recipient is not practicable, means must be provided for some alternative method of bridging the gap. This is given by the response indicator (called in earlier papers the 'semantic matrix') which accompanies the message. The indicator contains the key-points arranged in random order in a matrix. Usually, 20 such points 5 x 4 matrix proves adequate The author then proceeds to amplify his message by making statements-and posing problems or by asking questions to which the recipient responds by selecting the appropriate key-points from the response indicator (which has been provided by the author and so is common to both). Each problem is constructed to correspond to one of the sub themes.

We now have the equivalent of a dialogue. The recipient's selection is compared with the author's intention and if there are any discrepancies further explanations are available as they would be in a rehearsed tutorial. The communication is not regarded as complete until all the stages have been carried out. The entire cycle consists of four sets of operations.

1.4.4.1. The Presentation of the message.
This is usually in the form of a condensed verbal statement using diagrams or pictures as required. A tape-recording or a TV screen or any combination of sight and sound images may be used. The recipient may be asked to use his eyes in looking at paintings or his hands in carrying out an experiment or constructing a mechanism. In other wards, any means may be used to put the recipient in touch with the theme. The reader does not attempt to memorize the message or any part of it; but seeks to recognize the main theme and its structure as a combination of sub-themes. The presentation may include a preamble indicating the purpose or objective of the message followed by a statement and even a re-statement of the message. In some cases, the message may be placed in the context of an extended memorandum, a specified passage in a text-book or recommended reading. The author has to decide upon the audience be is addressing and estimate the extent and difficulty of the message that will not over-tax the recipients.

1.4.4.2. The Challenge.
This will usually take the form of a series of distinct operations to be performed in response to instructions questions or problems in each case stated in such a way as to require a grasp of the author's intention if they are to be performed successfully. This is called the 'investigation'. The operation to be performed consists in selecting, from the 'response indicator' that accompanies the message, the items that in the recipient's opinion should be included in a reply to the question put to him. The response indicator has a three-fold significance: (1) it schedules the notes or key-points in the author's intention, (2) it provides the reader with a field of action that corresponds conceptually to the actual situation described in the presentation and (3) it enables the recipient to express his grasp of the problems by means of a small group of numbers. For this latter purpose, the indicator is constructed as a matrix labelled from one to whatever may be the number of key-points needed. The investigation and the response indicator form a single whole.

1.4.4.3. The Response.
The set of numbers selected forms the link between the recipient and the author. If the method is used for personal teaching. the numbers feedback to the teacher a picture of the pupil's grasp and enable him to spot the best way of helping him to go further. This has been found to work remarkably well in practice and is one of .the ways in which the method is applied. The most economical and effective way of using the method is with the Systemaster teaching machine which is designed to carry out the logical operations required. Recent developments in extending the range of operations so that the machine simulates the flexibility of a tutorial will be described in the next section. With a text-book, the reader is told how to evaluate his own responses. In each case, the message is reinforced and the recipient's grasp strengthened, by the prepared dialogue of investigation, response, discussion and comment.

1.4.4.4. Integration.
After mistakes have been corrected, ideas ampIified and understanding tested, it remains to bring the communication into one integral whole. For this purpose, the author writes a final statement called 'Viewpoints' in which he brings out some of the finer or more profound aspects of the message. In doing so, he has in mind the dialogue in which the theme was first stated, then discussed and when he is satisfied that the main point has been grasped, he adds some more difficult and suggestive ideas which his interlocutor will take away and ponder over.

The addition of supplementary notions is a permissible but not an indispensable part of the fourth and last stage. The author must look over all he has conveyed and ask himself whether it forms a well-balanced and integrated structure. A pithy re-statement of the whole message in slightly different terms is often the best way of completing the action of communication.

1.5. STUDENT PARTICIPATION
Involvement on the part of the recipient is an essential factor if understanding is to arise through any communication. In a good novej i this is provided by the act of the writing alone. The novelist, should his objective be clear, has to anticipate the possible misunderstandings of his message and ensure an active involvement of the reader by the vividness of his imagery. Although the great novel can provide conditions for the development of understanding, it is limited by the scarcity of people capable of writing it and by the type of human experience for which it can serve as a channel.

In structural communication a formal mechanism is provided for the involvement of the recipient. This mechanism gives scope for the author to write different episodes to be read by the student according to the kind of viewpoint or misunderstanding which arises in him. Thus the writing becomes adaptive to the recipient's own train of thought. The way in which the author is able to anticipate the recipent's reactions is by inviting him to investigate certain specific problems and indicate his views through a selected vocabulary provided by the author. The correctness of the author's anticipation is enhanced by the trial of the material on typical samples of recipient population and revising this material in the light of this trial. This is part of the rehearsal process for the communication.

So far in our explanations of the method we have adopted a necessary restricted version of structural communication in the book form. This version. although limited, has enabled specific writing exercises on various topics to be undertaken. As our work progresses we shall develop and extend this version to include face-to-face groups. books, the 'Systemaster' machine, and computer aided systems. To prepare the way for a more detailed description of the operational. involvement of the student we shall give a description of the form structural communication now in use; but with the reminder that a far more general form is under development. A specific example is shown in Appendix IV.

1.6. THE FORM OF A STUDY UNIT
Each structural communication is taken to have a fundamental unit which is indivisible if the structural character is to be preserved. Since this unit is provided as a means for active participation of the student it is termed a STUDY UNIT. Each study unit has six inter-dependent sections which are not restricted to being used in any single sequence, or to being studied only once. Each section, besides differing in content relating to the theme of the study, has a particular psychological role in providing conditions of challenge and response. The sections are named so as to indicate to the student what they are intended to convey, although it is difficult to find single words which adequately describe them. Each section is set out below with an attempt to formulate its psychological role.

INTENTION - Facilitates the recognition by the recipient of the indentity of the theme of study. From the recipient's previous experience it should enable him to anticipate the goal of the study through already established understanding.
PRESENTATION - Facilitates the interaction of the recipient's mind with the perceptual or conceptual material specially associated with the theme.
INVESTIGATION - Demands the expansion of the recipient's appreciation of the theme by involving him in problems.
RESPONSE INDICATOR - Demands the concentration of the recipient's enquiry on interrelationships within the limited field delineated for the theme of study.
DISCUSSION - Assists the ordering of the recipient's perceptions or reflections around the theme by the introduction of 'dialogue' material in response to his response.
VIEWPOINTS - Assists the freedom of choice of the recipient to adopt his own viewpoint towards the theme (not necessarily that of the author) by the statement of the author's position in greater depth.

All of these sections are variable in length and can be interpreted fairly broadly by a given author. The arrangements of these sections in the book or the machine is largely dictated by technical and economic requirements, but one overriding principle governs their organization. This principle is that the sections are non-linear. Technical restrictions of both the written word and mechanical display require some linear or sequential layout of the sections. but the participant in the communication is involved in these sections in a recurrent or cyclical way. not restricted by written word or mechanical display. Another way of describing this principle is to say that structural communication adopts a strategy which leaves open to the student a great variety of tactical movements rather than prescribing one fixed sequence of movements.

1.6.1. INVOLVEMENT IN A STUDY UNIT
The study unit is not inherently limited to verbal/conceptual studies, although this is the application we have investigated most thoroughly to date. For example, the sections could be proposed thoughts in the mind of a group discussion leader which have never been written down. Alternatively, the interaction in the presentation could be brought about by practical exercises (e.g. in laboratory or studio). However, most of our work to date has been concerned with verbal/conceptual material in mathematics, science and history. Some tentative experiments in art appreciation have indicated the successful operational involvement of students in perceptual/judgmental studies.

Essentially, the art of involving the recipient consists in getting him to do something which is as near as possible to what the successful exponent of a discipline does. If the study is mathematical, then the study unit should provoke mathematical reflection. If the study is in scientific theory the study unit should provoke conjecture on the organization and application of concepts. If the study is artistic then the study unit should provoke visual scrutiny and aesthetic judgment. Survey of results from field trials of study units indicate that the more effective communications have had this quality of active involvement,. Perhaps the most successful units are typified by the report that the recipient has found no new knowledge in a study unit, but has found exciting insights provoked by that study unit into knowledge be already possessed.

The approach, then, to the design and construction of a study unit is one of simulation. In the absence of the real thing the attempt must be made to get as near to the real thing as possible so that the probability of the reality emerging is as high as possible. For example, in a study of physical geography, the facilities to make a field expedition to remote but interesting geographical features may not be available. However, the best possible perceptual equivalent (e.g. good colour photo-graphs) is supplied and the various sections are used to challenge the student into observing and interpreting these features with the concepts of the geomorphologist. In attempting to meet this challenge the recipient may come to a better understanding of the way a geomorphologist works than he would if simply given a text-book presentation of the conclusion geomorphologists have come to.

Some of the more detailed ways in which we have tried to elicit the participation of the recipients are now described. It must be borne in mind white these descriptions are being read, that study units are not isolated experiences for the students. They take place in an environment of work and interest and cannot be evaluated objectively without taking the quality and relevance of this environment into account.

1.6.2. INVESTIGATION
The primary tool for student involvement is the investigation. As indicated in the part of this paper on the psychological points of departure, reading efficiency is always well under 100% for the recognition of meaning and the seeing of interrelationships. This efficiency is only increased to a limited extent by repetition, since at each repetition the mind is more likely to recognize only what it is now 'conditioned' to recognize. We consider that the administration, after limited repetition. of a 'shock' has a more expansive effect on the recognition of meaning. The theory of this procedure is discussed in Section 2.3.3. below as the Principle of Challenge. The shock is administered by means of a set of problems in the investigation section. The student is required to offer solutions to those problems which are constructed in a special way in relation to the theme.

Any theme chosen for a study unit should form an integral whole. However, the significance of the whole is rarely immediately apparent to the student. The theme must be approached from various sub-themes within this whole so that the viewpoint on the whole can emerge clearly. Sub-themes are made the subject of the problems. So far we have standardized on four problems per study unit, but this again is an arbitrary limitation imposed by the cost/effectiveness of books and machines. What is inherent to the method is that there must be more than one problem, and that any problems should bring out the qualitative importance of a sub-theme, rather than cover it quantitatively.

Each problem sets a challenge to the recipient's understanding of a subsidiary theme. In meeting this challenge he is required to discriminate which areas of the general presentation are relevant and see how these areas fit together. The vocabulary provided within the response indicator section in order to register his reaction to the problem covers the whole structure of the theme. Thus the recipient is required to make a statement which indicates his grasp of the domain of relevance of the problem. On the basis of this response the dialogue can take place since different types of response lead to different rejoinders or comments from the author. These comments also have a particular role in the involvement of the recipient.

The discussion section, which comprises all the comments and the ' information' as to how and when they are studied, simulates an exchange which is essential to any communication procedure which Permits exploration followed by progressive approximation. Having committed himself to a composite statement the recipient, we shall suppose, finds himself presented with a comment which either denies the compatibility of some of his sub-statements with the problem, or suggests he may have omitted certain key factors, or encourages him to go on looking in the same vein. Any such comment, rightly constructed, returns him to the problem to reflect again and make a fresh or revised attempt at the problem. Thus, if the recipient does not immediately see the demands of the problem, he can follow up hunches, even guesses, and refine these in the light of the comments.

This progressive approximation is clearly distinguishable from simple right/wrong situations and from direct correction situations. The indications supplied by the author are indirect food for reconsideration, not a direct statement of grounds for error.

Moreover, the recipient is not debarred at this point from access to the remaining three sections-intention, presentation and viewpoints. The recipient can use his discretion to return to the presentation to interact with the basic information, to return to the intention to remind himself of the objective, and to look up the viewpoints to clarify the position the author is taking in relation to the theme of study. Thus there is a balance between the prescribed structure of events between the investigation, response indicator and the discussion sections and the open structure of events between the intention, presentation and view-points sections.

The operational involvement of the recipient depends, in structural communication, on the necessity for him to interact intelligently with the conditions provided by a study unit, and to direct his own looking or thinking in a tactical way derived from the interaction of his own interests and insights with the dialogue anticipated and even rehearsed by the author. His response to this challenge is his total experience of enquiry throughout the study unit, and this is reflected in his more limited response to the particular investigation put before him.

1.7. APPLICATIONS
In the collaborative work with schools and universities we are concerned with two things. The first is the validation, with typical student groups, of Study Units, in various areas of the curriculum. The second is experimentation with typical student groups on methods of using structural communication to increase the efficiency and flexibility of the educational process. What we mean by the validation of Study Units is described in the next section. Some details of collaborative experimentation are given in Appendix II.

Arising from this collaborative work a number of applications are emerging where we feel confident in saying structural communication can do a valuable practical job. A brief description of these applications now follows.

1.7.1. UNIVERSITY LEVEL
Two applications seem of immediate value. Both have bad some verification in practice, although it is too early to evaluate the full implications since experimentation is still under way.

1.7.1.1. Levelling up and Grounding.
Freshmen to any undergraduate course may have very diverse academic backgrounds. Although all students will have the requisite formal qualifications for entry to a course of study, there is frequently a considerable uneveness in their knowledge and experience. Since the passing of an examination requires only. say, 50% knowledge, and not the same 50% for each student, it is found impossible to make facile assumptions about the background of the students.

University teachers often find that their initial courses misfire since the foundation upon which to build are not present, and yet this is not discovered until much of the first academic year has passed. Another factor is that understanding should become more important in undergraduate studies and often 'A' levels can be passed by memorization rather than insight. Thus the teacher cannot rely upon 'A' level results to tell him the whole story of the student's approach to his studies.

Structural communication is being used to tackle both the diagnosis of student potential and the levelling up of basic studies, in the same operation. The students are given a course in which Study Units on basic themes are given to work on in their own time. The tutor collects in response information which is processed in time for him to gain an impression of the student's strengths and weaknesses before the next tutorial. In this tutorial both students and tutor are better prepared to make the most of the time, especially in dealing with open questions not dealt with in the Study Unit. In working through the Study Unit on their own the students have already got some feedback on their understanding which can help them in their wider reading.

1.7.1.2. Digesting Lecture Courses.
The efficiency of intake of material presented by one-way communication is very low. Even if students work hard at digesting lectures, they often end up trying to memorize them rather than trying to understand them.

Structural communication is being used to provide a stimulus to study and reflection as an integral part of a lecture course in science. Two modes of operation are under trial. In mode A, after a preliminary lecture, the students are given Study Unit 1 to work through during the week. At the next lecture the lecturer discusses the material and answers Students' questions arising from the Study Unit. Study Unit 2 is handed out and the process repeated. Meanwhile the lecturer processes the response information on Study Unit 1 and gets an impression of how the subject is getting over. Used in this way students have reported the Study unit a valuable generator of work programmes for using the reference literature. In mode B, a lecture 1 is given which is equivalent to an expanded presentation. The students then work through Study Unit 1 in which the presentation section is equivalent to a precis of the main points of the lecture. They then return to discuss outstanding difficulties at the commencement of lecture 2.

Where students are not on completely new ground and have adequate reference facilities in their private study, then mode A is effective. Where the ground is completely new and reference material is not easily available, then mode B is preferred. Either method has proved to be a dynamic stimulus to the students.

1.7.1.3. On-line lecture feedback. Another application which has been tried in the university field is the organization of the lecture event itself as a structural communication. The lecture is divided into three parts. The first part is a presentation of material. In the second part the students are given an investigation and response indicator on the theme of the lecture. The third part is a live discussion of the responses of the students by the lecture, rectifying difficulties in the communication.

1.7.2. SECONDARY SCHOOL LEVEL
Most of the applications so far in the secondary school are in the 16 to 19 age range of the sixth form. We are commencing with experiments in using structural communication with the 13 to 16 age range during this academic year.

1.7.2.1. Third year sixth form work-special and scholarship work.
Very often this does not easily fit into Vl form teaching schedules since numbers are small and staff are not available for very much time. At this stage the student is capable of a much more self-directed study but still lacks guide lines on how to digest material available in text-books. If this transition is difficult the student may content himself with memorization alone or merely drift. It has been found that structural communications helps the student to develop a sense of method and direction in his studies. He becomes less dependent on the teacher's direction and more able to think and criticize for himself. At the same time, be gets an automatic verification of his progress.

1.7.2.2. Tutorials and discussions.
The sixth form teacher rarely has time to prepare tutorials and probe the obstacles to progress in his students. It has been found that Study Units may be used as a focus for group tutorials. Although students work through the investigation and response independently, in pairs or as a group, they are able to, chat about what they are thinking with the tutor who can elaborate at; will on the theme. The tutor can quickly diagnose which way. the students' minds are working and can discuss difficulties with the students, as a whole. If he wishes to deal specially with one student the others get discussion support from the Study Unit itself. It has been found that this method of working has beneficial effects on students' discussion ability in other situations.

1.7.2.3. Private Study and Revisions.
One of the students' main difficulties is coming to see that he has blockage in his understanding. It has been found that structural communication helps the student to realize more quickly where he needs to develop his grasp. He can theN follow this up within or outside the Study Unit. If not satisfied he is in a better position to formulate questions to ask his teacher. Thus revision and private study spent working in this way accelerates revision and relates it more to meaning than to memorization.

1.7.2.4. Feedback through homework.
The setting of homework and marking it to get impressions of the students' work can easily degenerate to a time-filling routine in the pressure of contemporary school organization. Ideally home-based exercises should be a real challenge to the student and provide a motivation to understand a subject better. Ideally the marking of student homework should be a means of assessing their capabilities, individually and collectively, and using this assessment to re-adjust their programmes of work. To some extent structural communication fulfils this ideal. The Study Unit is a reasonable 'package' for homework and through study of the response information the teacher gets a much quicker assessment of the students than with conventional marking procedures.

1.7.2.5. Special Problems in Understanding.
In all curriculum areas there arise at different stages critical transitions of insight which the teacher is unlikely to cope with in a classroom situation. Especially with average and below average pupils the provision of a technical aid for fostering the insight is desirable. Structural communication can be effective here as an additional tool for the teacher to present a carefully graded challenge in such a way that he can move about the class giving individual diagnosis and attention where necessary. The writing of study units and topic books for these special problems in understanding could be a very important role for structural communication.

1.7.2.6. Further Applications in Preparation.
Although these are the main applications in which collaborative experimentation is going on, there are many other applications there are many other applications which are at the planning stage. One important experiment with the teaching machine is the use of the machine for computer guided tutorials (C.G.T.). Even the sell-contained Systemaster has a computational / data processing capacity which is usually only associated with digital computers. Thus the presentation and response console can be in a form available to a group. The group as a whole have to work through the material, endeavouring to reach agreement on a response which they wish to be evaluated. This procedure has been demonstrated to a group of headmasters and heads of departments, and was highly commended.

Another application is the association of study units with practical work in studio or laboratory. For instance, in science one of the difficulties is in getting students to study and consider the implications in an experiment and resultant data. In this case the presentation section can be a practical study and the investigation present problems in the interpretation of results. The discussion comments could include further practical operations to be done to clarify a problem.

More elaborate experiments can be attempted when adequate data processing facilities become available. At the moment the feedback from processed response information is too slow and too simplified for much work on adaptive teaching strategies and progressive assessment.

1.7.3. MANAGEMENT AND ADMINISTRATION
The interdisciplinary workshop (Appendix II ) demonstrated that. public administrators, business executives, medical practitioners and engineers can readily grasp the principles of the method once they have actually used it themselves. It was agreed by the participants that many of the situations in which a grasp of the author's intention arise hi management were suitable for the structural method. A list of communications experimented with is given in Appendix II One or two instances are developed below.

1.7.3.1. Operational Instructions.
It is often necessary to convey a general appreciation of the problem, including the description of various situations which may arise and instructions as to the appropriate action to be taken in each case. The presentation will include a statement of the main, overriding objectives and subsidiary objectives to be attained if circumstances permit. If the operation is well-planned, the whole will form a coherent system in which it is vitally important to recognize' the way in which each partial decision and operation will affect the attainment of the main objective. The recipient of instructions requiring not only readiness to adopt alternative paths, but also ability to allow for consequential adjustments of the main plan, must be able to recognize a structure of intention. It is by no means enough that he should learn the instructions by heart and be able to describe exactly what is required of him in each set of circumstances. He must be able to see the connections between the alternatives. In order to verify successful communications he must show that he has recognized the structure of the operation as a whole and in all its parts. The author of the instructions provides for this by constructing a matrix with as many elements as are needed to identify every significant factor in his presentation. He then specifies a set of problems and asks the recipient to select out of the response indicator, firstly, those elements which are directly relevant to the operations to be performed to solve the problem and secondly, the elements that though not directly involved in the action will nevertheless be substantially affected or changed by what he has to do... Usually the communication is addressed by a superior to a number of subordinates each with executive responsibility.

1.7.3.2. Testing of Ability.
The method can be used in the reverse] order to evaluate various characteristics for such purposes as appointment, promotion, selection for training and allocation to specific duties. In this application the message is designed to be a challenge to the candidates' ability to make broad judgments. The time taken can be specified and the challenge can be graduated in such a way as to bring out, in quantitative terms, a range of personal characteristics that are missed in most of the assessment techniques now in use. This is an important point as there is widespread agreement that the intelligence and capacity tests used in industry psychology are proving inadequate and must be supplemented by personal interviews to an extent that largely nullifies their value. The advantage of using a structural communication test lies in the understanding that it engenders. The interviewer is provided with an assessment that gives indications of the strategic ability as well as the mental and other training the candidate has received.

1.8. CURRICULUM TASKS
The technique of structural communication has been used by several authors to produce communications in different disciplines. including mathematics, chemistry, physics, biology, history and art appreciation. These have been designed for a variety of uses by teachers and students working at the '0', 'A' and 'S' levels, and by first-year undergraduates. In each case the authors have found that the method of structural communication has brought something specific to the teaching of their particular discipline. Each of the following contributions has been written by one (or more) of the authors concerned with that discipline.

1.8.1. MATHEMATICS
The pursuit of mathematics is the investigation of structure in itself. This viewpoint has recently begun to influence the teaching of mathematics, resulting in a study, named 'modern mathematics', which is characterized by a new awareness of fundamental assumptions and explicit methods. Thus, for example, modern algebra teaching will investigate how different structures can be associated with a set group, field, vector space, etc., and then use these as tools to investigate the different number-systems and geometries.

Traditionally, the teaching of mathematics has proceeded by presenting the student with certain materials and then leaving him to work out for himself a multitude of 'examples'. In this way he is supposed to come to an understanding of the particular mathematics involved. What, in fact, he gets from this practice is a grasp of the notation and mechanism of manipulation-which are both very necessary-but not an insight into structure. The traditional approach is very useful for teaching, say, methods of integration, methods of solving differential equations, etc., but something more seems to be required for teaching, say, the corpus of modern algebra.

The method of structural communication has been applied to the teaching of modern algebra, with the intention of providing a supplement to existing teaching methods. The aim has been to help the student come to an unnderstanding of the variety of structures which can be' associated with a set by the introduction of specific axioms and operations upon sets. The procedure can best be understood through an illustration.

In a communication on 'Rings and Fields' the student is presented with the requirements for a ring structure and is given examples' of different kinds of rings; he is then introduced to the extra requirements for a field. Throughout the presentation there are 'self-teaching exercises', designed to facilitate a grasp of terminology, notation, and mechanism. The student then proceeds to an investigation which consists of the following four problems:

1. What are the requirements of an additive group?
2. Having ascertained that a set and + are a group, what must also be true before ring structure is established?
3. What items of the response indicator are necessary for a field but not essential for a ring?
4. There are some unusual features of Boolean algebra (for example complementation) that prevent it from qualifying as a field.

Find the items of the response indicator which
(a) Apply to Boolean algebra but not to fields.
(b) Refer to exactly similar statements for both Boolean algebra and for finite fields (take X and * to be synonymous for this purpose).

The student attempts to answer these questions by selecting a number 9f statements from the response indicator, which is a chart displaying the following twenty items:

1. there are no divisors of zero
2. commutativity for +
3. 1 + 1 = 1
4. distributivity
5. * is a binary operation
6. a + bc = (a + b) (a + c)
7. a finite set of elements
8. unique inverses for * except for the zero element
9. idempotant law
10. unity element
11. + is a binary operation
12. associativity for +
13. unique inverses for +
14. regular elements
15. commutativity for *
16. De Morgan's Laws
17. associativity for *
18. zero element
19. law of absorption
20. an infinite set of elements

In responding to a challenge of this kind the student is required to build up mathematical structures for himself, and to appreciate the differences between the various structures. Thus he is guided towards an under standing of the nature and method of mathematics.

Mathematics teaching at the secondary level has lagged far behind the practice of mathematics. It has lagged not only in content but, and this is more important, in method. The way mathematics is taught must reflect the way mathematics is practised. This requires that the student becomes familiar with the meaning and use of operations, laws, definitions, axioms, primitive terms, relations, proof, theorems, etc. The method of structural communication provides a useful teaching aid to mathematics since, by means of the investigation and response indicator, these different features of the mathematical method must be brought together and interrelated by the student himself. In this way the method of structural communication contributes to teaching mathematics effectively, both as a tool for those who wish to use it, and as a thought. game for those who wish to participate in its creation.

H.B.

1.8.2. CHEMISTRY
Chemical science tends to diversify and increase its factual detail. Learning chemistry very easily leads to the compartmentalization in the student's mind, and the structural coherence of the discipline is lost. One major task of chemistry teaching, especially in the sixth form and first year at university, is to counterbalance this explosive tendency with a synthetic study of interrelationships.

Structural communication can provide one means for this synthetic study. The multi-perspective structure of a study unit enables one to communicate the interrelationships between ideas and data which are treated, in conventional text-books under quite separate headings. It is also possible to build a series of study units which permits the gradual build up in the student's mind of different models for thinking about the data. It is very difficult to communicate the role of model building in. say, the theory and interpretation of chemical structure. The discussion technique permits the origins and assumptions of theoretical pictures to be dealt with as and when they arise in the student. Problems requiring the operational use of mental models can stimulate the student's grasp of the use of theory.

Structural communication also permits one to bring out the judgmental area of chemical science. The fact that all views of chemical theory are based on human judgment and may and do prove to be fallible can be put over since an argument structure can be written into a study unit. Coupled with this is bringing the student to awareness of the non-universality of concepts in chemistry. The study unit enables one to get across the sophisticated notion that ideas have a limited domain of applicability.

In much chemistry teaching the theory is presented in far too dogmatic a manner. This is not so much a fault of the writers as defect in the conventional text-took as a means to develop the student's interpretive capacity.

A.M.H.

1.83. PHYSICS
The teaching of physics is effective only in as much as it issues in. increased manual and mental abilities, together with a critical appreciation of the 'mechanisms' of physical methods. This requires more than the unreflective absorption of information and the repetition of well known experiments. Among other things it requires that the student be able to think physically; in other words, to use the languages of physics.

This faculty is difficult to train; it does not seem to arise naturally from using the traditional text-book and, within the present organization of education, the teacher has little time to encourage its growth. As a result, the development of practical understanding is so often a 'hit and miss' affair. For example, students are introduced to the idea of energy at an early stage, yet years later they do not know (outside of the specific examples with which they are familiar) how and when to use the concept 'energy'.

Practical research projects, undertaken in conjunction with teachers and students, have shown that the method of structural communication provides a technique for developing the capacity to use the various languages of physics effectively in practice. In order to use this technique for communicating such an understanding of a particular topic, it is necessary that the student should have previously gained knowledge and information, both practical and theoretical, of the topic in question. This is usually the result of classroom and laboratory work, together with the assimilation of text-book materials. The method of structural communication does not supersede this lead up.

1.8.3.1. Optics.
The author of a series of communications on geometrical optics was concerned with the understanding of a coherent method of approximation for explaining, predicting and applying the optical phenomenon of image formation. For this purpose three different kinds of statement were distinguished:

(i) phenomenal statements: those are ordinary-language statements describing observations and operations; for example, 'the air-cell was rotated until a position was reached when the light was suddenly cut off'. Such statements are used to compose the description of an experiment presented in a laboratory report.
(ii) physical statements: these are specialized-language statements which are formed to interpret experiments and to construct physical models and theories. Such statements employ the concepts peculiar to physics; for example, 'the emergent light path is parallel to the incident light path'. Here the key concept is 'light path' which, in the context of the given statement, implies the notion of the propagation of light together with the principle of rectilinear propagation. Thus, any physical statement has a 'theoretical' component which is absent from phenomenal statements.
(iii) mathematical statements: these are the quantifications of a physical relationship and their consequences, together with the formal algebraic assertions of the appropriate calculus. An illustration is provided by the mathematical expression Snell's law, viz:

' n₁ sin ₁ = n₂ sin ₂ '

Much of the art of physics lies in finding correspondences between the phenomenal, physical and mathematical levels, and being able to form that coalescence of the three kinds of statement which is most appropriate to the state of affairs under investigation. To foster the development of this art in a student it is necessary to distinguish clearly the different kinds of statement with which he will have to work. This is not easily done by the conventional text-book where, because of its 'flat' and linear form, the different levels cannot be effectively distinguished, so that physical statements often appear to be the same in kind as phenomenal statements. This collapse of the physical into the phenomenal level is a source of confusion which drastically hinders the true understanding of the way physics works.

However, the necessary distinction between different kinds of statements can be communicated. Structural communication is operationally cyclic in form (in the best examples produced so far it is spiral), wand so provides an extra dimension which permits distinctions of kind and level to be made within the form of the communication-not merely erected as part of the content. This key point of the method can be grasped through an illustration.

In a communication, the message of which is 'The Behaviour of Light at a Boundary', the student is presented with heterogeneous material on the refraction and reflection of light. When he has familiarized himself with this presentation of material he enters into an enquiry which consists of four problems. In one of these, for example, he is given a description of an experiment performed together with the Observations made, and asked to propose an explanation of the crucial observation in the experiment, and how this can be used to calculate a particular physical magnitude (the experiment described is the use of the air-cell refractometer to measure the refraction index of a liquid), The student attempts to answer this question (the others are answered by the same method) by constructing a combination of statements from the response indicator, a matrix of twenty (or fewer) individual statements. He struggles to achieve a 'rightness of fit', both of the individual statements selected and of these statements with the description of the experiment. In this he is helped by the discussion guide and comments, which circulate him through the question and indicator, injecting pointers at each turn. The result is a spiral converging towards the answer. When this is attained, the student is presented with further material which he is now able to appreciate because of his own participation in the communication.

The description of the experiment provided in the problem consists entirely of phenomenal statements; whereas the response indicator contains only physical and mathematical statements. The mechanism; of the student response is to operate on the problem (phenomenal state-, meats) with the indicator (physical and mathematical statements) to produce, through the integrative action of the discussion, an explanation, i.e. an integral statement composed of individual physical and mathematical statements (six individual statements are required for the example cited). This procedure is summarized in the following diagram:

Operating on the problem with the indicator in this way simulates part of the 'real life' activity of providing a physical explanation of a state of affairs, which can then be quantized so that physical magnitude(s) can be calculated. In a sense, the mature physicist carries the indicator in his mind, though in a relatively unformalized manner. The students understanding can only evolve by practice in operating with statements of different kinds, and one way in which this can be achieved is by providing an external indicator. This only applies to the 'normal problems of physics where it is required to use a known and established physical language. The difference between a normal problem and a creative one is that, in the latter case, the indicator is empty. The first task is then to provide a new language for physical statements.

H.B.

1.8.3.2 Thermal Physics
From school teaching about 'heat' to thermodynamics and its role in modern physics as it is met in university a considerable 'jump' has to be made by the student. This jump is not simply the addition of concepts and information of an unfamiliar kind. What is needed is a break with old habits of thinking which is not confined to the subject-matter of thermal physics.

For instance, at school the student is largely trained to think in terms of a one-to-one correspondence between concept and experiment. This is more especially the case when the 'experiment' is a demonstration concocted for pedagogical reasons and the actual experimental foundations are glossed over. Actual experiments are presented in an unrealistic way. For example, most text-books treat Joule's experiments as 'showing' that 'heat is a form of energy' and render it difficult later on for students to appreciate the distinction between mechanical and thermal energies.

In physics, the basic situation is that systems of concepts are harmonized with systems of operation. The notion can be indicated by enumerating a series of points.

1. Concepts such as 'work', 'heat', 'temperature', etc., are useful by being components of a system. For example, 'work' and 'heat' have to be conceptually related by the concept of 'internal energy The conceptual systems are not mutually exclusive, but overlap, and are also integrated in fundamental principles. Thus the conceptual relationship of 'heat' and 'work' is established in the first law of thermodynamics The difference in kind between thermal and mechanical energy is expressed in the concept of 'entropy' and it is on this basis that 'temperature' can be exactly defined, etc.
2. Systems of Operation include both mathematical structures and experimentation, In both these cases, the basic situation is that some component is strictly controlled while another is left free to vary. Thus, in practice, the 'isolated system' which is a fundamental concept in thermal physics, can be realized by intelligent control over the environment of the system and not over the interface with the environment.
3. Neither of the two kinds of systems are closed.
4. The two systems coalesce in the case of measurement. Measurement is one of the biggest stumbling blocks in the teaching of physics. Part of the reason for the difficulties is that the coalescence involved in a measurement complex is brought about by a structure of intentional acts.

The situation in physics has to be simulated in the learning of physics. At the end of the sixth form, students can begin to strip concepts of their spurious associations and come to see them in combinations. At the same time, they can pay greater attention to the complexities and problems of actual situations. The present situation in sixth form physics teaching is very far from realizing these things. In the Text on Thermal Physics material is provided that can give a student a taste of new ways of thinking. In trials, it must be noted, some students have disliked the taste. Some of the techniques involved are:

1. The Response Indicator is cast as an array of items which indicate concepts of varying generality. This array is confronted by a series of problems which demand conceptual interpretation of actual or typical situations.
2. Where the concern is with measurement, the array of items may combine both conceptual and operational elements. The problems are then cast in a practical form which require the right combination of concepts and operation.
3. In cases where the student has to be brought to a new awareness of some element in Thermal Physics, to which he has a conditioned attitude, the Presentation is a combination of conceptual and operational material which can only be properly connected together by answering the problems.
4. The student is encouraged to 'speculate' in fields unfamiliar to him in a context where 'failure' is taken as a positive means to discussion.

A.G.E.B

1.8.4. BIOLOGY
Much of the understanding of descriptive biology depends on understanding the significance of similarities and differences between different categories, e.g. groups of animals in taxonomy, different processes in biochemistry, or comparative studies in the areas of anatomy, physiology or embryology. The method of structural communication is an excellent tool for communicating data of this sort, for significances which may have been missed from a simple linear presentation of material can be followed up and emphasized in discussion comments. It is also possible to present 'research' situations, for example, an observation (presented in a question) may arise from one of several causes, or a combination of these: A student can be brought to think in this way with the help of the discussion comments which can be used, either to present possible ways of testing hypotheses, or to emphasize the emergence of hypothesis from the phenomenological situation. On the other hand, it is possible to turn this situation round so that he is given a hypothesis in the question, and required to select from the response indicator all items which would arise if this hypothesis held, and to test this against items which would arise if another hypothesis were presented. This method can also be used to advantage in producing an understanding of how a number of elements in a complex system (an organism, perhaps) may work together for a certain purpose, for example, excretion or respiration. This means that an awareness of the interactions and mutual compatibility of sets of elements of equal status in an integrated organism can be engendered, or an understanding of the relationship between actions occurring to the same purpose but at different levels of examination can be fostered. For example, there may be confusion or misunderstanding of the relevance of processes occurring at the molecular level to actions which proceed to the same end in a whole organism or in a social group. The author of a series of biological communications has been able to use the technique of structural communication as a flexible method of approaching a topic from a variety of standpoints including those described above.

J.B.

1.8.5. HISTORICAL STUDIES
The immediate apparent advantage that structural communication offers to the teaching of history is its capacity to draw the pupil to see a series of events, movements, ideas, as an interrelated whole rather than as a simple linear sequence of descriptive facts.

For 'O' and 'A' level pupils it is necessary that as many interrelationships as possible be seen and that they learn to allot significance to items in relation to a particular viewpoint or within a particular area being considered. In the Study Unit Presentation, as in a conventional text-book, it is necessary to lay out the topic being considered in a linear sequence to expose and explain it. Unlike a conventional textbook, however, using the structural communication method it is then possible to structure an investigation in such a way that the pupil is drawn to recognize interrelationships between what may have appeared from the linear presentation, to have been quite distinct items. By careful use of the investigation it is possible to communicate insights into the period, into motives, causes, results. etc., that might well have been completely unnoticed otherwise. For a conventional text-book to outline all the connections, which it is possible by this method to get the pupil to make for himself in a relatively short time, would take up an enormous amount of space, be extremely cumbersome and, almost inevitably, extremely boring.

In one Study Unit, for example, it is possible to give an outline of the Reformation in England, and then extract twenty essential elements for the Response Indicator. Four separate problems can then be posed concerning such issues as-How the Reformation strengthened the King; What caused the attack on the Church; Why the clergy did not put up a better fight against the lay-dominated Reformation; Why Henry dissolved the monasteries. From working through such problems and re-assessing the items in the matrix for each problem separately the pupil learns that 'facts' have varying significance in relation to whatever is being considered and learns the essential lesson that 'History' is not a static, settled thing, but is like a kaleidescope, changing and moving continually.

1.8.5.1. The Challenge
The particular nature of the 'Challenge' which it is possible to set the pupil in a Structural Communication provides an ideal vehicle for getting him to make the imaginative leaps which are necessary to understand the differences in the meaning attached to ideas, institutions, words, etc., which are current today and the meanings and implications they had in the period being considered (e.g. 'Liberty,' 'Parliament', 'War').

Again the method can call for similar imaginative leaps over a wider area. It is possible to build up an almost dramatic involvement of the learner with, for example, the problems facing Henry VU on his accession to the English throne. By presenting the pupil with a challenge which puts him in Henry's place, and calls for a response which asks him, for example, to decide how it would be possible to impose and1 enforce royal law at a time when royal authority was at a low ebb, and the throne bankrupt, it becomes possible to give the pupil a more complex and meaningful picture of just what sort of difficulties Henry VII actually overcame.

1.8.5.2. Independence of Judgment
It is becoming clearer that structural communication can be valuable in helping pupils to bridge the gap between their reliance on the teacher or text-book for facts and interpretation and the realization that their own view of the world, and thus of History, is as valid as anyone else's. This usually takes place in the years immediately before '0' level. Normally the pupil lacks the simple mass of evidence which is necessary to be able to successfully defend his viewpoint and has not yet developed a sophisticated method for marshalling evidence to back up an argument. In a structural communication, however, he is on equal terms with the author since there is in the Response Indicator a restricted language which is common to the pupil and the author. The simple quantity of facts in the Presentation frame is restricted and so again puts the student on equal terms. He can respond to the challenge .of the problems set, combine items according to his personal assessment, and his responses can be handled in the Discussion Comments in such a way that the validity of his personal view is respected and indeed encouraged, while at the same time teaching the basic historical discipline of respect for' facts'.

Further advantages of the method which are incidentally becoming clear as writing continues are:

1.8.5.3. Essay Writing
For examinations the art of essay writing is still necessary, and children still fail History exams, for reasons which are in no way connected with their historical knowledge and understanding, but rather matters of organizing that knowledge and showing that understanding in a written examination. It is possible to organize a Study Unit in such a way that the Investigation will ask the pupil to assume that he is to write an essay on e.g. 'The significance of the career of Thomas Cromwell'. Then he is invited to use the Response Indicator to compose a group of items around which paragraphs could be written. By this means the pupil also learns a method which he can apply, apart from the Study Units, for constructing essays which include all the items he considers relevant.

Because of the flexibility which a structural communication insists on in considering problems from a number of different viewpoints and the practice it teaches of looking at each problem in a variety of ways, it becomes less likely that in exams pupils will miss out questions they have the knowledge to tackle but misunderstand because of the way the question is worded. For example, pupils who have learned about Henry VII have been known to miss a question asking why it was that Henry VIII succeeded to the throne without trouble. A pupil used to working at structural communications would clearly see that this question is concerned with the measures taken by Henry VII in strengthening his kingship, whereas a pupil used to the ordinary routine of learning in a linear fashion without synthesizing what has been learnt is more likely to think that because the question asks about Henry VIII, then it has to do with the things that in his mind come under the category of 'Henry VIII'.

1.8.5.4. Understanding History

It is not uncommon for pupils who are good at assimilating information and instructions to do very well at 'History' by simply repeating factual accounts of what has been heard or read, without necessarily understanding what it is all about. Moses Maimonides describes such as 'People whose learning is not their own'. To such pupils History is largely useless, except as a memory exercise. It is normally hoped that whatever pupils learn from conventional History text-books will be synthesized, and clearly some synthesis almost invariably takes place. But in the Structural Communication method this synthesis becomes a central part of the activity, increasing understanding and communicating insights.

K.E.

1.8.6. AESTHETICS
When I was first approached to try my hand at writing a script in the field of art appreciation according to the structural programming method, I told myself that this simply could not be done.

Nevertheless, I made an attempt, and I immediately encountered a strong feeling that even if it were possible, it still should not be done. When I examined this inner objection to the task I found that it really lay in a reluctance to influence the aesthetic judgment of another. I still think that it is of paramount importance that there should be no final direction of that judgment. An individual's judgment in this field can be clarified and informed, but it must not be subjected to dictation or persuasion. if it should succumb to either of these it will become insincere, and an insincere aesthetic judgment is no longer an aesthetic judgment at all; it is then at best a product of ratiocination, at worst of fashion or snobbery.

On the simplest level, if someone likes a picture and you do not, he may be persuaded, if he is unaccustomed to making aesthetic decisions, that your judgment is better, that it is supported by 'informed opinion', or even by market value, and that his own judgment must therefore be abandoned as untrustworthy. His judgment has been invalidated, his capacity for aesthetic response has been weakened.

This notion that there is a fixed scale of accredited values must be put aside. Structural communication is a powerful device, not to be employed as a technique of persuasion in the turbulent area of judgments about 'good art ' and 'bad art'.

How, then, should it be employed; what is it able to do? I believe that what it can do very well is to present the student with a network of facts and ideas within which can be found the values upon which he can reach judgments for himself; in addition, it can help him to perceive irrelevancies for what they are.

It must be borne firmly in mind that the visual arts are visually based. The language that they speak is a visual language, a visual. communication. No amount of fact or theory is of any real significance in this field unless it is relevant to the faculty of sight, and not only sight, but insight. Knowing where to look for the significance of a work of art, and how to look at it, in order that it may make its visual communication, is a benefit that may be conferred by teaching.

For example, someone may be accustomed to perceiving the classical elegance of Greek sculpture; he knows how to look at it, assisted by some knowledge of the culture which gave rise to it. Presented with a piece of African carving for the first time, he may simply be unable to perceive the aesthetic elements in it. Looking for the aesthetic ingredients of the Greek, he will be blind to those of the African. Judicious information and questioning may equip him with the discernment to appreciate the African also. Furthermore, perceiving the properties of both, he may deepen his insight into the essential nature of the art of sculpture in general.

I purposely included the phrase 'judicious questioning', for this is a highly important part of the technique, not only in order to test the result of teaching and understanding. but to assist them both.

It is possible, for instance, to tell a student that the Impressionists were primarily concerned with the properties of light and colour. He may understand this and correctly answer verbal questions about it without having really perceived how this is so when looking at their pictures. It is possible, however, so to frame a question that the answer can be obtained only by looking for it in the pictures themselves, and looking for the actual visual evidence. Furthermore, he will sharpen his perception of these qualities by the very act of seeking them.

But where does the particular advantage of Structural Communication come in? While I have been writing these comments and framing them in my mind, I have been uncomfortably aware that I have been faced with the necessity to place them in a certain order, and that by doing so I may be giving the impression that this implies an order of importance, or an order necessary for a single logical sequence, whereas I would be happier if I were able to present, as it were simultaneously, a number of considerations, so that the relationships between them could be more readily seen and discussed.

Something much more like this is possible, I believe, in the Structural Communication method. In particular, the use of the response indicator matrix in the technique provides a simultaneous presentation of a number of concepts which can be grouped in various ways according to one's understanding of the whole field.

This is not only valuable as a teaching and learning technique, but is of very great assistance in the whole thinking process. The actual Construction of the matrix is an aid to thinking, and by its means the significances of concepts may be weighed against each other; groups or constellations of ideas may be formed, and the constellations as readily dismantled.

This is much more like the kind of perception which has to be brought to bear on a work of art. A work of art has to be perceived in its wholeness. It has its aspects, but it is in its totality, in its unity, that its ultimate significance lies.

H.W.

1.9. EVOLUTION BY VALIDATION
Structural communication is designed as an open-ended rather than a closed activity; hence there is no fixed 'terminal behaviour' (result) nor 'learning path' (process of communication) to be compared with a standard or ideal behaviour. This form of communciation is designed to enable different types of students to come to grips with its content or message in their own way.

However, there are certain decision points which must be reached in the process of communication and these are common to all recipients. The most important and indicative of these is the point where the student decides what selection of items from the response indicator will constitute his first response to a problem (see Section 1.5.). Further decision points are reached as a result of the diagnosis of responses (provided by the discussion guide in the book form which sets up certain pathways to help the student sort out misunderstanding, confusion and so on.

When a structural communication is first written, the author cannot gauge its effectiveness until he has some feedback from a sample of the population for which it was intended. Experience has proved that the stage of revision after validation is as important as the first writing itself. As an experienced teacher knows, every subject has its characteristic difficulties; but the way of dealing with these difficulties cannot be completely foreseen because it will depend on the particular situation and the particular ways in which students come up against them. The creative teacher needs a live situation for students to bring them out, so that he can 'see what is behind' their viewpoint; he can then use their misunderstandings as a way of getting to grips with the subject. But these points of confusion and misunderstanding are usually very difficult to pinpoint unambiguously.

The response indicator provides a ready-made 'language' and the way in which students use this language enables their difficulties to be pinpointed directly. The main use of the first 'validation' after writing the communication is to provide a simulated interchange between the students and the author, which enables him to see where their difficulties lie. The author has to be as receptive to the students' real difficulties as does a teacher in the live situation; so the primary need is to provide him with a set of objective 'results' which he can scan and interpret as unambiguously as if the students were actually talking to him. At the same time, the author has the advantage that he can 'hear' all the students simultaneously, so that he will know what significance to attach to each 'comment', according to whether a particular difficulty is widespread and general or just isolated and individual.

On the other hand, the method is concerned with communicating wholeness rather than separate 'bits' so that, although the students' difficulties can be more or less accurately focussed, this wholeness has to be preserved by using a method of display and revision which will enable the author to have in his mind a total picture. This should simulate the 'situation as a whole' of the live teacher in a discussion, and include the student background as well as the subject background. Difficulties always arise in a context, part of which is definite, part indefinite and part unknown. Operating in such an open-ended context is part of the 'art' in the writing, just as in live teaching.

The technique of validation has itself to be modified continually as experience with the method is gained. For our first fairly large scale validation of the first four books, we used two forms-a response sheet and a questionnaire - which were designed as a balance between what we could reasonably expect to find out from students, and what we expected would be essential for revising the material. Together with these, we designed a system of display and summary which would present the crucial data clearly against a coherent background, so that the author could operate effectively with it.

1.9.1. ASSESSMENT INDICES

R is simply (sum of all r's for n students);

U is where p is the set of performance parameters derived from r

For each problem, the average performance A for the whole group can be displayed at the bottom (or for each sub-group separately, if the group is not homogenous). Deviation D (see text) can be given at the bottom of the columns for each response item.

All the first responses to a given problem are displayed on a single matrix. The students are numbered down the left hand side and grouped according to school and educational level, and their responses are shown as ticks in the appropriate boxes. Difficulty with a given response item is denoted by a circle in the box. The control data, which is used to set up the diagnostic pathways (see Appendix V), is superimposed as a simple colour coding. The particular logic that was standard for the trial editions of the first four books grouped the twenty items in the response indicator into three classes or sets according to their relevance: 'essential' (colour-coded red); 'additional' or 'subsidiary' (orange); and 'forbidden' (blue). (With the substantially more flexible logic which is being introduced in later editions, the colour coding is replaced by a spatial grouping of the items into sets on the form itself; these sets can be given whatever designation is appropriate).

Using these standard colour-coded matrix forms, an author can scan a set of first responses and pick out significant groupings or combinations of response items. In addition, a crude summary of students' 'performance' for each problem is worked out for him. This takes the form of two indices, called 'Marks ' and 'Clarity ' *. The first simulates a kind of test marking, and answers the question: 'To what extent did students agree with the author, in their first responses? ' The second gives a measure of the degree of confusion or clear-headedness in a student's construction of response items, and it answers the question (whether the student agrees or disagrees with the author): 'How clearly has he thought about the problem?

In the validation of a first edition, we use these indices to show us where communication has failed, i.e. where repeated 'disagreement' or seeming confusion indicates that there is some error or lack of clarity in the writing itself. Consequently the first thing that an author has to face is the average performance indices of all students for each problem. If they fall below a certain value, corresponding to about ' 5O% correct', he knows that revision is needed at that point. This is not the same sort of criterion as the Skinner 'pass' level of 95%, because the figure does not derive from one single yes/no response, but from a number of possible combinations; and a certain 'failure' rate is not only unavoidable with this technique, but even desirable, because it works with the kind of involvement in which 'mistakes' are made use of (as in the live teaching situation).

To provide the author with a clear idea of where difficulties occur, the degree of deviation of the student group for each response item (with respect to each problem) is summarized in the form of 'proportion of students deviating'. What they are 'deviating' from is taken to be the author's intention as expressed in the control data (e.g. deviation means including an item designated as 'forbidden' or 'irrelevant', or not including one designated as 'essential '). If the deviation is greater than 30% for a given item, it is taken as a significant clue to the source of confusion in a problem.

Thus the author is provided with definite and unambiguous feedback of information at the point where it is most needed. concerning the intelligibility (see Section 2.4.1.) of the communication. Experience has shown that his involvement with this information enables him not only to alter the content, but also to 'see behind' what he has written and clarify what might be called the intention of the message itself. The revision of the more 'passive areas' of the communication, concerning knowledge follows naturally from this.

1.9.2. HISTORY OF VALIDATION TRIALS
In March 1967 we were able to conduct preliminary trials in some local schools, with the help of the Surrey Education Office and the Goldsmiths Curriculum Laboratory. These first trials showed that the method had considerable promise in answering a real need that was felt as much by the students themselves as the teachers. After two months we felt confident enough to conduct a fairly large scale exercise, to test all the material we had available. At the end of June this consisted in three finished topic books of ten study units each (Chemical Structure, Thermal Physics and Modern Mathematics) together with five study units of Nutrition and two of Geometrical Optics. For the purposes of this exercise we made contact with twenty schools in London. Surrey, Leicestershire, the Manchester area and Oxfordshire and were able to arrange trials at short notice in nine of them. In all, about seventy VI form students took part.

Due to the short notice and the novelty of the technique, neither we ourselves nor the various teachers participating in the trials had sufficient opportunity to introduce the method to the students and put it in its proper context. Consequently some of the students were hostile to or bored by it; but a surprisingly large proportion recognized its usefulness. Whatever their attitudes, most of them made constructive comments and criticisms, two of these which frequently occurred were: (1) the method was ideal for revision or consolidation, but not for the introduction of new material, since it would be too condensed to be assimilated. (2) It was a refreshing challenge to be able to get to grips with material and really have to think about it.

Many of the teachers became interested in the technique and its potential applications, and they have continued with further trials. At the same time now contacts have been made and many of the schools which were unable to participate in July are now studying the technique and starting new lines of co-operation. Trials are now going on in the University of Salford, and in nine schools, and about 200 students are participating at the time of writing.

As a result of these trials, four of the books are being progressively revised and the first of these (Chemical Structure) is being used for experiments with the method itself. For example, at the University of Salford, it is being used for consolidating and 'levelling-up' among freshmen of many different backgrounds; and at a school in Berkshire it is being used in a comparison of two different kinds of tutorial approach. In three other schools it is already being used as part of the chemistry curriculum in preparation for the Advanced Level examination in the subject.

1.9.3. DEVELOPMENT

The processing of the results of validation trials-of which there will have been, up to December 31, about 3,600 student hours - has all been done by hand. This is a heavy work-load (equivalent to 800 manhours) and an obvious candidate for computerising, as long as a reasonable method of transferring data from student to computer can be found. The cheapest and quickest way is to cut out any data-handling at all and use a direct input system like an optical reader or a student-activated magnetic input (e.g. magnetic pen on a normal 80-line card). The Institute is at present studying how one of these methods can usefully be applied.

Computer methods also have obvious advantages for doing statistical operations (correlations, tests of hypotheses, etc.) on large amounts of data, and for the retrieval of specific combinations (e.g. of response items in an array) which becomes impossibly laborious when it is done 'by hand'.

Now that we have confidence in the method of validating first editions, the main line of research will be concerned with the use of Student feedback for the assessment of performance and capability. Such an assessment can be used to follow closely a student's progress through a line of study, and give a quick indication of where his strength and weaknesses lie; it is therefore useful for feedback to teachers in much the same way as 'validation' is for authors. A further aim is to be able to group all the feedback for a particular student from time to time', so as to give him a more general profile of his progress, so that he knows where he stands and can see how he needs to proceed.

One problem that this research is beginning to throw up is that we do not know how students set about understanding a problem; we only know what results they produce. The trouble here is that such detailed knowledge can only be obtained by closely following what the student actually does when he sets about it. At the moment this can only be brought to light in a carefully conducted tutorial session (such as is described in Section 1.7.). However, a very few sessions like this do promise to release a relatively large amount of useful data. Much more detailed analysis of student strategies will be possible with the Systemaster teaching machine coupled to a counter, or a computer which can trace the students' path through a study unit.

2.1. MENTAL OPERATIONS
The basic hypothesis upon which our work is based is that a plurality of levels of mental operation can be distinguished by qualitatively different modes of connectedness between the mind and its objects. In its simplest form the hypothesis predicates four levels of operation: automatic, sensitive, conscious and creative. In structural communication we are concerned primarily with the conscious and sensitive levels upon which understanding and judgment take the place of conditioning or automatic reaction.

2.1.1. THE PHENOMENA OF LEVELS - EXPERIMENTAL
2.1.1.1. Peaks of Activity.
In all spheres of human life, there are peak moments when what is done is done in a superior fashion. We speak of 'originality', 'creativity', and so on. Such moments clearly stand out when the structure of activity makes very complex demands. These arise in specialized activities such as science and art when those concerned are dedicated to achieving difficult goals; but such demands can also be made in suitably constructed exercises, whether these concern the mental or bodily powers. We conducted such exercises experimentally with many hundreds of people. The exercises were designed so as to show unmistakably when the desired result has not been achieved. The transition from a state of struggle and failure to one of steadiness and ability to do the task was always unmistakable. It was sudden, unrelated to the effort immediately preceding it and had the element of surprise in it.

We shall say more about the methods for attaining 'higher levels' of operation later. Here it should be noted that the struggle we referred to nearly always had a positive effect in unexpected directions. For example, people who did the mental exercises reported a greater ability to operate with concepts than before and that appreciation of works of art was given an added 'depth'. This 'spill over' into the fields of thinking and aesthetic appreciation is typical of the results gained with specially constructed exercises of this kind. Though a definite jump may not always be made, the lower levels are probably deconditioned - and hence perception becomes more fresh and direct.

2.1.1.2. Cut-outs.
According to our concept of levels, different Operations are possible on different levels. 'Absence' on a particular level means 'absence' of the corresponding kind of operation. In groups, a good deal of work was done on the observation of moments of the 'cut-out' of a particular operation. For example, one line of work was to observe the failure to perceive surrounding objects and events. At the moment of noticing some feature of the environment, one can see that it was 'unseen' before. The determination of what is noticed and unnoticed appeared to be governed largely by well-established habits, interspersed with unpredictable fluctuations. 'Direct contact' with an object was defined as the perception of the object in a way distinct from the associations evoked by the perception. Experiments showed that there was a time-lag before a direct contact could be made, which indicated the need for a build-up of intensity to a requisite threshold. The phenomena of build-up will be met again from many aspects.

Another line of work worth mentioning was on the effectiveness of decisions. Self-chosen tasks were taken up after a special 'decision-process' to see them through without omission or commission. Routine behaviour was by-passed so that the subject had to make a complete event. For this he had to draw on a basic synthetic power. When the 'decision-process' was inadequately executed, the task performance was poor (according to the criteria set up). The work with self-chosen tasks proved extremely useful in providing material for understanding the irreducible requirements of truly integrative operations.

2.1.1.3. Learning.
There is one highly relevant and widely known phenomenon in learning that provides important evidence for the concepts of levels. At the beginning of learning a skill or set of facts there is strain of some kind or other, clumsiness in execution or repetition and slowness. Eventually, however, there comes a time when strain has gone and the right speed is there. At this stage, there is no longer any awareness of what is going on in the process-e.g. in driving a car. The learning action on one level has resulted in something taking hold- being learnt-on a lower level. Similarly, moments of 'break through' into the higher levels can change the pattern of operations on the lower levels of ordinary effort.

2.1.2. LEVELS OF MENTAL OPERATION-THEORETICAL
The following summary sets out the principal human activities that operate characteristically at each of the four levels.

2.1.2.1. Creativity.
All truly creative steps, characterized by spontaneity, unexpectedness and the absence of any ratiocinative process, occur on the creative level. This is supra-mental in the sense that the subject is not conscious of the process by which the step is made. It is therefore true discovery, though discovery of another kind may be the results of search. It need scarcely be said that creativity occurs in all spheres of human activity, Science, art, religion, and practical affairs. There is no evidence that true creativity can be trained: but its practical value can be immensely enhanced by experience in recognizing when a creative step has been made and in bringing into use upon the conscious and sensitive levels. It is also probable that favourable psychological conditions can increase the frequency and strength of creative insights. At present, there is a vast wastage of the all-important property of spontaneous creativity inherent in man, wastage due mainly to the difficulty of communicating a new, and unpredicted insight.

2.1.2.2. Consciousness.
The study of conscious mental operation in man has been obscured by the general failure to distinguish between sensitivity, or the awareness of immediate presentations, and consciousness which is a synthetic awareness capable of seeing situations in their organized complexity. The conscious level of mental function is, according to our investigations, comparatively rare in operation and plays little part in our lives. It is, however, capable of being developed and its operations immensely increase the effectiveness of both thought and action. We use the term 'consciousness' in its strict etymological sense of integrative awareness or the power to apprehend and judge a multiplicity of separate ideas or presentations. Understanding, hypo thesis-formation, original - as distinct from creative - thinking, unbiased judgment and impartiality are characteristic features of mental function on the conscious level. Associated with the affective function, consciousness gives self-reliance and self-criticism, genuine taste and moral conscience distinct from moral conditioning leadership and-initiative. Inventiveness, adaptive skills and quick-wittedness are indications of conscious operation in the sensory-motor sphere.

Structural communication aims at the conscious level of the mind, though it may not often reach it. As a technique for training the mind, We have considerable evidence that those who have worked with the method for a few years do show an increased capacity for conscious mental work with all its attendant advantages in efficiency and speed.

The primary significance of the conscious level of mental operation lies in its being the seat of understanding as distinct from knowing. In our usage, the word 'understanding' always signifies the fusion of theory and practice. It is more than the combination of 'knowing what' and 'knowing how' because this combination does not ensure adaptability and judgment, whereas understanding implies both. True understanding is both inductive - leading from the known to the unknown - and intuitive-leaping directly into the unknown. Not all understanding exemplifies all these characteristics because the mind is capable of many from the lower limit of awareness - sensitivity - to the upper limit-consciousness. Effective working upon all gradations is fostered by the use of structural communication.

2.1.2.3 Sensitivity.
The ordinary waking state of man is characterized by a restricted awareness of the immediate objects of sensation, thought or feeling. The mental activity is predominantly associative. It can be trained to perform complex binary operations, but it seldom spans more than two distinct ideas in a single moment of awareness. This is what we mean by the sensitive level of mental operation; the term being chosen to indicate that in this state the self is sensitive to the existence of other, and is to that extent capable of objective experience. In the sleeping state, sensitivity is suppressed and there is no criterion for distinguishing hallucination from actuality.

Education today is directed mainly to training the mental and bodily operations that are associated with the sensitive level. Logical thinking, experimentation and the observation of nature, self-expression in word and symbol, adaptive and purposeful activity, are all possible on the sensitive level of operation: but do not generally require the, intervention of the conscious level except for self-appraisal.

In a very broad sense, it can be said that we know all that we need to know for the purpose of human existence by means of mental operations on the sensitive level. For this reason, we find that educational procedures are directed to this level once it becomes active in the boy or girl. It can be trained to perform operations of a high order; but if these are to be directed from the understanding. it is necessary to have access to the conscious level. This, though with a different terminology, is grasped by progressive teachers who employ various means to produce the challenge that awakens the conscious level and connects it with the ordinary-or sensitive-operations of the mind.

2.1.2.4. Automatism.
This level is below the threshold of mental awareness. Its operations sustain all animal and human activity. They include all instinctive processes as well as all those learned by imitation and repetition in early life: speech, walking, and using the hands; reading, writing and mechanical calculation and association; habits of all kinds mental, affective and bodily-all these operate normally without mental awareness upon the automatic level of function.

It has been demonstrated that many automatic functions can be acquired in sleep or by repetition without mental awareness. Much training is directed towards acquiring a wide and reliable repertoire of operations on the automatic level.

In many cases, the automatic functions cannot easily be acquired. unconsciously and must first be learned on the sensitive level. Then accordance with the rule of mental declension described below, the o operation passes out of awareness and continues on the automatic level This can very easily and convincingly be studied in adult learning of a foreign language or a physical skill, e.g. in a game.

We shall not go into further detail, as the automatic level plays little part in structural communication. It is necessary, however, to be clear about the four levels and to enquire into the way operations, mental and physical, pass from one level to another.

The four levels of function are implied in the distinction often made between sub-conscious, Conscious, and supra-conscious mind The 'mind' as commonly understood is the sphere of subjective experience varying in awareness from sensitivity at the lower threshold to consciousness at the upper threshold.

2.2. TRANSITION PHENOMENA

2.2.1. EXPERIMENTAL

2.2.1.1. Discontinuities.
As William James and many subsequent workers have noted, careful observation shows that the common assumption of a continuity in mental experience can he very misleading For example, observations of the thinking process lead to the noticing of discontinuities in the trains of association - that in ordinary circumstances would remain unnoticed since attention is within the associations themselves. Under exceptional circumstances, it can also be observed that each so called chain of association is a closed C:, cycle. Association chains are largely triggered by external impacts and internal impulses. In the transition from one cycle of thoughts to another there is an instant of pure distraction, or blindness.

The jump may not be externally caused, but result from some Spontaneous operation upon a higher level. The non-temporal character of these discontinuities makes description difficult and the conviction of one's own mental continuity is very strong. Consequently, only the actual practice of observation can lead to verification. Certain experiments have been done directly on this theme of the discontinuities in mind. One of these is described below because it also incorporates elements which have a bearing on some of the other points made and on things to follow.

2.2.1.2. The Reading of a Difficult Text
The experiment required of the subject a careful reading of a difficult text. When he noticed his level of mental operation in relation to the text, he made a corresponding mark on a time chart. After a set time, the subject had quickly to note down a set of key concepts by Which the main argument of the text could be constructed. This he then Compared with the original and made modifications. In conclusion, he had to write a critical discussion of the argument in a very restricted time. The terminal challenge was built into the experiment as an aid in developing a high intensity of concentration on the meaning of the passage. The results of it were also used in bringing out concrete evidence for or against the actuality of higher levels of operation marked in the time chart.

The experiment was tried with some subjects who had to some degree been familiarized with 'mind-observation' and with others who$ had not. The levels used in the time chart included a lower extension of complete inattention or phantasy. The other three were defined as:
(2) Simple attention. The words as sounds were being registered.
(3) Attention to meanings. Grasp of the implications of words and phrases and sentences as stages or components of an argument.
(4) Active attention. Simultaneously with grasping the argument, the reader is able to weigh up and assess its validity and construct his own independent viewpoint without obscuring the author's meaning. Critical judgment.

It was found here that less than a tenth of the time was spent on the highest level, although this was what the reader was aiming at. Usually at least a quarter of the time was spent below the threshold of meanings. In the middle range, the evidence pointed to a preponderance of preconceived ideas or an inability to begin afresh with the material.

On the subjective side, subjects reported on the moments of 'waking up when a transition was made from one level to a higher level. Thus, when a transition was made to recognizing meanings, subjects at that moment saw that they had only been taking in the sound of the words before. In the realm of meaning, periods of struggle occurred when there was discontent with the connections being made by the mind and a sense of confusion. Sometimes, these durations were superseded by an abrupt transition to a total grasp of the argument in which components fell into place and the reader felt able to meet the argument before him.

In both cases, the transition was followed by a sudden or a gradual decline of intensity. This is a case of the general law of mental declension discussed in Section 2.2.2. Subjects frequently reported on their need repeatedly to renew their attention by challenging themselves with what remained obscure in the passage they were reading.

The main immediate impact of the evidence we accumulated was that in reading, for most of the time there is no effective connection with what was being said. Our discussions with teachers confirmed this view of the predominance of inattention and the lack of active attention in all forms of verbal communication. The possible reasons for this state of affairs is a subject outside of the scope of these papers.

2.2.2. THE LAW OF MENTAL DECLENSION - THEORETICAL
Mental functions conform to the general rule that intensive magnitudes subject to random change tend towards lower values. This is familiar in thermodynamics and the corresponding psychological rule is the tendency of the mind to operate at the minimum intensity that the demands made upon it permit. In crude terms, the mind is naturally lazy.

The level of mental operation may be raised by a shock or a challenge that arouses our attention. We recognize the change in an enhanced sense of concern in what we are doing, and if the transition reaches the conscious level we become aware of aspects of the situation that previously escaped us. This state cannot be sustained. The activity of the mind plays the same role as energy exchanges in a thermodynamic system and the intensity of consciousness drops to some threshold value below which it ceases to influence the mental process. This may continue on the sensitive level, recognized by sustained attention and involvement. This in its turn declines and we pass into the automatic state where the mental process may continue, but below the threshold of awareness. We call this inattention or mind wandering. The process and the transitions are familiar but its significance for communication has not been sufficiently appreciated. Since this decline of mental level is found invariably to occur in the absence of an artificial stimulus, we have generalized the observation with the designation of the Law of Mental Declension. This proves to be far more than a statement of what is commonplace experience, for its provides us with a precise criterion for assessing mental processes. It has an important application in the theory of communication that can be formulated as follows:

To communicate with the lower levels of the mind the impact should be minimal and repetitive. To communicate with the higher levels of the mind the impact should be maximal and unique.

In the first case, the effect is to produce and establish an automatic response to a stimulus or group of stimuli. In the second case, the effect is to awaken the understanding and enable the intention or meaning of the communication to be grasped. In these two effects lies the distinction between programmed learning in all its forms and structural Communication as we have developed it. Because of this distinction, it would be entirely misleading to treat structural communication as a development of programmed learning. The aims, methods and requirements of the two are totally different and should not be compared. There is, in particular, no suggestion that programmed learning is inferior to structural communication or the reverse; they are different and serve different purposes. Whenever a reliable and reproducible response to a set stimulus is required-as, for example, in spelling or arithmetical calculation-the method of small steps and reinforcement works well and both accelerates and improves the learning process. When the aim is to arose the critical faculty and to make the recipient think for himself, the Law of Mental Declension must be allowed for and the necessary counter-stimulus provided. This is the key to the success of structural communication.

2.3. ENHANCEMENT OF MENTAL EFFECTIVENESS
The accepted distinction between 'knowing what' and 'knowing how' is a partial step towards the distinction between knowing and understanding. It is not complete because one can know how to do something without understanding it. In such a case, we have little leeway to meet and overcome unexpected difficulties. This suggests that under-standing requires an insight that goes beyond what is actually known or what can be deduced from what is known. This agrees with the use of the word that we seek to clarify.

We should also include some notion of practical utility and effectiveness in action. This leads on to the proposition that where as know-ledge can be acquired passively, understanding requires participation in an action.

Knowledge is the content of a framework or schemata. Consequently, knowledge grows by additions within a framework of what corresponds to that framework. The framework is established by some form of behavioural adaptation to the environment-material, biological or cultural.

In the acquisition of understanding, on the other hand, there must be a confrontation between the 'new perception' and the 'old schemata' that alters not only behavioural patterns but also the strategy of achieving1 aims and even the aims themselves.

Another aspect is: understanding tells one what one can do with one's knowledge, how it can be used and where to find useful knowledge. It can be viewed as the upper limit of 'transferability of training'. These notions can be illustrated by typical experiences in a line research with student groups.

2.3.1. EXPERIMENTAL
2.3.1.1. Understanding cannot be taught
The intention of the investigation was to try out various means of awakening the higher levels of mental operation in understanding the nature of scientific thinking. Discussion on the theme of scientific method was used only as the ground of the activity. The real work began when specific problems were taken and in considering these the students were called upon to notice and bring out their own mental activity. No theoretical conceptions of levels were introduced. Instead, a working notion of the 'mental image' as that which contains the results and possibilities of mentation on a subject was employed.

The ground was prepared by using the early discussion to show, practically, that verbal formulations were in correspondence with 'forms of thought' which can be variously analyzed. For example, there is the aspect in which a 'form of thought' is a group of assumptions, and a similar one in which it is a 'mental set'. An independent line of preparation was in the production and description of simple mental images. Conditions of communication were established in which the students were enabled to distinguish betwen the mental image as a 'replica of sensations' and the mental image as a 'form of thought'. At the same time, it was possible to demonstrate that mental work inevitably involves the holding together of disparate elements against habitual mechanisms of associations which tend towards disorder.

The particular discipline used in tackling the problems consisted in the following. After making a statement of viewpoint, it is possible to hold the attention still and the corresponding mental image becomes apparent. In conditions of ordinary discussion, no opportunity is given for the 'build-up of intensity' required. And until a certain intensity has been reached, the mental image remains 'subliminal ', though it is operative as a conditioning pattern. When the mental image or 'form of thought' begins to take shape, it is then possible to bring it under critical enquiry.

The discipline was applied, for example, in a discussion of the Einstein Gedanken experiment on the addition of velocities. By means of the discipline, it became possible for some of the students to grasp the bases of their conviction of the verity of the Galilean addition- empirical data of every-day life and the conceptual form of treating velocity as linearly additive because of its form of representation as a vector. The important end result was that the students recognized that they had a new attitude towards what was at stake in relativity. This attitude by-passed ordinary questions of truth and falsity and had to do with the change in method that relativity theory introduced.

In this and other exercises, hardly any knowledge was transmitted to the students. As far as possible, there was used only what was at hand. Neither was understanding taught. What was done was to create a Certain combination of conditions and lines of work that enabled moments of confrontation to take place within the minds of the students. In so far as they went through these confrontations, so their understanding developed. This was evident in the way in which they were able to examine more critically the material available to them in their standard courses in science. The procedure is a typical application of the Principle of Challenge referred to in Section 2.3.3. below.

2.3.1.2 . Closed Systems and Open Structures.
The line of investigation briefly discussed above showed very clearly the difference between two possible educational procedures. One is that concerned with setting up a closed system where boundaries are set by criteria of 'correct' and 'incorrect'. In the closed system, terms have an invariant meaning which can be defined separately. The other procedure is concerned with open structures in which meaning depends on the combinations of terms and the context, and definition is not effective.

Experiments with scientific concepts, such as we have done over the last few years with the aim of more effective communication, can demonstrate the difference. It is common ground that students are seldom able to use conceptual systems in front of new or complex situations. The reason is that the systems are closed and do not allow variance of meaning as is necessary for actual scientific practice. When students have received only one form of description for a concept they find it difficult to adapt to relevant ideas new to them. This can be easily seen with such a 'concept' as that of temperature. There are at least five quite different ways of looking at 'temperature', ranging from 'making thermometric readings' to the 'differential of energy with respect to disorder'. All these ways can be accommodated only within a dynamic open system of conception which allows for relativity of meaning. Right up to the sixth form, students have at most only two ways of looking at temperature (empirical scales, kinetic theory) which are never brought into fruitful confrontation.

In dealing with open structures, we have observed how mental operations are involved which are quite different in kind from those appropriate to closed systems. -

2.3.2. THE CONDITIONS FOR RAISING LEVEL OF MENTAL OPERATION
The evidence for the predominance of the lower levels of mental operation, and the difficulties attending any effort to reach the higher levels, is widely diverse but very convincing. We have found it of critical importance to hold before us the fact that in reading a printed page, the average reader will be largely 'reading' his own mental associations without verification of their relevance to what is being said. It is widely known that repetition of impressions in conjunction with each other sets up an invariant association between them. This is the way in which facts have to be memorized and skills acquired but they do not require the active participation of the recipient.

There is an increasing body of evidence (A. C. Abercrombie, Bion, etc.) that when a branch of study or a form of behaviour has been experienced only through the use of low level operations, it requires a severe deconditioning shock to break open the patterns of association- affective and sensory-motor as well as intellectual - which have been established. In our own work with groups, we have also gained some insight into the way in which affective factors are intertwined with intellectual factors in conditioning behaviour. Mental deconditioning is almost invariably accompanied by a release of emotion. Later on, we shall comment on the dangers involved in deconditioning.

2.3.2.1. The Indirect Breaking of Habits.
In our work with adult groups, many experiments were done on the 'changing of habits'. Typical habits chosen were: bad posture, dominant patterns of mental association, reactions to certain types of people. It was easy to verify that a simple desire to change the habit had no result. Simple wish is as ineffectual as admonition. What did prove a viable way was for the person concerned to concentrate not on the habit itself but on some associated area of experience. For example, habits of posture could be to some degree altered by concentrating on the associated feeling states habitual reactions to people could be 'softened up by concentration on the mental images associated with those people.

These findings concerning the indirect way in which habits have to be broken up, support the indications of certain psychologists such as those of the school of 'Gestalt Therapy'. They suggest the basis for the common practice of teachers who try to awaken their pupils' attention through seemingly 'irrelevant ' activities. They also belong to the sphere of the general principle (Section 2.3.3.) that change in level of operation cannot be produced by direct means, but needs a shock or challenge.

2.3.2.2. Deconditioning by Sensitizing.
Anything that is a break in routine arouses interest. This has attained the status of a principle in the well known 'Hawthorne Effect' and it is something at the back of every teacher's mind. But it is also widely known that the awakening of interest as such can turn into a chain of distractions. Only at the moment of the arousal of attention is there any change of level. What is needed to sustain attention or freshness is the combination of at least two quite distinct sets of impressions. It became a standard method in Our work to make a binary demand on people. The automatic working of the mind cannot cope with two different kinds of input at the same time. Any awareness of two possibilities, of ambiguity. uncertainty and so on means that there is activity on a level of mental operation above the automatic.

2.3.2.3. Deconditioning by Shock and Challenge.
The phenomenology of shocks has not been explored systematically and we are able at this stage only to indicate certain of its important features. Prominent amongst them is unexpectedness. A shock is unexpected because it arises from a direction in which one is blind. Another important feature is that it is not possible to 'make sense' of the impact involved. If one could make sense of it. the impact would not be a shock since it could. be assimilated into established frames of reference.

A great deal of work has been done on 'deconditioning', but not enough attention has been paid to the requisite conditions for ensuring that a more severe conditioning does not take the place of the original one. When the mind is shocked, it is very easy to make suggestions on ways of making sense of the situation that are eagerly assimilated. It is now well known that the best way to produce fanatical minds is first to produce a disorientation through shocks and then build upon hope of a special knowledge or power.

There is a dominant aspect of the human psyche that responds to suggestions of self-aggrandisement and the disorientated mind is very vulnerable in that respect.

In student groups subject to unstructured deconditioning, there is clear evidence of a reversal of the initial good intentions. For example, in art college, the lecturers took the students' design work at the end of the first year and deliberately criticized it violently as useless imitatory rubbish. The students were then set free from formal instruction and encouraged to experiment in as wild a fashion as possible. The end result was that the students were aping the styles of the lecturers themselves!

The distinction between shock and challenge is important. A shock is simply an impact that may or may not have any positive intention or result. A challenge makes an appeal to the intelligence of the person.

In our experiments with groups, we were careful to administer shocks within the context of a programme of constructive work.

2.3.3. THE PRINCIPLE OF CHALLENGE
A challenge is an obstacle placed, intentionally or by accident, in the line of activity that can be overcome only by increased determination and generally by a higher intensity of mental operation. The challenge can be accepted or refused. If there is no choice, the term 'challenge' is inappropriate and should be replaced by obstruction, difficulty, misfortune or other words which do not imply a positive response from the agent. Accepting a challenge is one of the ways in which the level of mental activity can be raised. The principle of challenge is the counterpart of the law of mental declension, for it asserts that it is possible to use the self-assertive instinct in man to overcome his mental laziness. It explains the apparent paradox that when an action is made easy its quality degenerates and when it is made hard its quality may-though not necessarily - be improved.

An accidental obstruction may be accepted as a challenge, but it seldom assists the process. The agent engaged in the activity may be stimulated to greater effort, but this will be directed to revising or overcoming the obstruction. It is different when the difficulty is deliberately raised in order to bring new content or significance into the process itself. Many successful teachers make use of the principle of challenge, but usually ad hoc and without any clear conception of how it works. A false application of the principle is made when we admonish or appeal. 'You can do better'. 'Try harder'. 'Accept the challenge this difficulty offers,' are all ways of deflecting interest without raising the level of mental operation and they seldom do any good.

To be effective the challenge must meet the recipient of the communication at the right time and in the right form. This has been determined empirically in a number of cases and it is found that it should consist in presenting problems to be solved with means that are not given directly in the presentation. The procedure can he visualized with the help of a kindergarten example. Children arc shown the rudiments of number and then allowed to play at shops, with real money and goods. The challenge to act as grown-ups is accepted and they apply their arithmetical gleanings to the handling of money. They are not told to 'try hard', but in practice they are seen to make a real effort to get their sums right.

2.3.4. THE RANGE OF COMMUNICATION
What we have been saying on the levels of mental operation suggests that communication is on more than one level. The higher the level of operation, the more the activity of the recipient enters into the construction of the communication. At the lowest level, there can only be imitation; whereas at the highest level there is an unique, independent insight. We have made various experiments in certain forms of communication which cannot be made effective by imitation. Two of these have already been discussed: the reading of a difficult text and the work on understanding science. Another enlightening experiment can be described.

2.3.4.1. Subtle meanings.
During the course of a certain line of research it became necessary to have available terminology capable of unambiguously indicating the various 'qualities' of people. Ordinary Usage of words proved utterly inadequate. We postulated a formal System in which 48 different basic 'forms of dynamism' can be distinguished and related together by means of a set of rules. This formalism was made available to people in four groups by specialists who had studied the interpretation of the formalism for some time.

Each interpretation previously made by the specialists had required an intuitive leap as well as the capacity to consider the inter-relationships of all the 48 forms. The way of working at the formalism could not have been described in simple terms, nor was there any way in which people in the groups could have acquired it purely through explanations.

Each member of the groups had to go through a process simulating that used by the specialists, and produce a corresponding result. Each member also had to listen to all the other contributions and ask himself why each of the formulations had been chosen. It was the work of the specialists to show the relevance of all the suggestions that had anything appropriate in them.

Thus both the specialists and the group members were involved in the same structure of activity. At its core was the integration of a wide spread of verbal meanings, including ones in contradiction with each other, under a single form. The specialists had the task of demonstrating this system, but only in the context of the equivalent activity of the group. It was found that this co-operative approach made possible a very rapid communication of the 'subtle meanings' involved in the formalism and also enabled the work of the specialists to be greatly accelerated.

2.3.4.2. Psychological insights.
We have already discussed our work with adult groups and it is clear that this involved problems of communication. The psychological notions were not put across theoretically, but conveyed by the interpretation of observations made by members of the group concerning their own work with various exercises and tasks. Explanations were made only in correspondence to the experience a person actually had. The main reason for this reticence was that it was imperative that the independent judgment and critical faculties of each person should be maximally brought into play. By having concrete experience to work with, members of the group were able to assess for themselves the relevance of any explanation.

This work brought to light an important principle, namely: if the aim is to awaken the higher levels of operation, explanatory material.' should never go beyond the range of personal experience of the recipient of the explanations.

2.3.4.3. Concordance.
Both these last two examples concern communication as grounded on a concordance between the people concerned. Concordance stems from the independent judgment of people and is not a matter of learning in the ordinary sense. Many human situations involving communication depend not on people memorizing the information given but on reaching a concordance on the meaning to be given to the information.

2.3.4.4 Subjectivity and intelligence
In the realm of communication by concordance, there is a real hazard; there is no guarantee that the recipient will see his way through to grasping the intention of the message involved. If the independent judgment of the recipient is allowed to play no important role in the communication, then the subjective prejudices, orientations of mind, etc., of the recipient can come into play. It should be noted that in our work on communication involving the higher levels of operation, there were always groups involved. If people have to share in a communication of this kind, then there is a mutual exposure of subjectivity that provides a means of attaining a real concordance in the end. This practice has a bearing on the line of research we have undertaken into the integral role of self-assessment in communication.

What we call 'subjectivity' is the counterpart of 'intelligence'. We do assume that in every person there is an intelligence which, however incoherent are its manifestations, is what has to be appealed to in higher levels of communication. Intelligence in people takes certain basic forms and these we call 'strategies', i.e. the way in which challenges are met and independent lines of action set up.

Only tentative work has been done so far on the question of strategies. We have, however, established that the same strategies can be discerned in all domains of behaviour of a person. It remains to be seen whether what we are calling a 'strategy' is really an invariant parameter of an intelligence. An interesting discovery is that whenever a person is set to do a task in a way which does not correspond to his usual strategies there arises in him or her a sense of revolt. To overcome this barrier of revolt does demand certain high level operations and in certain conditions is one of the most effective means of awakening these operations.

2.4. PRINCIPLES OF STRUCTURAL COMMUNICATION
Without implying any philosophical significance, we can usefully think of every situation as a structure having both a knowable and also an intelligible content. The knowable content is information. The intelligible content is the theme. In one sense, the distinction is the same as that we suggest in speaking of the letter and the spirit of the law. We know the letter, but we must grasp the spirit by an act of our intelligence. The letter binds us to a pattern of behaviour; the spirit liberates us to act freely and creatively. We certainly intend some such distinction in our insistence that structural communication transmits Understanding as well as knowledge. This is in accordance with the Conclusion previously reached, that information can be transmitted to a passive recipient, whereas understanding must be shared with an active partner. It also agrees with the observation that structural communication can go beyond the intention of the author and awake in the recipient capacities for thought and action that had remained latent.

2.4.1. A WORKING HYPOTHESIS
The two components of the structure are not separable in the actual situation, but they are apprehended differently and therefore they do separate in the mind. It follows that they require to be communicated in different-though complementary-ways. The method of structural communication is based on the hypothesis that the knowable and intelligible components of a structure are homologous though not identical. Homology in this context means that there is a basic correspondence of the essential elements, though there may be differences in the adventitious content. Thus, the construction of a motor car is homologous with its functional mechanisms, though there are many features in the body work that are irrelevant for understanding how a motor car works. Conversely the description of the mechanism will not convey the chemical, thermodynamic and mechanical principles on which a self-propelled vehicle operates. These latter can be regarded as referring to the 'intelligible ', as distinct from the 'knowable', automobile.

The hypothesis of structural homology explains why it is possible to convey knowledge without understanding and yet to use the knowledge in such a way that experience will lead to understanding. This is gaining understanding the 'hard way'. There is no alternative except by the sharing of experience in action. Structural communication is a bold attempt to achieve at the recipient's desk a transfer of understanding that might otherwise take months or years. The familiar difficulty encountered by university freshmen in making use of knowledge laboriously learned at school in the laboratory or classroom stems from the lack of penetration in depth from which conventional teaching methods suffer. This is what we seek to remedy.

We shall use the term subject-matter to express the informational content of a structure and the term theme for the ideational component. The subject-matter contains the letter, the theme, the spirit of the message. The subject-matter can be learned on the automatic level, whereas the theme is grasped at the conscious level of the mind. When both are brought together on the sensitive level we say that the recipient both knows and understands the structure in question.

2.4.1.1. Types of Themes
The distinction of subject-matter and theme is possible only when the structure is itself an integral whole. There are pseudo-structures which may verbally appear singular, but belong in reality to different realms of discourse. For example, the anatomy of the digestive system and the bio-chemistry of nutrition refer to the same situation, but they do not form a single structure. Any attempt to describe them in a single act of communication would lead to confusion. Again, the charge of the light brigade as an historical event and Tennyson's poem on the subject refer to the same event; but they must not be combined in a single communication. On the other hand, the physical scientist enters at least three distinct realms of discourse: phenomenal, physical and mathematical. He may wish to make a communication in which the theme is the interrelation of the three kinds of statement (Cf. Section 2.3.2. above), in which case the system is linguistic rather than physical, and he must bear this in mind.

In general, themes may refer to:
(a) Visible arrays: such as an animal body or a man-made mechanism.
(b) Situations or events, historical or hypothetical.
(c) Ideational systems, such as a scientific hypothesis.
(d) Subjective experiences: such as a work of art and its appreciation.
(c) Imperatives. i.e. instructions leading to action.

Every theme, whatever its type, is homologous with the knowable structure. The latter, by hypothesis, consists of inter-connected parts each of which is a sub-structure. It follows that themes must also be apprehended or composed of sub-themes so interconnected and interwoven as to be inseparable in the understanding and yet capable of being looked at separately. This characteristic of our experiences has been discussed in earlier sections. It is the ground upon which the second stage of structural communication is erected.

2.4.1.2. Sub Themes and Sub-Sets
Every structure is a more or less orderly assembly of parts forming a recognizable whole. We can organize any information we gain about it into an hierarchic system of superordinate and subordinate Components We shall call these sub-sets. First order sub-sets are those which are directly relevant to the structure as a whole. As we are seldom able to deal in a single communication with more than one hierarchic order, we shall omit such qualifiers as 'primary' or 'first-order' and speak simply of sub-sets. By the hypothesis of homology. there will be a sub-theme corresponding to each sub-set; but it does not follow that all sub-themes are significant. A sub-set may be required for an informational account of the subject-matter, while the homologous sub-theme may he trivial for understanding the structure.

Let us suppose, for example, that the structure is an aircraft and that we distinguish as first-order sub-sets; the air-frame, the power unit, the navigating system and the internal services. Any communication regarding the aircraft as a whole must take all four sub-sets into account. For maintenance purposes as an example, a schedule of all components and their inspection routine must be organized and used in the most efficient way. If we had to describe the construction process each sub-set must be specified both in itself and in relation to the other sub-sets and the aircraft as a whole. Instructions to the crew would single out the sub-set with which each is specially concerned, if we now turn to the study of aircraft design and the principles of heavier than air flight we should need a different set of concepts. This would be given us by the principile of homology as: the energy sub-theme associated with the power unit and aerodynamic design: the orientation sub-theme associated with the navigational sub-set and the structural sub-theme associated with the air-frame. We should probably ignore the internal services as irrelevant to the study of flight. Within the first-order sub-theme of orientation, come the sub-themes connected with speed, altitude, compass, rate of climb, etc. All these are relevant and necessary in the intelligible world of aerial navigation.

If, on the other hand, we were concerned only with instructions capable of being performed automatically, we should not need to go deeper than the informational content of the subject-matter.

It will be evident that everything turns upon the intention behind the communication. The intention is both general and specific. There may be a general intention to deal with navigation and there may be a specific intention to help the recipient to grasp the principles involved. With the same general intention the specific direction of interest may be to ensure correct inspection of the pilot system or the compasses. As our concern is structural communication as understanding rather than simple information, we must examine more closely what is implied in the identification of the theme and sub-themes.

2.4.1.3. Detection of the Theme
Since, by hypothesis, the theme is grasped by a synthetic mental act-or an act of intelligence-it can be found only if the conditions for such an act are present. The author of the communication must be an expert, i.e. having practical experience of tM structure, and be must also be trained in this kind of mental operation. These appear to be formidable requirements, but as has been shown in Part II of this paper they can be facilitated by a right approach.

The first step in setting up a system of communication is to decide what shall be transmitted. The theme can be approached in two ways:

1. The informational content of the structure can be scrutinized and placed in an orderly array from which the thematic pattern can be discerned. This is the extensive approach.

2. The theme may be grasped intuitively in the conscious experience of the author-or possibly communicated by someone more expert than himself in the field. This intensive approach leads to a mental image of the theme and sub-themes. By the hypothesis of thematic homology it is assumed that the two approaches will lead to the same result, though with different emphasis. We can say that the extensive approach will tell us what the message is to say and the intensive approach what it refers to.

It is usually possible to construct an aide-mémoire consisting of a set of notes on the sub-themes that will later be converted in the Response Indicator.

The material assembled for the extensive approach will be the starting point for constructing the Presentation and Commentary.

2.4.2. LOGICAL OPERATIONS
It is useful to distinguish sub-themes from the homologous components of the information system by designating the latter as sub-sets within the assembly of information. The total information content forms a main system S within which there will be a number of primary sub-sets A, B, C . . . etc. As before, we shall omit the qualification primary and ignore all second order structures. Let there be M sub-sets that jointly embrace the total informational content. These sub-sets are not partitions of S because there may be overlap of content between one sub-set and another.

We shall assume that the theme can be expressed-and indeed cannot be expressed otherwise than-by a number of ideas or thematic elements. A large number of such elements may be needed to express adequately the intelligible content of the theme.

We shall use the convention that capital letters refer to information and small letters to thematic elements.

Consider a main system S having a number of sub-sets of information A, B, C. . . respectively. Let M be the number of sub-sets, n the number of thematic elements, and x the sum of thematic elements necessary to express p. q, v. . . etc. Then:

2.4.2.1. Extensiveness and Complexity
The number m of sub-themes in the theme t will, by the principle of homology, be the same as the number M of sub-sets in S. t is also built up from n elements. We define the extent of t by reference to the number of elements and its complexity by the number of sub-themes.

We have to recall that t is a mental construction whereas S is a knowable situation or object. The extent of S may have no relationship to that of t. To illustrate the point let S be a time indicator, with four dials marked in seconds. minutes, hours and days of the year. There are 55,188,000 distinct states of the information system, but only two sub4hemes in the theme; the dial system and the input. If we turn to the operating mechanism there are about 200 parts, but several subsets: the power supply, the driving mechanism, the wheel train and escapement and the pace-setting regulator. The sub-themes of constant speed and number counting are all that require an intelligent understanding.

In structural communication limitations are set by the permissible extent of the information complex-the Presentation-and the complexity of the theme. The latter depends on the experience and mental powers of the recipient; but normally does not exceed five sub-themes.

2.4.2.2. Discernment of intention
Communication is the arousing in another mind of an experience homologous-in the sense above defined-with one's own. This is not hard in the domain of information relating to sense data. It is very much more difficult to transmit an intention which is an orientation of the will. Since understanding is always intentional, the core of the technique of structural communication lies in the way it seeks to solve this problem.

Let us suppose that the recipient R wishes to ascertain the intention of the author A by asking him questions to which he must reply in terms that can be coded and transmitted as a set of numbers. For this the questions must be grouped according to their logical type in order to avoid any possible ambiguity. The types of questions might be:

1. Is this item essential for conveying your intention?
2. Is this incompatible with your intention?
3. Does this help to clarify your intention?
4. Is this combination of two or more ideas compatible with your intention?
5. Does this combination fully express your intention?
6. Does the inclusion of this element vitiate your intention?
7. Does the inclusion of two or more of these items seriously confuse the expression of your intention?

These and many other questions can be reduced to terms of the algebra of operations and so coded for transmission and processing. In structural communication, the most effective set of questions is selected and used as the basis for constructing the INVESTIGATION, which is the link between information and understanding.

2.4.3. A TYPICAL PROCEDURAL SYSTEM
The following scheme illustrates one combination that has been extensively tried in practice.

Let us consider a sub-set A within the total structure S in terms of the elements set out in the response indicator. The demands of the group are wholly relevant and essential to a correct grasp of what the author intends in connection with A. We shall call these E-elements. Those of a second group are wholly irrelevant and their inclusion is incompatible with a correct understanding. We shall call these forbidden or F-elements. A third group, called desirable or D-elements, though not essential, reinforce the description and indicate a broad grasp of the theme of the sub-set A. A fourth kind of situation arises when the E, F and D-elements have been eliminated and a residue remains of elements irrelevant to the sub-theme. These we shall call G-elements. The accidental inclusion of one or two of these in the response may not indicate failure of understanding, but if some stated maximum is exceeded the attention of the recipient must be drawn to the need to be more selective. Finally, we may have elements that in themselves are harmless, but if joined to a group of others show a serious failure to grasp the theme. These J-elements may have a very important diagnostic value and for certain communications they must be included. The five cases do not exhaust the operations above outlined, but they will serve to exemplify the operational logic.

Each kind of operation implies a different decision. The logical operations in the machine to implement this decision fall into a limited number of specialist categories.

The construction of the discussion guide in the text-book and the circuitry of a teaching machine provide for the operations required. They do not transmit understanding but they serve a dual purpose.

1. They verify the degree of agreement between the author's intention and the recipient's mental image.
2. They provide the opportunity of bringing additional information to the notice of the recipient in such a way as to build up and extend his grasp of the author's intention.

Every response of the recipient evaluated by means of the appropriate algebra operates as an instruction to furnish him with additional information This constitutes the DISCUSSION AND COMMENTARY. There can be several cycles of presentation, investigation, response, evaluation, discussion and comment. If well constructed, the total set of operations will lead the recipient to the optimum grasp of the author's intention. This can never be complete, because two minds are never identical. The 'optimum' is the best obtainable within the limitations of time and space imposed by the degree of importance of the communication.

The operations described can be performed in various orders. They can be repeated and the results resolved and revised. Throughout the procedure the recipient R should picture himself as engaged in a dialogue with the author A. Through this dialogue the two components: the informational content and the intelligible theme takes shape in the sensitive mind of R. They coalesce by an action that involves the conscious mind. Since most people do not operate consciously without some shock, the moment of understanding of each sub-theme is likely to come soon after the recipient has seen for himself the challenge to his intelligence that is offered by each problem in the Investigation. Theoretically, his insight into the entire theme and its fusion with his knowledge of the informational content should come when he studies the summing up given by VIEWPOINTS. This is by no means necessary. The entire exercise is to be conducted in a spirit of open enquiry. It should be enjoyed as an authentic achievement not to be accomplished by any passive conditioning. The recipient is, and. must feel himself, free to take charge of the operation and conduct it in whatever way he prefers, his choice of strategy should be the expression of the natural working of his own intelligence. He may ask for guidance, but he must make his own decisions.

2.4.3.1. The Structure of a Transmission Cycle
The construction of a teaching unit comprises several distinct elements. From the theory of structures developed in the discipline of Systematics, we should expect to find that each cycle of transmission would consist of a number of steps linked together in a characteristic and definite manner. The steps are set out in the following scheme in which the left hand column gives the general character of each operations and the right hand column the corresponding elements in the study unit as at present constructed.

The transition from knowing what the author has written to understanding what he intends by it, is made possible by the challenge between Stages 3 and 4. The scheme here outlined lacks one dimension: that which gives both the author and the recipient a sufficient measure of freedom in selecting the strategy to be followed to ensure that the recipient makes his own necessary contribution to the build-up of the communication. This is achieved partly by the variation inherent in the response chart and partly by the facility open to the recipient to Study the material in any order that he prefers. All the stages must be traversed, but not necessarily in a set order and certainly not once only. The dialogue between author and recipient turns upon the challenge. The communication begins to be effective when the recipient sees what he is up against. This is precisely what is very hard to achieve in the straightforward reading of a consecutive text. Some of the most significant and philosophical writings from Aristotle's metaphysics to Kant's Critique of Pure Reason present the required challenge to the understanding by reason of their obscurity and repetitiousness. These features, commonly regarded as defects, are the means by which the student, determined to come to grips with the master's insights, makes the transition from knowing to understanding. In our psychological researches, we made use of some of the most pregnant utterances of the great philosophers and were able to verify the importance of the 'challenge of obscurity'. This challenge is too severe for most readers and in structural communication is tempered to the capacity of the recipient who is given the further aid of access to the supplementary explanations in the discussion and comments.

We are confident that the method can be used successfully from the age at which the sensitive mental level begins to operate independently. This must not be confused with conceptual thinking which is only one mode of mental operation. The lower limit of age is determined by the ability to accept and respond to a challenge. Evidently, this depends on the way in which the material is presented. We should not expect children from 12 to 14 years of age to meet a challenge to grasp an abstract principle concealed in a set of factual statements. Nevertheless, a few exploratory experiments that have been made indicate that young children can use structural presentations - both oral arid visual - providing the response indicator can be grasped by visual, imagery.

It is evident that the author of a communication must be able to gauge reliably the level of mental operation of the population to whom it is addressed. He must also know the mental skills that the recipients have acquired in order to give form to the challenge and response. He must not only be expert in the subject-matter but also be capable of penetrating into its creative potential. All these requirements are hard to satisfy, but they can be approached by progressive stages. This corresponds to the rehearsal of a drama used as an illustration in the introduction ( C.f. Section 1.3.1.4.). The author builds up the; communication by working on it with an experimental population and tests it with recipients who have the same characteristics as those who will use it independently. The need to 'rehearse' is not a sign of inadequacy on the part of the author, but rather a feature inherent in the method.

During 1967, the term 'Structural Communication' has been introduced to replace structural programming mentioned in earlier papers. This was done to avoid the misleading suggestion that we are concerned with some form of programmed instruction of the kind developed from the work of B. F. Skinner, J. G. Holland, N. A. Crowder, S. L. Pressey and others. Programmed instruction in all its forms is based on the principle that learning by small or large steps is a carefully designed sequence with frequent verification and encouragement can reduce the time and increase the efficiency of the learning process. This principle is valid for the transmission of knowledge and skill, but it has little bearing on the communication of understanding which involves a mental process entirely different from those required to produce a predetermined pattern of behaviour. The distinction is very hard to convey by description; but it has been found that teachers and students, who have used the method, recognize its affinity with face-to-face group tutorials.

For the reasons stated it is not easy to define or even to describe structural communication. This is not because it is obscure or complicated but because it contains an intuitive element similar to a judgment of taste or quality. "He who has not tasted cannot know" applies to the communication of understanding. It is not always necessary to reproduce the total experience in order to reach a shared understanding, but it is indispensable to participate in an act. Knowledge can be conveyed to a passive or almost passive recipient, but understanding requires an act in performing which, the intuitive leap is made. This agrees with the observation often made that "one does not understand something until one has done it."

The concept of structural communication has evolved through several stages. At first, it was treated as a closed-loop in which the author and the recipient of a message interact by way of a feed-back that indicates whether or not the recipient has grasped the theme and if not, where he has gone wrong. When the method began to be applied in schools, it became evident that the response of the students had a far wider significance than verification alone. The entire operation was an act of communication which was not complete unless several stages of two-way interaction had been worked through. The preparation of a text or a programme is a matter of selection, formulation, testing, correction, validation and revision. These steps are progressive and the result is used in accordance with a procedure prescribed in detail by the originator of the message. Structural communication is subtly but decisively different from this. Just as a group tutorial is not confined in its action to the period when the leader and the group meet and discuss their problem; so also is structural communication built up through the rehearsal stage through the presentation and response to the subsequent activity of the participants. This is clearly brought out in the reports of teachers who have used the method and find that it arouses a spirit of enquiry and debate that spills over into the classroom teaching and experimental work of the students.

The basic difference between the acquisition of knowledge or skill and the development of understanding is that the first is possible by a conditioning process whereas the other is possible only by a deconditioning action. Deconditioning comes by challenge and response. It is open-ended and therefore never wholly predictable. It cannot be studied or tested by the same means as serve for elucidating any kind of conditioning or learning process.

The recognition that understanding has entered one's own or another's mind is an act of judgment that is possible only for someone who has experience of the matter in hand. It might seem that this precludes any kind of predetermined test. This is not the case, because tests can be devised that are themselves acts of judgment. Such tests can never take the form of simple or multiple choice. They must permit and require a synthetic combinatorial assessment. This is achieved in structural communication, by employing a special technique. This consists in assembling the material in such a way as to make the intended theme unambiguous and yet not obvious. The procedure is described in section

2.4. The communication begins from the moment the recipients of the message are involved in studying it and continues-if successful-after they have performed the operations required of' them. We emphasize this to bring out the evolving character of the act whereby understanding is shared. A situation for shared experience must be created. This is the necessary factor, without which communication cannot get beyond the transmission of information.

It may be asked whether occasions for sharing understanding often arise in the process of education or other situations in which we advocate the use of structural communication. It can be argued that understanding is a private matter and that it must and can be acquired exclusively by the personal effort of the person who wants to acquire it. If this were correct, teaching would be mainly a matter of transmitting information accurately and economically; but we know that this is not the case. The practical value of any item of communication depends upon understanding how to make use of it. In many cases, this comes by experiment and trial; and the aim of structural communication is to achieve the same result more rapidly and effectively by a mental operation.

DISCUSSION PROCEDURES
In a live intelligent discussion there is considerable scope for logical and intuitive dialogue which adapts to the interest and difficulties of the participants. From such discussion the shared understanding evolves. The only constraints on what can be said arise from the breadth of vision of the participants and from their habitual 'sets' of mind. These constraints may be severe, however, in all but the most inspired discussion. This point must be remembered by readers who may dismiss structural communication as a means of mind broadening simply on the grounds that it imposes formal restrictions on what can take place in interchange between author and student. In the hands of a creative author, these formal constraints may prove to be much less restricting than a live conversation.

The key research problem in this field of the discussion section concerns the marriage between educationally effective dialogue and a flexible logical structure. In our work so far we have not found it possible to make this marriage effective by developing the study of educational dialogue and logical structure in isolation. Constant interchange between educational specialists and the logic designer has been necessary to reach the present stage and we anticipate much greater steps forward arising out of this interdisciplinary collaboration.

A third factor is required to bring together educational dialogue and logical structures. This factor is diagnosis. The evolution of a diagnostic technique has been one of the main concerns of our work to date. Although this work is in its infancy a clear step has been made and verified in validation trials. In order to report on this it is necessary to describe the formal mechanism of structural communication in more detail.

The response indicator generally consists of twenty items which may be data, concepts or statements. (See Appendix IV for an example). The number twenty was chosen for cost/effectiveness reasons in that twenty "bits" of information are convenient to handle in the standard computer word of twenty-four bits. From the educational point of view there need to be more than ten or so items according to the complexity of theme and age/ability group of student.

By combining together a sub-set of these items the student is able to indicate his response to any problem in the investigation. He writes down the code numbers of these items or indicates which they are by pressing code switches. Let us suppose a response to a problem contains the items (1, 7, 9, 10, 12, 14). This is all the information that the author has to go on in designing and writing his discussion dialogue. The student has selected this set of items and he has not selected (2, 3, 4, 5, 6, 8, 11, 13, 15, 16, 17, 18, 19, 20).

The formal logic of structural communication is based on the following logical unit.

(1) All twenty items are scanned in order to see which are present and which are absent.
(2) This information is compared to other information on inclusion or exclusion of items specified by the author as requiring specific action.
(3) If this comparison indicates a positive correlation then that specific action is provided in the form of a discussion comment.
(4) Otherwise the items are re-scanned for some other specification in the same manner or the student proceeds to the next problem.
(5) If the specific action is required then the student must revise his combination of items and then repeat the procedure of evaluation.

From this basic operation a quite complex sequence of diagnostic tests can be devised. In our work so far we have restricted ourselves to three or four basic operations of this kind for each problem. Since each basic operation requires a discussion comment and there are four problems, this means that study units so far written usually have between twelve and sixteen comments.

CRITERIA FOR DIAGNOSIS
In order to construct a diagnostic technique certain criteria have to be chosen so that a precise specification of the logical decision points can be made. Every criterion must reduce to a set of binary decisions yes/no. Where the subject-material of the communication is itself logical this is fairly easy, but where the material is more intuitive or aesthetic this reduction to a binary criterion has to be made very carefully. A summary of the kind of criteria we have been using is given below.

(1) The criterion of misconception. The assumption here is that the inclusion of certain items as a response to a particular problem are found to correlate with, the hypothesis that the student Is approaching the problem with some notion incompatible with the solution. This can be expressed colloquially as going off the track, getting things upside down, making false assumptions, confusing separate ideas, etc. This criterion enables the author to specify certain items as forbidden'. The inclusion of any such item routes the student to a comment which indirectly points out the possible difficulties which have arisen.
(2) The criterion of key points. The assumption here is that the omission of certain items as a response to a particular problem are correlated with the student failing to grasp essentials necessary to the approach to the problem. This can be expressed colloquially as missing the point, not noticing significance, not grasping the priorities, etc. This enables the author to specify certain items as 'necessary'. The omission of any such item routes the student to a comment which indirectly gives clues as to the features not noticed or not given sufficient importance by the student.
(3) The criterion of subsidiary points. This assumes that a further set of items, though not key points, are relevant to the problem and hence are permissible. These items correspond to being within the domain of relevance of the problem. This criterion enables the author to specify either the requirement for complete recognition of his viewpoint, or for sufficient recognition. The items are specified and in the first case if only one is omitted the students is routed to a comment. In the second case if any two or three items are omitted the student is routed to a comment.
(4) The criterion of adequate points. This is similar to criterion (2) but rather than the key points being relevant to the understanding in depth of the material they are specified on the basis of relevance to an adequate step for that particular student problem.
(5) The criterion of solution in depth. This is similar to (3) but instead of the items specified being subsidiary they are items indicating insight and depth. The conditional specification is changed to inclusion rather than exclusion and the comment is an encouragement to pursue this insight.
(6) The criterion of loose thinking. This is usually a terminal criterion which routes the student to a comment if he includes any one, two or three of those items which are not forbidden but nevertheless irrelevant.

Although there are the six criteria which have been exploited in study unit construction so far, it will be seen from the total logical System associated with structural communication in Part II of this Paper that an enormous potential has not up till now been exploited. The obstacle to this is observational and evaluative research problems in seeing the form in live dialogue in education situations and spotting the appropriate reduction to formal logic. It should be noted that the more usual forms of reward/punishment, right/wrong dialogue are only appropriate to conditioning situations.

Any of these six criteria may be divided into sub-classes. For example, there may be two distinct classes of pitfall that the student has to avoid and so each of these requires a distinct comment. Thus two classes of forbidden item can be specified provided the order of treatment is not important. Sub-classification of this type is limited in practice at the moment, however, by the cost/effectiveness restrictions on the number of unit operations that can be made available for the discussion of each problem. We hope that the wider scope theoretically available will be made available in practice as the project develops.

THE VALIDITY OF DIAGNOSIS
Assume an investigation with four problems. Suppose problem A concerns people, problem B concerns animals, problem C concerns vegetables and problem D concerns minerals. A response indicator could be constructed with five personal items, five animal items, five vegetable items and five mineral items. In such a situation there would be a degenerate challenge since the obvious classification into four sets would reduce effective discrimination to within five items for each problem. At a glance fifteen items could be ruled out.

Assume an investigation with four very similar problems which have very similar and many common response items. In this case, since almost all items would be relevant to any problem, almost no items need be ruled out.

These two situations typify two extreme degenerations of the structuring of a study unit which have to be overcome if the challenge to the student and the reliability of the diagnostic procedure are both to be sustained.

One approach which we have used extensively to overcome this difficulty is to frame each problem as a perspective on the total theme. From each perspective, some common features may be visible but they will have different relevance. Some features will only be apparent from one particular perspective. Also, features which are essential to one View may be incompatible with an opposing view. Yet, from the overall view they are equivalent in importance. If this construction is adopted then each of the twenty items is 'right' in the context of the study unit. They contain no intentional misinformation.

If this multi-perspective construction is achieved then greater reliance can be placed on there being an objective equivalence between the formal criterion for routing the student to a comment and the actually experienced difficulty or interest of the student.

One of the features we have tried to build in to the validation of the study unit is a comparative study of the difficulties registered in the actual response and the difficulties as reported by the students. The interpretation of this comparison is not straightforward since various situations can arise.

(1) The responses indicate difficulty and the student experiences difficulty, i.e. difficulty objective.
(2) The responses indicate difficulty but the student does not indicate difficulty. Either the criterion is wrong or the student is unaware of the difficulty.
(3) The responses indicate no difficulty but the student experiences difficulty. Either the writing, or the writing and diagnosis, are inadequate.
(4) Neither the responses nor the student record difficulty, but the communication still does not lead to understanding.

The analysis and data processing involved in sorting out these alternatives is considerable and until we have available the use of a computer and programming expertise we can only use the elementary non-statistical approaches outlined in Section 1.9. Nevertheless, we find that the case (1) is often experienced and is confirmed by the discussion comment actually helping. Eventually we hope that more comprehensive analysis of data in cases (2), (3) and (4) will also assist us to derive recommendations for improvement of the writing. At present we have to rely on a combination of analysis and intuition in sorting inadequacies out.

A more direct approach to validating the diagnosis is being attempted. In this approach the author actually takes a tutorial group in face-to-face circumstances through the study unit and discusses directly with the students what they find. This more intimate procedure, although subject to the additional activation factor of the author's presence, is proving a valuable means of experimenting with the diagnostic potential within a study unit.

The discussion procedure, then, is arrived at by a combination of educational insight, right structuring of material, logical test procedures and careful validation. Bringing about this combination appears to have in it an art as well as a slowly emerging science. Examples of discussion procedure can be seen in Appendix IV.

*1

*2

*3