The text presented here is not precisely as published by OUP, but modifications are minor. Illustrations are another matter. Where images used in the original book were not my copyright, I have in most cases been able to substitute links to coloured images on the web.

When this text was submitted as part of a PhD thesis in 1996, the Notes were greatly extended. Most readers may prefer to ignore them. They have been collected at the end of each chapter, with internal links leading to them and back to the text. They are a mixture of: simple page references; additional examples or quotations to justify generalisations; and some afterthoughts.

The subconscious process.

To the extent that the different phases can be separated, postulation of solutions occurs when a sufficiently clear definition of the problem appears to have been achieved. The 'creative act' has been described as a 'leap' of the imagination. It has been identified in science and art, as well as in engineering, as a bringing together by the subconscious mind of a number of ideas to form an original proposal. Poincaré's description of how he established the existence of a class of Fuchsian functions is often quoted. After weeks of fruitless effort he experienced a sleepless night during which, he reported,

"ideas rose in crowds: I felt them collide until pairs interlocked, so to speak, making a stable combination." [Note 1.]

Such introspective accounts by famous scientists contradict popular concepts of scientific discovery as a methodical and entirely logical process. (These have in any case been overturned in recent times by Popper (1963, 1968) and Kuhn, 1962).

In structural design the process is rarely so dramatic because of the limitations imposed by the traditional role of the engineer. However, it is a failing of many technologists that they greatly undervalue the capacity of their subconscious thought processes. As Candela, the Mexican architect and structural designer of innovative thin-shell structures, is reported to have said: "… the logical process is always an after-thought". [Faber 1963, p.15.]

The most common means of exploiting the subconscious is to devote a period of time to diligent conscious work and then to relax or work on a different problem for a while. A flash of inspiration is then liable to arrive at any place or time, as in the story of Archimedes.

Although the working of this process is hidden it is reasonable to hypothesize that new ideas must by synthesized from previously acquired elements of knowledge which appear relevant to the situation. This implies that a wide knowledge of material and structural behaviour is a prerequisite.

To generate a solution it must be necessary to search through this knowledge, assembling those elements that will fit together in a relevant and meaningful way and rejecting those that will not. To some extent this search must be a random process; there must be an element of chance in arriving at a satisfactory or an optimum solution in a given time. The time required to find a solution will vary from one individual to another for this reason alone.

What is it, though, that causes one individual to be consistently less creative than another? Apart from insufficiency of basic knowledge, there are a number of factors which inhibit the search for solutions. These may be classified under the following headings:

• conceptual and perceptual blocks
• emotional blocks
• cultural blocks

One of the more practical approaches to improving creativity is the identification and removal of these obstacles.

Conceptual and perceptual blocks to creativity.

Conceptualization can be thought of as the means by which the mind 'makes sense' of the infinite variety of natural phenomena; observing, classifying and describing. Experiences are grouped into coherent units or 'concepts' which the mind can grasp with reasonable ease. It is then possible to 'understand' a phenomenon by identifying the applicable concepts and studying their interrelationships as a part of that phenomenon. Conceptualization has much to do with language because it involves the process of naming things.

The context in which knowledge was acquired is rarely identical to that in which it is to be applied. It must therefore be broken down into its conceptual elements and the mental associations between them altered to build up an extensive system of 'cross-reference'. The problem itself will be perceived in terms of engineering concepts with which we are all familiar such as 'force', 'strain', 'failure', etc. To conceive a solution it is necessary to match elements of acquired knowledge with the elements of the problem.

Conceptualization is thus of extreme importance in creativity. Without concepts, thought would be impossible. They thus provide the potential for creative activity. However if they are too rigid we may be imprisoned by them. It is this latter aspect which now concerns us.

Concepts which are nowadays taught to undergraduate students in a one-hour lecture were developed with difficulty over a period of hundreds of years by men of the stature of Newton and Leibnitz. Unless the designer is of similar stature he must depend for most of his thinking on such concepts, transmitted through cultural, educational and professional contact, while developing a few of his own as a result of personal experience. However, concepts are discrete, whereas nature is a continuum. Therefore all concepts must be approximations.

The main lesson to be learned from these considerations is the need for a flexible attitude towards concepts, a full awareness of their advantages and limitations, and a wide store of basic facts and concepts ranging from the most abstract to the most concrete, and from the most generalized to the most particular, together with a well-developed system of cross-referencing.

One means of increasing flexibility is to employ more than one concept to 'understand' the same phenomenon, as illustrated in Fig. 15.1 where the stress trajectory concept of beam action is compared with the stress-resultant concept. While neither gives a true picture of the stresses in a real beam, it is obvious that a designer who was unaware of one of these two quite different concepts of beam action would find himself at a disadvantage.

Fig. 15.1(a). The state of stress of a thin beam represented by plotting the trajectories of principal stresses.

Fig. 15.1(b). The more common representation involving bending moment and shear force diagrams and diagrams showing the variation with depth of horizontal normal stress σ and shear stress τ.

A wide range of conceptual levels from the general to the particular also improves strategy in the search for a solution. The problem of supporting the point loads in Fig. 15.2 could be tackled by employing 'arch action' or 'axial compression' rather than 'beam action'. If the overall characteristics of such broader concepts are known, some may be eliminated without detailed analysis down to the level of stress. On the other hand, because of unusual circumstances, the assumptions implicit in these familiar categorizations may be invalid. There might be another means of support which could only be envisaged by the synthesis of more basic concepts.

Fig. 15.2. Alternative concepts of structural action applied to the problem of supporting two loads across a gap. (a) beam action, (b) pure compression, (c) arch action.

To further develop flexibility in perception, associations between concepts should always be questioned, as should the fundamental concepts themselves. Another person is often in a better position to question one's assumptions, but all of us are equally subjected to certain cultural habits of perception. It is therefore necessary to develop the habit of analyzing the validity of concepts and observing lack of fit with reality.

Consider the concept that 'concrete is brittle'. In the conventional theory of reinforced concrete beams, it is stated that bending causes the concrete to crack under low tensile strain, and in the minds of many students this is transformed into the concept that any beam made of concrete must be extremely rigid. This is, of course, borne out by observation of conventional buildings and bridges. Some years ago, in Italy, a canal with a parabolic cross-section was lined with precast prestressed concrete planks which were picked up by their ends so that they bent into a shape suitable for placing in the excavation (Fig. 15.3). This idea could hardly have been conceived by a designer who thought of concrete as a material which cracked as soon as it was bent. [Note 2.]

Fig. 15.3. Preconceptions of the 'rigidity' of concrete are shattered by the sight of these flexible prestressed planks, used to line a channel supplying a power station at Pontecorvo, Italy. (Engr: S. Zorzi, c.1962.)

The opposite of flexibility in the use of concepts is known as 'set'. This can be described as the tendency to continue to use concepts and strategies which have proved satisfactory for previous problems despite their unsuitability to the problem in hand. [Note 3.]

We are indebted to the design methodologists for their identification of a particular form of perceptual block known as 'analogue take-over'. An example of this in architecture occurs when the designer becomes more involved in producing beautiful drawings than he is in devising habitable buildings, so that the nature of two-dimensional representation employed in design is reflected in the nature of the resulting buildings. [Note 4.] Unfortunately, many design methodologists themselves fell into the trap and became more concerned with fascinating diagrams and mathematical symbols than with the complexities of reality.

Blocks associated with personality traits.

There are a number of blocks associated with the designer's psycho logical functioning and emotional responses to the work environment and to his colleagues.

The phenomenon of 'set' has already been mentioned. The concept of set is not approved by all psychologists but there is evidence that certain types of person find it harder to break out of acquired habits in problem-solving than others.

A somewhat related phenomenon is the tendency of most people to imagine constraints on their behaviour which do not in fact exist. A very common illustration of this is the puzzle which requires that nine dots on a piece of paper be joined by drawing four continuous straight lines through them (Fig. 15.4) without lifting the pencil from the paper. For some reason most people limit their attention entirely to the enclosed region and do not realize that for a solution two intersection points must lie outside the area bounded by the dots.

Fig. 15.4. The nine-dot problem. Asked to join all the dots by four continuous lines without lifting their pencils, subjects often fail because of self-imposed constraints.

The most common emotional block is the fear of appearing foolish in the eyes of one's colleagues. This may inhibit the expression of new ideas unless the originator is sure they will stand up to analysis. Since thorough engineering analysis often takes days or weeks, ideas may be suppressed because they cannot be quickly proved right, rather than because they are likely to prove wrong.

A second personality factor has been described as the 'intolerance of ambiguity'. This is a sense of unease which is felt while a situation is unresolved and results in a desire to bring the design process to a speedy conclusion.

Both these factors limit the range of the search for relevant elements of knowledge and lead to the premature adoption of any practicable solution.

Further personality factors worthy of mention are an extreme degree of caution, distrust of colleagues and a lack of application and initiative.

It is debatable whether such personality factors can be altered significantly without some form of psycho-therapy. Simberg (1964), however, is confident that the designer can alter his own behaviour by 'being honest with himself' and, when he meets an intractable situation, checking against a list of mental blocks to see which may be interfering.

Cultural blocks.

The type of thinking necessary to solve the nine-dot problem has been popularized by de Bono (1967) as 'lateral thinking'. Longitudinal, or logical thought is highly valued in our educational system and rigidity may be partially a cultural phenomenon.

Social custom also values conformity, yet non-conformity is, after all, what creativity is about. Creativity is enhanced by a sense of playfulness (see below) and this is at variance with the general attitude that 'work' must be taken seriously.

Cultural constraints on problem solving also include preconceived minimum standards of aesthetics, ethics and environmental conditions. We have already seen how businessmen resisted the use of space-truss roofs for shopping centres because they reminded them of factories.

Although suspended roofs, inflated membranes and 'stressed-ribbon' footbridges do have their practical problems, the slowness with which they have been introduced must have much to do with attitudes in the community and the professions towards sag, flexibility, and perceived impermanence in structures. The fact that rope bridges have been in use for centuries in Asia and Africa suggests that this is a matter of cultural conditioning.

Simberg (1964) provides a useful list of mental blocks to creativity which is worth checking as an exercise in self-diagnosis.

Formal techniques for avoiding blocks.

The various formal design methods that have been proposed aim to avoid mental blocks by taking the process to some extent out of the hands of the human designer. Alexander's prime motivation was to transcend familiar concepts which locked the designer into conventional modes of response. To use the language of structural engineering, concepts such as 'beam' or 'truss' carry with them associations and preconceptions which may be irrelevant to the forms of structures required by a changing society. As Alexander saw it the designer's main task is to reduce the problem to its fundamental elements, transpose these and their relationships into mathematical symbols, and then stand back whilst the mathematics produced the optimum solution.

Although Alexander has joined in the rejection of his earlier ideas, we are indebted to people like him for the idea of formalized analysis of design problems using more elementary concepts than we would otherwise. [Note 5.]

Systematized methods fall into four categories:

• those such as Alexander's which attempt to reduce the problem to fundamental requirements and means of satisfying them so that totally new solutions are synthesized;
• those such as AIDA which reduce the major problem to a series of substantial sub-problems for which standard 'sub-solutions' can be readily chosen;
• those such as System Transformation and Functional Innovation which seek to improve existing solutions by evolutionary modification; and
• those which attempt to remove mental blocks directly by the use of key words or phrases to stimulate association of normally 'remote' concepts and ideas (that is, concepts which would not normally be considered to have any relevance to each other). [Note 6.]

Inevitably, the first three techniques, which employ such aids as matrices, decision trees, nets and morphological charts may also be classified under 'Problem Definition and Clarification', because of the intimate link between that phase and the postulation of solutions.

All such techniques aim at indicating to the designer new relation ships between the elements of a problem and combinations of sub- solutions which he would not otherwise have envisaged.

When such methods are listed side-by-side, as in Jones (1981) it is difficult to determine what makes one of them really different from the others except perhaps the degree of breakdown recommended. The reader is referred to Jones for a detailed description and an assessment of the value of each method.

It was too much to expect these techniques to really take over from the human designer. Their major failing is that at least as described in Jones they place insufficient stress on cycling and (except for Boundary Shifting) fail to allow for modification in the original definition of the problem.

It may be possible to develop computerized techniques which compensate for these defects as both hardware and skills in programming improve. The best that can be hoped for in the near future is that formal methods will provide the designer with a certain stimulus as much by altering his attitude to design as by suggesting actual ideas to him, and enable him to manipulate these ideas and communicate them to others somewhat better than he would have done otherwise.

Perhaps greater success may be expected from methods which have the more limited objective of stimulating the production of new ideas (without directly guaranteeing their worth) in the hope that some will prove, on investigation, to be of great value.

Methods for increasing the quantity of new ideas.

All the various breakdown and re-assembly methods, particularly morphological charts, may be used for this purpose because they may suggest associations of practical worth between remote elements. However, the act of breaking the problem into its elements requires a certain amount of imagination if it is to provide any worthwhile recombinations.

In this, the use of a 'check-list' may be helpful. In Chapters 16 to 19 of his renowned book Applied Imagination, Osborn (1957) provides a lengthy check-list intended to stimulate creativity. Typical 'idea-spurring questions' are (condensed):

(1) What is like this? (2) Does the past offer a parallel? (3) Is there something I could copy? (4) Should it be stronger, bigger? (5) What if it were blown-up or miniaturized? (6) What if it were lower, shorter, lighter? (7) What about a change of pace or sequence? (8) What is the opposite? (9) What if I say it in reverse? (10) How about up-ending it?

Questions (2) and (3) are standard engineering practice but the others appear strange and perhaps irrelevant. However, (7) could have inspired the suspended multi-storey building or the lift-slab technique. Osborn quotes the way in which the Kaiser company constructed the deck-houses of World War II Liberty ships upside-down as an example of (10). [Note 7.] (This approach speeded construction because the welders could work down-hand.) The upside-down suspension bridge (such as the Rio-Colorado, Chapter 3) could also have been inspired by such a question.

One drawback is that many of Osborn's questions are more suited to the entrepreneurial aim of 'inventing anything for profit' than to the specific problem-solving inventiveness required of the structural engineer. More specific lists may be prepared for any more specialized field, and examples are provided by Jones (1970, pp.362-9). All the available checklists reflect the preponderance in design methodology of mechanical engineers and architects and there is a need for the development and testing of such methods in the area of structural engineering.

There are a number of rather exotic techniques which aim to free the subconscious from cultural and personality-based blocks. The two most well-known attempt to overcome such blocks by establishing an atmosphere in which unusual ideas may be expressed without fear of hostility or ridicule and in which pride is based on originality of ideas rather than proof of practicability. The accent is on separating the 'production-of-ideas' stage of the process from the 'cautious-appraisal' stage.

'Synectics', developed by Gordon (1961), is based on two mental tricks; "making the strange familiar"; and "making the familiar strange". The second may be done by using Personal Analogy, Direct Analogy, Symbolic Analogy and Fantasy Analogy.

Personal Analogy is to imagine oneself to be the item under con sideration. The classic example of this is due to the chemist Kekule. In trying to envisage the construction of the benzene molecule he imagined himself to be a snake swallowing its own tail and realized that the molecule must take the form of a ring rather than a linear chain. [Gordon 1961, p.23.]

Direct Analogies are drawn from other disciplines or from nature. The idea of the caisson is supposed to have come to Brunel when he observed the shipworm lining its tunnel as it progressed through the wood. [Gordon 1961, p.24.]

Symbolic Analogy is the expression of the desirable features of the hoped-for solution in a compressed poetic phase. This may lead to verbal associations which spark off new trains of thought.

Fantasy Analogy is based on expressing the fulfilment of the wish to solve the problem without any regard for the limitations of 'reality'. Thus the designer might say 'If only the law of gravity were abolished we could …'

The state of mind induced by such procedures is far removed from that in which the traditional sober engineering graduate has learned to approach his tasks. It is more akin to the mental play of which Einstein wrote (see below) and is thus frowned upon by our society as incompatible with the work ethic (cultural block). However, its proponents claim to have demonstrated its effectiveness.

'Making the strange familiar' comes quite naturally (and usually prevents a true appraisal of the strange). It therefore receives less attention from Gordon than the opposite process.

The other well-known technique is 'Brainstorming', developed by Osborn. This is based on the principle of deferred judgement and may be used by the individual, though it is usually practised in groups. The idea is to note down or verbalize ideas as quickly as they are generated, so that premature rejection on the grounds of strangeness, or fear of looking foolish, is avoided. The difference between a group brainstorming session and a normal committee meeting is entirely in the atmosphere; the receptivity to new ideas and the refusal to criticize or assess. It is obviously very easy for the former to degenerate into the latter and some skill and experience is called for on the part of the leader and participants.

Both Synectics and Brainstorming seem to benefit from some intervention on the part of the leader to stop discussion at appropriate times, change the stimulus, concentrate on one particular idea, or call for convergence to a solution. Neither is the fully automatic procedure it might appear to be at first sight.

Lessons from psychological research into creativity.

There has been much psychological research into problem-solving and creativity. Descriptions of problem-solving technique bear a close resemblance to formulations of the design process. Wallas (1926) identified four phases: preparation, incubation, illumination and verification. [Note 8.] The difficulties of 'set' and self-imposed imaginary constraints can also be demonstrated. However research into problem-solving has concerned goal-centred problems usually with only one 'correct' answer and, although it has much interest for the engineer, will not be included in this discussion. Research under the heading of 'creativity' has addressed the wider phenomenon of the production of original ideas. The price paid has been that the results are much less conclusive. [Note 9.]

The first problem was to decide whether such a phenomenon as creativity existed at all as a definable trait, separate from, but just as real as 'intelligence'. The debate has been at times quite bitter and still continues. It has been recognized however that brilliant people in even the most rigorous sciences actually produce their best ideas in very illogical ways, even though they later explain and prove (or reject) them using logical processes. Poincaré's and Kekule's accounts have already been mentioned. Einstein wrote that he thought in terms of non-verbal non-symbolic images:

"The physical entities which seem to serve as elements in thought are certain signs and more or less clear images which can be 'voluntarily' reproduced and combined … this combinatory play seems to be the essential feature in productive thought … The above mentioned elements are, in my case, of a visual and some of muscular type." [see Hadamard 1945, p.142.]

Difficult as this passage is to understand, it proves that conscious rational thinking of the type involved in intelligence tests is not the only way of synthesizing new ideas. The use of the word 'play' is in keeping with the view of psychologists that creative effort is generally enjoyable and not always purposeful. Most psychologists now feel that creativity is a faculty which all people possess to some degree.

The definition of creativity is closely related to the method adopted for measurement. Gilchrist (1972) listed three main types of measurement by identification of creative products; identification of creative behaviour; and assessment of creative potential using psychological tests. [Note 10.] Each contributes something to the understanding of the phenomenon, but each has considerable disadvantages.

In spite of these a surprisingly consistent picture of the personality characteristics of people identified as 'creative' does emerge from the research. These will now be considered in some detail because the findings, particularly those relating to architects, are of special interest in engineering practice, and to some extent conflict with conventional preconceptions about 'creative' people.

Observed characteristics of creative people.

This summary owes much to Gilchrist (1972) and Vervalin (1962).

All creative subjects exhibit great intellectual curiosity. Intelligence is high, but shows a bias towards the subject's own field. Creative writers score especially high in verbal intelligence while creative scientists score in spatial relationships and mathematical concepts. Subjects are especially discerning and observant, alert and able to concentrate, but are also able to shift attention appropriately.

Creative people possess a wide range of information which they are able to combine freely, to sort and to extrapolate. They are sensitive to their own psychological intricacies and have few repression or suppression mechanisms. They are emotionally responsive. They almost unanimously report unhappy childhoods. As a group they tend to be introverts, but extroverts amongst them are no less creative. They are not particularly conscious of what others think of them and are thus freer than the average person from conventional restraints and inhibitions. They are not conformists, but are not non-conformists either, being genuinely independent in their ideas.

They do well when independence in thought and action are called for, but less well in situations requiring conforming behaviour. They are less interested in simple facts than in their meanings and implications. They are intellectually verbal and communicative and are not interested in policing their own images or impulses, or those of others. Barron (1962) found that creative people have a preference for complexity in figures and drawings. [Note 11.]

MacKinnon, who made a major study of creativity in architects, summarized his results under three headings, or 'personality domains' in which creative people differ from the average. [Note 12.] These he identified as (a) socialization; (b) richness or complexity of psychological development; and (c) psychological health or adjustment.

Under (a), MacKinnon found that creative architects

"displayed less inhibition of impulse, had more unusual and unconventional views, were less constrained by conventional expectations, and were less concerned with making the right impression on others."

This was not to be taken as meaning that they were socially irresponsible but that their behaviour was guided by aesthetic values and ethical standards which they had set for themselves. They were still described by the investigators as "genuinely dependable and responsible". Most architects were found to be introverted but creative ones were even less concerned with social activity than those in the control group. On the other hand, they were more dominant in their social relationships.

Under (b) creative architects were found to be above average in flexibility, psychological insight concerning others, aesthetic sensibility, preference for complexity and "feminine" interests. (Note that the latter was combined with a high level of dominance, self-assertion and independence of judgement.) All were found to think "intuitively".

Under (c) several other studies besides MacKinnnon's have suggested that creative people have more psychopathology (indicated by, for instance, more hysterical, paranoid or schizophrenic responses), less restraint of impulse, and more openness to emotional drives than average. [Gilchrist 1972, pp.71-2.] However, they also have high scores on 'personal effectiveness' and ego strength and are more able than the average person to cope with their internal tensions. Some investigators have suggested that their higher scores on psychopathology in questionnaire-type tests are simply due to a greater readiness to allow their impulses access to the conscious mind.

Whichever is correct, the general picture is one of a lively tension between impulse and control at the level of either personality or conscious mental functioning. This ties in with a development of the Freudian theory which holds that creativity is 'regression in the service of the ego'. The subject is able to call on his 'childlike' impulses by an intentional temporary lessening of control. [Note 13.]

The engineer differs from the architect in the greater degree to which he is subject to discipline by the limitations of theory and ultimately by experimental fact. In view of the scarcity of studies of design engineers it might be worthwhile studying what has been established about the scientist, who is in a similar position. [Gilchrist 1972, pp.75-85.]

Unfortunately, the difference between the creative and non-creative person is much less evident in science than in architecture and art. Several studies suggest that initiative, dominance and motivation - qualities which are not associated with everyday concepts of 'creativity' - are the clearest measures which differentiate the creative scientist or engineer. Another suggested that wide reading habits were the only factor which distinguished creative industrial scientists when creativity was measured by contribution to successful patents. Results such as this are possibly due to the problems of isolating technological creativity from organizational ability and determining individual contributions to team work. In some cases 'creativity' has simply meant the number of papers published by a research scientist. It must also be remembered that industrial research scientists are a highly select group and differences amongst them would be much smaller than between the group and the general population.

Drevdahl (1956, p.26) has summarized the evidence as suggesting that radicalism in the scientist is likely to be more intellectual than personal. A scientist must have a reasonably conventional approach to master the knowledge and methods of his discipline, and his originality and unconventionality are likely to be expressed predominantly within the socially acceptable framework of the scientific field.

Stimulating creativity.

Probably the major aim of any engineer who takes the trouble to study creativity is to improve his own ability in this regard. The intuitive reaction of many people is to assert that 'creativity cannot be taught'. The validity of this statement depends to a large extent on what is meant by 'teaching' and 'creativity'. However, many organizations, particularly in the United States have sent employees to courses intended to increase their creativity. These are based on psychological techniques such as Synectics and Brainstorming, on prescribed strategies and on more formal methods such as Value Analysis. A reasonably available self-instruction course is the Parnes Creative Behaviour Course (Parnes, 1967a and 1967b) which is a development of the work of Osborn (1963).

A great deal of research has been done in an attempt to check the effectiveness of such courses. Many investigators claim to have found improvement in performance on divergent thinking tests and in problem- solving techniques.

It is perhaps appropriate to repeat here that at least one investigation has suggested that traditional courses in Engineering actually reduce the creativity of students passing through them. This view has also been expressed by some eminent engineers and academics.

All of these conclusions are of course open to challenge on the basis of experimental design and interpretation of statistics and most physical scientists would feel unable to reach a firm opinion on the evidence available.

Creativity research and structural design.

What, then, is the relevance of psychological studies to the design process in structural engineering?

For the critic, attempts to define and measure creativity assist in analyzing the quality of structural design. Findings concerning the process of creative thinking and problem solving offer some guidance to those attempting to reproduce this behaviour with the digital computer.

For the practising designer, there are lessons about the nature of creativity and the creative personality which may help him improve his own attitude. He may, for instance, be able to identify himself amongst the four groups into which Westcott (1968) divided the subjects from his problem-solving experiments. Westcott's detailed descriptions are too lengthy to reproduce here but the four classes are: successful intuitive thinkers; wild guessers; successful problem solvers; and cautious, careful failures.

If he does not like what he sees, the designer may find a role for himself amongst the different styles of personality which Gough and Woodworth (1960) identified within teams of research scientists. These included the Initiator, or 'ideas man'; the Diagnostician, who gets projects out of trouble spots; the Scholar, who is well-informed but indecisive; and the Artificer, who takes barely formulated ideas and makes them work.

A familiarity with the basic concepts evolved for research into problem-solving and creativity should help the designer to analyze his own creative process as well as the contribution of those around him. It is, however, the insights afforded into the personality of the creative person that are of most potential value.

Many engineers must be surprised to find that introversion is no bar to creativity, and that what our society sees as the 'feminine' traits of creativity can co-exist with qualities such as self-assertion, dominance, and lack of concern for the conventional wisdom. Realizing that others see us as servile and self-effacing (Chapter 8) we might be encouraged to cultivate these more extreme characteristics in ourselves and be more willing to tolerate and even appreciate them in others.

Perhaps the most significant trait revealed by psychological research and by a study of successful engineers from Brunel to Nervi is a level-headed courage which in the past would have been described as 'strength of character'. It has been said that: "any engineer who has achieved anything notable must expect to have his share of … 'painful interviews' where he must return to the client to admit to mistakes". [New Civil Engineer 28 Sept. 1978, p.34.]

While engineers should maintain a cautious approach where the safety of human life is involved, there is ample evidence in the literature that innovation stems from those willing to take risks in terms of finance or reputation.

Creative people also tend to score highly in the Asch test which measures the ability of an individual to stick to the evidence of his senses when group pressure is applied to the contrary.

Notes.

Note 1. This phrase first appeared in Poincaré's Science et Méthode, Flammarion, Paris (1908). It appeared in English in Poincaré, H. (1913) The foundations of science (trans. G.B.Halstead), Science Press, New York. It is reproduced in Vernon (1970), p.81, which contains a large extract in pp. 77-88. [Return.]

Note 2. For the Italian canal lining see Leonhardt (1964), p.569; Gregotti (1968), p.101; and Engineering News Record 24 May 1962, pp. 40-2.
Another example of rigidity in thinking is given by Cassie and Napper (1966), who point out the limitations of the conventional idea that forces 'flow' from the top of a building down through the structure (increasing in magnitude as the weight of the various members is added to them) and down into the foundations. They suggest it is more useful, and more correct, to imagine the structure as being squeezed between the loads pressing down from above, and the reaction pressing up from the foundation. [Return.]

Note 3. See e.g. Freeman, J., Butcher, H. J. and Christie, T. Creativity: A Selective Review of Research (2nd edn.), Society for Research into Higher Education, London, (1971), pp. 47-52. [Return.]

Note 4. Branner (1963) suggests that the change in the facades of Gothic cathedrals from a sculpted three-dimensional form to an almost flat plane with linear detailing corresponds with the introduction of scaled design drawings on parchment.
Nervi warns designers against the danger of analogue take-over in Nervi (1956), p.28. [Return.]

Note 5. Cited in Broadbent (1973), p.289 where Alexander is quoted as describing his early approach as a "painful and drawn-out misunderstanding". Bazjanac (1974), p.8 states that Alexander "acknowledged the complete failure of the concept of decomposition in the improvement of the quality of design solutions". Both these references are to an interview reported in the Design Methods Group Newsletter, Vol.5, No.3, (Mar. 1971) pp. 3-7. [Return.]

Note 6. Alexander's method is set out in Alexander (1964). Jones (1971) describes 'AIDA' on p.310, 'System Transformation' on p.316 and 'Functional Innovation' on p.331. [Return.]

Note 7. Osborn (1957), Chapter XIX, p.281-2. [Return.]

Note 8. Excerpts are reproduced in Vernon (1970). The relevant passage occurs on pp. 91 and 92. Wallas states that the first three phases were defined by Helmholtz in 1891. [Return.]

Note 9. The main source consulted for research on Problem-Solving was Newell and Simon (1972). The categorisation of problem solvers was found in Westcott (1968), 140-7. [Return.]

Note 10. Gilchrist (1972), p.16. A concise account of typical creativity tests appears in Barron (1969), Chapter 3. [Return.]

Note 11. Reported by Gilchrist (1972, p.69) who gives several sources for this statement. See also Barron 1969, p.96. [Return.]

Note 12. MacKinnon (1962). Excerpts are to be found in Vernon (1970), pp.289-311. The reference to personality domains is on p.309. See also MacKinnon (1964). [Return.]

Note 13. A more sceptical view of creativity research is expressed in Broadbent (1973), Chaps, 1, 3, 13, 14, 16, 17. Stein (1967) provides an interesting though very wide-ranging summary. Readable introductory texts consulted included Gilchrist (1972); Davis and Scott (1971); Barron (1969); Parnes (1981); and Davis (1981). [Return.]

[Top.]

The art in structural design navigation:
[Chapter 14.] [Chapter 16.] [Contents] [References]

Site navigation:
[Aesthetics of built form.] [Papers.] [Work of Jörg Schlaich.]
[John Monash's early engineering.]