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. The sources are listed under Image Acknowledgements.

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.

Introduction.

The immediate purpose of the foregoing chapters has been to introduce to the reader a number of topics on the fringe of what is conventionally taught (or thought of) as 'structural engineering'. The aim has been to show how these may affect the work of the structural engineer and the form of his structures. Some examples of these effects have been provided, and more are to be found in the source material for the assignment questions.

A more indirect, though no less important, aim is to see what lessons can be learned by the structural engineer from the approaches adopted in related disciplines. Certainly the average engineer should see himself less as a blind servant of those who present him with well-defined questions. He should join the other members of the project team in taking a wide view of the entire project.

However the most striking impressions to be gained by crossing for a while into other fields are of the comparatively low level of interest amongst structural engineers in developing a conscious philosophy for design, and of the low level of interest in criticism in structural engineering. These two factors are of course inter-related.

In recent decades the situation has been improving, with increased discussion of the philosophical aspects of design by authors such as Billington in the United States and Arup, Dunican, Happold and Harris in the United Kingdom. In the 1950s there appeared a number of classic texts by the famous engineers Torroja (1958a,b) and Nervi (1956, 1965a). On the more strictly technological side there has been lively debate about Codes of Practice; continuing discussion (mainly academic) about the definition of safety factors and performance criteria; and increasing interest in the art of mathematical modelling of structures to suit available theories and computer programs.

Nevertheless there is still a need for a comprehensive treatment which discusses the appropriate weight to be accorded to the many influences which affect the form of structures, from aesthetics through economics to construction methods, material properties and site conditions. This book is only a contribution towards that effort, but it is appropriate in the concluding chapter to discuss some views on the philosophy of structure and examine the possibility of developing a discipline of worthwhile criticism in the field of structural engineering.

Factors which influence the nature of structures.

The factors which influence the form and nature of a structure may be summarized briefly as follows:

(a) The structure should conform as far as practicable to the requirements for space, circulation and support defined by the function for which it is intended (education, worship, retailing, power generation, ore handling, etc.).

(b) It should provide protection, if required, against cold, heat, wind, rain and snow, and be proof to some defined extent against earthquakes.

(c) The form of the structure should take into account the limitations imposed and opportunities offered by the nature of its particular site. These include the topography, the nature of the soil and rock, sub-surface impediments such as under ground services, adjoining buildings and environmental constraints such as requirements for light and lines of sight.

(d) The form of the structure should be appropriate to the efficient use of the mechanical properties and methods of fabrication of the chosen material.

(e) The structure must efficiently support the estimated loads, providing an adequate factor of safety against buckling or material failure.

(f) It must support these loads without excessive deflection or vibration which might alarm the occupants or interfere with serviceability.

(g) The structure must have a sufficient degree of resistance to fire to permit escape of occupants and access of fire-fighters.

(h) The material must not suffer significant loss of strength, within the projected life of the structure due to corrosion or other effects.

(i) In service, the material must not show signs of distress due to excessive strain or movements incompatible with those of other materials.

(j) The structure should be aesthetically pleasing, at least visually, and preferably in the full sense of the word.

(k) The structure should have the minimum compounded cost of construction and maintenance required to achieve the desired standards negotiated between the client and design units.

(l) The material or standard components chosen must be commercially or economically available in the required quantities at the chosen location.

(m) It must be possible to fabricate and construct the structure efficiently using existing techniques or practicable developments thereof.

(n) It must be possible to predict the behaviour of the structure using present theoretical concepts and computational modelling techniques.

(o) The operation of all these strictly legitimate factors may be affected by politics both within and outside the design team, problems in human relationships, attitudes and communication, and the level of creativity, motivation and interpersonal skills of the designer.

Many questions have been begged [evaded; side-stepped] in formulating this list and these are the main concern of the philosophy of structural design.

Item (a) raises the question; to what extent should the structure be distorted beyond the needs of efficient transmission of force in order to cater for functional requirements? What price in terms of structural inefficiency and cost should be paid for a given advantage in functional efficiency?

In designing against earthquake forces (b) it is impossible to guarantee complete resistance to the most severe earthquake on record and would certainly be uneconomic. The philosophy adopted is to accept the inevitability of some damage but to bear the cost of ensuring that important structures such as hospitals and fire-stations, and dangerous ones such as oil storage tanks remain functional even though perhaps somewhat bent. Similar considerations apply in the design of jetties, dolphins and bridges where there is a danger of accidental impact from ships.

The influence of site conditions on form (c) is well illustrated in engineering journals and the choice of form to accord with material properties (d) is a favourite topic of writers on the philosophy of structural design.

Items (e) to (i) raise the questions of estimation of loads, determination of safety factors and choice of criteria for performance. Again a great deal of material is available in the engineering literature and in books such as Blockley (1980).

Live loads on floors are determined by surveys and their statistical distribution is fitted to a standard curve such as the Gaussian. Design is based on a value having only a small chance (perhaps 5%) of being exceeded. Traffic loads on bridges are determined by a combination of regulation and observation. Loads in silos and bunkers are more a matter of complex mechanics. Loads due to wind, waves and earthquakes are fluctuating and are influenced by the vibrational response of the structure. A great deal of thought has to go into devizing mathematical representations of their effects which result in the design of adequate structures.

There are two major philosophical considerations in determining safety factors. One is to assess the amount of risk in terms of life or money which the public and the owners are willing to accept. This can be done to some extent by looking at the risks involved in travelling by various means of transport or in crossing the road or being killed in sports, on the assumption that these chances are acceptable to those who take them. However people have much higher expectations of the safety of buildings.

The second consideration is to decide how the desired safety is to be ensured. The approaches adopted have included the 'permissible stress' method which places an upper limit on the ratio of ultimate strength to design load, and lately the 'limit state' method which extends this approach to criteria other than strength and emphasizes the probabilistic nature of design.

The art of choosing criteria for design against overwhelming hazards has already been mentioned. Even for such apparently straightforward phenomena as deflection and the connection between cause and effect; between design parameters on the one hand and physical damage and observer reaction on the other; is so tenuous that limits must be imposed in a relatively arbitrary manner.

The influence of aesthetics (j) and economics (k) has already been discussed at some length.

Conventional methods of fabrication and construction, which are determined partly by economics and partly by the historical development of the industry have a considerable effect on form (m) although the designer is often unaware of this because he is so accustomed to convention. This aspect is also well represented in existing literature on the philosophy of design.

The process of devizing theories of structural behaviour (n), which involves the art of idealization and simplification, is less well represented. Even text-books on theory tend to gloss over the question of what factors have been omitted from the formulas, what approximations have been made, and what is the accuracy of prediction [Note 1.] Some studies have been done to compare the predictions of different computational methods and to compare theoretical predictions with actual behaviour [Note 2.] These should be common knowledge in the philosophy of design.

No theory is satisfactory in its raw state for application to design. The relationship between classical plate theory and code rules for the design of reinforced concrete floors is a prime example. The process of devising code rules on the basis of a theory is a very complex task. Further approximations and simplifications must be made, in the light of conventional practice and typical geometry of practical structures to arrive at a routine design procedure. These must to some extent allow for the practical factors which the original theory omitted (such as creep in concrete, imperfections in the geometry of struts).

A significant number of engineers see codes of practice as an unnecessary (and insulting) limitation on the initiative and skill of the designer. Others currently feel that the codes have been taken over by academics and made unnecessarily bulky and complex. The resulting debates have generated a further contribution to the philosophy of design [Note 3.]

Thus, subjective decisions must be made concerning a wide variety of questions, before design can proceed. Admittedly many of these answers are determined for the individual by convention and many are laid down in codes of practice. The sum total of these answers constitutes the philosophy of design.

The literature of the philosophy of structural design.

In the literature the ultimate concern of the structural engineer is understandably seen to be the determination of the appropriate nature and form of the structure. The problem of force transmission is seen as central, and questions of economics, aesthetics and communication within the design team as peripheral. It cannot be said that any of the existing texts on philosophy covers the entire field set out above, although Torroja's Philosophy of Structures comes close, and their titles often refer to 'Form' rather than 'Philosophy'.

Many authors are content simply to describe forms, either at the level of basic geometry (e.g. Gheorghiu and Dragomir, 1978), or at that of elementary mechanics (e.g. Engel, 1967). There have also been interesting studies of form in Nature, such as D'Arcy Thompson's classic On Growth and Form (1917). More recent examples are listed in Note 4.. Not surprisingly, these rarely go into detail regarding those factors such as engineering economics, function and problems of fabrication and construction, which distinguished technology from nature.

Another approach is to trace the historical development of form in structure over recorded history or during the 20th Century. Although these books provide the reader with an appreciation of the rich selection of forms available, and the better ones stimulate his appreciation of the influence of the various factors by examining their effect in different ages and locations, they are mostly wide-ranging and often intended for the lay reader or architect.

A further category of books and papers does offer guidance to the practical architect or engineer. This ranges from prescriptions concerning the shape and dimensions of individual members to philosophical discussions of the factors listed in the previous sections of this book and the extent to which they should be allowed to influence the overall form of a structure.

Much of the more basic information occurs as individual chapters near the beginning or end of conventional text books on structural 'design'. The following treatment will therefore summarize briefly the more philosophical material.

Torroja's Philosophy of Structures.

Probably the major classics of the philosophical discussion of form are Torroja's Philosophy of Structures (1958) and Nervi's Aesthetics and Technology in Building (1965a). Their authority stems from the fact that both authors established highly successful practices, so that their ideas have been put to the test in the commercial arena.

The largest part of Torroja's book consists of a review of the characteristics of the main engineering materials and of the various elementary forms and a discussion of their relative merits. He then comes to the load-bearing function of the structure which he describes as: "In principle - the primary criterion by which the designer must make his selection of structural type and must plan the completed structure" (1958a, p.230).

At the same time he recognizes the complexity and diversity of the subject matter of his previous chapters and the fact that in considering load-bearing alone he is eliminating "constructional, aesthetic and functional" factors. Disregarding these, for the time being the structure should achieve the transmission of forces to the ground with a minimum of material, or a maximization of safety, and with a "sensation of unforced stability". All constructional elements should contribute whenever possible to the strength of the structure. [All quotes from p.231.]

Later he reaffirms the desirability of bringing the loads to the foundation by as direct a path as possible (p.202-3), arguing in favour of "a clean and clear structure composed of a small number of elements whose respective functions are well defined" (p.239) though this is not to preclude structures such as composite bridge decks where the top surface performs two functions simultaneously.

After a brief review of the various categories of structure he goes on to discuss the relative merits of tension, compression and bending, pointing out that while tension is theoretically the most efficient means of resisting load it is very rarely that it can be introduced without the inclusion of members acting in the other modes (p.231-5).

Torroja then emphasizes the importance of rigidity in structures both as a means of controlling deflection and preventing buckling, and as a means of organizing the load-sharing between individual members. He points out that while rigidity in a structure, like continuity, has great advantages it also leads to increased stresses due to thermal effects and foundation settlement. The importance of joints in influencing form and behaviour is also emphasized (pp.235-7).

He then reaffirms the importance of material properties in determining structural form and reviews their major effects (p.240-4). In a later chapter he also mentions the cost of fabrication and construction and makes a plea for simplicity in these areas also (p.325). In discussing their influence on form, Torroja points out the difficulty of deducing abstract principles and contents himself with a number of case studies.

In the later Chapters, Torroja turns his attention to the general philosophy of design, especially to what might be called the 'aesthetic purpose' of the designer, and points out that money should not be the only criterion. Other things to be considered are: a "clear distribution of stresses"; a "good balance between loads and strength" (Factor of Safety); an opportunity to provide "greater strength and safety at a relatively small increase in cost"; and "a better ability to withstand abnormal and unforeseen loads" (p.323). (This last concept is nowadays covered under the heading of 'sensitivity').

At the less concrete level Torroja sees a need for "a complete compatibility between the technical and aesthetic properties" and later states: "The functional purpose and the artistic and strength requirements must be considered integrally from the initial conception of the project (p.327)". Although the structure is an essential part of any building, "the degree to which each part … should be made dependent on the others is a matter that should be approached with a broad and open mind". By the same token, the structure should not be forced to fit a preconceived arrangement of space which has been arrived at without due consideration of structural necessities. He points out that often a slight change in the initially selected functional arrangement will simplify the structural problems.

Torroja also warns against a basic approach which values originality for its own sake or expresses an immature desire to impress (p.283). This viewpoint is modified somewhat by another statement that our present competence in the field of structures provides unrivalled opportunities for "personality and variety" in form (p.276).

In considering the question of aesthetics Torroja discusses at length the concept of Truth as a criterion of Beauty. He distinguishes between "obvious" and "hidden" structure. [pp.270-1, 276.] The 'hidden' structure may sometimes be designed purely on mechanical and economic principles but more often it has an indirect influence on the outward form and hence its design cannot be isolated from aesthetic and other considerations.

He considers it wrong and offensive to the observer if the designer goes to special lengths to conceal the nature of structural action or to consciously mislead (p.271). However, in many cases, such as the skeletal frame of a curtain-walled building, it simply happens that the true working of the structure in is indiscernible. Torroja feels that in these cases there is no need for the designer to make a special attempt to reveal it; for instance, it would be ludicrous to paint the line of the reinforcement on the side of a concrete beam to help the observer understand how it worked (p.276).

He sees a subsidiary problem in the ignorance of the general public about the way in which any but the most simple structures operate (1958a, pp.270-1). He quotes the case of the 'multi-lobular shell' (two adjacent cylindrical shells spanning between end walls) which the layman is likely to interpret as a pair of vaults from which the central support has been mysteriously removed (1958b, pp.31-40). (He is referring here to his Frontón Recoletos, Fig. 18.1.)

Fig. 18.1. The roof of Torroja's Frontón Recoletos, Madrid (1935) spanned longitudinally. However laymen interpret it as a double arch spanning laterally which lacks a necessary line of support at the central cusp. [Photo: (Univ. Stuttgart).]

The task of interpretation is made even harder by the influence on form of the other factors, particularly the method of construction, of which no trace is usually left, and the possible combinations of live load patterns which are not self-evident to the observer.

Torroja adopted a largely uncommitted attitude to 'Functionalism', describing it as the spirit of the age (1958a, pp.279-80), but gives the impression that he welcomed the way in which it raised the importance of structural action as a factor in architectural aesthetics.

However, Torroja's constant theme is of moderation and the 'middle-path' and in his consideration of visual aesthetics; the play of light and shade and the beauty of form, surface and outline. He allows for the artistic sensibility of the designer to modify the raw mechanically-derived form to enhance its appearance. He is obviously fascinated by the interaction of mechanical requirements with the visual appeal of geometrical form. An interesting feature is Torroja's enjoyment of "architectural jokes" (1958a, p.276) such as twisted clusters of columns (Fig. 18.2) which looks back to Mannerism but also foreshadows Post-Modernism.

Fig. 18.2. An architectural joke of the twelve century. Twisted columns in the cloister of San Pedro de la Rúa, Estella, Spain. (Photo: Howard Thomas.)

Chapter 19 of the book is devoted to the "genesis of the structural system". It addresses the question of 'attack' in design. Torroja recognizes the importance of the personality of the designer, recalling his previous reference to the engineer's "tedious technical training … which often malevolently deforms the spirit"! (1958a, p.269)

The first step recommended is to divide the initial requirements into three categories: the essential, the desirable and the supplementary. Any compromize or sacrifice made at this stage should be fully conscious. Care should be taken not to neglect any implied requirements. Then the limiting conditions, including the properties of the available materials and the foundation conditions should be examined. After checking the conventional solution to the problem and any individual precedents, a rough outline of a proposed solution should be sketched (1958a, p.314).

At this stage inherent conflicts will become apparent. In trying to solve these the designer should always consider the option of 'cutting the Gordian knot' by adopting a completely novel approach or by redefining the problem. "How very seldom can one be sure" Torroja states "that there is no alternative functional arrangement that will satisfy the requirements at least as well as the initially proposed solution" (p.318).

He also warns against blindly adopting a standard solution to a seemingly standard problem. There are no standard problems; each contains some elements which make it unique and raise the possibility of a particularly appropriate unique solution.

The solution should then be subjected to the most stringent criticism. This is aided by putting it to one side for a few days so that the designer's ego is less involved. "When we turn again to the problem … it is probable that we shall realize that it was no more than a foolish idea" (p.318). Torroja also emphasizes the importance of imagination in the critical phase.

Where several people are involved in design, Torroja stresses the need for close collaboration and condemns the system "still typical in some countries" whereby architects choose a form and expect the engineer to make it stand up. Equally "the artist should not be required at the last moment to give artistic appearance to what is already completed" (p.327).

Finally, in Chapters 20 and 21 Torroja has some things to say about the role of computations and the place of the engineer in the organization that are of relevance to earlier Sections of this book.

Torroja's philosophy is also revealed in an interesting manner in his book The Structures of Eduardo Torroja (1958) in which he discusses the reasoning which led to the form of many of his important structures. These accounts possibly have a more direct and informative impact than the abstract discussion of philosophy. They certainly underline a much less obsessive regard for structural logic than Nervi's.

Nervi's philosophy.

In his book Structures (1956, p.13) Nervi wrote "I have been both championed and challenged by people who did not understand my thoughts". This summary is therefore offered with some trepidation particularly as there appears on the surface to be some inconsistency in his writings. [Note: because of variations in different editions, references to page numbers may not be precise.]

Like Torroja, Nervi starts with a modest disclaimer, offering us his "few certainties and many doubts" (1956, Preface). In all his examples, showing the way in which he evolved his own designs, the paramount importance of competitive 'design and construct' tenders and limits on construction time is evident. Nervi sees these as providing the stimulus necessary for innovation and the discipline which leads to excellence of achievement. His principal motivating philosophy is the drive towards "correctness" in building under the impetus of "absolute imperatives" and beyond this towards the integration of art in building.

"The loftiest and most difficult problems arise in architecture from the need to realize a synthesis between opposing sets of factors: harmony of form and the requirements of technology, heat of inspiration and the coolness of scientific reason, freedom of imagination and the iron laws of economy". The aim is thus the "simultaneous satisfaction" of "functional, economical and aesthetic" requirements. [1956, p.12.]

The study of historical works should be used to show that "form is a consequence and not a determinant of function and structural needs" (p.28). Nervi warns against "irrational" formalism, or planning a building from the outside in, giving prime consideration to its surface.

The aspect of Nervi's philosophy which has been seized on by the proponents of the 'machine aesthetic' in the Modern Movement has a fundamentalist ring. He states that "an honest product is also aesthetically satisfying". As examples of this though, he chooses a golden horse-drawn coach (Fig. 18.3) and the streamlined automobile of today. He contrasts the latter with the early cars, which were basically horse-drawn carriages with engines attached. "All these untrue products are, without exception, ugly." (p.26)

Fig. 18.3. Nervi includes a golden coach in his examples of 'honest' products. [Photo: 'Places to go'.]

The evolution of cars, steamships and planes from their initial forms is held up as the perfect example of the way in which natural laws and economics impose so rigid a discipline on a given type of high-technology structure that it evolves into a unique and perfect 'form-type'. This form is thereafter basically unchangeable although it will continually become more refined. For some reason, which Nervi does not claim to know, we instinctively approve of such forms even though we can have no innate comprehension of the complex laws of mechanics with which they accord. "Every improvement in the functionality and technical efficiency of a product brings about an improvement in its aesthetic quality" (1956, p.26). [Note 5.]

Nervi calls the result the "Truthful Style" and says its essential characteristics are: a structural essence; a necessary absence of decoration; a purity of line and shape. "All true solutions … are also those of maximum efficiency. (1965a, p.186-7)"

In Aesthetics and Technology in Building (1965a) the concept of "correctness", the minimum that should be achieved in building, is further defined as: stability, durability, use of materials in accordance with their natural properties, and functional and economic efficiency (p.8). The latter two are defined as a proper proportioning of spaces, degree of ornament and value of materials.

"To search for an economic solution in the structural field means to find the most natural and spontaneous solution … to find the method of bringing dead and live loads down to foundations in the most direct way and with minimum use of materials." (p.24)

Generally speaking, technical correctness coincides with the naturalness and comprehensibility of the static scheme (p.8). The aim of the scheme must be to solve a specific problem in the most efficient manner (commensurate with economic efficiency) and with a clear expression of the materials of which it is built.

These are the elements of Nervi's philosophy which have been given most publicity by the Modern Movement and have obviously been greatly influenced by it. However, the reality is much more complex.

Nervi's choice of the 'golden coach' as an example of a 'form-type' suggests a willingness to separate ornament from basic conformation. If the latter is found to be functional, the former may be accepted as valid 'styling'. The only justification for the gold and the baroque ornamentation would be to display the owner's wealth. If Nervi were, perhaps, willing to include this as a genuine criterion, the way is open to all sorts of interpretations of his functional determinism.

In fact he definitely states that beauty is something more than what has been described above. He reassures those of us who are dismayed by the idea of living in a world of unchanging 'form-types' that there will always be a margin of freedom which will permit art to flourish.

In general, "the objective data of the problem, technology and statics … suggests solutions and form; the aesthetic sensitivity of the designer who understands their intrinsic beauty and validity, welcomes the suggestion and models it, emphasizes it, proportions it, in a personal manner which constitutes the artistic element in architecture". (1965a, p.10)

Nervi provides an example of this in his account of the design of the Florence Stadium (Fig. 17.2) which was conceived at a time where there were only two other covered stadia existing in Italy. [1965a, pp.23-5 and 38-43.] The basic form was selected after "inevitable and patient research work":

"Soon it also became evident that a notable aesthetic expression could be achieved if one were to interpret with care the technical and economic facts and to refine the curve of the supports and their connections."
"I still remember the long and patient work to find an agreement between the static necessities which imposed a determined resisting section at the various points, and the desire to obtain something which for me would have a satisfying appearance. I must say, however, that basically his work resulted only in small refinements …" [p.24]

Thus for the sake of argument Nervi separates the design process into two phases. "The first is objective in character and answers only to the technical, while the other is completely personal, and cannot be controlled by rules or reasoning" (p.101). In reality he sees a dialogue taking place as these two phases constantly alternate in the mind of the designer or in discussion between engineer and architect.

Nervi's concept of Truth in aesthetics could therefore be described as 'Truth Plus', despite the fact that some of his definitions of 'Truth' appear to leave no room for anything else. On the other hand he sees no need for wilful flouting of the rules of mechanics (e.g. 1965a, p.187). His attitude is thus similar to Torroja's principle of "moderation in all things", but there is a significant difference in degree.

Nervi's philosophy could be summed up by saying that function, and in particular load-bearing function, should be the inspiration and driving force in determining the form of a structure. This form may then be modified slightly in accordance with an intuitive feeling for visual aesthetics. The average designer, if he does this competently, taking into account problems of fabrication, construction and economics, is guaranteed a 'correct' and praiseworthy structure. The occasional genius is able to produce an inspired combination of correctness and art.

Truth in structure is basically "correctness" and this is "a necessary and sufficient condition of satisfactory aesthetic results" (1956, p.26). A correct work, satisfying the "functional, statical, constructional and economical needs and the creation of a well-balanced organism" may be "aesthetically insignificant or expressively beautiful, but will never be aggressively annoying" (p.27).

A selection of other authors on the philosophy of structural form.

Many other writers have made equally interesting, if sometimes less extensive, contributions to the debate. Howard in his book Structure: An Architect's Approach (1966) pleads for the re-integration of structural technique into the art of architecture. "Every aspect of the ultimate building must be included in this first stage of creation if the building is to measure up to the requirements society will make of it" (p.6). Beauty is not something which can be stuck on after the practical considerations have been solved, and similarly the engineer should be included in the preliminary studies.

What might be regarded as Howard's main contribution is his analysis of the relationship between structure and form. He divides structure into four categories: 'Minimal', 'Adequate', 'Formal' or 'Sculptural' and 'Pretentious'.

Minimal Structure corresponds roughly to the engineer's ideal of optimization of weight in the efficient and direct transmission of forces. Howard agrees that this can lead to "very interesting buildings, the validity of whose forms we recognize subconsciously and intuitively" (p.12). However he quotes the example of Gothic cathedral architecture to illustrate that even when efficient structure is used as an "essential and visible design element" the result is just as dependent on the symbolic and functional requirements of the situation (p.13).

Adequate structure is the most common and is the result of taking into consideration all the factors discussed earlier in this book such as fabrication methods, theoretical limitations and space requirements, which result in the structure being, one hopes, of nearly minimum cost for the required job, but by no means of least weight or maximum load-bearing efficiency. He quotes the example of an arched roof for which the minimum material is achieved by a rise-to-span ratio of one-third. However this generally results in too much space for efficiency in heating and ventilating. Hence "what is 'best for the structure' is not necessarily best for the building as a whole" (p.14).

He defines two sub-categories; "visible" adequate structure in which columns, arches, etc. can be used to provide scale and interest or to demarcate space; and "hidden" adequate structure of which his prime example is the framing of the conventional American and Australian timber house. The latter, while it leaves the architect with wide freedom to choose surface materials, makes it "hard to achieve that unity of material, structural principle, and form so helpful to good architecture" (p.15).

Formal or Sculptural structure is typified by the TWA building, Ronchamp, the Yale School of Art and interestingly, the work of Mies. The elements are exaggerated, or the forms reflect non-efficient use of material just for the sake of emotional impact. The structure is often chosen in spite of the limitations of material, skills, etc. (This must mean at greater expense than necessary.) The extremes are the mannerisms of twisted columns, broken pediments and pendant vaulting which delighted Torroja as "structural jokes". Howard strongly defends these against the purists. He describes their criticisms as "Wasted imprecations! One might as well blame a rose for not being a daisy …" (p.16).

Pretentious structure is defined as " 'structure for structure's sake', when aesthetic sensitivity is lacking and when novelty of form is the only claim for attention". It also results "when a sculptural structure is copied by a less gifted designer" or when a structurally efficient system is employed in a situation where it is inappropriate "just for the sake of fashion" (p.17).

After a brief summary of the essential elements of structure, Howard subjects a number of individual buildings to relatively detailed analysis and criticism. This portion of the book cannot be summarized but provides good case-study material.

Billington (1977) in his paper Structural Art and Robert Maillart argues that the design of structures, as practised by Maillart, Nervi, et al., is a new art form which is neither architecture nor sculpture. He too recognizes that despite strong claims to the contrary, often on the part of the designers themselves, the forms they employ are by no means inevitable outcomes of the natural laws of mechanics. Although these forms are often pushed to the limits of their potentialities, the original choice of form is fairly arbitrary and there is no guarantee that a completely different form could not have been refined to the same degree of efficiency with an equally pleasing appearance. He points out that "Maillart's thin Stabbogen bridges are rational within the limits that he sets for himself but there is no rationale for those limits themselves". As Billington puts it; "Maillart had a passion for thinness" and this guided his choice of a particular form from amongst the many possible ones which would appear potentially suitable within our ability to predict cost. "Only when people begin to sense the emotion, the passion in a work of structure do they begin to recognize it as art".

Fig. 18.4. Maillart's 'Stabbogen' type of bridge. A rational optimum within the constraints defined by his 'passion for thin-ness'. My photo shows the bridge at Schwandbach, Switzerland.
[In the stabbogen bridge, a rigid deck provides stability for a thin 'arch'. The two are connected by thin lateral walls which have the appearance of props in the side elevation.]

All the authors quoted so far have argued strongly against 'formalism' which is to take the sculptural or space-defining qualities of a form as the starting point for choice, rather than functional requirements of mechanical properties. The opposing view is put in a readable paper by Evans and Houghton-Evans (1964) the first author of which is, ironically, an engineer.

The authors feel, like the others, that architecture is a practical art and quote the longer of Vitruvius' definitions. They also agree that a decline set in when designers ceased to be masters of their own materials. They too point out the arbitrary nature of initial choice of form, and quote in particular the decision of the Greeks to exclude the arch from public architecture (p.264), since many protagonists of honesty in structure base their argument on the alleged inevitability of classical forms. They are in favour of mannerist works of history, provided the observer is aware of the preceding logic against which they were in rebellion. They too enjoy the joke of the twisted column. [p.271]

However they are concerned with attacking the basic premise of the purist argument. The authors feel that engineers have been too ready to accept the Modern Movement's doctrines of structural honesty and functionalism because they seem to correspond to the engineering outlook. However they point out that the Movement has had "few, if any, engineers amongst its pioneers" and that it is "from the ranks of the artists, architects and literati generally, that its leaders have been recruited" (p.276).

The point to the many formalist works of history, particularly of the Renaissance, including domes which are permanently propped from within (Fig. 18.5) and ask; must we condemn these works "in the name of doctrinaire structural honesty?" Having decided that such a result would be absurd they consider it proved that we must reject the theory rather than the reality.

Fig. 18.5. Dome of St Paul's Cathedral, London. The main structural member is the brick cone supporting the lantern. It is concealed from within by a domical ceiling with oculus. A timber framework resting on the cone supports the familiar outer casing. Should these have been omitted in the name of honesty?
[More photos: Structurae.]

The authors then suggest that, "having failed to establish a simple causal relationship, in which beauty results from honest structure, we may attempt an approach from the contrary direction … and see to what extent beauty results from form itself, not taken as the result of structure, but contrived and composed for its own sake".

They point out that what Howard would describe as "adequate structure" is governed by a multitude of factors besides the need to transmit force efficiently and thus rarely acts as a prompt or a constraint to the architect in the sphere of aesthetics. The architect is therefore free to communicate to the observer, particularly in public buildings, using a language of historical associations and psychological responses with which the general public is familiar. In this he needs "not only a vocabulary, but a grammar". To communicate "one must arrange one's ideas in an intelligible way, present them in an orderly fashion … and where shall order be found if not in those formal qualities we have already discussed: unity, pattern, contrast, and harmony, related proportion and so on?" (p.282).

Finally, by pointing out that "fitness for purpose requires much more than mechanical efficiency" in that an object and its purpose must be intelligible to the user; and by stating that "all art is useful" in that it provides delight, refreshment, satisfaction and inspiration; the authors re-define 'functionalism' to include their concept of form (p.286).

Ove Arup was another engineer whose views approached formalism tempered by structural logic as evidence by his willingness to supply architects with the external forms that they required. He saw no need for 'hidden structure' to conform to abstract ideals of honesty. While he liked to see 'Art' and "sensible building" united in the same project he was willing to see the structure sometimes "strained to comply with the architect's aesthetic requirements" to "serve the ends of architecture" (p.355).

Leonhardt (1976) seems to see a similar disjunction between mechanics and aesthetics. After condemning the many engineers who "behave like slaves in collaboration with architects" and noting that shells, folded plates and cable structures must be shaped by the engineer according to the laws of mechanics rather than by the architect, he states that "for many types of structures" such as these, "it is not easy to combine good engineering and good appearance. This is true for all new forms and shapes." (1976a, p.27)

Like Torroja, he is confident that there is an as yet undiscovered science of aesthetics of structures the secrets of which will be revealed by analyzing the characteristics of those few which are generally recognized as aesthetically appealing.

Notes.

Note 1. There were exceptions, even at the time of writing, one of which was Rutten (1973). See the more readable parts of Chapter 1. [Return.]

Note 2. Examples located were: the work of the ASCE Research Council on the Performance of Structures and Feneron and Rodin (1976), Ockleston (1955), RILEM (1955), Swamy and Potter (1977), Green and MacLeod (1976), Chung, Lee and Ho (1978) and Lord (1971). [Return.]

Note 3. Examples of the debate on codes of practice were located in Siess (1960); Matthews (1976); Moffatt and Dowling (1976); Bulletin of the Comité Européen du Beton No. 114, p. 93 ff.; Civil Engineering-ASCE Mar. 1979, pp. 56-61; NCE 15 June 1978, p. 30; 7 Sept. 1978, p. 49; 24 Aug. 1978, p. 29; 14 Sept. 1978, p. 56; 27 July, 1978, pp. 20, 21; 10 Aug. 1978, p. 30; 6 July 1978, p. 32; 15 June 1978, p. 30; 3 Aug. 1978, p. 30; 27 July 1978, p. 38; Struct. Engr., Aug. 1978, pp. 209-215; Dec. 1981, p. 392; Oct. 1982, p. 320; Concrete, Mar. 1979, pp. 24-25. [Return.]

Note 4. These were: Alexander, R. McN. (1968). 'Animal Mechanics'. Sidgwick and Jackson, London; Alexander, R. McN. (1971). 'Size and Shape'. Edward Arnold, London; Pearce, P. (1978). 'Structure in Nature: A Strategy for Design'. MIT Press, Cambridge; Steadman, P. (1979). 'The Evolution of Designs'. Cambridge University Press, Cambridge; and Wainwright, S.A. et al. (1976). 'Mechanical Design in Organisms'. Edward Arnold, London. [Return.]

Note 5. Nervi (1965), p.186. The concept of the 'form-type' or 'objet type' is discussed in Banham (1960) pp.205-13, and 221. Nervi does allow that there are sometimes several equally suitable solutions for a given problem in structural design. In 1965, pp. 9 and 20 compares the bridges at Paderno d'Adda and Truyère. However, this can only happen when our technique is not being pushed to its limits (1965, p.187). [Return.]

Image Acknowledgements. Linked images, Chapter 18.

Grateful thanks to the following organisations and people who have made it possible to provide links to full colour photographs and descriptions of buildings for this chapter.
University of Stuttgart, Fachgebiet für Grundlagen der Planung und Konstruktion im Hochbau. Link.
Places to go around London and the UK. Link.
Structurae. Link.

[Top.]

The art in structural design navigation:
[Chapter 17.] [Chapter 19.] [Contents] [References]

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