Symmetry:Art and Science

the by ture M. d e c t edi rchite ELGIU e u rA sB Iss ial ent fo ussel c e Sp partm as Br De t-Luc Sin

The Quarterly of the International Society for the Interdisciplinary Study of Symmetry (ISIS - Symmetry).

Volume 2 (new series), numbers 1 - 4, 2002.

Symmetry: Art and Science (formerly Symmetry: Culture and Science) is the journal of the International Society for the Interdisciplinary Study of Symmetry (ISIS-Symmetry). The views expressed are those of individual authors and not necessarily shared by the Society or the Editors.

Regular Editors: George Lugosi R&D&I (Research-Develop-Invent) 2 Union Street Kew, Victoria 3101 Australia Fax: +61-3 9852 8344 Email: g.lugosi@hfi.unimelb.edu.au Dnes Nagy Institute for the Advancement of Research Australian Catholic University Locked Bag 4115, Fitzroy, Victoria 3065 Australia E-mail: d.nagy@patrick.acu.edu.au

ISIS-Symmetry. No part of this publication may be reproduced without written permission from the Society.

ISSN 1447- 607X

Cover layout: Gunter Schmitz. Image on the front cover: Atomium Anamorphosis by Phillip Kent. Images on the back cover: An artistic design by means of an algebraic structure, by F. Ruiz and M. Peas; 3D model of Schindlers Braxton House, by Jin-Ho Park. Ambigram on the back cover by John Langdon (Wordplay, 1992).

Book Production: D. Gillis, Brussels, BELGIUM. Tel +32 (0)2 522 39 69; fax +32 (0)2 520 03 78; dirk.gillis@gillis.be.

Symmetry:
Founding editors: G. Darvas and D. Nagy

The Quarterly of the International Society for the Interdisciplinary Study of Symmetry (ISIS-Symmetry) Published by the International Symmetry Foundation.
Volume 2 (New Series), Numbers 1 - 4, 2002.

SPECIAL ISSUE: Papers presented at the Mat mium Euro-workshop, a Regional Congress of ISIS-Symmetry, supported by the European Commission. Guest Editing: Department of Architecture Sint-Lucas Brussels, BELGIUM.
1

EDITORIAL
This issue contains papers presented at the Euro-workshop Mat mium in Brussels, held 9-13 April 2002 at the Department for Architecture Sint-Lucas in Brussels, Belgium. The Congress was co-presided by Dnes Nagy with the collaboration of Slavik Jablan. The papers are arranged in order of participation. The cover followed the ISIS format, and thus it was printed on soft paper. A hard cover would have been more suitable, but then the larger ISIS network would not have been accessible for its distribution. Eric Blanckaert took most of the photos on the different locations. An apology is owed because of the delay in the publication of these proceedings. However, there is a very good reason: after a heavy traveling schedule, Dnes Nagy fell ill. Actually, while this issue was prepared, he was still recovering from treatment. We express our thanks to all participants for their patience, and to all authors for their assistance and understanding in preparing this issue. The Euro-workshop Mat mium was supported by the European Commission, Research DG, Human Potential Programme, High-Level Scientific Conferences, under contract number HPCF-2001-00377. The information provided is the sole responsibility of the authors and does not reflect the Community's opinion. The Community is not responsible for any use that might be made of data appearing in this publication.

CONTENTS
Event Day 1, At the Atomium, 2002 April 9 Presentation of the speakers Special Feature: TV-Brussels broadcast. 1.Metadesign, LAB[AU] Laboratory for Architecture and Urbanism. 2. Footprints Literacy: The Origins of Art and Prelude to Science, Tsion Avital. Day 2, At the Institute Architecture, 2002 April 10 Special Feature: Fractal art exposition. 1. Architecture, mathematics, and a symmetric link between them (From the Atomium building to the Mat mium project), Dnes Nagy. 2. The Role of Mathematics in the Gothic Architecture Structural Analysis, Javier Barallo. 3. From Ittens Tower to Virtual Towers: a generative Algorithm, Elena Marchetti and Luisa Rossi Costa. 4. Programmed Design: The Systematic Method and the Form of Pattern, Karen Y. Li. 5. Defining Basic Design as a Discipline, William S. Huff. 6. Design and Cognition: Contribution to a Design Theory, Claudio Guerri. 7. Growth, Curvature and Computation, Chaim Goodman-Strauss. 8. Sliceform, Surfaces and a Serendipitous Discovery, John Sharp. 9. Art of Anamorphosis, Phillip Kent. 10. Basic Crystal Symmetries Generated by Molecular Dimers, Alajos Kalman, Laszlo Fabian and Andrea B. Deak. Day 3, At the Africa Museum of Tervuren, 2002 April 11 Special Feature: conference movie 1. Folded Structures, Tibor Tarnai. 2. Symmetry and Ornament, Slavik Jablan. 3. How Plato Designed Atlantis, Leslie Greenhill. 4. The Mnemonics of the Cretan Labyrinth, Tessa Morrison. 5. The Parthenon Design Measurements, Anne Bulckens. 6. Artistic Designs by Means of Algebraic Structures, F. Ruiz and M. Peas. 7. Form World Generated by Integer Permutation, George Lugosi. 8. Visualization vs. Verbalization, Insight into the Morphology of Polyhedra, Irit Wertheim and Nitsa Movshovitz-Hadar. 9. Virus Model, Florian Kovacs. 3 Page 5 6 16 19 23

25 26 31 65 75 85 91 99 111 123 131 135

139 142 147 161 163 203 219 231 251 255 265

Day 4, At the Horta van Eetvelde Hotel, 2002 April 12 Special Feature: conference cartoons 1. Pattern Design by Improper Use of Mathcad, Patrick Labarque. 2. Growing Visible, Nonexistent Trees and Building four-Dimensional Polytopes, Virpi Kauko. 3. Virtual and Real States The Structure of Things and Objects, Tomek Michniowski. 4. Tessellations of Euclidean, Riemannian and Hyperbolic Plane, Radmila Sazdanovic and Miodrag Sremcevic. 5. A Remarkable Horta Type Spiral, Annie Van Maldeghem. 6. Rudolph M. Schindlers Braxton House: The Fibonacci and Lucas Series, Jin-Ho Park. 7. An Unexpected Encounter with the Mathematician E. B. Christoffel, Helena Alexandra and Robert Willem Van der Waall. 8. SPACE PRODUCTION, 51N4E. 9. Frustration: Source of Complexity, Tohru Ogawa. 10. Digital Shaping of Spatial Structures, Janusz Rebielak. 11. Proportions and Dissections in Polygons, Encarnacion Reyes Iglesias. 12. Gothic Town Halls in and Around Flanders, 1350-1550: A Geometrical Analysis, Han Vandevyvere. 13. Nonperiodic Selfsimilar Tilings and Symmetry, Dirk Frettloeh. 14. Symmetry Groups in Mathematics, Architecture and Art, Vera Winitzky de Spinadel. 15. Useful Mathematics: Advantages of Decentralized Electricity and Heat Supply for Buildings, using Fuel Cells, Erico Spinadel. Day 5, Bruges, 2002 April 13.

269 270 275 281 287 299 305 313 325 333 351 353 357 365 379 385 403

415

Day 1 At the Atomium, 2002 April 9. Location.

The location for the first days talks was the Atomium, the lasting symbol of the 1958 Brussels World's Fair and Belgium's answer to the Eiffel Tower or the Statue of Liberty. Architects Andr and Jean Polak realized the project together with engineer Andr Waterkeyn. The Atomium became Belgiums landmark, and one of the most visited attractions in the country.

The first session at the Atomium started with the presentation of the participants, in order of appearance, by the chairpersons of each day: - Day 1: ATOMIUM: Dirk Huylebrouck, Brussels. - Day 2: W&K ARCHITECTUUR: Patrick Labarque, Brussels. - Day 3: AFRICAN MUSEUM TERVUREN: Dnes Nagy, Melbourne. - Day 4: HOTEL VAN EETVELDE, VICTOR HORTA: Slavik Jablan, Belgrade.

PRESENTATION OF THE SPEAKERS

51N4E Space Producers "SPACE PRODUCTION" Arch. Peter Swinnen Sint-Lucas Brussels, Architectural Association London; editor for Financieel Economische Tijd, de Architect; coordinator Stichting Stad & Architectuur Leuven. Arch. Freek Persyn Sint-Lucas Brussels, collaborator Xaveer de Geyter Architecten. Arch. Johan Anrys Sint-Lucas Brussels, tutor architecture design Campus Faydherbe Mechelen. Brussels BELGIUM 51n4e@pi.be. Tsion AVITAL "FOOTPRINTS LITERACY: THE ORIGINS OF ART AND PRELUDE TO SCIENCE" Holon Academic Institute of Technology Department of Design and Art, Holon Campus, Holon ISRAEL avital@hait.ac.il www.hait.ac.il/staff/Avital/Avital.htm Javier BARRALLO "THE ROLE OF MATHEMATICS IN THE GOTHIC ARCHITECTURE STRUCTURAL ANALYSIS" University of the Basque Country, San Sebastian SPAIN mapbacaj@telepolis.com www.sc.ehu.es/mathema1/BelVie.htm www.sc.ehu.es/mathema1/Anglet.htm www.sc.ehu.es/mathema1/fas.html

Anne BULCKENS "THE PARTHENON DESIGN MEASUREMENTS" Dr in Architecture Jakarta INDONESIA. annepaul@cbn.net.id Luisa Rossi COSTA "WORKING WITH AFFINE TRANSFORMATION: THREE VIRTUAL TOWERS" Dipartimento di Matematica "F.Brioschi" Politecnico di Milano Milano ITALY. luiros@mate.polimi.it Vera W. de Spinadel "SYMMETRY GROUPS IN MATHEMATICS, ARCHITECTURE AND ART" Director of the Center of Mathematics & Design MAyDI Faculty of Architecture, Design and Urban Studies University of Buenos Aires Buenos Aires ARGENTINA. vspinade@fibertel.com.ar Erico de SPINADEL "USEFUL MATHEMATICS: ADVANTAGES OF DECENTRALIZED ELECTRICITY AND HEAT SUPPLY FOR BUILDINGS, USING FUEL CELLS" Argentine Wind Energy Association AAEE University FASTA Mar del Plata ARGENTINA. vspinade@fibertel.com

Laszlo FABIAN "BASIC CRYSTAL SYMMETRIES GENERATED BY MOLECULAR DIMERS" Junior research assistant Chemical Research Center Hung. Acad. Sci. Budapest HUNGARY. Dirk FRETTLOEH "NONPERIODIC SELFSIMILAR TILINGS AND SYMMETRY" Dipl. Math. FB Mathematik Universitaet Dortmund Dortmund GERMANY. Dirk.Frettloeh@udo.edu Chaim GOODMAN-STRAUSS "NEW MODELS OF GROWTH AND FORM" Mathematician, graphic designer, artist. Dept. Mathematics Univ. Arkansas Fayetteville USA. cgstraus@comp.uark.edu www.delojo.com/be.html Leslie GREENHILL "HOW PLATO DESIGNED ATLANTIS" Dip. P.S.P. Educated at the Royal Melbourne Institute of Technology Background in public education and public affairs in State government in Victoria, Australia. He presented early findings on ancient design systems at the World Mathematical Year 2000 Conference at the University of Melbourne. AUSTRALIA. lesgreenhill@yahoo.com.au

Claudio GUERRI "DESIGN AND COGNITION: CONTRIBUTION TO A DESIGN THEORY" Architect Buenos Aires ARGENTINA. claudioguerri@fibertel.com.ar William S. HUFF "DEFINING BASIC DESIGN AS A DISCIPLINE" Professor Emeritus State University of New York at Buffalo USA. wshuff@earthlink.net Slavik JABLAN "SYMMETRY AND ORNAMENT" Mathematical Institute Belgrade YUGOSLAVIA. jablans@mi.sanu.ac.yu . www.mi.sanu.ac.yu/~jablans/ Alajos KALMAN "BASIC CRYSTAL SYMMETRIES GENERATED BY MOLECULAR DIMERS" Ph.D., D.Sc., FASc President of the Hungarian Chemical Society Chemical Research Center Hung. Acad. Sci. Budapest HUNGARY. akalman@chemres.hu

Virpi KAUKO "GROWING VISIBLE, NONEXISTENT TREES AND BUILDING FOUR-DIMENSIONAL POLYTOPES" Department of Mathematics and Statistics University of Jyvskyl Jyvskyl FINLAND. virpik@maths.jyu.fi Phillip KENT "ART OF ANAMORPHOSIS" School of Mathematics, Science and Technology Institute of Education London UNITED KINGDOM. p.kent@mail.com www.anamorphosis.com Patrick LABARQUE "PATTERN DESIGN BY IMPROPER USE OF MATHCAD" Architect Hogeschool voor Wetenschap & Kunst Departement "Sint-Lucas" Architectuur, campus Brussel Brussels BELGIUM patrick.labarque@archb.sintlucas.wenk.be LAB[au] "://> project - sPACE, navigable music" LAB[au] = laboratory for architecture and urbanism M. Abendroth, J. Decock, A. Plennevaux, G.Verhaegen. BELGIUM GERMANY. lab-au@lab-au.com www.lab-au.com www.lab-au.com/space

10

Karen Yuqing LI "PATTERNS OF SYMMETRY AND ANTISYMMETRY A PROGRAM DESIGN" Designer Architect Pittsburgh USA Karen_yli@yahoo.com George LUGOSI "FORM WORLD - GENERATED BY INTEGER PERMUTATION" Chemical Engineer, Electronic Engineer, Patent Attorney Director of Forarc Pty. Ltd., R&D&I (Research, Development, Inventions) trading. Former Computer Scientist at Howard Florey Institute, University of Melbourne Melbourne. AUSTRALIA. g.lugosi@hfi.unimelb.edu.au Elena MARCHETTI "WORKING WITH AFFINE TRANSFORMATION: THREE VIRTUAL TOWERS" Dipartimento di Matematica "F.Brioschi" Politecnico di Milano Milano ITALY. luiros@mate.polimi.it Tomasz MICHNIOWSKI "VIRTUAL AND REAL STATES - THE STRUCTURE OF THINGS AND OBJECTS" Surname: Tomek. pool@kul.lublin.pl

11

Tessa MORRISON "THE MNEMONICS OF THE CRETAN LABYRINTH" School of Fine Arts The University of Newcastle Newcastle AUSTRALIA. c9520975@alinga.newcastle.edu.au Nitsa MOVSHOVITZ-HADAR "VISUALIZATION VS. VERBALIZATION, INSIGHT INTO THE MORPHOLOGY OF POLYHEDRA" Ph.D., Head of Kesher Cham - National Center for Mathematics Education, Director of the Israel National Museum of Science, Planning, and Technology Technion ISRAEL. nitsa@techunix.technion.ac.il Denes NAGY "INFORM-A-TOMIUM" Nagy is the president of the ISIS Society. d.nagy@patrick.acu.edu.au

Tohru OGAWA "FRUSTRATION: SOURCE OF COMPLEXITY" Japan. He renewed upon a paper written together with Y. Nakajima: Frustration, Degeneracy, and Forms: A View of the Antiferromagnetic Ising Model on a Triangular Lattice, Progress of Theoretical Physics, Supplement No. 87 (1986), pp. 90-101. JAPAN. ogawa-t@koalanet.ne.jp

12

Jin-Ho PARK "RUDOLPH M. SCHINDLERS BRAXTON HOUSE: THE FIBONACCI AND LUCAS SERIES" School of Architecture University of Hawaii at Manoa Honolulu USA. jinhpark@hawaii.edu Maria PENAS "ARTISTIC DESIGNS BY MEANS OF ALGEBRAIC STRUCTURES" Department of Didactics of Mathematics University of Granada Faculty of Education Campus de Cartuja Granada SPAIN. mtroyano@ugr.es Janusz REBIELAK "DIGITAL SHAPING OF SPATIAL STRUCTURES" Professor at Wroclaw University of Technology Department of Architecture Wroclaw POLAND j.rebielak@wp.pl Encarnacion REYES IGLESIAS "PROPORTIONS AND DISSECTIONS IN POLYGONS" Mathematician. Departamento de Matemtica Aplicada Fundamental E.T.S. Arquitectura University of Valladolid Volladolid SPAIN. ereyes@maf.uva.es

13

Francisco RUIZ "ARTISTIC DESIGNS BY MEANS OF ALGEBRAIC STRUCTURES" Department of Didactics of Mathematics University of Granada Faculty of Education Campus de Cartuja Granada SPAIN. fcoruiz@ugr.es Radmila SAZDANOVIC "TESSELATIONS OF EUCLIDEAN, RIEMANNIAN AND HYPERBOLIC PLANE" Faculty of Mathematics University of Belgrade Belgrade Yugoslavia seasmile@galeb.etf.bg.ac.yu John SHARP "SLICEFORM SURFACES AND A SERENDIPITOUS DISCOVERY" Watford Herts ENGLAND. Sliceforms@compuserve.com www.counton.org/explorer/sliceforms , www.nmsi.ac.uk/visitors/surfaces/ Tibor TARNAI "FOLDED STRUCTURES" Member of the Hungarian Academy of Sciences Budapest University of Technology and Economics Budapest HUNGARY. tarnai@ep-mech.me.bme.hu

14

R.W. VAN DER WAALL "AN UNEXPECTED ENCOUNTER WITH THE MATHEMATICIAN E.B.CRISTOFFEL" KdV-Instituut, Universiteit Amsterdam Amsterdam THE NETHERLANDS. waallr@science.uva.nl Han VANDEVYVERE "GOTHIC TOWN HALLS IN AND AROUND FLANDERS, 1350-1550: A GEOMETRICAL ANALYSIS" ir.-arch. Research group caad and design methodology Departement architectuur, stedenbouw en ruimtelijke ordening K.U.Leuven Leuven BELGIUM. asro.kuleuven.ac.be. Annie VAN MALDEGHEM "A REMARKABLE HORTA TYPE SPIRAL" Hogeschool voor Wetenschap & Kunst Departement "Sint-Lucas" Architectuur, campus Gent Ghent BELGIUM. annie_van_maldeghem@yahoo.com Irit WERTHEIM "VISUALIZATION VS. VERBALIZATION, INSIGHT INTO THE MORPHOLOGY OF POLYHEDRA" Architect, Ph.D. Technion Israel Institute of Technology Department of Education in Science and technology Haifa ISRAEL. Main research: morphological approach to 3-D geometry. weririt@techunix.technion.ac.il.

15

Special Feature: TV-Brussels broadcast.

TV Brussels is the local TV-station in Brussels. Some sites are: http://www.tv-brussel.irisnet.be/ http://tv-brussel.vgc.be/ . This TV station broadcasted briefly on the Mat mium event, during its news broadcast. Below are some images from that broadcast.

General view on the conference room, inside one of the spheres of the Atomium.

D. Nagy and T. Tarnai, attentively listening to the introduction.

D. Nagy during his introduction.

D. Nagy linked Horta to mathematics and the Belgian environment.

16

Next was LABaus turn, with a musical performance inside the Atomium.

It took a lot of preparation to get everything installed in the 9 spheres building with its many stairs and escalators.

What we demonstrate here tonight is called: Space Navigable Music.

The project links architecture, music and images.

The project was initially meant for the Internet 17

... but it has grown to a live-act with musicians.

The project is based on a very simple principle.

One has a three-dimensional space where movement is free.

By moving though that space, you make music.

You can record your music and let your friends listen to it through the Internet.

LABaus presenter became part of his own show. 18

Spectators came in large numbers, and even had to sit on the ground

METADESIGN.
LAB[AU] LABORATORY FOR ARCHITECTURE AND URBANISM

Name: LAB[au] is Manuel Abendroth, Jrme Decock, Alexandre Plennevaux. Address: 19, Quai au Foin, B-1000 Brussels. E-mail: lab-au@lab-au.com Fields of interest: Architecture, urbanism, interfaces, new technologies of information and communication. Awards: 2002, Culture2002 first prize in category Art for the project Worldebt, you can count on it http://www.encorebruxelles.org/jk 1999: 'Prix de la Brique Belge', 'Lightscape(s), displacement maps' - light plan study for the Heizel plateau http://www.lab-au.com/lightsc 'Tech-art prize', Vlaamse Ingenieurs Kamer. Publications and/or Exhibitions: A+ magazine, nr. 168 April-May 2001 'http://mind space', A+ magazine, nr. 172 Dec.2001/ Jan.2001 'http://e.motional space', A+ magazine, nr. 173 Feb./March 2002 'http://soundspace' Keywords: Hypermedia sonic space soundscapes - inFORMation processes connectivity - eSPACE CONSTRUCTionS.

Abstract: The transposition of inFORMation processes, transmission and computation, in textual, graphical bi-dimensional, three-dimensional and biomorphic (autogenerative; n-dimensional) forms explores new constructs proper to the electronic medium and outlines the spatial and semantic mutation provoked by technologies on the perception and conception of our environment. Metadesign thus can be understood as a technology determinism that constitutes the main vector/thought in the concern of networked, information-based societies.

1 METADESIGN
A technology is not an independent or alien object, it complements integrally our sensorial and cognitive system; as a medium, it conditions not only communication modes but also the way we perceive and conceive our environment. The increasing implication of communication and information technologies in the process of production and knowledge leads to the fundamental re-thinking of the organization and definition of space. Technology based on the transmission and computation of information influences organization models (modes of production, work and knowledge) and affects the communication process (code, symbol) and the social relations as well as their spatialization. The affectation of traditional articulations 19

between information, space and time leads to the augmenting need to flatten the electronic realm into the concrete space. If, as all communication systems, new technologies induce a transmission channel (signal-medium), a message (information) and a code, their property is to operate on any kind of information, even space, a reduction in a sequence of elementary information coded in a binary language, 0/1 or bit/second. However, contrary to its analogue counterparts within which information was materially fixed on a medium, the digital media celebrate the loss of inscription; it is the transposition of all stable FORM into transmissible and editable inFORMation, processes. As a consequence, the investigation in information space constructs shows the shift from traditional architecture into a metadesign, exploring new spatio-temporal structures as well as their representation practices such as architecture and urbanism. New technologies therefore perform a transformation on semantic and spatial structures (architecture) as much on the level of language (code, style) as on other levels such as social/spatial/economical/political relations. e-SPACE CONSTRUCTionS display the theme of new space constructs relative to information processes, as the formalizations of communication and computation processes. In relation to >INFORMation< processes, metadesign is Information architecture, related to the structuring of information, its textual, graphical, spatial and biomorphic transcription and interfacing grounded on the inherent logics of computation and communication technology in networked societies. Metadesign deals with the setting of new senses as components of language, while improving, increasing our cognitive capacities and influencing in a major way our psychic state (consciousness), our emotional and social behavior and thus participate as much in the individual project as to the collective. Consequently, in the field of new medias, it is important to understand the relation, which is established between perception (the use of senses), recognition, comprehension and the representation (the extraction of sense/meaning), and the action that results from it (production of sense/meaning). In this manner, information architecture deals with intelligible electronic constructs not only as modalities of perception and cognition, but as mental and psychic settings of behavior, ontological concerns, as well as the production of active and functional space settings, spaces of intervention within the constitution of e.SPACE CONSTRUCTionS. Metadesign thus deals with information as programming and meta-inscription, versus as an output of interpretation - and data as objective reality versus information as narrative and simulation. Metadesign displays the theme of new space constructs relative to information processes, as the formalizations of communication and computation processes according to social, semantic and spatial structures (architecture) as much on 20

the level of language (code, structure) in order to build up connectivity and effectiveness.

1.1 Space, navigable music project

Figure 1: screen of the online project space, navigable music (http://www.labau.com/space). sPACE, Navigable Music is an online project investigating the impact of IC technologies and particularly, 3D Real Time modeling languages (such as VRML) in the construct of space. According to the objectives of LAB[au] the project constitutes as much a space for theoretical research as a space of experimentation on the forms of interactions in networked systems exploring the possibilities of space settings in shared processes in order to build up connectivity. In sPACE, navigable music, the object or architecture is generated in real time according to the position and movements of the user (mix color, mix image, mix sound). Operating on structural parameters, the integration (recombination) of spatial (x,y,z), temporal (t-movements) sonic (frequency, pitch) and generative image sequencing functions, each interaction by the user, displacement, transforms this visual and sonic environment. In addition, the recording of movements allows users to produce a traveling according to camera movements, montage and image sequencing. The 21

established relation between the spatial, visual and sonic formalization processes and the editable interactivity of users lead to an experience combining architecture, music and cinematic techniques through movement patterns. The Navigable Music thus constitutes a space, in which the user experiments cyberspace by dropping sounds into space, mixing music throughout space and navigation, record its movements to produce an animation, a traveling in its sonic space architecture, a kinetic music clip. As such, inFORMation processes, computation and communication through codes / language, VRML, thus describe programmatic relations between these different media fusing them into a hypermedia, which can be experimented through networks, extending the construct of space to the digital matrix (mixed reality), where the multi-user space even more enlarges this experience to shared and collaborative processes based on sound and e.space. Year the Work was created: 2001 Project URL: http://www.lab-au.com/space Technical requirements: VRML, Blaxxun 5 plug-in, Flash player.

22

FOOTPRINTS LITERACY: THE ORIGINS OF ART AND PRELUDE TO SCIENCE.

Tsion AVITAL

Abstract: Art is a far too complex cultural phenomenon to have been invented ex-nihilo. However, no adequate explanation has so far been given regarding the graphic and cognitive skills, which preceded prehistoric art, and made its actual emergence possible. This essay proposes that prehistoric art was preceded by a more primitive kind of pictorial literacy, namely footprints literacy. The obvious attribute common to many early prehistoric paintings and footprints is that both represent their subjects by contour and negative. A deeper analysis of these two kinds of visual literacy reveals many other common attributes: connectivity-differentiation, classification, abstraction, generalization, signification, visual class-names, symmetry-asymmetry, schematization, complementarity, induction, deduction, hypothetical thinking and others. Thus, it is probable that footprints are the proto-symbols from which figurative art evolved. It is striking that the same attributes which appear in footprints literacy about 4 millions of years ago and in figurative art since its beginning about 40.000 years ago, appear much later as basic attributes of modern science, but at a much higher level of sophistication. Possibly, these three domains represent successive stages of noetic evolution. Probably, this finding points to fundamental cognitive attributes or "mindprints" that are basic not only to these areas, but also to human intelligence itself and probably to all other phases of Being. Pointing out the origins of art might be a substantial contribution to the lifting of the veil from the most fundamental attributes of art since its very beginnings. This may provide a new key to the delineation of the demarcation lines between art and non-art, which seems to be the most haunting problem of modern art. It can be shown that works of nonrepresentational art do not share those unique fundamental attributes or mind prints shared by footprints literacy, figurative art, science and other branches of culture. Hence, there is room for doubts if works of nonrepresentational art are culturally relevant and if they are works of art at all.

23

This contribution was related to the publication Footprints Literacy: The Origins of Art and Prelude to Science, Symmetry: Culture and Science. Vol. 9. No. 1. pp. 3-46 (1998), and to a special volume of the ISIS journal (Vol. 1, No, 2, 1999). An electronic reference is Visual Mathematics, at the Internet address members.tripod.com/vismath/ . The talk served as "opening act" to mark the aimed level of abstraction, scope, interconnectedness and directness. It gave the participants a feeling they were allowed to really speak their mind, and to the best of their ability. Avital worked all his life to develop new ideas, rather than repeating old stuff. He was really involved in his own research, intellectually and emotionally. Therefore, he states, simple-minded conformists get angry because they have never read about it before and they think it is outrageous that one dares to speak his mind. The truly intelligent and open-minded people get excited because they are exposed to new ideas. For sure, one way or the other his audience is never indifferent: those who hate his ideas and those who love them, feel that he talks about something that many feel but do not dare to express or do not have the words and understanding to express.

24

Day 2 At the Institute for Architecture, 2002 April 10. Location.

The location for the first full day of the Mat mium conference, and the base location of the organization, was the W&K Institute for Architecture, W&K being short notation for the Dutch words Wetenschap (science) and Kunst (art). The contributions of this chapter were presented there. The participants were embedded in the Brussels environment by an additional evening visit to the Grand-Place of Brussels, a mandatory tourist spot for every dedicated artist or architect.

25

Special Feature: Fractal art exposition.

At this location, during the time of the Mat mium workshop, a mathematical art exposition was simultaneously held at the W&K Institute for Architecture. The initiative for this exposition was due to Prof. Javier Barallo, of the University of the Basque Country (Spain), or, in the Bask language "Euskal Herriko Unibertsitatea Donostia or else, in Spanish, "Universidad del Pais Vasco San Sebastian. Barallos group The border between Art and Science made its first exposition in September 1997. Since then, it realized numerous exhibitions in Germany, Argentina, Austria, Brazil, Spain, France, Japan and Serbia-Montenegro (Yugoslavia). The paintings are grouped under the denominator The Frontier between Art and Science. After the Mat mium workshop, the exposition moved to another Belgian city, Ghent, where it was shown in a major high school for a broad audience. This survey exposition consists of works by different authors. They all share the use of mathematics in their work. All canvasses were made with computer programs, in which mathematical formulas, based on fractals, were especially designed for the graphical representation. The mathematical form and the used parameters give to each image a unique and singular structure, form and color. Each author works meticulously on his oeuvre by trying different values for the equations, formulas and parameters. In this way, the final image was obtained. Such a canvas should not be seen as a synthetic, cold and mechanically generated image set up by a computer, but as an artistic expression capable of transferring emotion and sensitivity. For more information, see: www.sc.ehu.es/mathema1/Anglet.htm, www.sc.ehu.es/mathema1/BelVie.htm, or www.sc.ehu.es/mathema1/fas.html. If you would like to have the exposition at your university or hometown, please do contact Prof. Javier Barallo (mapbacaj@telepolis.com). Below, the 12 images shown in Belgium are presented, but of course infinitely more fractals are available. This artwork is copy right protected, and thus shown here in very reduced format, just to give the reader a taste.

26

The announcement of the exposition. Javier Barallo in front of one of the fractal exposition images. Barallos explanations about how the fractal artworks were obtained were artworks too. Below right shows Kedaja or triptych by Kerry Mitchell (left), Damien Jones (center) and Janet Preslar (right).

27

"Faeries" by Domenick Annuzzi.

"The Coral Reef" by Linda Allison.

"Volcano" by Javier Barrallo.

"Avalanche" by Sylvie Gallet.

"Alien Blood" by Earl Hinrichs. 28

"Spade" by Damien Jones.

"Weathered" by Daniel Kuzmenka.

"Polygon 1" by Luke Plant

"Big Bang" by Kerry Mitchell.

"Taupensky" by Janet Preslar.

"Jewelry" by Frederik Slijkerman. 29

"The Mysterious Conjunction" by Mark Townsend.

30

ARCHITECTURE, MATHEMATICS, AND A SYMMETRIC LINK BETWEEN THEM.

(From the Atomium building to the Mat mium project)

DNES NAGY

Dedicated to - Jnos Bolyai, the co-discoverer of non-Euclidean geometry, who was born 200 years ago in 1802, and - Victor Horta, the pioneer of Art Nouveau, who designed the Tassel House, his landmark building, 110 years ago in 1892.

1 Architecture and mathematics? (from the dome of the Eskimos to the St. Paul Cathedral)
There is an intricate relationship between architecture and mathematics since prehistoric times. Making an artificial dwelling obviously required some sort of planning and inspired geometry-related ideas. We may call this knowledge as proto-mathematics or ethnomathematics. Gradually, some people developed special skills in such questions, and they were able to initiate more complex designs. Such an interaction between protoarchitecture and proto-mathematics led to the development of both sides and to various ingenious ideas. For example, the Eskimos invented the structure of the dome in the form of igloo, their typical dwelling, and widely used it perhaps earlier than Roman architects. Turning to African culture, the studies of Paulus Gerdes demonstrated the importance of various mathematics-related ideas in design, while Ron Eglash pointed out the appearance of self-similarity and even fractal geometry in traditional architecture. We may believe that some of these geometrical structures are very old and kept invariant by generations of craftsmen. We also should pay a special attention to wooden buildings that were developed in many regions in different forms. Although the ancient versions of these buildings do not survive, the lashing technique used in Oceania and the art of joinery mastered in China, Korea, and Japan give an insight into the possible methods that were used for fixing these structures. Both of these are associated with mathematical problems, including knot theory and solid geometry. Turning to the 31

architecture with stone blocks, the art of cutting stones and fitting them together, often without using binding materials, required a high-level of geometrical accuracy. This craft is called stereotomy, but in mathematics we may characterize it as constructing 3dimensional tilings. Geometry (geo + metron, i.e., land-measuring) was obviously necessary for marking out the foundation of a new pyramid, palace, or temple, and, during the completion of the buildings, further mathematical ideas were used. The Rhind Papyrus, for example, includes a mathematical problem that is associated with the angle of a pyramid. Around the 4th century B.C., the birth of deductive mathematics with axioms, theorems, and proofs marked a split between mathematics and architecture. In Euclids Elements (Stoicheia, perhaps around 300 B.C.) there is no reference to architectural problems; this is a work of pure mathematics. In Vitruviuss Ten Books on Architecture (De architectura libri decem, 1st century A.D.) we see some mathematics, but there is no reference to Euclid. I even suggested speaking about Euclidean (or deductive) mathematics vs. Vitruvian (or craftsmans) mathematics (c. f. Mathematics and Design 98, San Sebastian, 1998, pp. 17-25). However, the two types of mathematics were not totally separate. For example: - a part of the so-called 15th book of Euclid was written, with great probability, by a pupil of Isidorus of Miletus, the architect of the Church of St. Sophia in Constantinople (now Hagia Sophia or Aya Sofya in Istanbul), in the 6th century A.D., - the builders of the Milan cathedral consulted with the mathematician Gabriele Stornaloco, whose sketch and calculations survive form the late 14th century, - the mathematician Al-Kashi included problems related to architecture in his mathematical treatise in the 15th century, and - major figures of the Renaissance dealt with both architecture and mathematics. In most cases, however the master masons or the architects constructed their buildings without mathematical assistance and used their own skills. Interestingly, they solved higher-degree equations and simulated complicated curves and surfaces without the utilization of advanced mathematics and long before the age of computers. Their usual secret was a set of simple geometric rules and the method of trial-and-error. Indeed, many Gothic churches collapsed several times before the master masons determined the suitable shape of the structural elements. In some cases, geometric scale models helped the work, including the problems of constructions and stability. A famous example was the brick and plaster model of the Church of San Petronio in Bologna in the 14th century. Note, however, that the questions of strength cannot be studied in scale models, as Galileo pointed it out in 1638. He illustrated the problem by the different shapes of leg-bones of small and big animals. Instead of geometric similarity, we see much thicker bones in the case of large animals. Specifically, the mechanical strength of a structural element varies with the cross-sectional area, hence with the square of its linear 32

dimensions, while the volume varies with the cube of its linear dimensions (square-cube law). Perhaps masons also used an original method of simulating the optimal curves and surfaces by mechanical analogies. For example, by hanging a cord between two points we get a curve that can be used, after inverting it up-side down, as the shape of an arch. Although we do not know too much about such methods in the Middle Ages, perhaps these were kept secret by the free masons, we have an insight into the importance of hanging experiments from later periods. Here we may mention Sir Christopher Wrens design of the St. Pauls Cathedral in London, and more recently Antonio Gaudis workshop at the Sagrada Familia in Barcelona. With the names of Galileo Galilei and Sir Christopher Wren, we also reached a point where applied science started to replace the Vitruvian mathematics and the medieval traditions in architecture. Galileos referred to work contributed to the birth of structural design, and Wren was originally a professor of astronomy and made important contributions to mathematics, including the rectification of the cycloid (On finding a straight line equal to that of a cycloid and the parts of thereof, Philosophical Transactions of the Royal Society, November 1673). Characteristically, the second edition of Newtons Principia (1713) mentioned Wren, together with Wallis and Huyghens, as the leading geometers of his age, while the general public recognized him as the person who rebuilt many churches after the Great Fire of London in 1666. However, the specialization of mathematics made less probable the personal union of the work of the architect and of the mathematician. The rapid developments led to separate fields, which are manifested in Departments of Mathematics and Schools of Architecture at the universities. These institutions rarely cooperated, although the Euclidean distance between them on the campuses became smaller than that one in the Hellenistic age.

2 From the handcuff of historic styles to the freedom of form in Art Nouveau (from a gardeners contribution to an architectural revolution in Belgium)
Before turning to the modern relationships between architecture and mathematics, we should make an outlook to some interesting developments in the history of design and architecture. In the 18th and 19th centuries, most architects still focused on the revitalization of some historic styles and the decorative art also followed this tendency. However, the political changes that were associated with the French Revolution in 1789 and later with Napoleons empire opened the door for new ideas in art and architecture and made Paris the center of influence for a long period. Interestingly, a revolutionary movement in French architecture did not follow the events, but became a forerunner. Shortly before the revolution, tienne-Louis Boulles and Claude-Nicolas Ledouxs presented their utopian designs, and later their student Jean-Nicolas-Louis Durand introduced his composition principle based on the combination of basic geometric units. Although most of their designs, including Boulles enormous spherical building, the Cenotaph for Sir Isaac Newton (1784), were never executed, their works and 33

geometrical ideas became influential for later generations via Durands textbooks. Their fresh approach, however, did not prevent the majority of architects from continuing the application of historic styles. It is true, that this historicism was not always a way of simply looking backwards. Eugne-Emmanuel Viollet-le-Duc studied the Gothic construction principles as an example for rational architecture and his goal was to create a modern parallel of it by using iron. As a prolific writer, he devoted a ten-volume dictionary to this topic (Paris, 1854-1868). He was also a leading, although controversial, figure of restoration of ancient monuments and wrote a monograph on the geodesic and geologic construction of the Mont-Blanc and its glaciers. In the mid-19th century, another and more radical inspiration came from outsiders who introduced a new structural rationality into architecture and design. The rapid development of industrial production provided new materials and new possibilities. The gardener Sir Joseph Paxtons iron-and-glass Crystal Palace for the Great Exhibition in London in 1851, which was also marked by the 1851-foot length of the building, contributed to the birth of a new aesthetics. Later the engineer Alexandre-Gustave Eiffel continued this process. He designed a 300-meter-high wrought iron Tower for the Worlds Fair in Paris in 1889. (The earlier invention of the elevator was also essential for the Eiffel Tower, while the utilization of steel was, surprisingly, missed.) The changing focus of these expos is also characteristic. While the central part of the Crystal Palace was a gigantic hothouse for a horticultural exhibition with a hidden drainage system that utilized the hollow cast-iron pillars, the later Paris Worlds Fairs gave more emphasis to machines and industrial products. The machine age, however, gained not only support, but also a strong opposition with social criticism: the Arts and Crafts Movement, which was initiated by John Ruskin and William Morris in England, promoted the value of handicrafts.

34

Figure 1.Victor Hortas Tassel House, Brussels (1892-93). This tradition was also continued in the New Art or Art Nouveau, which suddenly appeared in architecture, first in Belgium in the 1890s. Its innovator, the architect Victor Horta, ignored the traditional styles and introduced a new form-language for architecture, which is dominated by curves, organic motifs, and the interplay between symmetry and asymmetry. His Tassel House in Brussels (1892-93) was a Gesamtkunstwerk, the synthesis of the arts, where all details, from architecture to internal decorations, from metalwork to furniture, were carefully executed according to his plans. Hortas building is a celebration of the freedom of form, the imaginative use of geometry, and the harmony among various fields of art, as well as the promotion of the use of iron as structural material. Remember that Viollet-le-Ducs also suggested the latter. Horta made this move in a period when even the machines were often decorated with Greek columns or Gothic arches. This stylization, however, became immediately nonsense and the designers of Art Nouveau were able to concentrate on the optimal form of objects, which is required by their function. Horta was an architect who became involved in the design of almost everything inside the building, while another Belgian 35

artist, Henry van de Velde took the opposite direction from paintings to design and architecture. This was the time when the role of Paris as the global center of new art movements was challenged: an interesting anti-globalization and de-centralization was marked by pioneering achievements in architecture and design in Belgium and many other countries. In Germany, Velde moved there in 1897, the new style was first called as Veldean (Veldische) or Belgian (Belgische) and later simply Youth-Style, Jugendstil, which reflected not only the young age of the artists, but also the related interest of the Munich-based magazine Jugend. In Austria and Hungary, however, Secession-style or simply Secession (Sezessionstil in German, Szecesszi in Hungarian) became the conventional expression. There were about a dozen different names for very similar movements. Some of these new centers were directly related, while others worked independently, from Catalonia (Antonio Gaudi) to Hungary (dn Lechner). In Art Nouveau, the main inspiration of a new form-language came for organic nature, especially from botany. Samuel Bing, who opened his gallery and shop LArt nouveau in Paris, claimed that nature is the infallible code of all laws of beauty (in the journal The Craftsman, Vol. 5, October 1903). Some artists published botanical and horticultural articles (Emile Gall, Eugne Grasset), while Ernst Haeckel, a leading biologist with a special interest in evolution and morphology, released his beautifully illustrated album Art-Forms of Nature (Kunstformen der Nature, Leipzig, 1899-1904). Note that in the late 18th century, Johann Wolfgang von Goethe, beyond his literary and political activities, was also involved in morphological studies and wrote an essay on the spiral tendency in vegetation. In the 19th century, there were important mathematicalbiological works on spirals in nature, including the discovery of the spiral phyllotaxis by German and French scholars in the early 1830s. It is a tendency that the number of left and right spirals on the surface pinecones, sunflowers, and cacti are neighboring Fibonacci numbers: 1, 1, 2, 3, 5, 8, 13, 21, ...; each term is the sum of the previous two and the ratios of the neighboring numbers approximate the value of the golden section (a/b = b/(a + b) = 0.618). Sir Theodore Andrea Cooks books Spiral in Nature and Art (London, 1903) and The Curves of Life (New York, 1914), and, especially, Sir DArcy W. Thompsons monograph On Growth and Form (Cambridge, 1917) attracted a special attention. The latter was written originally for biologists, but influenced rather designers and geometers of the forthcoming generations. Last, but not least, Art Nouveau was also inspired by the discovery of Japanese art, following the end of seclusion of that country in the late 19th century, and by a growing interest in objects of art from Africa, Asia, and Oceania that were collected by enthusiastic expeditions and displayed in museums. Japanese architecture offered many interesting features for designers: the use of organic motifs, the striking geometric simplicity, the delicate interplay of the inside and the outside of the building, and the modular system of units. The half-size model of the Katsura Imperial Villa was exhibited at the World Exhibition in Chicago in 1893 and attracted a great attention among visitors, including the American architect Frank Lloyd Wright. The influence of Japanese art had two levels: partly reinforced an interest in floral motifs, partly contributed to a new level of organic art where the nature of the materials and geometric simplicity dominates without 36

depicting flowers and other motifs. Later we shall return to Wright and the importance of his organic simplicity, which was also inspired by his reading of Herbert Spencers biological-philosophical works. In Europe, Rudolf Steiner, originally a Goethe-scholar and a supporter of Theosophy (ancient divine wisdom in East and West), later the founder of Anthroposophy, initiated a form of organic architecture and an interdisciplinary educational movement. (From the 1960s, the Hungarian Organic Architecture, the schools of Gyrgy Csete and of Imre Makovecz, gave a new emphasis to organic forms and materials, as well as to Hungarian folk art.)

3 Industrial design and geometric abstractions with style in the period of war and piece (from the Werkbund in Germany to verhouding in the Netherlands, as well as beelding instead of bleeding)
The popularity of new objects dart, from furniture to wallpaper, from doors to lamps, was growing, but their prizes remained high. The obvious question was the following: do you need the expensive object made by handicraft or the cheap industrial product that lacks the artistic quality? Since many people preferred the advantages of both, some artists initiated a compromise, first by making multiple copies of the same object, and later by cooperating with the industry to mass-produce it. Van de Velde, in his 1897 paper A chapter on the design and construction of modern furniture, explained his principle of systematically avoiding designing anything that cannot be mass-produced (in the Berlin journal Pan, Vol. 3, pp. 260-264, 1897; English trans. in Tim Benton at al., Form and Function, London, 1975, pp. 17-19). Indeed, a new field was born, which is called, by a later term, industrial design. One of its early representatives was Peter Behrens, a German painter turned designer and architect, who become the artistic advisor of the large electric company AEG in 1907. Interestingly, Behrens office employed Mies van der Rohe, Walter Gropius, and Le Corbusier, three architects who later played crucial role in the 20th century developments. The cooperation between artists and engineers also required some exact mathematical methods in order to describe the parameters of the artists original design and to organize its industrial multiplication. The keyword was types and even the mass-produced furniture was named as type-furniture (Typenmbel). The central figure of integrating art and technique, as well as modernizing the system of art education in Germany, was Hermann Muthesius. Earlier he studied the arts and crafts movement in England as a diplomat, wrote a book about The English House (Das englische Haus, Berlin, 1904-05), and returned to work in the Prussian ministry. In 1907, he founded the German Werkbund (Work-Federation). This organization, which institutionalized the cooperation among architects, artists, engineers, and industrialists, contributed to the extensive standardization and mechanization. Sadly, this was utilized not only by the civil sphere, but also by the military industry. 37

The new freedom that Art Nouveau gave to designers, together with the influence of Japanese and African art, paved the way to a more aggressive geometric interest in art in the early 20th century. This was first manifested in Cubist paintings and Futurist sculptures, then in abstract compositions where the figurative motifs totally disappeared and simple geometric figures dominated (Kandinsky, Malevich). This period became the age of manifestoes and the foundation of new artistic centers in Italy (Futurism), Russia (Rayonism, Suprematism, and the early form of Constructivism), Hungary (Activism), and the Netherlands (Neo-Plasticism, Elementarism). The number of isms rapidly grew and started to approximate the number of significant artistic groups. In architecture, the preference for more simple geometric form replaced the Art Nouveaus extensive usage of floral motifs. The Austrian architect Adolf Loos, who had a great influence on the next generation of architects (including Andr Lurat, Eric Mendelsohn, Richard Neutra, and Raymond Schindler), did not hesitate to speak about Ornament and Crime, which is actually the title of his book (Ornament und Verbrechen, 1908). We should not forget, however, the debt of modern architecture to Art Nouveau and its revolution against historic styles. The tensions before and during World Word I (1914-1918) led to various contradictory feelings with illusions and disappointments, enthusiasm and nihilism, but also to the search for new ideas in small groups. Shortly before the war, the artists of the Futurist group had a great hope in technology. They also liked to refer to mathematics, at least metaphorically, as Filippo Tommaso Marinettis Manifesto of Geometrical and Mechanical Splendour, and the Sensibility of Numbers (1914) demonstrates, while Antonio SantEllia predicted in his designs many aspect of the new city. As a numerical obsession, they published the majority of their manifestoes on the 11th day of the actual month. Their interest in military technology, however, led to a tragic direction: the leaders of the movement volunteered to serve in a motorcycle-battalion during the war and Boccioni and SantEllia were killed. The disillusionment with the war and the military machine also marked the end of the movement (although Marinetti remained active as a writer and later joined the fascist leader Mussolini). The case of the Netherlands needs a special attention. Since the country remained neutral in the war and the only military goal was to defend its borders, as well as to help the occupied Belgium economically, the developments in Dutch architecture and design were not interrupted. In the late 19th century, the architect Jan Hessel de Groot and some of his colleagues generated a special interest in mathematical aesthetics, especially in proportional systems. Perhaps this was associated with the idea of rational architecture that Viollet-le-Duc initiated earlier. De Groot summarized his views in many papers and books; here we refer just to one of his later work, a monograph on Form-Harmony in Dutch (Vormharmonie, Amsterdam, 1912). For him the importance of proportion was to create a whole, to make from, say, twenty forms, one form. J. L. M. Lauweriks suggested combining the cosmic significance of mathematical proportions and the ethics of socialism. In Dutch both terms, the international proportie and the native verhouding, can be used in the sense of proportion, although the latter also means relationship, including a love affair and the relationship with God. Perhaps this was 38

suitable for Lauweriks, who had a love affair with proportions and also headed a lodge of the Theosophical Society. (Indeed the title of his paper on proportion is Verhouding in the journal Architectura, Vol. 5, pp. 175-176 and 178-180, 1987.) The case of Lauweriks is interesting to us, because he linked many ideas that we discussed earlier: started with Art Nouveau (Nieuwe kunst) decorations, in 1904 moved to Germany, following the invitation of Peter Behrens, and became an important figure of applied art and also designed various houses, in 1916 returned to the Netherlands and worked as a professor. With great probability, he was the designer of the villa at Bremen, which inspired Le Corbusiers interest in mathematical proportions. (Corbus data on the villa are fuzzy in his book Modulor, but Reyner Banham later pointed out that it must have been Lauwerikss building, cf., Theory and Design in the First Machine Age, London, 1960, p. 142). Hendrikus Peter Berlage, another influential figure of Dutch architecture, was also associated with the 19th century rational movement, as well as the Dutch school of proportions. He visited Frank Lloyd Wright in 1897 and again in 1911, and, following his reports, the American architect became an influential figure in Europe, too. The young generation of architects in the Netherlands, Belgium, and some other countries were confused with the sharp criticism against Art Nouveau and against the use organic decorations, even Horta switched to a more academic style. The helping hand came from Berlage who distributed information on Wrights works that offered organic simplicity. It was new, but also gave a feeling that the earlier organic approach had a meaning. Berlage, in his textbook in German, pointed out that the role of the architect is the creation of space, not the sketching of faades (Grundlagen und Entwicklung der Architektur, Berlin, 1908). The importance of geometry and creation of space, while requiring simplicity, provided the intellectual basis for the painter Theo van Doesburg and some young artists and architects who founded the journal De Stijl (The Style, 1917-1928 and 1932). Even the title was associated with the views of de Groot and Berlage who often referred to Gottfried Sempers book The Style in Applied Arts (Der Stil in den technischen Knsten, 2 vols., 1860-63) and his approach to practical aesthetics. The Dutch phase and the international phase of De Stijl, with a special emphasis of architecture, are discussed by Reyner Banhams referred to book on theory of design in detail (pp. 138-200). I would say, however, that the first phase was also international in some sense. Their group included, beyond Theo van Doesburg who edited the journal, not only the Dutch painters Piet Mondrian, Bart van der Leck and the Dutch architects J. J. P. Oud, Gerrit Rietveld, Rob van tHoff, but also the Hungarian painter Vilmos Huszr, the Belgian sculptor and painter Georges Vantongerloo (van Tongerloo), and the Italian ex-futurist Gino Severini. Around 1922, at the beginning of the international phase the group included just two non-Dutch members, the German Hans Richter and the Russian El Lissitzy, while Theo van Doesburg and two architects, Gerrit Rietveld and Cornelis van Eesteren, were the Dutch participants. Mondrian and some other members of the original group advocated total abstraction, without any subjectivity, and called their approach Neo-Plasticism (Nieuwe beelding). Mondrian was influenced not only by 39

cubism, but also by the spiritual-mathematical works of the Thesophist M. H. J. Schoenmaekers, who wrote a book about the Principles of Plastic Mathematics in Dutch (Beginselen der beeldende wiskunde, 1915). Although Mondrians interest in Theosophy is often mentioned, Robert P. Welsh presented some important new details more recently (see, e.g., his paper Mondrian and Theosophy in the book The Spiritual Image in Modern Art, Wheaton, Illinois, 1987). While a large part of Europe was bleeding during World War I, the peace in the Netherlands led to remarkable studies in beelding, plasticism. The main meaning of the noun beeld is image, but it is also associated with beauty; its derivative is used in beeldende kunsten plastic arts. (In German, both van Doesburg and Mondrian rendered the Dutch expression as Gestaltung.) What is Neo-Plasticism? Mondrian reduced everything to horizontal and vertical lines and three primary colors, red, blue, and yellow. Later van Doesburg also allowed diagonal lines in his counter-compositions and called his new approach as Elementarism, which move led to a split with Mondrian. As a counterpoint to the extensive decorations and the use of curved lines in Art Nouveau, strong geometrical restrictions took command. Note that modern music made a similar step in Arnold Schonbergs serial music. The scientific interest of van Doesburg and his colleagues is well represented by the fact that the subtitle of De Stijl was changed from Monthly Magazine for Plastic Arts and Crafts (Maandblad voor de beeldende vakken), the last word means not only arts and crafts, but also professions, trades, subjects of study, and even, as a nice play on words, squares and panels, to Monthly Magazine for New Art, Science and Culture (Maandblad voor nieuwe kunst, wetenschap en kultuur). This reference to science was not an empty word: the journal published, for example, excerpts from a paper by Henri Poincar, a leading figure of mathematics (Why does the space have three dimensions?, De Stijl, Vol. 6, No. 5, pp. 66-70, 1923). The movement associated with the journal De Stijl had an important impact on architecture where geometric simplicity had an advantage. Although modern technology does not exclude curved surfaces, but economical reasons may limit the structure. The architects demonstrated how to enrich the world of form, how to make new harmonies in the restricted universe of right angles, straight lines, and interlocking planes. These new ideas from Cubism to Neo-Plasticism, together with the needs of rebuilding many parts of Europe after World Word I (1914-1918), contributed to the popularization of modern design and the establishment of two movements with educational institutions: - the Constructivists (1918-1922 and, in a restricted form, 1922-1934) in Russia, soon after the socialist revolution in 1917, and - the Bauhaus (1919-1933) in Germany, soon after the establishment of the Weimar Republic in 1919.

40

In both countries, the dramatic political changes contributed first to the birth of these movements and later to their suppression. However, their influence remained essential via their achievements and via the activities of some of their members who continued their works in other countries, as well as via their failure. (It is a very important that we should also consider the failures and learn from these; the method of trial-and-error has two sides and both are important.). Let us see these two brothers (or sisters), the Constructivists and the Bauhaus in the context of design and mathematics.

4 Constructivism: art industrial design (an architect turned professor of geometry)

architecture (

art)

The Russian Constructivists interest in scientific and technical questions were immediately emphasized in the names of some of their institutions that were quickly established after the Socialist Revolution in October 1917. In Moscow, they united two schools of art and architecture as the Free State Art Workshops or shortly Svomas in 1918, with the leadership of Vladimir Tatlin, but soon renamed as Higher State ArtTechnical Workshops or Vkhutemas (the abbreviation of Vysshie gosudarstvennye khudozhestvenno-tekhnicheskie masterskie, 1920-1926), which became the Higher State Art-Technical Institute or Vkhutein (1926-1930). In Petrograd, later Leningrad, now St. Petersburg, they organized similar institutions and Tatlin himself moved there for a longer period. Some constructivists and avant-garde artists also participated at the foundation of the Russian Academy of Artistic Sciences (Rosiiskaya akademiya khudozhestvennykh nauk or shortly RAKhN) in 1921. Since the expression artistic sciences is puzzling, we may illustrate this diverse field with two examples. The painter Wassily Kandinsky (Kandinskii) became the head of the Physico-Mathematical and Physico-Psychological Department of this new Academy, and he initiated various research topics from the basic elements of art to the psychology of aesthetics. At the same department, Leonid Sabaneyev, who earlier studied both music and mathematics, conducted his research on the possible importance of the golden section, the proportion a/b = b/(a + b), in the temporal organization of musical works and analyzed 1,770 works by 42 composers. Characteristically, some of the constructivists had training at engineering schools. Thus, El Lissitzky (Lazar Lisitskii) was educated at the Technische Hochschule Darmstadt (1909-14), while Naum Gabo, one of the Pevsner brothers, studied medicine (1910), science (1911), and engineering (1912) in Munich. Constructivism originally benefited from the various geometrical and abstract tendencies that were popular among Russian artists (Larionov, Goncharova, Malevich, Gabo, and others) and even from the combination of abstract art with a psychological and spiritual interest (Kandinsky). However, these tendencies of pure art were soon overshadowed by the emphasis on industrial design and production, including such diverse fields as textile design and typography. Down with guarding the traditions of art. Long live the Constructivist technician, claimed Aleksandr Rodchenko and Varvara Stepanova in their manifesto of the Productivist Group in 1920. These constructivists or artist41

engineers believed that they are able to build a new industrial culture and participating in the social changes. (Perhaps we may find some similarities between the views of the Russian Artistic Left in the 1920s and the British Scientific Left in Cambridge in the 1930s; indeed, John Desmond (Des) Bernals representative book The Social Function of Science, London, 1939, frequently refers to science and industry in the Soviet Union, although not to the artists.) Constructivism was started as a new form of art (or an anti-art), became associated with industrial production, tried to serve a socio-political system, but its main stream was politically destroyed from 1922 by the socialist-realist artists. The Declaration of the Association of Artist of the Revolutionary Russia claimed that their real goal is to depict the life of the Red Army, the workers, the peasants, the revolutionaries, and the heroes of labor. (Obviously, there were enough volunteers to pose for such pictures and sculptures!) On the other hand, the constructivists also contributed to this attack against them by their mistakes. They did not try to defend the autonomy of art, but became involved in agitating propaganda with poster design and street decorations, then in production, without even checking that the politicians like or dislike their works. In fact, the leading political figures, with the exception of Lunacharskii, the Peoples Commissar for Enlightenment (1917-1929), had little sympathy with their art or even opposed it. They also provoked other artists by claming that they represent the only form of artistic expression. The first sign of troubles were well demonstrated by the fact that even Kandinsky and Gabo left Russia in 1921 and 1922. However, constructivism did not die immediately, but survived in various fields of applied art and even advanced in architecture where the goals of depicting the revolutionaries had no meaning. The architects partly responded for the great demand for building new industrial complexes, houses, and other objects, partly tried to gain support for some advanced building. For example, the Vesnin Borthers proposed a transparent glass-iron-concrete building for the Leningrad Office of the newspaper Pravda with the elevators outside (1923), El Lissitzky initiated skyscrapers (Cloud Hengers) where the offices or the dwelling units were on the top of elevator-towers (1924), and I. I. Leonidov designed a spherical auditorium with 4,000 seats for the Lenin Library (1927-28). Most of these futuristic ideas had the same fate as the earlier French plans: they were not built, but influenced later generations. This period also produced many important theoretical works where architecture and mathematics were linked, including - Nikolai Karsilnikovs quantitative method for testing all the possible arrangements of given elements in architectural design (in the journal Sovremannaya Arkhitektura, i.e., Modern Architecture, 1927), - Yakov Chernikovs book Construction of Architectural and Machine Forms (Konstruktsiya arkhitekturnykh i mashinnykh form, Leningrad, 1931), which was written in Russian, but also advertised in a brochure in English. 42

Perhaps the swansong of constructivism was another book by Chernikov: Architectural Fantasies (Arkhitekturnye fantazii, Leningrad, 1933). Ironically, this title immediately hints that the goal was not anymore real architecture, but play with geometrical forms. John E. Bowlt, in his book Russian Art of the Avant-Garde (London, 1988, p. 156), concluded that constructivism both began and ended as art. We may add that this final form of constructivism is such an art that has a special openness toward mathematics. Using compass, ruler, and even mathematical formulas was not unusual, as it is demonstrated by El Lissitzkys picture Tatlin at Work on the Third International (pencil, gouache, photomontage, ca. 1920). In Tatlin s hand there is a ruler and at his eye a compass. It is true that the mathematical formulas written next to Tatlins body are meaningless, but these were obviously workable for the general public as a metaphorical reference to higher mathematics. However, Chernikov went far beyond basic geometry and even anticipated the field of shape grammar that we should discuss later. Characteristically, Chernikov earned a degree in Architectural Sciences in 1935 and left Leningrad to take up the position of the head of the Department of Descriptive Geometry and Graphics at the Institute of Engineering Economy. Indeed, the constructivists lost their last hopes to continue their work as artists. In 1930, the movement had a second death by renaming its main educational institution, the Vukhtemas/Vukhtein (1920-1930), which pioneered a strong cooperation between art and crafts and also initiated the links to industry. The new name, Moscow Institute of Art, had no reference to technique (-te- in the earlier names) anymore. Finally, in 1934 the third and final death of constructivism came with the political declaration that all forms of art and literature should be based on Socialist Realism, which ideology soon occupied the field of architecture and design by rejecting the luxury formalist ideas. However, constructivism retained an influence via some of its members who moved to other countries and via the help of its younger brother in Germany.

5 Bauhaus: art industrial design architecture ( basic design) (Emaille by email order, or else the roots of computer-aided manufacture)

The Bauhaus (1919-1933), building (Bau) a house (Haus) where architects, sculptors, and painters work together and there will be no barrier between artists and craftsmen, is perhaps the best known movement and educational institution in the field of modern architecture and design. However, its history was not so straightforward as it is widely believed and there were many struggles in order to find its direction. The Bauhaus was established, similar to the Vukhtemas in Moscow, by uniting two existing art colleges in the German city Weimar. It is interesting to note that the director of these institutions was the same Henry van de Velde who contributed to the birth of Art Nouveau in Belgium and had a strong interest in industrial design. The founding director 43

of the Bauhaus was the architect Walter Gropius, who was nominated by van de Welde some years earlier, but no move was made until the end of World War I and the birth of the republic system in Germany. Gropius, in his manifesto of 1919, immediately announced the establishment of a new department of architecture, the reunification of sculpture, painting, handicrafts, and the crafts as inseparable component of a new architecture, and promised constant contact with the leaders of the crafts and industry (Programme of the Staatliches Bauhaus in Weimar, April 1919). However, no real emphasis was given to industrial design and functionalism until the arrival of the Hungarian self-taught artist Lszl Moholy-Nagy in 1923 and no department of architecture was established until 1927, which was also the last year of Gropius at the school. These steps show a striking similarity with the Russian constructivists initial interest in pure art, then a shift to industrial design and later to architecture and applied arts. Incidentally, some authors call the very first period of the Bauhaus, until 1923, as expressionist, which was followed by a constructivist phase. What factors did delay an immediate interest in industrial design at the Bauhaus? What was the reason of interrupting those links between art and industry that were available in Germany via van de Velde and, more intensively, via the German Werkbund? Obviously, the horrors of Word War I and the importance of standardization and mechanization for the military industry gave some negative feelings and did not encourage the artists to rush to the industry. The second reason could have been the personal attitude of the first instructors. Gropius hired creative artist, painters and sculptors, of great ability (Lyonel Feininger, Johannes Itten, Gerhard Marcks, later Paul Klee, Oskar Schlemmer, Vasily Kandinsky, and others), but they had less interest to move towards industry. Still, the early period was very important to work out a progressive structure of education with a half-year, later one-year, Preliminary Course (Vorkurs) and then a three-year dual system of instruction in form and in crafts problems (Formlehre and Werklehre) in various workshops organized according to the used materials, specifically stone, wood, metal, clay, glass, colors for wall-painting, and textiles. The idea was that two instructors should lead each workshop: an artist and a craftsman whose title is form-master (Formmeister) and craft-master (Werkmeister), respectively. The Swiss painter Johannes Itten was the most influential instructor of the first period. He organized the Preliminary Course, with an intention to liberate the students creative power, and also headed some of the workshops. The students created individual pieces of art-works; these were carefully crafted, but without too much consideration of the possibility of industrial reproduction. Since Itten also favored a kind of Oriental mysticism and introduced breathing exercises, diets, and prayers, while Gropius had a growing interest in social questions and the reestablishment of the links to industry, a growing tension appeared between them. The move towards the constructivist phase of the Bauhaus was also encouraged by the private seminars of Theo van Doesburg, the founder of the journal De Stijl, who spent a longer period in Weimar in 1921-22. Finally, Itten left the school in 1923 and Gropius view became the dominating force: Art and technology: a new unity. 44

The appointment of Lszl Moholy-Nagy, who was associated with the Hungarian Activism, a Constructivism-related group, became a very successful step into this direction. In 1922, shortly before his arrival at the Bauhaus, Moholy-Nagy introduced the idea of telephone-pictures (Telefonbilder), which were made using vitreous enamel (or, in German Emaille or Email) on steal, and the description of the figures were such exact that, theoretically, a factory was able to produce it via a telephone order, without seeing its sketch. This method included, in preliminary form, those ideas that later led to the algorithmic description of figures, their presentation by computer graphics, and producing by computer-aided manufacture. (Nomen est omen: we are able to order Emaille pictures not only by telephone, but also by email.) Moholy-Nagy was in charge of the new one-year Preliminary Course, with the participation of Albers, Klee, and Kandinsky, and he also headed the metal workshop where many prototypes were developed for industrial production. Perhaps the best-known objects of the Bauhaus were those chairs that Marcel Breuer, a Hungarian student and later a leading architect, designed in the furniture workshop. On the other hand, Moholy-Nagy did not follow those Russian constructivists who favored production art and rejected pure art, but suggested a balanced view. He intensively dealt with the use of new media, from photo and film experiments to works with kinetic light and mobiles. Moholy-Nagy also had a special interest in graphic design, which lead to his asymmetric typography for the Bauhaus books. In this series, edited by Gropius and Moholy-Nagy, 14 books were published between 1925 and 1929, including not only the works of the Bauhaus members (Kandisnky, Klee, Schlemmer, and the editors own books), but also representative works of the French cubism (Gleizes), the Russian suprematism (Malevich), and the Dutch De Stijl (van Doesburg, Mondrian, J. J. P. Oud). The Bauhaus brotherhood with the Russian constructivists and other avant-garde movement was also reinforced with the appointment of Kandinsky in 1922 and the short visits by El Lissitzky and Malevich. Although the Bauhaus did not have such a member as the Russian Chernikov, who continued his career as a professor of geometry, the importance of mathematics was directly emphasized by Gropius in his second manifesto The theory and organization of the Bauhaus in 1923. Specifically, he pointed out that the dual system of instruction in form and crafts problem needs a general coordination (Harmonisierungslehre), and he included here, among others, mathematics, physics, and mechanics, as well as the synthetic study of space. The program of the instruction in form (Formlehre) also had a section of descriptive geometry. In 1926, the Bauhaus moved to Dessau into a new building designed by Gropius. This glass and reinforced concrete building, with an asymmetric shape, symbolized the Bauhaus interest in functionalism and demonstrated the relationship between design and modern technology. Finally, in 1927, the Bauhaus established its Department of Architecture with the chairmanship of the Swiss architect Hannes Meyer, who also become the new Director in 1928 when Gropius, together with Moholy Nagy and some other masters, left the school. Meyer represented a Marxist ideology and was strongly productionoriented, while reduced the teaching commitments. In 1930, he was removed from office 45

and replaced by the German architect Mies van der Rohe. The Bauhaus, similar to the Russian constructivists workshops Svomas/Vukhtemas/Vukhtein (1918-1930), had three different full names: Staatliches Bauhaus in Weimar (1919-1925), Bauhaus Dessau: Hochschule fr Gestaltung (1925-1932), and Bauhaus Berlin (1932-1933). The lack of founding in Weimar did not cause too much problem and the Bauhaus moved to another place. However, the closure in Dessau and later in Berlin were politically motivated by the Nazis.

6 Beyond the Bauhaus (Briefly about the International Style, two schools, and two personalities)
The Bauhaus-idea survived in various forms, including some remarkable works of the classical period (the Bauhaus chairs are still available commercially) and the later teaching activity of its former members (for example, Gropius at Harvard University, Mies van der Rohe at the Illinois Institute of Technology in Chicago, Albers at the Black Mountain College, North Carolina). The Bauhaus impact on education of design is significant, although not without controversy. Peter Collins remarked, with some irony, that at Harvard University, where Gropius was teaching from 1937, virtually all elements of the Bauhaus curriculum, with the exception of the Preliminary Course, were abandoned. Moreover, the Bauhauss only graduate who furthered the ideal of architectonic art in the case of buildings was Marcel Brauer, who studied only furniture design there (Changing Ideals in Modern Architecture, London, 1965, p. 269). Perhaps the best legacy of the Bauhaus in the field of education is those Basic Design Courses or Foundation Courses that became popular almost elsewhere at schools of architecture and design in order to develop the students creative skills. Here we should pay tribute to Johannes Itten who designed the first Basic Design courses before his forced departure. The fact that such type of courses may also contribute to mathematics is demonstrated by the Rubiks Cube, which was originally a simple tool for design students in Hungary, later became an international obsession, and ended up as a research topic for mathematicians. Interestingly, the Bauhaus had a very strong impact on Japanese design education: many schools introduced similar foundation courses and all Bauhaus books were translated into Japanese. In the history of architecture, the expression International Style was introduced for those achievements that the European architects represented from Gropius to Mies van der Rohe, from Oud to Le Corbusier, although they never formed a coherent group in such a broad scale. The name was obviously inspired by Gropiuss book entitled International Architecture (Internationale Architektur, 1925), the first volume in the series Bauhaus Books, but he himself never referred to a new style. In 1932, the title of the first architecture-related exhibition in the Museum of Modern Art in New York, just three years after its opening, and the book published for this occasion, canonized the name International Style.

46

There were two institutions that emphasized the Bauhaus-traditions in their names, although none of them tried simply to copy it. The first institution had three names (this number became magic in this field): the New Bauhaus in Chicago (1937-1938), which became School of Design (1939-1944), then Institute of Design (1944-), which was later incorporated by the Illinois Institute of Technology in Chicago. Moholy-Nagy was the first head of this institution and his commercial works was marked, among others, the designed of the fountain pen Parker 51 in 1941. (Note the conflicting interest of Hungarians in the world: Moholy-Nagy designed the best-known fountain pen, but Biro invented, with a little help of some friends in the Budapest cafs, the Biro pen, the original form of the ball pen.) Sadly, Moholy-Nagys wide-ranging activities were stopped by his early death of leukemia in 1946. Between 1946 and 1951, Serge Chermayeff was the new director. His name is known as both a successful industrial designer, including various chairs, and the author of theoretical works. Gyorgy Kepes, a close coworker and compatriot of Moholy-Nagy, became a leading figure of building bridges between art and the newest results of modern science and technology, first in Chicago and later, as the founding director of the Center for Advanced Visual Studies at MIT, in Cambridge, Massachusetts. Kepess books, with contributions by very many artists and scientists, attracted a special attention (Language of Vision, Chicago, 1944; The New Landscape in Art and Science, Chicago, 1956; Seven volumes of the series Vision + Value, New York, 1965-72). The other institution, the Hochschule fr Gestaltung in Ulm (1953-1968), was initiated soon after World War II, by a foundation for the memory of Hans and Sophie Scholl, who were members of the White Rose resistance group and executed by the Nazis. Similar to the Bauhaus and the Constructivist Movement, the founders were eager to contribute to the solution of social and intellectual problems after the war. Note that the German name of the institution is identical with the secondary name of the Bauhaus in Dessau, while it is usually referred to in English as Ulm School of Design. As an emphasis on the traditions, Henry van de Velde, Walter Gropius, and Mies van der Rohe were among the early visitors. We should discuss this institution in more details because of an interesting controversy in connection with mathematics. The Ulm School of Design was first headed by the Swiss artist Max Bill, a former student of the Bauhaus, and later by a collective leadership including the Argentine designer Toms Maldonado, the German artist Otl Aicher, and others. During the first years Johannes Itten and Josef Albers, former masters of the Bauhaus Preliminary Course, participated in the teaching of the new schools Foundation Course (Grundkurs), and later, among others, the mathematician Hermann von Baravalle and the architect William Huff (who is a Honorary Member of ISIS-Symmetry) continued this work. In the meantime, the curriculum was changed from a Bauhaus-type teaching to a new one with two types of courses: (1) design practice, and (2) complementary sciences, which included both natural and social sciences. The school had four departments: visual communications, information, industrial design, and building departments. While the Bauhaus attracted some professors with great ability in intuitive mathematics, the Ulm School of Design 47

made a further step and regularly invited mathematicians, crystallographers, and engineers to give courses or lectures (Z. S. Makowski, Frei Otto, Paul Schatz, K. L. Wolf, and others). The school was eager to introduce the newest results of communication and information theory, cybernetics, and semiotics into the curriculum and attracted a remarkable group of people from these fields as instructors (Max Bense, R. Gunzenhuser, Abraham A. Moles, H. Stachowiak, and even a visit by Norbert Wiener). This scientific component of the curriculum was never challenged, however, the question how much mathematization led to a conflict. Horst Rittel, originally a mathematician who was in charge of design methods and also a member of a threeperson team that headed the school between 1959 and 1961, preferred the use of more mathematics in all fields of industrial design, for example modular systems and complicated grids, while the pragmatists rejected the worshipping of some mathematical methods. They suggested concentrating on the links to industry. The battle was won by the latter group. We should emphasize that not just the Russian Constructivists and the Bauhaus made contributions to linking architecture and mathematics in the machine age, the first decades of the 20th century. For example, Le Corbusier and Ozenfants aesthetic program Purism (ca. 1918-1926) in Paris and later Buckminster Fullers Dymaxion house in America (from the 1920s) also initiated such links. In the case of both Corbu and Bucky, this interest remained long-lasting, which is well-illustrated by the goldensection-related proportional system Modulor and the theoretical works on Synergetics, the cooperative forces (energies) in tensile structures, respectively. (We should not confuse, however, Buckys Synergetics with the later usage of the same expression by the physicist Hermann Haken and his interdisciplinary group.) We should stop at Bucky for a while. He was trained as an engineer, not as an architect. Thus, his Dymaxion house and geodesic dome, which became popular first at expos and later at many places, reiterate the 19th century story with Eiffels contribution to the development of architectural design. Playing on words, Bucky used not the analogy of the handcrafts, but immediately the aircraft. (Much later Frank O. Gehry used a computer program for shaping the surfaces of the Guggenheim Museum in Bilbao, which was originally developed for designing the French Mirage fighters; indeed, the shape immediately hints some relationship with aircrafts and the design encouraged some fight around the building, fortunately without fighter jets, just verbally.) Bucky himself was a fighter for his own ideas, which were often not fully his, but adopted from others and popularized with a remarkable efficiency. One may say that the hexagonal shape of the Dymaxion house was also an adoption, but it was legal because the bees did not patent it. However, Bucky went far beyond the economy of nature and introduced prefabricated high-tech components made by aircraft construction methods. The structural components were prewired, preplumbed, and equipped with various devices. Thus, the house became fully operational immediately after assembling. Fullers design took advantage of the tensile strengths of materials and later used the same idea in the case of his geodesic domes, which were shaped with grids based on the great circles (geodesic 48

lines) on spherical surfaces. In the same time, Bucky made a sharp criticism against the Bauhaus and the architects of the International Style, claiming that they never went back of the wall-surface to look at the plumbing. He accused the Bauhaus with designblindness and the ignorance towards modern technology. Reyner Banham, in the conclusion of his book Theory and Design in the First Machine Age (London, 1960), agreed with Buckys view and declared that the architects of the First Machine Age did not understand the contemporary science and technology and, in this sense, they were wrong. Fortunately, we have here just a smaller problem: architecture and modern mathematics. Do architect utilize modern mathematics? Do mathematician care those problems that are important for architecture and urban design? The answer is: usually not, but there are exceptions...

7 Mathematical architecture (?): the systematic usage of modern mathematics in architecture and urban design (a Cambridge center with some encouragement from constructivism)
I am not quite sure that the expression mathematical architecture is the best one, but, using the analogy of mathematical crystallography, mathematical physics, mathematical biology, and even mathematical aesthetics, why not to try this combination. Of course, this is not about the architecture of mathematics or the foundations of mathematics, which usually refers to the system of axioms, definitions, theorems in mathematics and to the related questions of logic, but a field of architecture and design or, using the idea of the Ulm School, a complementary science to architecture. From the 1950s structural engineering made important advances. Beyond the geodesic domes (Buckminster Fuller reinvented in the U.S.A.), the thin shell structures (Flix Candela in Mexico and Pier Luigi Nervi in Italy), and its inverse, the suspended roofs (Frei Otto in Germany) became very popular. Of course, these structures provided further mathematical questions. We may consider mathematical architecture as a topic associated with building science or architectural science, which became a wellestablished field with strong links to applied physics and engineering. Let us see the partnership of mathematics and architecture from the mid 20th century. As we have seen, various groups of architects and designers turned to mathematical methods. Most of these people, however, used elementary mathematics, for example proportional systems, modules, basic geometric transformations. A few others applied higher mathematics, but just occasionally in a few concrete cases. The complex problems in architecture and urban design, however, led to the reappearance of mathematiciansarchitects. (We should not forget that Sir Christopher, as we have seen in Chapter 1, was both a mathematician and an architect.) Some architects decided to learn the necessary mathematics or, in some fortunate cases, they were trained in both fields. Since we spent a longer period in continental Europe, it is time to visit the United Kingdom and then the U.S.A. again. Earlier we have seen the birth of Art Nouveau, Jugendstil, Sezessionstil in continental Europe, but we should add that the Glasgow style 49

in Scotland and the Tiffany style in the U.S.A. represented similar developments and their representatives had regular interactions with other movements. Remember that the Belgian Horta was interested in the Arts and Crafts Movement in England (Chapter 2) and the German Muthesius published a book about the English house (Berlin, 1904-05), before he founded the Werkbund (Chapter 3). About a decade later, in 1915, the British architect W. R. Lethaby wrote a paper entitled Modern German architecture and what we may learn from it, which is available in his book Form in Civilization (London, 1922, pp. 96-105). In another paper, he wrote about the need of systematic research on the geometry of architectural structures, as Lord Kelvin investigated the geometry of crystalline structures. The first help came, however, from the side of Constructive art. The architect Leslie Martin, later Sir Leslie, met the Constructive artist Naum Gabo, later Sir Naum, who left Russia in 1922, and these two and the English painter Ben Nicholson, edited a collection of papers with almost 200 plates: - Circle: International Survey on Constructive Art (London, 1937). Among the contributors, we may find, among others, Gropius, Le Corbusier, MoholyNagy, Mondrian, and, as Lethaby dreamed earlier, a leading British crystallographer, J. D. Bernal (earlier we mentioned his name as a representative of the Scientific left in Britain). Fortunately, J. D. Bernal remained active in building bridges between crystallography and architecture and he wrote two more papers, with almost symmetric titles, Architecture and science and Science in architecture for the Journal of the Royal Institute of British Architects (Vol. 44, No. 16, 1937 and Vol. 53, No. 5, 1946). The latter was based on a lecture given at an informal meeting of the same organization. All of the mentioned three papers were reprinted in Bernals book The Freedom of Necessity (London, 1949, pp. 185-213). However, these contributions were brief survey papers, without going into technical details. Their importance was to identify two mathematical topics that are useful for architects: symmetry and topology. If we are looking for the systematic usage of mathematical methods in architecture and urban design and the establishment of an educational institution where such problems became prominent, we should remain with the intellectual circles of Sir Leslie Martin. After World Word II, he faced a situation that was similar to Sir Christopher Wrens circumstances after the Great Fire of London in 1666: he helped the post-war reconstruction of London as the Chief Architect. From 1956, he taught at Cambridge University, where he became the first professor of architecture, then, in 1967, the Director of the new Centre for Land Use and Built Form Studies at the same university, which was later renamed after him as the Martin Centre for Architectural and Urban Studies. Martin declared at the very beginning of his professorship that it is necessary to establish a bridge between faculties, between the arts and the sciences. His institution became, beyond dealing with social and economical problems of architecture, a center for the application of mathematical methods to architecture. Later some of the members of his circles founded the Centre for Configurational Studies at the Open University in 50

England and started a cooperation with the School of Architecture and Urban Planning at the University of California Los Angeles (UCLA). The desire to synthesize new ideas in Europe and America, architecture and modern mathematics led the Austrian-born Christopher Alexander to graduate in both architecture and mathematics at University of the English Cambridge, then earning a doctorate in architecture at Harvard of the American Cambridge. His thesis was later published as a book, while he moved to University of California, Berkeley: - Christopher Alexander Notes on the Synthesis of Form (Cambridge, Mass., 1964). From graphs theory to statistics, he applied various fields of modern mathematics, together with some results from social sciences. His next work attracted a special attention: A city is not a tree, claimed Alexander in the title of his paper, which refers not a green tree with leaves, but to a type of semi-lattice with mathematical inspirations (Architectural Forum, April and May 1965, pp. 58-62 and 68-71). He concluded that if we develop our cities like trees, they will cut our life into pieces. This paper became a widely discussed document in the field of city planning and was reprinted several times (Design, February 1966, pp. 46-55; Jeffrey Cook, editor, Architectural Anthology, Tempe, Arizona, 1969, pp. 580-590; etc.). Alexander also worked with Serge Chermayeff (Community and Privacy: Towards a New Architecture of Humanism, New York, 1963) and contributed a paper to the series of books edited by Gyorgy Kepes (From a set of forces to a form, in: The Man-Made Object, New York, 1966, pp. 96107). Note that earlier we referred to both Chermayeff and Kepes in the case of the history of Moholy-Nagys New Bauhaus in Chicago (Chapter 5). In 1967, Alexander became the president of the Center for Environmental Structure in Berkeley, California, and worked on a pattern language that considers verbal criteria for generating architectural designs. Interestingly, Alexander decided not to use mathematical weighting or valuation of the criteria, the often conflicting requirements by the client, the local council, and so on, but to consider these as verbal statements and the achieve the best possible solution with a concrete pattern. (Some problems of this methodology, and a sympathy with Bruce Archers alternative approach based on operations research with weighted criteria, were discussed in Lionel Marchs paper The logic of design and the question of value in the book The Architecture of Form, Cambridge, 1976, pp. 140). Alexander and his coworkers continued refining their methodology and published a large number of works on this topic, partly in the book series of the center, partly at other publishers, for example: - Christopher Alexander at al. A Pattern Language: Towns, Buildings, Construction (New York, 1977).

51

The interested reader may find the list of more recent publications of Alexander and his colleagues at the home page: patternlanguage.com. Note that Alexander united the theoretical works and the practice of the architect: he designed more than 200 buildings. Returning to the circles of Leslie Martin and the activities in the triangle of Cambridge University, Open University, and UCLA, a group of scholars published many works on mathematical architecture (note again that it is not their term, but my trial the name an important movement), where they applied a variety of mathematical methods, from symmetry groups to matrices, from graphs to networks. Their activity is marked with monographs and collections of papers, for example: - Lionel March and Philip Steadman The Geometry of Environment: An Introduction to Spatial Organization in Design (London, 1971), - Leslie Martin and Lionel March, editors, Urban Space and Structures (Cambridge, 1972), - George Stiny Pictoral and Formal Aspects of Shape and Shape Grammars: On Computer Generation of Aesthetic Objects (Basel, 1975), - Lionel March, editor, The Architecture of Form (Cambridge, 1976), - Philip Steadman Architectural Morphology: An Introduction to the Geometry of Building Plans (London, 1983), - as well as, by papers in the journal Environment and Planning, B [series] (1974-). Note that the first item in this list, March and Steadmans book, also has an American edition (Cambridge, Massachusetts, 1974), while its Hungarian translation has a modified title Geometry in Architecture (Geometria az pitszetben, Budapest, 1975). Independently of this, William Blackwell used the same English title for his book (New York, 1984). The second item of the list, which was edited by Martin and March, also inaugurated a new series of books Cambridge Urban and Architectural Studies, which is, of course, not limited to mathematical architecture. Perhaps it is necessary to discuss the idea of shape grammar, which features in the title of George Stinys book. The word grammar was frequently used and even overused in architecture and design, starting with Owen Joness monumental work The Grammar of Ornament. (London, 1856) and Charles Blancs Grammar of the Arts of Drawing in French (Grammaire des arts de dessin, Paris, 1867); both books were frequently used by architects. In most cases, including these two examples, grammar is just a metaphor: we have a set of geometrical patterns or designs, these are somehow arranged, and we should feel that the arrangement is based on a grammar. In Stinys case, however, grammar is not a metaphor, but it actually manipulates shapes, which are defined by sets of lines. Shape 52

grammars are very useful to generate all the possible designs in the framework of a given grammar. We may compare it with a corner stone of mathematical crystallography: finding all the possible types of periodic patterns (crystallographic symmetry groups). Incidentally, the 17 two-dimensional or wallpaper patterns are useful for analyzing ornamental art and urban patterns: March and Steadmans book devotes a chapter to this topic. Although some of the mathematical methods in the listed books were worked out prior to the micro-electronic revolution, the research in the field of mathematical architecture, old and new, contributed to the computerization and also befitted from it. Even more methodologies were worked out for using computers in the design process. Here we mention just one monograph: - W. J. Mitchell Computer-Aided Architectural Design (New York, 1977). We separated this book from the others since it represented an emerging new field. Incidentally, Mitchell is a frequent collaborator of Stiny, who initiated the application of shape grammar in the field of computer-aided design. On the other hand, the same methodology can be used not only for generation of shapes, but also for formal analysis of objects of art, as Terry W. Knight demonstrated in her book Transformations in Design: A Formal Approach to Stylistic Change and Innovation in the Visual Arts (Cambridge, 1994). It is a special benefit that modern methods can be applied for dealing with historic topics. For example, Stiny and Mitchell coauthored papers on such diverse topics as The Palladian grammar, which is continued by Counting Palladian plans, and The grammar of paradise: on the generation of Moghul gardens (Environment and Planning, B, Vol. 5, pp. 5-18 and pp. 189-198, 1978; Vol. 7, pp. 209226, 1980). In the first two papers, they dealt with the layouts of the villas of Andrea Palladio, the 16th century Italian architect, and enumerated all the possible plans of fixed sizes. Returning to the early inspiration from the Russian constructivism, the names of Gabo and Martin appeared together again in 1957, exactly twenty years after the publication of the book Circle: International Survey on Constructive Art (London, 1937). Specifically, Gabos album was published by Martins preface (London, 1957). Moreover, the publisher of Circle came out with another interdisciplinary and international collection Data: Directions in Art, Theory and Aesthetics (London, 1968), which was edited by the artist Anthony Hill. Although data here is an acronym of the subtitle (it is better to write it in all caps), we may find a nice metaphor here for the changing time: from the to utopian discussions in a circle to modern data processing, from the perfect geometric figure to the information age. The theoretical works of Russian architects (Chapter 4) were also rediscovered, March started his 1976 book with a reference to Karislnikov, and Stiny applied Chernikhovs drawings for illustrating a shape grammar. The growing in interest in Chernikhov and Krasilnikov is represented by many publications, including Arthur Spragues paper Chernikov and constructivism (Survey, No. 39, pp. 69-77, 1961), Catherine Cookes book Chernikhov: Phantasy and Construction (London, 1984), and her paper Nikolai 53

Krasilnikovs quantitative approach to architectural design (Environment and Planning, B, Vol. 2, pp. 3-20, 1975). We do not refer to more recent works in the field of mathematical architecture, because the home pages of the referred to institutions and some related groups provide detailed information. We also confess that our survey is very far from being comprehensive, we missed many important developments. Our main goal was to demonstrate the beginnings of a new field and its early inspirations.

8 A symmetric link (and two notes on the golden section)

In modern times, no architect and designer can graduate without completing courses in mathematics. It is true, that this statement is not symmetric: most mathematicians graduate without taking any course in architecture and design, but it would be not a bad idea to offer such courses in the framework of general studies programs. More recently, computer aided architectural design (CAAD) is a field where mathematicians, computer scientists, and architects should cooperate. Obviously, there are many problems that require discussions in broader circles of architects and mathematicians. For example, modern architecture may need new transformations and new algorithms that are available in mathematics. In the same time, some questions in architecture and design may inspire mathematicians. The historic reconstruction of building is another problem where we may have fruitful cooperation among representatives from various fields. Even history of mathematics may provide some help in special cases (I will give an example later).

In this paper, we mentioned the topic of symmetry as a link between mathematics and architecture several times. Thus, the crystallographer J. D. Bernal recommended it for architects, March and Steadmans referred to book The Geometry of Environment: (London, 1971) has a chapter on the basic symmetry operations and another one symmetry groups in the plane, Steadmans book Architectural Morphology (London, 1983) deals, among others, with the symmetries of rectangular plans. We may give further examples, including a mathematical contribution - J. A. Baglivo and J. E. Graver Incidence and Symmetry in Design and Architecture (Cambridge, 1983), and a video program made by the Open University in Great Britain,
- P. Steadman Symmetry for the course Design: Principles and Practice (1992).

54

If we are interested how the topic of symmetry became a part of the curriculum in design education in a systematic form, we should go back to the complementary science courses at the Ulm School of Design (Chapter 6). The leaders of the school invited the chemist K. L. Wolf to give lectures on symmetry. He coauthored two interdisciplinary books on symmetry in German:

- K. L. Wolf and D. Kuhn, D. (1952) Gestalt and Symmetry: A Systematic Presentation of Symmetric Bodies (Gestalt und Symmetrie: Eine Systematik der symmetrischen Krper, Tbingen, 1952), - K. L. Wolf and R. Wolff Symmetry: An Attempt towards an Instruction in Seeing Gestalt and Meaningfully Creating Gestalt, Systematically Described and with Numerous Examples Explained (Symmetrie: Versuch einer Anweisung zu gestalthaftem Sehen und sinnvollem Gestalten, systematisch dargestellt und an zahlreichen Beispielen erlutert, Mnster, 1956). Note the presence of the term Gestalt in both tiles, which also featured in the name of the Ulm School of Design (Hochschule fr Gestaltung). The second book has not only an extraordinary long subtitle, but also an interesting structure: the first volume gives the main text, while the second one is an exciting selection of illustrations. The American architect William S. Huff, who was the guest instructor of the same schools Foundation Course between 1963 and 1968, made the next step, a symmetric response. While the chemists wrote a book on symmetry that is useful for designers, too, the architect published a series of booklets as a visual introduction into the subject, which is also interesting for people with scientific background. The series includes not only many rarely seen illustrations, but also presents interesting documents from the history of the related scientific fields: - William S. Huff Symmetry: An Appreciation of its Presence in Man's Consciousness, Parts 2-6, Designed by Toms Gonda (Pittsburgh, 1967-1977). This publication was distributed in Northern America for those universities that had design programs in that time, while the Oppositios: A Journal for Ideas and Criticism in Architecture reprinted three parts of the series (Nos. 3, 6, 10, 1974-1977). Another step was due to a duo, a designer and a mathematician, who wrote a textbook jointly:
- Sadao Kumagai and Yasuaki Sawada Ornamental Patterns and Symmetry (Moyou to shinmetorii, Kanazawa, 1983). This book presents a visual approach to symmetry groups in the plane (wallpaper groups), including black-and-white and colored ones. Although the book is in Japanese, it is dominated by illustrations and tables that are given in English. A similar

55

interdisciplinary cooperation between an anthropologist and a mathematician led to a comprehensive book with the symmetry-analysis of very many ornaments: - Dorothy K. Washburn and Donald W. Crowe Symmetries of Culture: Theory and Practice of Plane Pattern Analysis (Seattle, Washington, 1988. Polyhedra with various symmetry properties are frequently discussed in both mathematics and design; these are very important tools for the development of 3dimensional thinking and also can be applied in practice. Beyond a large number of specialized mathematical books on polyhedra, many authors wrote non-technical, or less technical, books that favor visual approach and consider the needs of designers. The modern pioneers of this approach, beyond Leonardo, Drer, and some Renaissance artists, were H. S. M. Coxeter, who added a chapter on polyhedra to W. W. Rouse Balls book Mathematical Recreations and Essays (London, 1939), and H. M. Cundy and A. P. Rollett, who presented many useful illustrations and data on polyhedra in their book Mathematical Models (Oxford, 1951). Later, both designers and mathematicians contributed to the growing number of such works, including Ugo Adriano Graziottis Polyhedra: The Realm of Geometric Beauty (San Francisco, 1962), Magnus Wenningers Polyhedron Models (New York, 1971), Anthony Pughs Polyhedra: A Visual Approach (Berkeley, 1976), Peter and Susan Pearces Polyhedra Primer (New York, 1978). We should also refer to three more complex books where the study of polyhedra play a central role: Keith Critchlows Order in Space: A Design Source Book (London, 1969), Alan Holdens Space, Shapes, and Symmetry (New York, 1971), and Robert Williamss The Geometrical Foundation of Natural Structures: A Source Book of Design (New York, 1979). As a clearly interesting development, some scholars with backgrounds in architecture and design achieved not only interesting structures, but also produced new mathematical results: - A. Wachman, M. Burt, and M. Kleinman Infinite Polyhedra (Haifa, 1974), - Haresh Lalvani Transpolyhedra: Dual Transformations by Explosion-Implosion (New York, 1977), - Koji Miyazaki An Adventure in Multidimensional Space: The Art and Geometry of Polygons, Polyhedra, and Polytopes (New York, 1986),

- Michael Burt The Periodic Table of the Polyhedral Universe (Haifa, 1996), as well as, some papers in the Canadian journal Structural Topology (1979-). Of course, the more recent works are available at home pages. We suggest starting with the mathematician-artists George Harts web site. 56

Last, but not least the concept symmetry is also associated with proportions, which attracted, from Vitruvius to Le Corbusier, many architects, including serious theoretical works and non-sense proportion hunts. Here I would like to discuss just one special proportion, the golden section, a/b = b/(a + b), which attracted, unfortunately, very few serious works and very many non-sense trials for gold digging, although these people always claimed that they found some gold. We referred to some of the more serious type of works: the discovery of the spiral phyllotaxis in the 1830s and the survey in DArcy Thompsons book in 1917 (Chapter 2), the Russian musicologist Sabayevs study in the 1920s (Chapter 3), and Le Corbusiers modulor (Chapter 6). We may add that Klee and Moholy-Nagy, two masters of the Bauhaus (Chapter 5), also used the golden section in their teaching practice. It is widely believed that the golden section (or, in Latin, sectio aurea) was often used in ancient and medieval art and architecture. Some scholars even used the argument that the preference of the golden section was associated with an interest in gold, which was used in decorations or occupied the minds of medieval alchemists, and so on. However, there is a serious problem with this argument: the golden section was not related to gold until the 19th century, but was called the extreme and mean ratio (since Euclids time), the divine proportion (Luca Pacioli), the continuous proportion (German mathematicians), and the medial section (British mathematicians). Incidentally, the mathematical-historical works on the subject give credit to Max Ohm, the brother of the physicist Georg Simon Ohm, for the first known printed usage of the expression golden section (der goldene Schnitt.) in 1835. This information is repeated in many etymological dictionaries in English, French, German, and Italian. Some years ago, I reported an earlier example of 1833 and now, for the first time, I present a further one of 1830, which is available in the first edition of Ferdinand Wolffs textbook of geometry (Lehrbuch der Geometrie, Berlin, 1830). Interestingly, Wolff was a professor of the Gewerbe-Institut (Trade Institute), which played an important role of popularizing industrial design about 70 years later (Chapter 3). His textbook, however, deals with pure mathematics and there is no evidence that artists and craftsmen contributed to the new terminology. Interestingly, the sectio aurea is not an ancient or medieval expression, but a 19th century Latinization of the German term; it demonstrated that the mathematical concept, not the expression, is ancient. I still maintain my earlier view that the expression golden section was coined by educators of mathematics. Although we moved backwards just by five years, from 1835 to 1833, then to 1830, this result is still significant from the point of view of an interesting question. Did the discovery of the botanical importance of the golden section in the case of the spiral phyllotaxis inspire the new term golden section? The answer is negative: the recently discovered example of 1830 shows that it was available shortly before the discovery of the phyllotaxis by German and French scholars, Braun, Schimper, and the Bravais brothers, in the early 1830s.

57

I should refer to a new finding in connection with Leonardo, too. Many authors claim that Leonardo was interested in the golden section: he illustrated the polyhedra in Luca Paciolis book Divina proportione (Venice, 1509), which title refers to the golden section, and introduced it into his proportional system of the human body. The latter is best represented in Leonardos Vitruvian man, which was associated with a proportional system that Vitruvius (1st c. A.D.) described in an unclear paragraph. The first part of this statement that Leonardo worked with Pacioli is correct, but the referred to book is not about the application of the golden section in art. This proportion is discussed in the mathematical parts of the book, while Paciolis separate essay on proportion in art suggests using simple ratios of integers. Even more problematic is the statement that the Vitruvian man is based on the golden section. The usual view is that the navel divides the height of the man according to the golden section. Let us disregard the possible chronological problem that the Vitruvian man was drawn around 1490, many years before Leonardo met Pacioli. Indeed, one may argue that the dating of Leonardos manuscripts is also problematic or Leonardos interest in the golden section could have started earlier. The Vitruvian man with outstretched arms and legs is inscribed into both a square (homo ad quadratum) and a circle (homo at circulum), which are not concentric. This was Leonardos important contribution, all earlier trials to illustrate Vitruviuss text considered concentric figures. In Leonardos case, the center of the square is the point where the penis begins, while the center of the circle is the navel. On the other hand, the distance between these two points is not given, and this missing particular prompted various studies to reconstruct Leonardos canon. These trials usually introduce new regulating lines and other figures, and many of these are associated with the golden section directly or indirectly. Here I offer an alternative approach, which is based on the accompanying text and does not require the extensive use of new regulating lines. Let us start with the square. It is drawn on the basis of the following observation: The span of a mans outstretched arms is equal to his height (I quote Edward MacCurdys translation with some corrections). Of course, a side of the square is equal to the mans height. Leonardo described various lengths of the body by simple ratios of integers: 1/2, 1/4, 1/6, 1/7, 1/8, 1/10 lengths of the mans height. I projected the marked maximum width of the shoulders, which is a fourth part of the man, to the top horizontal side of the square, and calculated the distance between an end-point of this projected line-segment to the point where the circle intersects this horizontal line. The point of intersection is marked by the tip of the middle finger of an outstretched arm. I calculated the described horizontal distance by considering the righttriangle where the other two sides are: (1) the vertical distance from the top of the breast to the crown of the head, which is the sixth of the man, and (2) the length of the arm from the elbow to the tip of the middle finger is the fourth part; from this elbow to the end of the shoulder is the eighth part, that is, all together, 1/4 + 1/8 = 3/8 height of the man. Using the Pythagorean theorem, the horizontal distance in question is 65/24 = 0.335...times the mans height. If we know the exact position of this point, it is easy to calculate the radius of the circle as the function of the height: (325 + 365)/576 = 0.606... times the height. Since the center of the circle is at the navel, this number also 58

gives the ratio of the height of the navel to the full height of the body. It is significantly less than the golden number, (5 1)/2 = 0.618... (Incidentally, the 19th century German scholar Adolf Zeising was the pioneer of considering the navel as the goldensection-point of the height; he tried to explain almost everything with his new theory of human proportions.) I also noticed that Leonardos ratios included a redundant one: the distance from the top of the breast to the crown of the head can be calculated from earlier data and it is 47/280 (1/8 + 1/7 1/10 = 47/280), but later he gave it as 1/6, which is a good approximation of the other one. This means that Leonardo did not hesitate to round the complicated ratios in order to have simple ones. Consequently, we may believe that our key ratio of 0.335... should be 1/3. I am less enthusiastic to round 0.606... as 3/5 = 0.6. Moreover, by fixing 1/3, we determine the entire system. Thus, the ratio 1/3 is either a well kept secret of Leonardo or, at least, the key to a method to approximate his system. If we mark this point on the top horizontal side of a given square, it is very easy to construct the center of the circle on the vertical midline of this square, which also marks the mans height at the middle. First, we connect our point with the midpoint on the bottom horizontal side of the square (where the circle and the square should have a tangential point), and then the perpendicular bisector of this new line-segment will intersect the midline of the square at the necessary point. I do not go into further details, but will publish this approach elsewhere. This result is a further piece of evidence that the books on basic design should not rush to explain Leonardo s Vitruvian man with the golden section or other sophisticated methods. The importance of the golden section in the history of art is strongly overemphasized in the literature, as we discussed in the paper Golden section(ism): From mathematics to the theory of art and musicology (Symmetry: Culture and Science, Vol. 7, 413-441 and Vol. 8, 74-112, 1996-97). Although computer-aided design programs are able the handle proportional systems based on the irrational value of the golden section (earlier it was not so easy), I do not recommend introducing this proportion for the analysis or reconstruction of historic monuments in an uncritical way. Since the golden section was much less popular among artists and architects of the past than it is believed, we should use this proportion only in those cases where we have some evidence that it was really used. We may end up with less gold, but a few relevant cases may shine better...

59

9 The Symmetry Society and Brussels

Although the cooperation between architects and mathematicians may have various advantages, the meetings between them are relatively rare. Most symposia deal with specialized questions and there are very few broadly interdisciplinary meetings where representatives of science and of art may actually meet. The International Society for the Interdisciplinary Study of Symmetry (ISIS-Symmetry), or shortly Symmetry Society, pioneered such a forum with a triennial congress and exhibition: (1) Budapest, 1989; (2) Hiroshima, 1992; (3) Washington, D.C., 1995; (4) Haifa, 1998; (5) Sydney, 2001; (6) planned in Brussels, 2004. The concept symmetry is very useful to symbolize the interdisciplinary co-operations since it is widely known that it has both scientific and aesthetical meanings since ancient times. Interestingly, the very term symmetry was strongly associated in particular with mathematics and architecture in a complex way (cf., D. Nagy The 2500-year old term symmetry in science and art and its missing link between the antiquity and the modern age, Symmetry: Culture and Science, 6, No. 1, 18-28, 1995): - the original Greek expression was first a mathematical term in the sense of commensurability, but later it was also used in aesthetics as good proportion (the term was not associated with mirror symmetry!) - it was adopted into Latin as symmetria, but was rarely used since the mathematical meaning of the Greek term was rendered in Latin as commensura or commensuratio, while the aesthetical one as proportio; the expression still survived because Vitruvius frequently used it by making a small distinction between the theoretical vs. practical aspects of proportions as symmetria vs. proportio, - the expression symmetry became an international word in architectural texts and was used in many languages following the Renaissance period, when many artists-scholars translated the Vitruvian text and they needed not only the equivalent of proportio, but also symmetria (the term was not yet associated with balance in architecture) - it was readopted by mathematicians as mirror symmetry since the original Greek meaning can be associated with the common measure of two equal parts; this modern understanding of symmetry became the everyday meaning of the expressions and it was also utilized by architects; - architects utilized the new meaning of symmetry and used it to describe arrangement with a mirror-reflection; however, many of them emphasized that we need a more sophisticated balance, not just geometric symmetry 60

- mathematicians, crystallographers, and biologists generalized the concept by considering not only mirror reflection, but also other operations that transform a geometric figure or figure-system into itself (isomorphism); thus we may consider, among others, rotational symmetry, translational symmetry, crystallographic symmetries, color symmetries, similarity-symmetry, biosymmetries, etc. (later physicists continued this process and considered symmetry as invariance of a property, not definitely a geometric one) - it became clear that the idea of crystallographic symmetries and biosymmetries is also useful in the theory of architecture, partly as a composition principles in the case of individual buildings, partly as an organizing principle in town planning. Although the main interest of the Society is symmetry and the related concepts (proportion, balance, equilibrium, invariance, etc.), as well as the relationship between symmetry and asymmetry, it is not limited to these. We may understand the concept also metaphorically as the search for symmetry or bridges between concrete fields of art and science. Last, but not least, we should remark that the some of the major figures of the historic developments discussed earlier are associated with the Symmetry Society directly or indirectly. The mathematician Heinrich Heesch, an Honorary Member of the Society, discovered various tilings in the 1930s and 1940s, which were later commercially produced. He established connections with both Peter Behrens, a pioneer of industrial design (Chapter 3), and Johannes Itten, the initiator of the first Preliminary Course of the Bauhaus (Chapter 5). Earlier we referred to the architect William S. Huff (Chapters 6 and 8), another Honorary Member of the Society, who gave the Foundation Course of the Ulm School of Design and published a series of booklets on symmetry. He is a regular contributor of the Society and attended all of the five congresses and exhibitions in four continents. Indeed, we are very happy to have him and his fresh ideas at our events, which links architecture and mathematics. There is a symbolic meaning that Brussels became a European center of the Symmetry Society in the symmetric year of 2002 (we should wait until 2112 for the next one). Almost fifty years ago, Brussels hosted the World Exhibition of 1958. Interestingly, but not surprisingly, some people who featured in our survey were also there. The architect Le Corbusier and the composer Edgar Varse presented their joint multimedia show entitled Electronic Poem in the Philips Pavilion, which was designed by the composer Iannis Xenakis who worked as an architect in the office of Le Corbusier in that time. The designer Toms Maldonado presented the new curriculum of the Ulm School of Design during a lecture at the Brussels Exhibition. The text of this lecture was printed in the journal of the school; it was the first declaration of some distance from the Bauhaus. The Exhibition was also special for the engineer Frei Otto. At the beginning of this paper (Chapter 2), we discussed how the expo architecture, the Crystal Palace and the Eiffel Tower, contributed to the development of modern architecture and design. Now the suspension structures, which were suggested by Frei Otto in his book of 1954, 61

became dominant. The main symbol of the exhibition, however, was the Atomium, a building representing an iron molecule enlarged 150 billion times (designed by Polak and Waterkeyn). Eight spherical rooms (atoms) were arranged, according to the bodycentered cubic crystal-system, around a central one. The cubic structure stands on one of its vertex (the bottom spherical room) in a position where one of the diagonals is vertical (the elevator shaft is there). Earlier we discussed various utopian buildings that were never built, including Chernikovs designs and Leonidovs spherical library (Chapter 4). As a further coincidence, Ulrich Conrads and Hans G. Sperlichs book The Architecture of Fantasy (New York, 1962, pp. 90-91) put the Atomium, the sketch of Chernikovs Observation Station, and Leonidovs referred to design after each other after. The Atomium, however, is not only fantasy, but became reality... The first Mat mium conference of 2002 was opened in the Atomium building, in the presence of the Brussels Television, and concluded in a Horta Building...

62

Appendix: MM = Manifesto on Mat mium (in the third millennium)

The Atomium Building symbolizes modern architecture, the beauty of mathematical structures, the importance of chemistry and crystallography, and, directly or indirectly, many fields of art, science, and technology. Since it was built for the Brussels Expo and preserved some of the original exhibitions, it also gives an insight into history and its meaning for the future. Last, but not least the Atomium is a symbol of Brussels, the capital of the united Europe, which is just before a historic expansion by reuniting Western and Eastern, Northern and Southern Europe, including such regions that were divided by political walls. This historic time is useful to express our wishes to reunite some efforts in different disciplines, symmetrically from all directions. Of course, the goal is not the unification of these disciplines, but to provide an informal forum where some representatives of different fields of art, science, and technology may come together and discuss mutually interesting topics. Another important goal is to contribute to the modernization of education. The discussions in design about a century ago focused on the question: standardization or individuality. The U-turns in the history of design show that we need both. The too much standardization, for example, contributed to the military machine before World War I, while the too much individualism led to very many isms and movements in art and design, which disappeared very soon. The Constructivists and the Bauhaus tried to unite the positive aspects of all of these, with a strong social commitment and, indeed, produced some remarkable results. They tried to unite various forms of arts and crafts, as well as to educate artists-craftsman in a unified system, but some of their idea remained nave utopia, while they failed to make a real unity between modern art and modern science and technology. There are many examples in our history that too much unification leads to failures. Thus, our goal is not to find some artists-craftsmen, as the Bauhaus tried, or the few people who can be considered as an architect-mathematician, a painter-biologist, a sculpture-chemist, a musician-physicist, and so on. Our actual goal is to find artists and scientists, including architects and mathematicians, painters and biologists, and others, who are interested in the dialogues and have a social commitment. Instead of a unified global view, we should consider the different opinions and contributions from all regions. We need unity in plurality, using a phrase that was used from ancient philosophy to modern architecture by various groups of people. In the age of globalization, we should make a special effort to listen to the views in different countries and regions, including the smaller ones. Both design and mathematics provide good examples where something totally new came from small countries. For example, the initiator Art Nouveau was Victor Horta in Belgium, while one of the codiscoverers of the new non-Euclidean geometry was Jnos Bolyai in Hungary. We are 63

facing complex problems and the search for solutions should be also a complex interdisciplinary and international effort. For the preparation of this, however, we need regular meetings for dialogues, trialogues, quadrologues, ... and a parliament of the disciplines of art and science... D & D in B & B (and A & A): Dnes in Budapest (and Australia) Dirk in Brussels (and Africa)

64

THE ROLE OF MATHEMATICS IN THE GOTHIC ARCHITECTURE STRUCTURAL ANALYSIS

Javier BARRALLO

Name: Javier Barrallo, Address: University of Basque Country, San Sebastian, Spain. E-mail: mapbacaj@telepolis.com

Abstract: For centuries the restoration works made on gothic heritage buildings were based on experience and intuition, as there was no methodologies or mathematical models to simulate accurately the complexity of Gothic structures. Nowadays, the development of new technologies like the Finite Element Method, the computer monitoring of structural elements or the stress measurements techniques provide the necessary data to establish the intervention criteria based on scientific results. The geometry of the Gothic structure must be carefully measured using techniques like topography or photogrametry. The data obtained from the measurement process, the experimental techniques and the mechanical analysis is used to create a computer model that represents the structural properties of the structure. This work will show the mathematical processes involved in all these processes. We will discuss the mathematical idea philosophically, with an emphasis on presenting a wide range of graphical applications. Below is the entire Power Point presentation of Prof. Barallos contribution.

65

67

68

69

70

71

72

73

74

FROM ITTENS TOWER TO VIRTUAL TOWERS: A GENERATIVE ALGORITHM

ELENA MARCHETTI LUISA ROSSI COSTA

Name: Elena MARCHETTI, Mathematician, (b. Milan, Italy, 1948) - Luisa ROSSI COSTA, Mathematician, (b. Casatisma, Pavia, Italy, 1947). Address: Department of Mathematics F. Brioschi, Milan Polytechnic, Piazza Leonardo da Vinci, 32, 20133 Milano, Italy. E-mail: elemar@mate.polimi.it, luiros@mate.polimi.it. Fields of interest: Functional and Numerical Analysis, Geometry, Architecture and Arts. Publications: Numerous papers about Functional or Numerical Analysis in specialized journals. Several papers about didactics of Mathematics in Engineering and Architecture Faculties. Same papers in Nexus Journal referred to Mathematics and Architecture or Art.

Abstract: This paper deals with the mathematical reconstruction of different models of Ittens Fire Tower, which appear in his Tagebcher. A strong geometrical spirit characterizes the configuration of the towers: consequently, we decided to describe them using affine transformations. Vectors and matrices are the basic elements of linear algebra, which allow us an appropriate and straightforward description of the mathematical interpretation of the towers. Using generative algorithm to build up virtual different towers enables us to compare them and decide which is more practical and more pleasing to the eye. Speaking about the towers planned from other artists around the beginning of the twentieth century, we try to remember some examples, underlining that the spirals are involved as crucial lines.

1 INTRODUCTION
We were inspired by the numerous towers or monuments planned around the beginning of the twentieth century to create a connection between Mathematics and Arts. With this aim, from the point of view of Geometry, we described and built virtually The Fire Tower of Johannes Itten (Marchetti, Rossi Costa, 2002), a monument realized only as a prototype, known by only few photographs dated around 1919-20 (Fig.1a). Carrying out our investigations, we found in Ittens Tagebcher drawings and sketches related with other configurations of towers, never realized (Fig.1b, c). We were intrigued by the idea to reconstruct virtually, not only the known prototype, but also other different models, to 75

compare them and discover analogies and differences, in order to establish how far mathematics unconsciously or consciously influenced Ittens aesthetic choices.

Figure 1: a) Ittens tower prototype, b) and c) Sketches in the Tagebcher. In this paper in Section 2 we recall some information about Ittens production and we briefly describe other typical towers planned at the beginning of the twentieth century. In Section 3 we underline the different configurations of the Fire Tower presenting their mathematical description and a generative algorithm. In Section 4 we mention some points of discussion, which came out during the Mat mium talk, and we give some final conclusions.

2 THE ARTISTIC CONTEXT 2.1 Johannes Itten and the early Bauhaus
The Bauhaus School, founded by Walter Gropius in 1919 in Weimar, involved many important artists of the beginning of the twentieth century, like Feininger, Itten, Kandinsky, Klee, Marks, Moholy-Nagy, Muche, and subsequently students who were particularly clever became teachers, like Albers, Bayer, and Breuer. The School had as a purpose the mass-production of common-use things but aiming at a fine design, as in the English Arts and Crafts movement. Therefore teachers and students were engaged not only in the artistic field but also in mathematical or geometrical studies, in analyzing forms and colors, in preparing models and respecting proportions. Among the teachers, Johannes Itten as teacher of form was one of the most important, bringing in that context his artistic and mathematical formation.

76

The members of the Bauhaus School were also involved in the philosophical and metaphysical field and made choices characterizing their way of life. In this sense Johannes Itten, very close to Gropius at the beginning of his stay in Weimar, had a big clash with the founder and he left the school in 1923. During the Weimar period Itten concluded an interesting study on towers, already begun in Vienna before 1919. He had probably different motivations: he planned a bell-tower or a beacon for the Weimar airport as it is evident reading his Diary (Tagebcher) (Badura-Triska, 1990). He realized also a prototype now lost, but well known by few photographs. The promoters of two important Bauhaus Exhibitions in Nuremberg (1971) and in Milan (1995) gave two different reconstructions of the prototype (De Michelis-Kohlmeyer, 1996). In our paper The Fire Tower (NNJ- Spring 2002) we examined in detail the photograph of the prototype realized by Itten and were led to accept the Milan reconstruction as being the more faithful to the original idea (Fig.2b). Some of Ittens drawings in the Tagebcher (Fig.1c) may well have been used for the Nuremberg model reconstruction (Fig.2a), which is however marked differently from the one presented in Ittens photographs. Other sketches, abandoned by Itten, suggest different forms of the tower, which he never developed beyond this primary stage (Fig.1b).

Figure 2: a) Nuremberg reconstruction, b) Milan reconstruction.

77

We decided to build virtually three different models - Model A: based on Ittens prototype and the Milan reconstruction; Model B: the Nuremberg; Model C: an unrealized plan using the same method, contained in the Tagebcher.

2.2 Towers around the twentieth century

As we mentioned before we discover numerous towers planned as monuments by artists around 1900. Some of these contain characteristic cubic or prismatic parts, anticipating the artistic movement called Cubism; others present the spiral, an important line full of significance. We think it is interesting to present some examples in order to underline the two different leit-motif, frequently connected in the same monument, as in Ittens towers. Probably most of these new monuments were inspired by very ancient or less ancient examples, because all the History of Art of any age is rich in buildings having analogous form. We only give examples dated around the turn of the twentieth century.

Figure 3: a) Rodins tower, b) Obrists tower, c) Tatlins monument. The most famous monument is certainly The Tour Eiffel (1889), but we can recall another work of the French artist Rodin, who planned in neoclassical style, at the end of the nineteenth century, the tower named Tour du travail (1893-94), in which the spiral appears evident (Fig.3a). It is also important to mention the prototype of Obrist for a monument (1898-1900), a strange cone with oblique axis and spiral motif (Fig.3b). Quite unusual is Tatlins Monument for The Third International (1919-20): structure conceived in steel bars, forming a spiral developed around an oblique axis, evoking the one of the earth (Fig.3c). The steel bars are connected in a pattern remembering a cage, 78

like the Tour Eiffel, but in Tatlins tower it is not possible to recognize any of the symmetries, which characterize the Paris symbol.

Figure 4: a) Gropius monument, b) Habliks Exhibition Building, c) Klints spiral tower. Gropius and Itten themselves used polyhedra in planning, respectively, the March Dead Memorial (1922) (Fig.4a) and the Composition with dice (1919) (Marchetti, Rossi Costa, 2002). Among the students of the Bauhaus School we recall two artists involving polyhedra or spirals: Wenzel Hablik for an Exhibition Building (1919) (Fig.4b) and Hilma af Klint for a Spiral Tower (1920) (Fig.4c).

3 MATHEMATICAL GENERATION OF ITTENS TOWERS

Itten studied basic Mathematics and Geometry but certainly he never applied linear algebra techniques. The three projects for the towers presented in the previous section have in common a strong geometrical formation. The virtual prototypes we constructed on the base of different drawings and descriptions follow the same mathematical procedure. Our method was to generate towers by adopting affine transformations of only a few geometric elements. In fact basic elements can be seen to appear in the tower many times - always reduced, rotated and translated. We can synthesize our mathematical process by matrix operators. The points of the structure, intended as points of the three-dimensional space R 3 , are represented in a suitable homogeneous Cartesian system Oxyzu by four real component vectors v = [ x, y, z ,1]T .

For l 2 x, y l 2 , 0 z l the vectors v describe a cube of side l as the basic element of the towers interior. We apply to vectors v a (4,4) matrix M , which synthesizes the three fundamental transformations recognizable in the construction of the tower: rotation, scaling and translation.
79

The matrix M is the product of the three matrices related respectively to each single transformation. The product of the matrices R , S and T , related to these transformations, gives the (4,4) matrix used in the process:

k cos k sin k sin k cos M = TSR = 0 0 0 0

0 0 0 0 k l 0 1
2 2 k < 1 is the scaling

where 0 < < 2 is the rotation angle around the z axis, factor, l > 0 the translation factor and

cos R = sin 0 0

sin cos 0 0

0 0 1 0

0 k 0, S = 0 0 0 1 0

0 k 0 0

0 0 k 0

0 1 0 , T = 0 0 0 1 0

0 1 0 0

0 0 1 0

0 0 l 1

are the corresponding matrices. The vectors w = Mv describe the cube transformed from the basic element. By applying this process eleven times, we obtained the twelve cubes of the tower. A similar procedure can be followed to generate the two supports and the conic surfaces decorating the sides of the tower. In Fig.5 you can see the projections on the Oxy plan of the cubic system and of the two supports (like in Fig.6 and Fig.7, from left to right, the drawings are referred to the Ittens prototype, the Nuremberg reconstruction and the virtual model with = 4 , respectively).

Figure 5: Square projections. 80

From the detailed mathematical study of the bi- and tri-dimensional figures obtained, we have deduced that the vertices of the squares or cubes belong to arcs of logarithmic spirals, whose equations can be easily evaluated. In Fig.6 the 3D spirals and their corresponding plan projections are shown. The detailed mathematical description, the particular vector-equations and the matrices involved are in (Marchetti, Rossi Costa, Internal report, 2002).

Figure 6: Spirals in 3D. From the mathematical point of view the different towers are characterized by a different choice of the parameter , being the value k depending on . A generative algorithm can give all the infinite models; in all of them the initial idea is recognizable at every step. In this way we have generated (Fig.6) the three towers proposed by Itten, corresponding to the value = arctan(1/3), = arctan(1/2), = /4, respectively.

81

Figure 7: Elements of the virtual reconstructions. We can observe that the shortest tower is for = / 4 , corresponding to k = 2 2 ; increasing function h = h ( ) symmetric with respect to = / 4 , having its minimum in = /4. In analogy with the ancient question known as cube duplication, connected with Delos altar, during the discussion in the Mat mium we were asked about the value of giving at every step a cube of half- volume. We verified that it is possible to solve this problem with two angles symmetric with respect to = / 4 , corresponding to k * = 1 3 2 . It must be noted that in this case the limitation 2 2 k < 1 is satisfied; on the contrary if you would like to have a sequence of cubes with a volume a third of the previous, you cant do it applying this process, as the corresponding k is out of the range.

between 0 and 2 the height h of the tower can be represented as a

4 DISCUSSION AND CONCLUSIONS

Our talk gave us the opportunity to examine different aspects of the subject, at the end of the presentation of the work and during informal talks. We were encouraged to give more historical information about the production of artists around the change of the century and about the Bauhaus period. In the previous section 3 we answer to an interesting question related to the mathematical study of the tower, that it is possible to halve the volume of the initial 82

cube. We appreciate the friendly atmosphere and the easy approach to Mat mium participants. Going back to the towers, we underline once again that all of them have the same leitmotif. Thinking to the different versions, we built up virtual examples, which tend to be squat (as model B and C), so we can conclude that Ittens prototype (model A) is graceful, soaring towards the sky. Not only is it more pleasing to the eye, but also it is more practical, if as it is thought, it was designed to function as an airport beacon. The mathematics is always the same; the change of angle gives the aesthetically more pleasing model. The choice of the angle makes the difference! References
Altamira, A. (1997) Il secolo sconosciuto, Rossellabigi Ed., Milano. De Michelis, M. and Kohlmeyer, A. (edited by) (1996) 1919-1933 Bauhaus, Fondazione Mazzotta, Milano Badura-Triska, E. (edited by) (1990) Johannes Itten- Tagebcher, 2 vols, Vienna. Bogner, D. (edited by) (1994-95) Das frheBauhaus und Johannes Itten, Weimar-Berlin-Bern. Marchetti, E. and Rossi Costa, L. (Spring 2002) The Fire Tower, Nexus Network Journal, [http://www.nexusjournal.com/MarRos-it.html], 4, no. 2. Marchetti, E. and Rossi Costa, L. Un algoritmo generativo per particolari trasformazioni affini, (Internal report) Quad. Dip. Mat. Politecnico di Milano, n. 516/R, (2002). Ray, S. (1987) LArchitettura dellOccidente, La Nuova Scientifica, Roma.

83

84

PROGRAMMED DESIGN: THE SYSTEMATIC METHOD AND THE FORM OF PATTERN

KAREN Y. LI

Name: Karen Y. Li
Address: Pittsburgh, Pennsylvania, U.S.A. Email: Karen_yli@yahoo.com

Abstract: A programmed design project I did in school after studying in Professor William Huffs Basic Design Studio has resulted in my interests in pattern study and the systematic method. What I found particularly fascinating is, how complex form can be generated by a simple system, and how the rigorousness of logic and the variability of form, or simplicity and complexity can work so perfectly together, and further, how this can be applied to architectural design and urban study, or other artistic creation. A Programmed Design: A group of four elementary squares, each subdivided by triangles in an asymmetrical manner, are arranged in a 9-square square each dictated by three regulating subsquares. This square is continuously transformed as it translates four times across and six times down while the three regulating subsquares shift in the numerical patterns of 123/412/341/234 across and 123/432/134/243/142/324 down this closed group forms one panel. By rearranging the position of triangle in the elementary squares or changing their sequences, a different panel is generated. Through combinatorics, a total of 24 panels are generated each displays areas of mirrored symmetry and mirrored antisymmetry.

85

Programming cells and a closed panel. The above programmed design is presented here both as an example of how to generate a complex pattern by a mathematic system, and a case study of pattern. But the application of systematic or structural methods is not limited in the design field. There have been many works from various disciplines that are created in this approach, such as the music of Arnold Schoenberg and Steve Reich; or the literary writings of Jorge Luis Borges and Italo Calvino; or the serial constructions and wall drawings by Sol LeWitt. For example, Sol LeWitts Variations of Incomplete Cube is a juxtaposition of all the open cubes of 3-edge (minimal sides of a cube) to 11-edge (12-edge will be a complete cube). His wall drawing All Combinations of Arcs From Corners & Sides; Straight, Not Straight, And Broken Lines is another such kind of construct through combinatorics. It is a sequence of all the 190 combinations of any two basic elements listed on the title an exhaustion of all the possibilities of a preset system. For Sol LeWitt, the appeal of this approach lies on being the way of creating art that did not rely on the whim of the moment, but on a consistently thought out process that was interesting and exciting, and from which, the realizations of straight-forward ideas often turn out to be unexpectedly beautiful. (Sol LeWitt and Andrea Miller-Keller, from Sol LeWitt: Twenty-Five Years of Wall Drawings, 1968-1993, p39.)

Sol LeWitt: Variations of Incomplete Cube (left) and All Combinations of Arcs From Corners & Sides; Straight, Not Straight, and Broken Lines (right).

86

Part I of the final pattern. Combinatorics, or ars combinatorial, also takes a prominent role in the formalism of Italio Calvinos writings. Take his Invisible Cities as an example. The book consists of a series of short stories of eleven subjects (memory, desire, etc.), and is structured in the order of number 5 in a music-like sequence. If we letter the 11 subjects as A to K, then Chapter One the introduction will be in the order of (A1/A2, B1/A3, B2, C1/A4, B3, C2, D1). As soon as the number goes up to 5, the book enters into Chapter Two of (A5, B4, C3, D2, E1) and then Chapter Three of (B5, C4, D3, E2, F1). As such, the same pattern is repeated from Chapter Two through Chapter Eight, while the subject/letter continuously fading in and out. Then in Chapter Nine, the order becomes: H5, I4, J3, K2/I5, J4, K3/J5, K4/K5 as an ending, formally if not thematically, appears both as an echo and an inversion of Chapter One. And by this, all combinations of numbers and letters are complete. Through this compositional method, what is presented in the book is not a conventional storyline, but the interweaving of multiple layers and the complex texture of a city. Compare these examples with the programmed design at the beginning, there are some common characteristics of the projected patterns, either visual (3D & 2D) or textual or tonal, that need to be further noted and studied. First is the contrast and interplay between the rigorous order and the apparent randomness of the pattern, and the simple repetition of basic elements and the complex variation of the results. What this observation tells us is: on one hand, the phenomena may seem random or even chaotic, but a logic can be always assumed hidden behind even though it may not be easy to discern; on the other hand, the sum of the set of orders may be simple, but its working can be highly complicated. Further, what results from the generative system is not a composed figure, but a pattern by the autonomous operation of the system. In other words, the orders of the system may be conceptually or contextually referential, but the form they generate is completely selfreferential. Thus, the usual figure-field relation is irrelevant here. A pattern itself is a field without singular focus or favours to any privileged point. Contrasting with the classical tradition that parts are governed by the overall formal relations (proportion, axis, etc.), in a pattern, parts are simply parts and all parts are equal. Each part is indifferent to or unaware of the form of the whole. 87

Part II Yet, is defined only by its relations with its neighbouring parts. For example, in the programmed design at the beginning, a rotational relation defines each unit against its preceding and following unit without knowing the consequential symmetric or antisymmetric relations of the whole. We should also note that the form of each basic element (e.g. the elementary squares with triangle in the programmed design) is independent of the system: while the underlying structure may be the same, the form of the element is changeable, and each change will cause substantial changes to the pattern, and vice versa. To use architecture and urbanism as an example: absolute liberty is granted to the single architectural fragment, but this fragment is situated in a context that it does not condition formally: the secondary elements of the city are given maximum articulation; while the laws governing the whole are rigidly maintained. (Manefredo Tafuri: Architecture and Utopia, 1973, p38.)

Versailles (left, painted by Pierre Patel, ca. 1668) is an example of classical composition that emphasizes focal points and axes. Right: Bucuresti 2000 master plan by O.C.E.A.N. is an urban design example that uses systematic method emphasizing field conditions and patterns. Finally, similar to algorithms, the set of rules that comprise the system usually address the conditions and procedures rather than an intended form. Thus, we have seen more uses of diagram for representation in the practice of such an approach than perspective and illustration. Also, since there are various factors, individuals, forces, and operational methods that are addressed by these rules, each system in fact comprises a number of smaller systems. 88

Part III As such, the proceeding of the system that results in the unfolding of the pattern is a complex act of superimposition and interaction between each individual system. Therefore, each system is in fact an open system, and is capable of variation and adjustment by accepting any changes or interference from any additional elements. It is my belief that the understanding of pattern can help us to understand the complexity of nature and culture. What have been discussed above are just some basic observations, and I believe that the current cross-disciplinary studies on Complexity and System will provide a much boarder and stronger theoretical base. In a world that is both complex and dynamic, the static and homogenous nature of absolute control and unity has rendered the notion itself obsolete and inappropriate. However, a world without any order would prove to be too chaotic. It seems to be the balance and union between system and individuality, orders and chaos, logic and chance, structure and randomness, and repetition and variation that we would like to pursue. What is suggested here is, on one hand, we need to continue to find out the systems that are hidden behind the complex phenomena, on the other hand, we can apply these systems that we have understood to guide us, and help us to set up a minimal structure that can both maintain an order and allow the maximum possibilities and freedom. References
Allen, Stan: From Object to Field, Architectural Design 1997, May-June, P27. Hurtt, Steven: The American Continental Grid: Form and Meaning, Threshold, Journal of the School of Architecture, University of Illinois at Chicago. Calvino, Italio (1972-74): Invisible Cities, translated by William Weaver, Harcourt Brace & Company.

89

90

DEFINING BASIC DESIGN AS A DISCIPLINE

WILLIAM S. HUFF

Name: William S. Huff, Professor Emeritus (b. Pittsburgh, Penna, 1927). Department of Architecture, School of Architecture and Planning, State University of New York at Buffalo, Buffalo, N.Y., U.S.A. Home address: 1326 Murray Avenue, Pittsburgh, PA, U.S.A. E-mail: wshuff@earthlink.net Fields of interest: Basic Design (including symmetry, theory of structure, visual topology, Gestalt principles, color perception, function of color), Architectural Design Award: Honorary Member of the Board, ISIS-Symmetry, 2001 Publications: Ordering Disorder after K. L. Wolf, in: Forma, 15, Tokyo (2000), 41-47; The Landscape Handscroll and the Parquet Deformation, in: Katachi U Symmetry, Tokyo: Springer-Verlag (1996), 307-314; The Programmed Design: Probing the Discernibility of Properties of Symmetry, [Abstracts] Symmetry: Culture and Science, 6 (1995), 254-257; That Unordinary Mirror-Rotation Symmetry, [Abstracts] Symmetry of Structure, 1, Budapest, Hungarian Academy of Sciences (1989), 228-232

Abstract: Johannes Ittens original Bauhaus Vorkurs took from the great movements of Cubism, Constructivism, and Expressionism, all developed by the end of the second decade of the 20th century, creating thereby an introductory medley of Modern Art. Josef Albers revolutionized Basic Design by purging it of unrestrained expressionism and established it as a full-fledged discipline that deals only with universal and persistent, non-objective formal elements of the visual world. At the Hochschule fr Gestaltung, Toms Maldonado retained the Albers model of strict non-objectivity and linked it interdisciplinarily to abstract fields of knowledge, among which are symmetry theory, topology, and Gestalt principles. In the discipline of basic design, the opportunities for self-determined, non-applied research are copious.

1 BACKGROUND AND BASIS

Johannes Ittens original Bauhaus Vorkurs borrowed from the great movements of Cubism, Constructivism, and Expressionism, all of which had been fully developed by the end of the second decade of the 20th Centurythat is, by the time the Bauhaus opened in 1919. In his Vorkurs, Itten created, consequently, a sort of introductory mlange of Modern Art. Josef Albers revolutionized the Vorkurs at the Bauhaus by purging it of unrestrained expressionism and established it as a full-fledged discipline 91

that deals only with persistent and universal, non-objective, formal elements that inhabit the visual world. With the closing of the Bauhaus, Albers took his strict regimen of non-objective design studies to Black Mountain, N.C. The teaching of color in any manner had been reserved at the Bauhaus for Alberss elders, particularly Kandinsky and Klee; so it was at Black Mountain that Albers developed the most notable of his several basic design topics, based not on the theory of color (the teaching of which he depreciated), but on the perception of color. That study culminated at Yale with his ambitious book, The Interaction of Color. At the Hochschule fr Gestaltung, where Albers had twice served as a guest teacher, his model of strict non-objectivity was retained, first, by Max Bill, a Bauhaus student of Albers, and subsequently by one of a group of Argentinean vanguards of art concret, Toms Maldonado. Maldonado went one step further; he linked non-objective design interdisciplinarily to abstract fields of knowledgeamong which were symmetry theory, topology, and Gestalt principles. Laying this down as a background, I claim that basic design is a discipline unto itself so long as it deals only with non-objectivity. As a complete domainperhaps, a strange domainit is one that is all-empowered, self-contained, and possessed of all the prerogatives of sovereign sway. Another way of putting it, non-objective basic design does not involve issues that are specific to any of the disciplines of applied design: architecture, graphic design, or product design, and allied fields. I will not take the time to make that argument here. Permit me, at the same time, to introduce a consummate statement, regarding this point. It comes neither from Albers nor Maldonado, but from Iakov Chernikov of the Vhuetmus school of design of Moscow, a design school that predated the Bauhaus: Specific functions or subject matter as such, do not play any part in this course of teaching. Not once do we use real briefs and problems. The whole methodology is based upon the development of combinations and assemblages of lines, planes and volumes, independent of what the given elements may represent. Just as an appropriate assembly of sounds gives us musical products, so too we construct and assemble a representation in which lines, planes and volumes can be musically turned. Thus we create a skilled composer of new forms. The interesting result of this non-objective approach is that we produce an executant who is freely capable of handling tasks that are based on real subject matter, for the nonobjective and real are erected upon identical principles of form.1

Catharine Cooke, Iakov Chernikov: AD Profile, vol. 55, IC, 22.

92

Taking another cue from the writings of Albers (the discrepancy between physical fact and psychic effect2), my basic design studio (which I had recently preferred to call a formative design studiothough I now think it should be called a non-objective design studio) dealt principally with two issues: how structure actually is and how structure is perceived. By structure, I mean elements (or parts) of a design and how they are related or (arranged) in a design. In regard to the visual manifestations of design, the elements are basically shape and color; and to arrange such elements is to design. When concentrating on the actual nature of structure, its perception was always brought into consideration; and it was the other way around when concentrating on the perception of structure.

2.1 ACADEMIC RESEARCH

Working on non-applied research has the disadvantage that there is not much internal institutional funding and, often, even less external funding; and deans do not regard unfunded research to be of much importance. The flipside is that researching formal issues very often requires little or no funding; and it affords, therefore, relative freedom for the researcher to do exactly what he chooses to dothat includes not having to meet arbitrary deadlines. In my case, at least, it seems that research found me, perhaps haphazardly, but welcomely. My deep-rooted curiosity, companioned with a certain facility of observation and a decided visual orientation, had something to do with that, to be sure. When I studied at Yale, just about every one of my design critics practiced: Eliot Noyes, Philip Johnson, Louis Kahn, Paul Schweikher, Edward Stone, and Richard Neutra, among others. Frederick Kiessler practiced very little; rather, he sculpted and exhibited eccentric architectural projects. Two of my critics were in-house faculty, who did not have practices, even if they received an occasional commission; and they handled most of the perfunctory activities around the school to make up for that deficiency. At the same time, the curriculum in those days was put out by the departments chairman, who took counsel with his faculty rather sparely. When I began teaching in a school of architecture, research was still not a requirement for a sideline career of critiquing design students; the mode still was to have a practice. Then, non-practicing graduates of schools of architecture filtered into the academy; and they began to justify their fulltime faculty positions through research. Thus, I entered teaching at the cusp between design critics who were practitioners and design critics who were research academician.

Josef Albers, Search Versus Re-Search, (Hartford, Conn.: Trinity College Press, 1969), 10.

93

I was, in fact, doing architectural work on my own at the beginning of my teaching career; and I was one of the first (July 1965) to be awarded an Arts and Humanities Program (then NEH) grant, which led to the series of booklets, Symmetry: an appreciation of its presence in mans consciousness. The projects justification, however, did not hang on its own merits as an artifact of abstract knowledge; instead, it was framed as a tool for design education, as indicated by the title of my proposal, Issues of Symmetry in Design Education. I had been introduced to the topic of symmetry by Toms Maldonado at the Hochschule fr Gestaltung (HfG), mainly through the text of K. L. Wolf and Robert Wolff; and it is my recollection that Maldonado had encouraged me to develop such a studywith one intention being the inclusion of other members of the HfG family. Excepting for a small supplemental grant on the same project, this was the only research funding that I ever received. I might add, this was the only research project on which I ever had to meet deadlines; and an absence of deadlines has allowed projects to mellowto improve with age. Since, then, a large part of my research, but not all by far, has come out of the classroom.

2.2 RESEARCH IN THE BASIC DESIGN STUDIO

To conduct research deliberately in the undergraduate classroom can dilute and abuse pedagogic purposes. Students should not be shortchanged for the sake of research. The fact is, however, research topics can just turn up in the design studio (the design studio being a rather different environment from the typical academic classroom) in many an unpredictable instance and does so, almost inevitably, as a byproductwhen the curriculum itself is solid and forward-looking. A large part of the studio way is the facilitating of the students inventiveness. But the student is often not equipped to judge what is truly inventive, in respect, for instance, to solutions that other students had already hit on. It takes the discerning eye and seasoned experience of the instructor to catch what is really new and significant. Furthermore, the grist for research can occur when the experiences of previous studios are built into subsequent ones. Unforeseen solutions often create unforeseen questions. Without a doubt, I can make the case that my students were never deprivedeven while material that was destined for research was ensuing from my assignmentsunawares to myself, often, at the time of its occurrence. My objective with my students was mainly to train them to become proficient and self-sufficient in aesthetic judgment; their inventionsinnovations, which would not have happened without the context of their instructors studio taskwere more of a dividend than a demand. (I should make clear that the identities and contributions of individual students have been scrupulously acknowledged in all publications that have included their workso long as editors honored my watch over that.) 94

In my design studio I assigned two types of tasks: the exercise and the project. Exercises tested the students acquisition of new information, attuned them to the basic formal issues of design, trained them experientially in perceptual acuity, and implemented them with the called for skills of presentation. Exercises are precisely prescribed. At the same time, their drill tends to have a timelessness that works across the ages. Projects challenged students to the extents of their inherent and newly acquired formal abilities, having been awakened, it was hoped, by the exercises, and educed their ingenuities. Projects allow great latitude within the framework of a challenging geometrical proposition or perceptual ambiguity. Though their topical particulars become dated, they make openings for creativity. Exercises, it is submitted, are to design what scales and etudes are to music; projects are preludes to composing.

3 FOUR THREE-DIMENSIONAL PROJECTS

Periodically, I offered basic designs studios that were conducted in a three-dimensional format, rather than in a more usual two-dimensional one. The topics are ones that have only three-dimensional physicalities. Both types of studios covered the same general formal issues of design. Four of the three-dimensional projects that I developed are presented here; many remarkable results have come out of them. 1. The Twofold Mirror-Rotation Project [Fig. 1] came out of my one funded project, Issues of Symmetry in Design Education. A grasp of the concept of mirror-rotation, even in its simplest twofold form, was made difficult by K. L. Wolfs German texta text that Germans found difficult to penetrate. This intriguing type of symmetry, initially baffling me, led to its exploration and exploitation in the design studio. If its properties took inordinate time for our research team to learn, how many others in our midst still do not know what it is? The basic design studio seemed to be a good place to make demonstrative, didactic models of twofold mirror-rotation, as well as to try to bring out some of its aesthetic potential. Of a sudden, while trying to find a good way to explain mirror-rotation to others, I realized that I had a simple didactic example of this symmetry operation in my own two handsliterally: I mirror my hands (like Drers Praying Hands)then, I rotate one hand in respect to the other by 180now, I have 2-fold mirror-rotation. This is what Wolf called a coupled coverage operationsimultaneously mirroring while rotating, or simultaneously rotating while mirroring.

95

See: <http://members.tripod.com/vismath/exbud/bud1.htm> 2. The Fourfold Mirror-Rotation Project [Fig. 2] came to light while I was familiarizing myself with crystallography. At the beginning of the funded project, no texts, which I had referenced, made clear that symmetry theorythat is, a whole system of symmetrycame from crystallography, or that molecular chemistry was decades behind crystallographymainly because the nature of the molecule was not well understood at the time that the 32 finite Crystal Classes had been established. It was from papers of Federov, translated into English through David Harker, that I had learned about famed Auguste Bravaiss missing a thirty-second Crystal Class: the symmetry of fourfold mirror-rotation. For well over a century, the discovery of that class had been attributed to Hesseland, in fact, he is the one to whom all the other early crystallographers referred. It was relatively recently, in 1986 (only three years before the organizing meeting of ISIS-Symmetry in Budapest), that it became widely known about one Moritz Ludwig Frankenheim who, in 1826, had preceded Hessels publishing all 32 Classes in 1830and Frankenheim had done it in a brilliantly simpler way than Hessel, through combinatorics.3 While materializations of this symmetry type exist in some numbers at the molecular level, fourfold mirror-rotation is rare at human scale. It is known in crystals; but no manmade artifact is known to employ it, at least not known to me. Though some students in earlier studios produced a few objects with this property of symmetry, in 1986, when I taught some sixty beginning design students at the University of Hong Kong, the majority of them made fourfold mirror-rotation objects, though some made sixfold ones. May I claim that never in the history of the world were there so many differently designed and produced fourfold mirror-rotation artifacts in one place at one time? 3. The Project of the Trisection of the Cube into Congruent Parts [Fig. 3] has its own odd history. Already exploring for a decade the sectioning of the cube into any number of congruent partsand the congruent sectioning of other regular and semi-regular solids, I concentrated on the trisectioning of the cube with my students, after Martin Gardner had faltered in his presentation of this particular topic in his September 1980 column, Mathematical Games, for Scientific American. Since then, I have found a number of noticeably different types of trisections. Though I can surmise, not being a topologist, I cannot be certain which trisection types are unique in respect to others (there are a minimum of three, I believe) and which types are only variations of others.

J. J. Burckhardt, Die Symmetrie der Kristalle, (Basel: Birkhuser Verlag, 1988), 34-47.

96

4. The Project of the Non-orientable Surface [Fig. 4]. Max Bill had come to the HfG with a history of sculpting the Mbius band, more generically called the non-orientable surface. Bill produced such a band for the Breuer-Roth Doldertal Apartments in Zurich, but he did not know about the nature of its peculiar topological properties until a mathematician informed him. It was Maldonado, however, who introduced the topic to his Grundlehre classes at the HfG. When I continued studies of non-orientable surfaces in my own studio, our first question of the Mbius band was whether any properties of symmetry could be introduced into such a type; for upon the casual viewing of its usual depiction, it appears to be irregular. Being a topological entity, the Mbius has no fixed cross-ratio and is, therefore, subject to infinite deformation (as Bills original sculpture serves to demonstrate); however, in its usual presentation, a special case to be sure, it conforms to twofold rotational symmetry; thus, this property of symmetry can be imposed at will on such an artifact. One student demonstrated that, by starting with a flat pattern of a fourfold rotational nature, she could easily attain fourfold mirror-rotation in her configuring that pattern into a duplex non-orientable object.

97

Fig. 1. Twofold Mirror-Rotation Artifact; Basic Design Studio of William S. Huff; Student: Lisa Lu, SUNY at Buffalo, 1980.

Fig. 2. Fourfold Mirror-Rotation Artifact; Basic Design Studio of William S. Huff; Student: Sally Kilmer: SUNY at Buffalo, 1990.

Fig. 3. Array of Three Trisected Cubes; Basic Design Studio of William S. Huff; rows from right to left: Student: Vincent Marlowe, SUNY at Buffalo, 1978; Student: Fernando Polletta, SUNY at Buffalo, 1983; Student: Michael Cooke, SUNY at Buffalo, 1987.

Fig. 4. Non-orientable Surface (with surgical clamp); Basic Design Studio of William S. Huff; Student: Habib Khalafi, SUNY at Buffalo, 1980.

98

DESIGN AND COGNITION: CONTRIBUTION TO A DESIGN THEORY

CLAUDIO GUERRI

Name: Claudio F. Guerri, Architect (b. Rome, Italy, 1947). Department of Morphology and Communication, Faculty of Architecture, Design and Urbanism, University of Buenos Aires, Buenos Aires, Argentina. Home-Address: Olleros 2532 2A, (1426) Buenos Aires, Argentina. E-mail: claudioguerri@fibertel.com.ar Fields of interest: Theory of Design, Graphic Languages, Space and Visual Semiotic, Architecture, Graphic and Industrial Design Awards: Article of the Year (Best Art Critic Article) AICA Award, 1986; Cientific and Technological Production Award, SICyT-University of Buenos Aires, 1995. Publications: (1988) Article: Architectural, Design, and Space Semiotics in Argentina in The Semiotic Web 1987. A yearbook of Semiotics by T. A. Sebeok and J. Umiker-Sebeok (eds.), pp. 389-419. Berln: Mouton de Gruyter. ISBN 3-11-011711-8 (1997) Symposium paper: Deep structure and design configurations in paintings in Semiotics of the Media: state of the Art, Projects and Perspectives by Winfried Nth (ed.), pp. 675-688. Berln: Ed. Mouton de Gruyter. ISBN 3-11-015537-0

Abstract: The traditional graphic languages used for designing, orthogonal projective geometry and perspective, account for only quantifying or qualifying space. They do not allow for conceptualizing or specifically working with pure design, the morphological and aesthetic aspects in the practicing of design. TSD is a new graphic language that systemizes necessary and sufficient morphic and tactic dimensions of planar and volumetric forms, allowing all possible relationships of selection and combination, basic operations of pure design. Symmetry is a property of high isotopic synthesis and, therefore, affords an effective cognitive strategy in the practice of design that is widely employed in it, whether obvious or hidden in a designs final outcome.

99

1. INTRODUCTION
Although there is no certainty, sedentarism, durability of constructions and geometric thinking emerged at the same time. In his famous History of Architecture, Fletcher (1896 [1956]: 1) claimed that Architecture, with all its varying phases and complex developments, must have had a simple origin in the primitive efforts of mankind to provide protection against inclement weather, wild beasts, and human enemies. This explanation about the origin and also most writings about architecture stress function, highlighting construction as an inhabiting-protective device. The time has come to change the discourse and analyze the pure form related aspects, art, the artistic side, the esthetic concern, etc. from the angle of the prefiguring possibilities that enables the design operations through the available graphic languages in the framework of an incipient Grammar or Theory of Design. Architecture was born at a very complex crossroad. Architecture results from the application of design to the formal elaboration of constructions mankind builds to inhabit the world. Architecture is both the use of built space and the delimitation of inhabited space. On the other hand, Design is creation, delimitation, and articulation of form, articulation of form to take under control the construction and dwelling necessities. As autonomous knowledgefrom construction and dwellingDesign enables, through a projectmodifying the first anthropoids monotonous and utilitarian constructions. (Jannello 1983: unpublished). Geometry was probably the first development in human thinking about the world that had not emerged directly from sensory experience. Geometry belongs to the world of forms, and its application to inhabitable constructions allowed for its exact formal determination. The history of architecture, as one kind of applied design, has produced in five thousand millenniums a vast typological repertoire of combined shapes with their own syntax. Not much has been said about this syntax or combination. One of the first problems is to understand that Geometry was not conceived for purposes of Design, although since ancient times some of its intuitive and non-systematic properties and techniques have been used, such as geometrical tracings (Ghyka 1927: 221-322; Naumann 1930; Lawlor 1982 [1992]: 23-64). These still are and have always been the quintessence for appreciating the value of symmetry operations, generally hidden in constructions but present in tracings. Even though Geometry deals with forms only from an entitative viewpoint, geometrical tracings were generally used to explain the structure of symmetry. Designers, however, have conceptual and practical needs: to compare and establish relationships between forms and to select and combine them. 100

Graphic languages, such as Monge System (orthogonal projections) and Perspective (conic projections), traditionally used for designing, account only for quantifying or qualifying space, respectively space of construction or dwelling space. They do not allow for conceptualising or working with pure design (Jannello 1980; Guerri 1988: 394) understood as selection and combination of forms without pertaining to an object. Because of Jannellos original research context Argentina 1950-1985, pure design is some how related to the Bauhauss Formlehre and basic design (Huff 1990: 76-85). In turn, going to the geometry-symmetry relationship, we must acknowledge that a geometric figure as a whole, as an entity, as a problem of simple selection of differentiated recognition, lacks symmetry. Symmetry implies parts, which are in symmetry. It could be the parts of a figure or several figures. When there is symmetry, there must be a combination of parts. According to Nicolle (1950: 16), ce qui nous frappe, cest donc une certaine rptition, soit dun objet, soit dlments qui le constituent. A geometric figure considered as a whole has no parts. Its entitative consideration would not allow for a symmetrical approach. The only way of finding symmetry in a figure is to consider partial, syntactic, or combining aspects based on subdivision criteria. But Geometry does not refer to comparisons among figures, nor to combination aspects between figureseither flat or volumetricin relation to constant and comparable dimensions (Jannello 1984; Guerri 1988: 399). Geometry was not created specifically for designers.

2. GRAPHIC LANGUAGES
The traditional graphic systems, such as orthogonal and conic projections, were developed as methods of execution. They were never conceived as ideological systems implicit in any language, either graphic or verbal. The first graphic system explicitly defined as such is the graphic language TSD. TSD is the acronym for Theory of Spatial Delimitation used in honour of its first proponent Csar Jannello (1918-1985). This graphic language systematizes all possibilities of selection and combination of flat and volumetric figures. It establishes necessary and sufficient morphic and tactic dimensions to account for all possible relationships of selection and combination, the pure design operations. The morphic dimensions are Formatrix, Size and Saturation (Jannello 1984; Guerri 1988: 406-408). These three dimensions can be justified to be necessary and sufficient conditions from a geometrical point of view: they determine a point in space, in the infinitely high semi-cone of morphic selection possibilities (Guerri 1988: 398), and from a logical-semiotic (Peircean) point of view, because they are firstness, secondness, and 101

thirdness respectively (Peirce CP 2.235-2.241). The tactic dimensions are Tactrix, Separationin two or three directions, and Attitude (Guerri 1988: 409). TSD systematizes the execution of tracings and subsequent description in hierarchical tree structures of corresponding complex configurations of pure design. (Guerri 1988: 389-419). Thus, insofar as each graphic language supposes different types of analogies, each of them has its specificity: Perspective to speak about quality of space, Monge System to state the quantity of matter present in that space, and TSD to account for the relationships in which matter and space abide. Regarding Symmetry, a variety of conclusions may be drawn from the TSD proposal and its use through CA-TSDComputer Aided TSDas a specialized-graphic software: 1. Although Geometry is taken as its foundation, given morphic and tactic dimensions, considered necessary and sufficient for both flat and volumetric figures; what is considered a figure and a simple configuration for this graphic system is thus redefined. For example: a rhomboid and a trapezium are no longer figures for the system and will be considered as design configurations. A simple configuration is the combination of at least two figures. The tactrix of a simple configuration is defined by a description of the morphic relation of the intervening figures. 2. The morphic paradigm generates the figures as a continuous group of substitution possibilities. As to symmetry, figures should be considered on account of their lineal components and of the dimensions proposed by the graphic language TSD itself. For example: Saturation, insofar as it is a qualitative evaluation of figures, implies a problem of constance-variation, which is usually known as proportionbut that, is a concept that cannot be used to compare two irregular figures. Saturation also posits a problem of presence-absence, there will always be some kind of saturation-proportion in any figure, but there may or may not be saturation-symmetry. 3. The tactic paradigm generates simple configuration, as a continuous set of integration possibilities. The set of morphic and tactic paradigms constitutes theoretically exhaustive systems of all possibleflat and volumetricfigures and configurations. It demonstrates that symmetry operations are a specific convergence of morphic and tactic dimensions; and it demonstrates that given any two forms, any simple flat or volumetric configuration, all derived symmetry variables can be determined a priori. 4. For Design, Symmetry is an operation of a higher isotopic synthesis and, as a result, it is one of the most widely used design operations. Since its origin, it has been known as a concept linked to regularity that has been used to keep order, a way to skirt chaos, 102

monstrosity, and shapelessness. History has it that symmetry was patently used or hidden, denied or reconsidered, but has always been present in design processes.

3. SYMMETRY
In Western architecture, order is construed in an agitated dialectics between historical periods taken as concrete units- and the general structure of architectural design since its origin, up to current times. Thus, the concept of order in design (style) is not a constant concept but the manifestation of structured fluctuations. In this history of fluctuations, of permanence and change, the concept of Symmetry follows its own development, and will be considered by Western thought as a concept of regularity. In architecture, symmetry has been one of the pillars for designto the extent that all epochs have been ordered around it. Thus, symmetry anchors an abstract practice of design in a concrete design operation. As stated, it becomes the first design algorithm. In Greece, symmetry is ostensible, generally, mirror and twofold rotation. The Renaissance continued making it ostensible, but in a more complex way: there was a tendency toward higher degrees of rotation; Villa Rotonda, with fourfold rotation; Bramantes San Pietro in Montorio with multi-fold rotation. At the break of the XXth. Century, Modernism meant a significant change in the application of symmetry, insofar as it was not concretely built into the constructed object. This rupture in ostensible symmetry lead to the development of hidden symmetry in subjacent design operations, never shown on plans or in buildings. Post-modern architecture attempts to establish a new order completely different from the modern order: it counters functional and constructive principles that, in their discourse, backed the basic principles of Modernism, and once again welcomes symmetry; but symmetry becomes a reference and a parody of other symmetries. In post-modern architecture, there is a transposition, and designed or built symmetry is a reflex or reference to some historical symmetry. The history of architecture reveals that, as in any human aspect, there are no laws, but rather a series of different rules that have been diachronically occurring over time.

CONCLUSIONS
TSD is presented as a temporary improvement regarding control over pure design operations. Within pure design operations, the possibilities of symmetry, in any configuration, are predetermined a priori by the tactic paradigm. As shown in the computerized and transparency examples, the use of TSD for analyzing architectural designPalladio, Le Corbusier, graphic designEl Lissitzkyor pictorial 103

designVelzquez, Malevichallows for systematizing ones approach to aesthetic values, thus contributing to the understanding of both historical practices as well as new design possibilities. Putting aside the problems of perception, TSD implies a new code of graphically representing the problem of Symmetry; and thus, it is also a new cognitive approach.

References
FLETCHER, Banister (1896) A history of Architecture on the comparative method. London: Batsford [1956]. GHYKA, Matila (1927) Esthtique des proportions dans la nature et dans les arts. Paris: Gallimard. GUERRI, Claudio F. (1984) "Semiotic characteristics of the architectural design based on the model by Charles S. Peirce" in Semiotic Theory and Practice, M. Herzfeld and L. Melazzo (eds.), 347-356. Berlin: Mouton de Gruyter, 1988. (1988) "Architectural design, and space semiotic in Argentina" in The Semiotic Web 1987, T. A. Sebeok and J. Umiker-Sebeok, eds., 389-419. Berlin: Mouton de Gruyter. HUFF, William S. (1990) What is Basic Design and Basic Design Studios of William S. Huff in Intersight 1, 76-86, New York. ISSN 1049-6564 JANNELLO, Csar (1980) Diseo, lenguaje y arquitectura. Buenos Aires: Course lectures, Facultad de Arquitectura y Urbanismo-UBA (mimeo). (1984) "Fondements pour une semiotique de la conformation delimitante des objets du monde naturel" in M. Herzfeld & L. Melazzo, eds., Semiotic Theory and Practice, M. Herzfeld and L. Melazzo (eds.), 483-496. Berlin: Mouton de Gruyter, 1988. LAWLOR, Robert (1982) Sacred geometry. London: Thames and Hudson [1992]. NAUMANN, Hans Heinrich (1930) Das Grnewald-Problem und das neuentdeckte Selbstbildnis des 20 jhrigen Matis Nithart aus dem Jahre 1475. Jena: Eugen Diederichs. NICOLLE, Jacques (1950) La symmtrie et ses applications. Paris: Albin Michel. PEIRCE, Charles S. (1931-58) Collected Papers of Charles Sanders Peirce, vols. 1-6 by C. Hartshorne, P. Weiss, vols. 7-8 by A. W. Burks. Cambridge: Harvard University Press. PRIGOGINE, Ilya (1983) Tan slo una ilusin? Una exploracin del caos al orden. Barcelona: Tusquets.

On the next pages are parts of Prof. Guerris Power Point presentation.

104

105

106

107

108

109

110

GROWTH, CURVATURE AND COMPUTATION

CHAIM GOODMAN-STRAUSS

Name: Chaim Goodman-Strauss, Mathematician, (b. Austin, Texas., U.S.A., 1967). Address: Department of Mathematics, University of Arkansas, Fayetteville, Arkansas 72701, U.S.A. E-mail: cgstraus@uark.edu Fields of interest: Geometry, topology, symmetry and ornament. Publications: Goodman-Strauss, C. (1998) Matching rules and substitution tilings, Annals of Mathematics. 147, 181-223. Goodman-Strauss, C. (2000) Open questions in tilings, preprint. Goodman-Strauss, C. (2002) Regular Production Systems and Triangle Tilings, preprint.

Abstract: We discuss some new mathematical techniques for describing growth and form across a range of mathematical, biological and artistic applications. The essential idea is in the air these days: to describe emergent complex behaviour through simple local interactions. These interactions are shaped by local combinatorial restrictions, and in turn give rise to global geometric properties and tremendously rich behaviour is possible. New tools are proving to be quite powerful, producing a body of interesting mathematical, but too they seem to shed light on real phenomena, such as the way that tissue grows to produce complex geometric forms such as the human ear. We will discuss the mathematical idea philosophically, with an emphasis on presenting a wide range of graphical applications.

111

In this talk, Id like to present a model of growth and form that arises from local interactions among little combinatorial agents. These agents can be thought of as puzzle pieces which may fit together in certain ways; two pieces can be neighbors only if they are compatible. To visualize this, imagine a sea filled with copies of a few kinds of the organisms in Fig 1: these organisms have hands of certain shapes and sizes that can fit together in only certain ways with each other. Even though it is very simple to tell what is happening locally, the global behaviour of the system may be quite subtle. The study of such agents immediately touches on fundamental issues in the theory of computation, and rich, beautiful behavior arises.

Figure 1: A cartoon model of local combinatorial structure. We will focus on how curvature can arise through these interactions; in particular, we will see that there will be no method, a priori, of determining just what the global curvature of a given system will be. Conversely, all kinds of behaviors will be possible.

Curvature
Before getting underway, Id like to take a moment to discuss the curvature of surfaces. These ideas are well known and quite old, but may be unfamiliar to the general audience. A flat surface is a good place to start (Fig 2): a flat surface has the same local 112

metric properties as the Euclidean plane. In particular, as a circle grows on a flat surface, the circumference grows linearly with as the radius of the circle increases. The cylinder, for example, is in fact a flat surface. (An easy way to see this is simply to roll up a sheet of paper). This rasies an important point: flatness and curvature are intrinsic properties of the surface. It doesnt matter one bit that a cylinder is curvy: its flatness depends on the way we measure things when we are confined to the surface.

Figure 2: Two flat surfaces. This isnt so strange the earlier image of the plane was in perspective, and distance was certainly not accurately rendered in that image either. In Figure 3 we see two images of a checkerboard. The images appear quite different but they are intrinsically the same, measuring distances so that all the squares are to be of uniform size. An inhabitant would not be able to tell which of the two spaces he lived in.

Figure 3: Geometry is intrinsic; two more flat surfaces. We can contrast this with positively curved surfaces, such as the sphere, and negatively curved surfaces such as a crinkly piece of lettuce. 113

A sphere really is curved; there is no way to make a sphere out of flat pieces. On a sphere, the circumference of a circle grows much more slowly than the radius, and the amount of curvature can be measured by this deficit. As before, the particular rendering is unimportant; the intrinsic notions of distance on the surface define curvature. We are quite used to seeing this: the Earth is not flat despite the image at right in Figure 4.

Figure 4: Positive Curvature. On a negatively curved surface, the circumference of a circle grows exponentially as the radius increases. A physical object with negative curvature will typically be quite crinkly, or have numerous tubes. In Figure 5, at middle at right, we see two forms, drawn by Ernst Haeckel, with high negative curvature.

Figure 5: Negative Curvature.

114

Combinatorics and Curvature

Consider fitting together puzzle pieces to make a surface. For the moment, we will suppose we that the surface is growing outwards along some boundary. (Fig 6, again due to Haeckel)

Figure 6: Growth of a surface along a boundary. Locally the curvature of the resulting surface depends on how fast this boundary is expanding. If the boundary is expanding rapidly with each step, then the curvature will be negative. If the boundary is contracting, or growing very slowly, the curvature of the surface will be positive. This is precisely the effect we see here: as the shell accretes, the boundary is growing extremely rapidly and we have high negative curvature. This is precisely the effect we see in Figure 7: as the shell accretes, the boundary is growing extremely rapidly and we have high negative curvature.

Figure 7: Highly negatively curved surface. 115

One may enjoy performing a small experiment to see how local conditions give rise to curvature. Consider gluing many equilateral triangles together to make a tiling or polyhedron. If exactly six triangles meet at each vertex, the resulting surface will be flat. If fewer than six meet at each vertex, the result will be a surface with positive curvature (for example, if five meet at each vertex, we will obtain an icosahedron a spherical surface). And if more than six meet at each vertex, we will obtain a surface with negative curvature. In fact, this is all true regardless of the specific geometry of the triangles used what is important is how they meet, i.e. the nature of the local combinatorics. Now suppose the local curvatures are uniformly mixed: in some spots the curvature is positive, and elsewhere negative. What will the overall curvature be? If we can calculate the overall proportion of each local behavior we can easily calculate the overall curvature of the surface. For example, consider triangles that can meet in two kinds of ways: 113 and 331 (that is, in 5s and in 7s) There is in fact only one possible solution: the resulting surface must be flat: the two kinds of local curvature must balance out perfectly.

Combinatoria
Id like to turn now to the kinds of local combinatorial objects Im considering. Such gadgets can be thought of as little puzzle pieces that can fit with one another. One example of such gadgets are tiles in the Euclidean plane. Here, of course, we know a priori what the local curvature will be. But consider the following question: Given a set of tiles, can you tell whether you can cover the entire Euclidean plane with copies of these tiles? For example, can you cover the plane with copies of the tile at left in Figure 8? How about with copies of that at right? (These examples are due to Jarkko Kari).

Figure 8: Which of these two tiles can admit a tiling?

116

This question is known as the Domino Problem. The real question is: Is there a way to decide the answer in general? That is, Is the Domino Problem decidable? The Domino Problem is exceedingly subtle, even when we have only one kind of tile. (As a hint of this, even the two tiles in Figure 8 are tricky to analyze. The tile at left in Figure 8 can tile a large region but no more (Figure 9). The tile at right in Figure 8 can tile, but the simplest way this can be achieved is fairly complicated (Figure 10).

Figure 9.

Figure 10. In 1966, R. Berger showed the Domino Problem is undecidable in the Euclidean plane. That is: There is no general method to decide, for a given set of tiles, whether they can form a tiling 117

Any computation can be modeled by some set of tiles. A very nice corollary to this is that there exist aperiodic sets of tiles tiles that can tile but can never tile periodically. The Penrose tiles are the most famous example (Figure 11).

Figure 11: The Penrose rhombs. Now for our game here we are laying down layer after layer of these combinatorial objects; can the overall rate of growth be determined?

Symbolic Subsitution Systems

A first model is symbolic substitution systems. One begins with an alphabet, say 0, 1, and a set of replacement rules, say 01, 110 This defines a map on the set of all words in this alphabet. For example: 0110 1 10 10 1 and in turn 110101 1010110110 We can define superwords; these arise from applying this map repeatedly to our alphabet. So for example, here our superwords are 011010110110101101011011010110110 etc Now note:

118

a) Such symbolic substitutions can model the kind of growth weve been discussing. The letters describe our combinatorial objects. The words describe our boundaries. b) Classical theorems describe precisely the rate at which word-lengths grow under the substitution, and the overall distribution of letters in the superwords. In our example 01, 110 words tend to grow by the golden ratio and in the large, the ratio of the number of 1s to the number of 0s will also tend to the golden ratio. In particular, word length grows exponentially. The tiling of the hyperbolic plane shown in Figure 11 is precisely described by the system 01, 110. And indeed all symbolic substitution systems can be modeled in similar fashion.

1 1 0 0 1

Figure 12: A tiling of (the Klein model of) the hyperbolic plane based on the system 01, 110.

Regular Production Systems

However, the general situation is much more subtle. In brief, one considers an alphabet as before. However we restrict ourselves to a regular language of allowed words. For example, let our alphabet be 0, 1, 2 and take the language of words where: 0 can only be followed by 1, 1 can only be followed by 2, and 2 can only be followed by 0 or 1. So for example, the words 1201 and 012120 are in the language, but 1120 is not.

119

For each letter we take one or more replacement rules. In this example we take 012 112 220 121 201

Here the replacement is not deterministic. For two words W, V in our language, we write WV if there is some choice of replacements on the letters that takes W to V. But Note: a given word may be mapped to one, no or many words. For example the word 012120 can only be mapped to 12 12 01 21 20 12; the word 0120 cant be mapped to any word at all. And the word 1212 can be mapped to either 12012120 or 21201201. The point of these systems is that they can model any arrangement, with any curvature, of any set of combinatorial objects. The letters describe the pieces, the language how these pieces fit together in layers, and the rules how each layer may fit with the next. A cartoon of this is shown in Figure 13.

Figure 13: A cartoon illustration of how regular production systems can be used to model growth by the accretion of combinatorial elements along a boundary. However, quite unlike the classical substitution systems, it is likely that it is undecidable, for example, whether one may repeatedly substitute ad infinitum; it is quite likely that it is undecidable whether a given rule will be needed; it is likely that it is undecidable how frequently a given rule will be applied. In particular, it is certainly undecidable, given a particular regular substitution system, what the curvature of the corresponding surface will be. Or to put it another way, any desired behaviour can be attained.

120

Post Tag systems and growth

Id like to demonstrate this with an adaptation of an universal computer due to Emil Post, called Post Tag Productions One has an alphabet, a set of rules and a starting letter: aabc b a cba At each step, one has a word. You cross off the first two letters, and depending on the original first letter, add a word to the back. So for example: abbca bca abc (crossing off ab and adding abc) bcaabc aabca bcaabc etc. This particular system repeats quite quickly. However, a little experimentation shows it is difficult to predict whether the words in a given system will eventually repeat, grow forever, or shrink away to nothing. I encourage the reader to try a few more examples simply select different rules, a different alphabet, or a different starting word. In fact the long term behaviour is in general undecidable and indeed any computation can be modeled as a Post-tag system. In a manner that we cannot explain here, this implies that the curvature induced by a regular production system (that is, how fast the words in aregular production system grow) is undecidable; correspondingly, anything can be achieved. This is precisely, in fact, why nature has access to such a staggering variety of forms through simple growth along a boundary, using elementary rules. A complete bibliography and proper treatment of this material can be found in the papers Open Problems in Tiling and Regular Production Systems and Triangle Tilings both available at http://comp.uark.edu/~cgstraus/papers. abbca

121

References
Berger, R. (1966) The undecidability of the domino problem, Memoirs Am. Math. Soc. 66 (1966). Gardner, M. (1977) Extraordinary nonperiodic tiling that enriches the theory of tilings, Scientific American 236, 110-121. Goodman-Strauss, C. (1998) Matching rules and substitution tilings, Annals of Mathematics. 147, 181-223. Goodman-Strauss, C. (2000) Open questions in tilings, preprint. Goodman-Strauss, C. (2002) Regular Production Systems and Triangle Tilings, preprint. Grnbaum, B and Shepherd, G.C. (1987) Tilings and patterns, W.H. Freeman and Co. Minsky, M.L. (1967) Computation and infinite machines, Prentice-Hall, Englewood Cliffs, N.J. Queffelec, M (1987) Substitution dynamical systemsspectral analysis, Lecture Notes in Mathematics 1294, Springer-Verlag, New York. Robinson, R.M, (1971) Undecidability and nonperiodicity of tilings in the plane, Inv. Math. 12, 177-209. Thurston, W. (1989( Groups, tilings and finite state automata: Summer 1989 AMS Colloquim Lectures, unpublished notes.

122

SLICEFORM SURFACES AND A SERENDIPITOUS DISCOVERY

JOHN SHARP

Name: John Sharp, Chemist, Freelance Technical Author, mathematician and artist, (b. Kingston-upon-Hull, England, 1945). Address: 20 The Glebe, Watford, Herts, WD25 0LR England E-mail: sliceforms@compuserve.com. Fields of interest: Geometry and art Publications /Exhibitions: 1) Illustrator of David Wells' "The Penguin book of Curious and Interesting Geometry" Penguin1991. 2) Counton, the UK government website to promote mathematics at www.counton.org, sets of pages on Sliceforms, Morphing Tilings, GridWarps, and Anamorphic art. 3) Sliceform sculptures: part of the mathematical models collection at the Science Museum in London featured in their exhibition on Strange Surfaces 4) Contributor to Art and Mathematics 2000 exhibition Cooper Union, New York 5) Numerous articles and papers on geometry and Art and Mathematics in Nexus Journal, Times Education Supplement, Art Review, Mathematical Gazette and other educational mathematics journals.

Abstract: At the end of the nineteenth century, mathematicians created many models of geometrical surfaces which were beautifully crafted. They are as artistic as the many mathematical computer graphics created at the end of the last century. Among these was a group of models of quartic surfaces constructed as sets of circular slices. They were slotted together in such a way as to continuously deform. The artist Naum Gabo saw these models in Munich around 1910 and they influenced his construction of sculptural heads. My own work in constructing such surfaces I call Sliceforms has created a number of new surfaces. The technique has a wide range of possibilities, but one of the most interesting, resulting from a mistaken assembly of a model, is a new relationship between surfaces with an unusual change of symmetry. The work described here is work in progress.

1. SLICEFORMS
I have developed, but not invented, a method to create models of mathematical surfaces and to create collapsible paper sculptures which I have called Sliceforms [4].

123

At the end of the nineteenth century, mathematicians created many models of geometrical surfaces which were beautifully crafted. They are as artistic as the many mathematical computer graphics created at the end of the last century. Among these was a group of models of quartic surfaces constructed as sets of circular slices. They were slotted together in such a way as to continuously deform. They were discovered by a mathematician called Olaus Henrici and are often seen described in books on mathematical models [1] and three dimensional geometry [2] and were produced as a set for use in teaching German mathematician Alexander Brill. It is thought that the Russian constructivist artist Naum Gabo saw these models in Munich around 1910. He then used them as inspiration for his sculptured heads which are now in the Tate Gallery, Londons museum of modern art. [3] Mathematically, the idea does not generally appear other than as illustrations of quartic surfaces. I have taken the idea further in my Sliceform sculptures [4]. Some of these are shown alongside the Henrici/Brill models in the exhibition Strange Surfaces at the Science Museum in London.

1.1 Construction of Sliceforms

Pictures and displays of Sliceform models do not show their full beauty. Only by making and physically handling the models can their true dynamic qualities be fully appreciated. Their three-dimensional forms and surfaces are defined or suggested by two intersecting sets of parallel slices (see Figure 1). These intersections act as a multitude of hinges and as a consequence each model can be made to collapse flat in two different ways. Between these two extreme positions the surface passes through a host of different but related shapes. By using different colours for the slices in each direction, the patterns generated as the model is manipulated can be very attractive and unexpected. The play of light on them and the shadows they create as they are moved also offer more insights into the interplay of art and mathematics.

Figure 1, Construction method

124

Designing and constructing the sculptures offers insights for both artists and mathematicians. They can also be used as a starting point for other types of sculpture such as plaster casting since the gaps between the slices form hollow tubes.

1.2 Symmetry of slices

Surfaces can have many different types of symmetry and this affects whether the shape of the slices in each direction. With a surface like a sphere, the slices (like the left slice in Figure 2) are identical in both directions. In the case of surfaces of rotation which do not have spherical symmetry, the slices are identical in shape, but the slots are placed differently as in the two slices of an egg surface (like the right two of Figure2).

Figure 2, shape of slices

2. SLICEFORM REVERSALS
My own work in constructing Sliceforms has created a number of new surfaces. The technique has a wide range of possibilities, but one of the most interesting, resulted from a mistaken assembly of a model. By accident, in attempting to create a model where the slices were identical in shape, but had slots in different directions, I used two sets of slices for the same direction. Surprisingly, the slices fitted perfectly, although I was later to find that this is not always the case. Moreover, depending on the properties of the original surface, the resulting surface can have different symmetry properties. The resulting transformation is not a standard one since it involves a series of rotations and translations for each slice and not a transformation of the surface as a whole.

125

I am currently exploring this serendipitous discovery to find out the way different surfaces of revolution behave. Since I originally read chemistry, my scientific intuition is to experiment before starting to understand what is happening mathematically, in particular the constraints on the resulting surface. Since it is by no means a trivial transformation, the variety of results have vindicated this approach. This is still work in progress, and the results are more intriguing and surprising than I can understand at the present.

2.1 The hemisphere reversal

Figure 3 shows the result for the simple case of the hemisphere, which at first appears to be the developable surface formed by the union of two cylinders, but in fact shows subtle differences.

Figure 3, reversal of a hemisphere

2.1.1 Why the hemisphere reversal does not work

However, one must not take the evidence of ones eyes in attempting to determine the mathematics of such an object. Close inspection reveals that the slices may not match exactly. Normally, when flattened, the slices fit together so that there is no motion along an axis perpendicular to the slots. In this case, there is in fact some movement possible which distorts the Sliceform. That the two sets of slices do not match is shown as follows. By taking the central slice of one direction and then fitting all the slices in the opposite direction I calculated the expected shape of the other slices in the original direction. The

126

shape of the slices which are needed are no longer part circular sections. Figure 4 shows the result. The left diagram shows the shape of a slice from a hemisphere. The right slice shows how the true shape of the slices needed to match have a rounded shape.

Figure 4, shape of the slice

2.2 The paraboloid reversal

In the case of the surface of revolution of a parabola, the reversal of the Sliceform does give a set of slices that fit. Figure 5 shows the original surface and the reversal version.

Figure 5, paraboloid and reversal The slices when slotted together in the reversal, fit perfectly as can be seen in figure 6. Moreover, as with the apparent reversal of the hemisphere in figure 3, the resultant surface is one that has been created by the intersection of a pair of parabolic cylinders, where the axis of each cylinder is tangent to the peak of the other cylinder. The question then comes as to why the slices of the fit together perfectly. The answer, as yet unproved mathematically, is shown in figure 7. The slices of the surface of revolution of a parabola are all the same parabola. The diagram at the left of figure 7 shows the central slice which is rotated to form the surface. The one at the right is another slice. They are congruent parabolae.

127

Figure 6, views of the paraboloid reversal, showing cylindrical cross section

Figure 7, slices of the surface of revolution of a parabola

2.2 Other reversals

Another reversal of a surface of that appears to work is of an oval. This gives rise to an egg shaped surface. The central slice giving rise to the surface of revolution, together with the Sliceform is shown in figure 8. The result of reversing the slices is shown in figure 9. The surface has a pointed end which points up when viewed orthogonally to one set of slices as shown in the right view, but down when the surface is rotated. The half way view at the left of figure 9 shows the difference from the original egg of figure 8, although it is not possible to appreciate the full effect of the reversal in two photographs like these. You need to manipulate the object to appreciate its properties.

128

Figure 8, egg as surface of revolution of an oval

Figure 9, reversal of surface of revolution of an oval The accuracy of the reversal here has not been checked mathematically. When a surface will obviously not fit when reversed, the method is still a powerful one for the artist. Figure 10 shows a curve with five fold symmetry which has been used to create a surface of revolution. The reversal of the slices shown in figure 11 do not match at the outermost slots. However, this gives a visually interesting reversal where some of the slices can be moved to give a range of surface possibilities.

Discussion and conclusions

Many of the audience were intrigued by the beauty of the display of models shown. In answer to a request for help for working on the mathematics, no one was forthcoming. However, informal discussion afterwards showed that this type of symmetry change is

129

out of the normal stream and delegates were unsure how it fits in with conventional theories.

Figure 10, curve with 5-fold symmetry and its surface of revolution

Figure 11, curve with 5-fold symmetry and its surface of revolution

References
[1] H M Cundy and A P Rollet, Mathematical Models, Oxford University Press 1960 [2] W H McCrea, Analytical geometry of three dimensions, Oliver and Boyd, Edinburgh 1953 [3] Steven Nash and Jrn Merket, Naum Gabo, Sixty Years of Constructivism, 1985. A catalogue of a touring exhibition which began at the Dallas Museum of Art and ended in the Tate Museum in London, [4] John Sharp, Sliceforms, Tarquin Publications, Stradbroke,Norfolk, England, 1995

130

ART OF ANAMORPHOSIS
PHILLIP KENT.

Name: Phillip Kent. Address: School of Mathematics, Science and Technology, Institute of Education, London, WC1H 0AL, United Kingdom. E-mail: p.kent@mail.com.

This presentation will be based on text, artworks and software produced for an exhibition held in the United Kingdom in the spring of 2001. Anamorphosis is concerned with the creation of distorted images according to the mathematical rules of perspective and mirror reflection. The distinctive thing about these distortions is that the undistorted form of the image can be recovered by looking at the image from a particular location, or using an unusual shape of mirror (cylindrical, conical). The relationship between anamorphosis and mathematics is interesting in several ways. Anamorphosis is an application of mathematical transformations several of which have been implemented in a piece of free software, Anamorph Me!, that can be downloaded from the web site. This software can be used to turn any image into an anamorphic distortion, often with beautiful and fascinating results. Furthermore, the idea of anamorphosis developed in the European Renaissance, at a time when art and mathematics were deeply connected: it was artists who first developed the geometrical understanding of perspective on which anamorphosis is based.

Reference
See www.anamorphosis.com.

131

132

133

134

BASIC CRYSTAL SYMMETRIES GENERATED BY MOLECULAR DIMERS

ALAJOS KLMN AND LSZL FBIN

Professor Alajos Klmn X-ray Crystallographer, (b. Budapest 1935) and Dr. Lszl Fbin X-ray Crystallographer (b. Kecskemt, 1973), Institute of Chemistry, Chemical Research Center, Hungarian Academy of Sciences, Budapest, II. Pusztaszeri ut 59-67, P.O.B. 17, H-1525, Hungary. E-mail: akalman@chemres.hu (Chemical crystallography , isostructurality ) Academic Award, 1975; Golden Medal of the Labour Order of Merit, 1986; The Hungarian State Prize bearing the Name of Count Istvn Szchenyi, 1994. 375 original papers in the field of crystal structure analysis and chemical crystallography.

Abstract: In the last years numerous attempts have been made to recognize deterministic elements of crystallization among the stochastic factors. E.g. a series of crystals, formed by equal amount of mirror related cyclic molecules (CH2)n n = 58, bearing two vicinal hydrogen-bond donor/acceptor functions (e.g. OH, COOH) were found to represent seven of eight patterns of close packing. In each structure there are H-bonded homochiral () or heterochiral () chains assembled either in antiparallel or in parallel mode. We also revealed that these patterns of basic crystal symmetries can be deduced from the linear or lateral association of molecular dimers with either Ci, C2 or Cs symmetry.

1. BASIC PATTERNS OF CRYSTAL ARCHITECTURE

The crystal structure determinations of numerous 1,2-disubstituted cycloaliphatic molecules such as cyclopentanes, cyclohexanes, cycloheptanes and cyclooctanes resulted in the recognition of basic (almost canonical) patterns of supramolecular close packing (Klmn et al., 2001; 2002a). In accordance with the files of the Cambridge Structural Database (CSD) the patterns represent the most popular space groups assumed by organic molecules in the solid state: P21/c (35%), P1 (19%), C2/c (7%).

135

1.1 Early fact gathering

The simplicity of the compounds offered an opportunity to study intermolecular hydrogen bond interactions generated by two vicinal (cis- or trans) functions located on an alicyclic ring: (CH2)n n = 5, 6, 7 or 8. Each 2-hydroxy-1-cycloaliphatic-carboxylic acid (functions: OH and COOH) forms two intermolecular hydrogen bonds, while each 2-hydroxy-1-cycloaliphatic-carboxamide (functions: OH and CONH2) forms three. The hydrogen bonds HB1 (OHOC) and HB2 (XHOH, X = O or N) are independent of the stereoisomerism exhibited by these compounds. Every crystal is racemic, which means that each crystal has the same amount (1:1) of mirror related molecules in the unit cell. Overall, the crystal structures investigated (Klmn et al., 2001) reflected six of eight possible basic patterns of supramolecular self-assembly organised by the parallel or antiparallel arrangement of hydrogen bonded homo- () or heterochiral () chains of the molecules.

1.2 Deduction of the basic forms of close packing

The ninth pattern, observed first in trans-2-hydroxy-1-cyclooctanecarboxylic acid (hereinafter 8T), could not be inferred from the parallel/antiparallel arrays of the molecular chains. It was revealed only by the combination of the HB1 and HB2 bonds. They, depending on symmetry relationships between the molecules, may generate four motifs: translation forms homochiral tapes (T), screw axis generates helices (H), glide plane forms heterochiral meanders (M), while inversion center joins heterochiral rings (R). One of their independent combinations (R:R) corresponds to the linear association of heterochiral (Ci) dimers. Studying this pattern, it became apparent that the linear association of Cidimers, joined by twelve membered rings, generates automatically new dimers which are the links between them. These dimers differ in the acceptor group(s) of the hydrogen bonds. The OC dimers are formed by the OH...OC hydrogen bonds, whereas the OH dimers are formed by the XHOH hydrogen bonds. Along the parallel ribbons of 8T molecules these OC and OH dimers alternate. In other words, the heterochiral connection between two dimers of either type generates the other dimer, and this linear array is therefore unique. It can be regarded as the corner stone of close packing patterns recognized in our works. This recognition prompted us to generate all of the lateral and linear associations of the dimers maintained by twelve-membered rings with Ci, or C2 symmetry and assembled either in antiparallel or in parallel mode. They can be classified into six basic patterns of close packing as depicted in Fig. 1 (1-3) and Fig. 2 (4-6). 1. Linear assoc. of Cidimers (OC and OH dimers together): space group P1, 2. Lateral assoc. of Cidimers in parallel array: space group P1, 3. Lateral assoc. of Cidimers in antiparallel array: space group C2/c, 4. Linear assoc. of C2dimers polymerized into parallel helices: space group Pca21, 5. Linear assoc. of C2dimers polymerized into antiparallel helices: sp. group. P21/c,

136

6. Lateral assoc. of C2dimers polymerized into antiparallel helices: sp. group. P21/c. In addition, glide planes (Cg) may also form heterochiral dimers without ring closure: 7. Lateral assoc. of Cgdimers in parallel array: basic space group Pc, 8. Lateral assoc. of Cgdimers in antiparallel array: basic space group Pn, the alternative crystal packings: space group either P21/n or Pna21. Both glide plane-generated patterns 7 and 8 are hallmarked by 18-membered antidromic rings (Jeffrey & Saenger, 1991) which generate significant total dipole moment. Since dipoles must cancel out over the whole crystal, antiparallel stacking of the layers is compulsory. This can be done by screw axes (21) either perpendicular to (space group P21/n), or parallel (space group Pna21) with the best planes of the rings. The synthesis of novel compounds enabled us to demonstrate experimentally all dimer forms (except 7) by X-ray diffraction.

1.3 Future: Crystal engineering

From these works it was recognised (Klmn et al., 2002a) that (a) Pattern 1 could easily be deduced from pattern 2 if all HB1 bonds turn simultaneously from the respective homochiral chains to their neighbouring enantiomers and vice versa. (b) The carboxamide analogue (COX, X = NH2) of 8T (COX, X = OH) also crystallises with pattern 1, therefore they are isostructural (Klmn et al., 2002b). This prompted us to extend our symmetry versus structure investigation, termed combinatorial crystal chemistry, to novel chemical structures. In accordance with expectation, a superposition of patterns 1 and 2 was obtained by the replacement of OH function with NH2 moiety. The series of cis-2-amino-1-cyclopentane- to -cyclooctanecarboxylic acids is isostructural in space group P1. This is a unique case in which the common form of crystallisation is controlled by the deterministic formation of the expected hydrogen bond pattern, rather than by the stochastic effects of anisotropic forces. Support from OTKA (Grant No T034985) is thanked. References
Klmn, A., Argay, Gy., Fbin, L., Bernth, G. and Flp, F. (2001). Basic forms of supramolecular selfassembly organized by parallel and antiparallel hydrogen bonds in the racemic crystal structures of six disubstituted and trisubstituted cyclopentane derivatives. Acta Crystallographica Section B, 57, 539-550. Klmn, A., Fbin, L., Argay, Gy., Bernth, G. and Gyarmati, Zs. (2002a). Novel, predicted patterns of supramolecular self-assembly, afforded by tetrameric R44(12) rings of C2 symmetry in the crystal structures of 2-hydroxy-1-cyclopentanecarboxylic acid, 2-hydroxy-1-cyclohexanecarboxylic acid and 2-hydroxy-1-cycloheptanecarboxylic acid, Acta Crystallographica, Section B, 58, in the press. Klmn, A., Fbin, L., Argay, Gy., Bernth, G and Gyarmati, Zs. (2002b). Combinatorial crystal chemistry: Novel, predictable close packing similarities between cis- and trans-2-hydroxy-1cyclooctanecarboxylic acids and trans-2-hydroxy-1-cyclooctanecarboxamide. Acta Crystallographica, Section B, submitted for publication. Jeffrey, G. A. and Saenger, W. (1991). Hydrogen Bonding in Biological Structures, Springer Verlag, Berlin, Heidelberg.

137

P1

C2/c

P1

C2/c

Fig. 1. Linear (1) and lateral (2 and 3) associations of OH and OC dimers of Ci symmetry.

5a
Pca21 Pca21

5b

P21/c

P21/c

b
P21/c

6
P21/c

Fig. 2. Linear and lateral associations of homochiral dimers polymerized into parallel (4) and antiparallel (5, 6) helices.

138

Day 3

At the Africa Museum of Tervuren, 2002 April 11.

Location.
The location for the Thursday talks was the Africa Museum of Tervuren, nearby Brussels. Brussels counts several Leopold II style buildings, but this Institute for nonEuropean Art and Sciences was an inspiring choice. Built in 1897, there is a permanent exposition of animals, plants, ethnographical objects and sculptures linked to Central-Africa. After 1908, it became the Belgian Congo Museum, changing its name again in 1960 to become the Royal Institute for Central-Africa and one of the most important centers for scientific research about Africa.

Special Feature
At the Africa Museum of Tervuren Eric Blanckaert, a professional moviemaker shot a video and images reproducing the mood of the math-art conference in the Africa museum. Below are some of these images.

139

Impatient participants.

The participants, waiting to get started.

View on the conference room.

View from another angle.

Prof. Tarnai in action.

The projection system worked perfectly in the African environment.

140

Co-organisor Slavik Jablan explaining the mathematics of the surrounding African ornaments (and many other objects).

Chairman Denes Nagy, presenting Leslie Greenhill.

Les during his talk.

Anna Bulckens during her talk.

The traditional coffee break was really traditional.

Conference organization member Ann Ratinckx prevented elephants to enter the conference room.

141

Prof. Ruiz with junior researcher Maria Peas during their joint lecture. It combined expert research with refreshing ideas through a contagious energy. The quality of the movie images is lower, of course, and they do not give the correct impression of the video. Here are some images, in reduced format, but they hopefully will allow the reader to get an impression about the video.

142

143

144

145

146

FOLDED STRUCTURES
TIBOR TARNAI

Name: Tibor Tarnai, Structural Engineer, Appl. Mathematician, (b. Hatvan, Hungary, 1943). Address: Department of Structural Mechanics, Budapest University of Technology and Economics, Megyetem rkp. 3, Budapest, H-1521 Hungary. E-mail: tarnai@ep-mech.me.bme.hu. Fields of interest: Kinematically indeterminate structures, discrete geometry (packing and covering problems). Publications: Tarnai, T. (1996) Symmetry of golf balls, In: Ogawa, T., Miura, K., Masunari, T. and Nagy, D., eds. Katachi U Symmetry. Tokyo: Springer-Verlag, 207-214. Tarnai, T. (1996) Geodesic domes: natural and man-made. International Journal of Space Structures 11, 1325. Fowler, P.W. and Tarnai, T. (1999) Transition from circle packing to covering on a sphere: the odd case of 13 circles. Proceedings of the Royal Society of London A 455, 4131-4143. Tarnai, T. and Gspr, Zs. (2001) Packing of equal regular pentagons on a sphere. Proceedings of the Royal Society of London A 457, 1043-1058.

Abstract: The paper gives an overview of folding and its applications in some fields of natural and man-made structures, such as symmetric latticed cylinders, braced domes, deployable space structures, insect wings and swollen form of viruses. Application of folding to buckling of thin-walled box columns and expandable polyhedra are also investigated.

1. ORIGAMI
In traditional Japanese origami, folding of straight line is used, and in this way a great variety of shapes is produced. Origami with much more sophisticated curved lines is also available. Many lampshades are made with this type of folding. Origami occurs also in mathematics, however, the most striking constructions were produced not in Japan but in the US. Kaleidocycles and flexagons were invented in the 1930s. Nowadays a new 147

architectural style is called "folding" with a slight reference to paper folding, though, architectural forms based on origami have been used for a very long time.

2. SYMMETRIC LATTICED CYLINDER

The inextensional buckling form of a thin-walled circular cylinder is the Yoshimura pattern. Probably, it is not an accident that this buckling form was discovered in Japan, because this form of many boxes, lampshades and vases was known before. The Yoshimura pattern, in fact, is a column composed of equal antiprisms. The edge network of the Yoshimura pattern can be considered as a cylindrical bar-and-joint structure. If the lower joints are attached to a foundation and the upper joints are free, then for this latticed cylinder, the necessary condition of rigidity is fulfilled. However, it is rigid only if the number of nodes along the circular base is odd. For even numbers, the structure is not rigid (Tarnai, 1980). There are many examples of non-rigid latticed cylinders: braced cooling towers, interstage structures in multi-stage rockets, the timber octagon in Ely cathedral, children climbers, the Gemini telescope, etc.

3. BUCKLING OF THIN-WALLED BOX COLUMNS

Hollow columns with polygonal cross-section usually are made of thin metal plates by folding and welding. Their inextensional local buckling form is like a curved origami. A deeper insight into this buckling pattern is available by studying folding of uniform plane tessellations. Tessellation of Schlfli symbol (3,3,4,3,4) is used as a base of the investigation. The polyhedron obtained by folding leads to a curved origami that approximates the buckling pattern (Tarnai, 1997). Here, neither the polyhedron, nor the curved origami has complete straight-line folds. To maintain zero Gaussian curvature we need some theorems of differential geometry.

4. HIGHER-ORDER MECHANISMS
When folding a plane tessellation into a cylinder the whole structure behaves like a onedegree-of-freedom mechanism with finite displacements. There are bar-and-joint assemblies called higher-order infinitesimal mechanisms, which produce the feeling that they move with finite displacements, but in fact they have only infinitesimal free motions. This feeling is caused by imperfections. In the Madrid codex, there is a sketch of Leonardo about a chain-like bar-and-joint structure composed of T shaped units. The 148

horizontal and vertical segments of the T shape are composed of two parts connected to each other by a joint at an angle. This structure shows an exponential increase in forces in bars. On the other hand, if the member pairs are straightened, then the assembly becomes a higher-order infinitesimal mechanism, which shows an exponential decay of motion.

5. DEPLOYABLE STRUCTURES IN LIVING NATURE

Folded structures are common in nature. At plants, for instance, leaves of trees and petals of flowers are in folded position in buds. When they deploy, usually a growing process is coupled to deployment. At insects, the situation is different. For butterflies, wings deploy at the beginning of their adult period, and their shape is kept until the end of their life. For beetles, however, transparent wings under hard wings are deployed and retracted at every flight. Different species apply different folding patterns.

6. APPLICATIONS IN SPACE RESEARCH

Space structures are usually compactly packaged on earth, and deployed in orbit in space. Packaging of deploying masts, solar panels, membranes and antennas is a basic problem. There exist several different solutions to it. An inextensional folding system of rectangular membranes is the "Miura-ori", where the folded membrane deploys with one degree of freedom. The principle of this folding system came from the analogy with the large displacement buckling pattern of thin elastic plates that was experimented recently by metal-coated shrinking jelly layers. Miura-ori is a real origami that is used also in map folding. If a circular membrane should be folded, then folding about a central hub can be used by rotation.

7. BRACED DOMES
Nowadays many large domes are built for sport arenas. These structures are mainly constructed as space frames with large height. In order to reduce scaffolding, M. Kawaguchi has developed the "Pantadome" construction system, in which the main part of the structure is erected on ground, but some bars are deliberately left out the construction. The assembly in this phase is a mechanism that can be pushed up to the final position, then by adding the missing bars to it the structure becomes complete. This large-scale origami technique is frequently used in Japan and elsewhere. For temporary 149

use, transportable deployable latticed domes (tents) have also been developed. It is a requirement that a stadium or a sport dome is adapted to weather conditions. Therefore, retractable roofs are built. One group of these structures is the "Iris Dome". The latticed domes work as iris of an eye: increasing or decreasing the pupil, that is, the opening at the top of the dome. Hoberman has realized such a structure, but his structure requires very difficult nodes. Recently, F. Kovcs suggested a much simpler solution.

8. EXPANDABLE POLYHEDRA
There are several toys in which a polyhedron is transformed to another one by continuous motion. An early construction, the "jitterbug", is due to Fuller. It transforms an octahedron to an icosahedron then a cuboctahedron. Such a structure was constructed in the Heureka exhibition in Zurich about ten years ago. Not long ago, it was experimentally observed that certain icosahedral viruses have a swollen form if pH of the environment changes (Speir et al., 1995). The known expandable polyhedra having icosahedral symmetry, which were developed by engineers and mathematicians (Hoberman's (1991) expanding globe, Verheyen's dipolygonids, Yananose's Juno's spinner), do not provide proper model of the expanding virus. We have started a research to describe the motion of the expandable virus. Until now we could develop a simplified model, an expandable dodecahedron, which however has six additional degrees of freedom those are not required for the investigated motion (Kovcs and Tarnai, 2000).

9. CONCLUSIONS
At the beginning, origami itself, and folded configurations were considered only as toys. Later, it turned out that folded structures have serious applications in space research, medical research and architecture. Importance of folded structures will increase in the future. Acknowledgement. Research reported here was supported in parts by OTKA Grant No. T031931, and by OMFB and the British Council Grant No. GB-15/99.

References.

150

Hoberman, C. (1991) Radial expansion/retraction truss structures. United States Patent, Patent Number 5,024,031. Kovcs, F. and Tarnai, T. (2000) An expandable dodecahedron. In: Gerrits, J.M., ed., Proceedings of the 4th International Colloquium on Structural Morphology. Delft: University of Technology, 227-234. Speir, J.A., Munshi, S., Wang, G., Baker, T.S. and Johnson, J.E. (1995) Structures of the native and swollen forms of cowpea chlorotic mottle virus determined by X-ray crystallography and cryo-electron microscopy, Structure 3, 63-78. Tarnai, T. (1980) Simultaneous static and cinematic indeterminacy of space trusses with cyclic symmetry. Int. Journal of Solids and Structures 16, 347-359. Tarnai, T. (1997) Folding of uniform plane tessellations, In: Miura, K. ed., Origami Science and Art. Proceedings of the Second International Meeting of Origami Science and Scientific Origami. Otsu: Seian University of Art and Design, 83-91.

Below is the Power Point presentation of the contribution.

151

152

153

154

155

156

157

158

159

160

SYMMETRY AND ORNAMENT

SLAVIK JABLAN

Name: Slavik Jablan. Address: Mathematical Institute, Knez Mihailova 35, P.O.Box 367, 11001 Belgrade, Yugoslavia. E-mail: jablans@mi.sanu.ac.yu Web site: http://www.mi.sanu.ac.yu/~jablans/

Abstract: Throughout history there were always links between geometry and the art of painting. These links become especially evident when to the study of ornamental art we apply the theory of symmetry. Therefore, ornamental art is called by A. Spaiser "the oldest aspect of higher mathematics given in an implicit form" and the "prehistory of group theory".

The idea to study ornaments of different cultures from the point of view of the theory of symmetry, given by G. Plya and A. Speiser, was supported by the intensive development of the theory of symmetry in the 20th century. This caused the appearance of a whole series of works dedicated mostly to the ornamental art of ancient civilizations, to the cultures which contributed the most to the development of ornamental art (Egyptian, Arab, Moorish, etc.), and to the ethnical ornamental art.

Only in some recent works, research has turned to the very roots, the origins of ornamental art- to the ornamental art of the Palaeolithic and Neolithic. The extensions of the classical theory of symmetry- antisymmetry and coloured symmetry, made possible the more profound analysis of the "black-white" and coloured ornamental motifs in the ornamental art of the Neolithic and ancient civilizations.

This work gives the results of the symmetry analysis of Palaeolithic and Neolithic ornamental art. It is dedicated to the search for "ornamental archetypes"- the universal basis of the complete ornamental art. The development of ornamental art started 161

together with the beginnings of mankind. It represents one of the oldest records of human attempts to note, understand and express regularity- the underlying basis of any scientific knowledge. The final conclusion is that the most of ornamental motifs, which have been discussed from the standpoint of the theory of symmetry, are of a much earlier date than we can expect. This places the beginning of ornamental art, the oldest aspect of geometric cognition, back to several thousands years before the ancient civilizations, i.e. in the Palaeolithic and Neolithic.

The contents of his talk were published in Jablans recent book with the same title.

162

How Plato designed Atlantis

Leslie Greenhill 2002

Address: P.O. Box 314, Mentone, Victoria 3194 Australia E-mail: lesgreenhill@yahoo.com.au Fields of interest: geometry and architecture in antiquity; Neoplatonism; metrological systems of Ancient Egypt, Greece and the Roman Empire Some recent achievements: presenter at Mathematics 2000 Festival, University of Melbourne; divisional winner in Australian short story competition (2000)

163

Remarks on references and abbreviations

The works of classical writers often have ciphers printed in the margins of the translations to indicate a standard means of precise reference to passages. For example, in the case of Plato they indicate the pages in the edition of the philosophers works by Stephanus (Henri Estienne), Geneva, 1578. To accord with tradition, references to Platos writings are presented in this manner along with references to more accessible popular versions of his books. Hence, in the text of this exposition, references to classical works are presented in the following way: Plato. Lee, D. Timaeus and Critias, p. 137/S114 S114 indicates the page in the Stephanus edition. Herodotus. In the case of Herodotus (The Histories), the numbering system is that used by J. Marincola from the de Slincourt translation, e.g., Marincola, p. 88/H2.9 H means Herodotus and 2.9 indicates Book Two, passage nine. Vitruvius. For Vitruvius (The Ten Books on Architecture): Morgan, p. 27/V1.6.9 V means Vitruvius and 1.6.9 indicates Book One, chapter 6, paragraph 9. Plutarch. Lastly, for Plutarch (Moralia V), Babbitt, pp. 1757/S381 S381 indicates page 381 in the Books of the Moralia, the edition of Stephanus, 1572. Sometimes, for clarity, the letters BI are used to distinguish British imperial measures from ancient Greek and Roman measures: for example, 607.5 BI feet.

164

SPECIAL NOTES (1) Historians of metrology have noted the use of a range of feet of varying values in ancient Greece. In all important cases the present writer is able to demonstrate they are proportionally related in a specific geometric configuration to the so-called Greek/Parthenon foot, the ancient Egyptian remen and the ancient Egyptian cubits. The matter is comprehensively dealt with in a new paper tentatively entitled The Creation of Measure in Antiquity now in preparation (February 2003). The only Greek measures discussed at length in this account are the so-called Parthenon foot of about 12.15 inches or 308.6 mm, the Parthenon cubit (equal to 1 Parthenon feet), and the stade of 600 Parthenon feet. As an example of a proportional relationship, the so-called Doric foot is 36 the Parthenon foot multiplied by /35 twice. This measure, about 326.5 mm, had a 36 special design function, as did the /35 proportion; they are explicated in the new paper. (2) Since the Matomium Conference in Brussels in April 2002, when this paper was first presented, the measures in the design system described herein have been discovered, inventively encrypted, in a famous classical Roman text. Ancient Egyptian, Greek and Roman measures have been expressed in the text in terms of the so-called British imperial inch, with the values exactly the same as shown in this exposition because the same geometry was used. This far-reaching discovery (February 2003) is presently being incorporated into The Creation of Measure in Antiquity.

165

How Plato designed Atlantis

1. Preamble
1.1. The material and the sources. This exposition is extracted from a larger work entitled How Plato designed Atlantis and where Vitruvius obtained his model for Man. The two subjects are combined in the larger exposition because the designs are linked. In fact, in the larger work it is demonstrated that Vitruvius (fl. 1st century BC), author of The Ten Books on Architecture and an admirer of Plato (Morgan, p. 195/Introduction to Book Seven), has reformulated the Greek philosophers design for Atlantis in several ingenious ways, one of which is the mathematical design for a well shaped man. The larger exposition, though, contains more evidence to prove the case on Atlantis than is presented here. That said, there is sufficient evidence in this version to justify the titles contention.

Figure 1: Vitruviuss well shaped man , as realized by Cesar Cesariano in his edition of Vitruvius (1521).

1.2. Plato. 166

The philosopher Plato, 427347 BC, a disciple of Socrates, is a seminal figure in the Western intellectual tradition. Two of his late works Timaeus and the continuation Critias contain the earliest accounts of the mythical lost continent of Atlantis. Plato says the information about Atlantis was given to Solon, an Athenian and the wisest of the wise men, by an Egyptian priest (Lee, Timaeus and Critias, p. 337/S205). The principal physical description of the country is contained in Critias, which has been described by Desmond Lee, a translator of the books, as the first work of science fiction. The most detailed description of the place given by the philosopher, though, is of the capital city, in particular the inner citadel, a circular island surrounded by a concentric arrangement of rings of water and land. Amongst other things, the inspiring source of the design concept of the overall layout of the land is unveiled here along with its encrypted mathematical nature. 1.3. Plato and time measurement. In the Introduction to Timaeus and Critias Lee writes: Plato was aware of the close connection between time and time measurement. Can we speak of one without the other? And if not, are we not bound to say that time came into being with the heavens, that is, that time in the sense we use it cannot be conceived without the instruments, processes, and movements by which we measure it? (Lee, Timaeus and Critias, p. 11) The quote is from the Timaeus. extracted. The following is the passage from which it was

So time came into being with the heavens in order that, having come into being together, they should also be dissolved together if ever they are dissolved; and it was made as like as possible to eternity, which was its model. For the model exists eternally and the copy correspondingly has been and is and will be throughout the whole extent of time. (Lee, Timaeus and Critias, p. 52/S38) It will be demonstrated that the nature of time, arising from unexpected and surprising geometric sources, is inseparable from the nature of Atlantis. 1.4. Plato and The Laws. That twelve is a key time number is beyond dispute. In The Laws, believed to be Platos last major work, the philosopher sets out his ideas for the ideal state. This is his description of the ideal city:

167

After this, the legislators first job is to locate the city as precisely as possible in the center of the country, provided that the site he chooses is a convenient one for a city in all other respects too (these are details which can be understood and specified easily enough). Next, he must divide the country into twelve sections. But first he ought to reserve a sacred area for Hestia, Zeus and Athena (calling it the acropolis), and enclose its boundaries; he will then divide the city itself and the whole country into twelve sections by lines radiating from this central point. The twelve sections should be made equal in the sense that a section should be smaller if the soil is good, bigger if it is poor. The legislator must then mark out five thousand and forty holdings, and further divide each into two parts; he should then make an individual holding consist of two such parts coupled so that each has a partner near the center or the boundary of the state as the case may be. He should also divide the population into twelve sections, and arrange to distribute among them as equally as possible all wealth over and above the actual holdings (a comprehensive list will be compiled). Finally, they must allocate the sections as twelve holdings for the twelve [Olympian] gods, consecrate each section to the particular god which it has drawn by lot, name it after him and call it a tribe. Again, they must divide the city into twelve sections in the same way as they divided the rest of the country; and each man should be allotted two houses, one near the center of the state, one near the boundary. That will finish off the job of getting the state founded. (Saunders, pp. 2156/S745) Plato follows this up by stating: Now that we have decided to divide the citizens into twelve sections, we should try to realize (after all, its clear enough) the enormous number of divisors the subdivisions of each section have, and reflect how these in turn can be further subdivided and subdivided again until you get to 5040. This is the mathematical framework, which will yield you your brotherhoods, local administrative units, villages, your military companies and marching-columns, as well as units of coinage, liquid and dry measures, and weights. The law must regulate all these details so that the proper proportions and correspondences are observed. (Saunders, pp. 2178/S746) 1.4.1. In a commentary on Platos most famous work the Republic, James Adam, author of The Republic of Plato writes: We know from the Laws that Plato counted 360 days in the year. (Adam, p. 301) Adams footnote to this passage states: 168

The number of Senators in the Laws is 360: these are to be divided into 12 sections of 30 each, and each section is to administer the State for one month. The number 60 with its multiples and divisors is the dominant number throughout the Laws. 360 days is of course only an ideal division of the year: see 6. Plato elsewhere recognizes (with Philolaus) 364 days (Rep. IX 587 E ). (Adam, p. 301) The number 364 is half 729 which is 27 squared. The number 27 makes several notable appearances in this exposition. 1.5. Plato and the Pythagoreans. In the Translators Introduction to Platos most famous work, the Republic, Lee remarks: In 388387 BC Plato visited South Italy, perhaps in order to make the acquaintance of some of the Pythagorean philosophers living there. The Orphic-Pythagorean belief in the after-life and the Pythagorean emphasis on mathematics as a philosophic discipline certainly influenced him strongly, as can be seen in the Republic. (Lee, Republic, p. xvii) Given Platos familiarity with Pythagorean ideas there can be no doubt that he was familiar with the well-known symbol of the Pythagoreans: the pentagram.

Figure 2: the pentagram, the symbol of the Pythagoreans.

2. Central Atlantis
2.1. Platos description. 169

The philosophers description of the citadel and ring arrangement in Atlantis is as follows: The largest of the rings, to which there was access from the sea, was three stades in breadth and the ring of land within it the same. Of the second pair, the ring of water was two stades in breadth, and the ring of land again equal to it, while the ring of water running immediately around the central island was a stade across. The diameter of the island on which the palace was situated was five stades. (Lee, Timaeus and Critias, p. 139/S115) The full ring arrangement measures 27 stades in diameter: see figure 3 below. An ancient Greek stade measured around 607.5 BI feet (based on the so-called Parthenon foot of about 12.15 inches). The stade and the Greek/Parthenon foot are discussed later. Lee makes the following points in his commentary on Atlantis (Lee, Timaeus and Critias, p. 152): a) But what Plato is most interested in and spends most time describing is the capital city itself, and more particularly its inner citadel. b) The inner citadel is shown as a series of small rings at the center, Its basic form was determined by Poseidon who ringed the small hill where Cleito lived with two rings of land and three of water; but its equipment and buildings are the work of the inhabitants. c) The breadth of the rings of land and water is 3 + 3, 2 + 2, and one stade, and the central island is 5 stades across. (Is it significant that this gives a total of 27 = 33?)

170

Figure 3: the rings of water and land around the central island of Atlantis five stades in diameter (3 + 3 + 2 + 2 + 1 + 5 + 1 + 2 + 2 + 3 + 3 = 27 stades, the diameter of the full ring arrangement). 2.1.1. 3. Plato provides clues on the original design source for the illustration in figure

He [Poseidon] begot five pairs of male twins, brought them up, and divided the island of Atlantis into ten parts, which he distributed between them. The eldest, the King, he gave a name from which the whole island and surrounding ocean took their designation of Atlantic, deriving it from Atlas the first King. His twin, to whom was allocated the furthest part of the island towards the Pillars of Heracles and facing the district now called Gadira, was called in Greek Eumelus . (Lee, Timaeus and Critias, p. 137/S1134) Attention is drawn to four matters: the numbers five (notably the five-stade diameter of the central island) and ten, how the name Atlantis is derived from Atlas, and the mention of Heracles (Hercules). Atlas is frequently depicted in art bearing the globe of the cosmos or the world on his shoulders.

171

Figure 4: statue of Atlas in Collins Street, Melbourne.

3. Dissecting the Pythagorean symbol

3.1. Five and ten. In the Pythagorean symbol in figure 5 belowa five-pointed starreferences to five and ten can clearly be seen.

172

Figure 5: a five-pointed star, the pentagram. ABC above, one of five isosceles triangles, is a one-tenth segment of a decagon (see figure 6 below). Angle A is 36, angles B and C each 72.

Figure 6: a decagon inside a circle. 3.2. Angles. In the pentagram illustrated below the following angles can be discerned:

173

Figure 7: angles in a pentagram. Note that 36 is 6 squared, 72 + 72 (144) is 12 squared, and 108 + 108 (216) is 6 cubed. 3.2.1. If 36 is taken as base one then figure 7 above can be illustrated thus:

Figure 8. Instantly the source of Platos 3 + 3 + 2 + 2 + 1 ring arrangement becomes apparent: see figure 3 and the preceding description by Plato of the ring arrangement in 2.1.

3.3. Time and the world. What made Plato, the Pythagoreans and others so interested in the pentagram? An examination of the nature of time shows why. 174

A day contains 24 hours. Each hour contains 60 minutes and a minute contains 60 seconds. An hour contains 3600 seconds, twelve hours 43,200 seconds and a full 24hour period, 86,400 seconds. In antiquity, the 24-hour period was divided into twelve daylight hours and twelve nighttime hours. The latter arrangement has an ancient Egyptian origin. The Egyptians were the first to divide the day into 24 hours; there were twelve hours of the day and twelve hours of the night. (Gardiner, p. 206) 3.3.1. In figure 9 below it can be seen that from point A to B the sum of the angles is 432. A to C also adds up to 432. A to B plus A to C sum to 864. As stated, twelve hours contains 43,200 seconds and 24 hours contains 86,400 seconds.

Figure 9: (A to B or A to C) 36 + 72 + 108 + 108 + 72 + 36 = 432. Furthermore, the sum of the angles in the pentagon component of the pentagram is 540 (108 x 5) and the sum of the angles in the five isosceles triangles is 900 (180 x 5): 540 + 900 = 1440. A day contains 1440 minutes. 3.4. Connecting to ancient Egypt. In ancient Egypt, right back in the pyramid age, ancient Egyptians used the five-pointed hieroglyph illustrated below to represent, amongst other things, time. 175

Figure 10: the five-pointed star hieroglyph. The following information is from Sir Alan Gardiners opus, Egyptian Grammar. It has been extracted from a description of the star hieroglyph pictured above, which is N14 in his Sign-list. ideo or det. [ideogram or determinative] in star, teach, time as indicated by stars, month, hour. (Gardiner, p. 487) An important and far-reaching deduction can be made. As no star in the sky has ever looked anything like the five-pointed hieroglyph above, this geometric figure, abstracted from a pentagram (see figure 11 below) must have been chosen in antiquity as a symbol for time because it had an appropriate mathematical constitution. It is relevant to point out that an ancient Egyptian month of thirty days contained 720 hours, that is, 360 daytime and 360 night-time hours: cf. the angles in figure 5. Drawing lines from the apexes of the triangles to the center of the pentagon can also create the star hieroglyph.

Figure 11: the star in the pentagram. 3.5. Eratosthenes and his measure of the world. 176

Eratosthenes, who became head of the Alexandrian Library around 240 BC, is credited with producing the earliest sensible measure of the world: 252,000 Greek stades. Vitruvius tells us of the matter in The Ten Books on Architecture (Morgan, p. 27/V1.6.9). Curiously, a few paragraphs after discussing the measure of the world the architect remarks: Some people do indeed say that Eratosthenes could not have inferred the true measure of the earth. (Morgan, p. 28/V1.6.11) There is a reason for this comment. Vitruvius, Eratosthenes, Plato, Plutarch (ca. 45120 AD), Herodotus (ca. 490420 BC) and other notable historical personages knew the real measure of the world was 216,000 (60 x 60 x 60) stades. It was a designed measure, just like the French decided a few hundred years ago that the world would have a circumference of 40,000 kilometers, that is, 40 million meters. In antiquity, the real figure was disguised from enemies and the uninitiated. Proof of the matter follows. 3.5.1. The pentagram is further examined. The lettering in the diagram below is based on that in figure 5 with the addition of letters X, D, and E. X, like B, signifies an angle of 72. D and E are each 108.

Figure 12. The sum of B 72 + D 108 + X 72 is 252. The sum of A 36 + B 72 + D 108 is 216. (So is the sum of D 108 + E 108.) The fake and the real measures, in reduced form, are therefore located in one geometric figureand they are part of each other. A remarkable analogue of this is presented later. In the larger version of this exposition, it is demonstrated that both the fake and the real measures of the world are encrypted in the mathematical constitution of Vitruvian Man. 177

The Greek (sometimes called the Periclean or Parthenon) foot is estimated to have measured around 12.15 inches (Zupko, p. 6 and Petrie, MW, p. 5). As a stade contained 600 Greek feet, the stade was estimated to be around 7290 inches or 607.5 feet (Zupko, p. 6).1 Multiply the latter number by 216,000 and the product is 131,220,000 feet which 3 is 24,852 /11 miles. The modern polar circumference measure of the world is 24,859.82 miles. Most contemporary encyclopedias provide this information. The difference is about 7.5 miles or about 12.1 km. The Parthenon foot is readily found. It is generally agreed that the top step of the Parthenons base was designed to have a length of 225 Greek feet (about 2735 inches) and a breadth of 100 Greek feet (about 1215 inches). Contemporary measurements (Encylopaedia Britannica, Vol. 9, p. 173) show the length of the top step to be 2737.7 inches (228.14 BI feet/69.54 m) and the breadth to be 1216 inches (101.34 BI feet/30.89 m). The rate of accuracy for the layout is better than 99.9%. Some remarks about the Great Pyramid are pertinent. The base of the structure was originally 440 royal cubits square and the height 280 royal cubits (Petrie, TPTG, p. 183). In Number and Divinity in Antiquity, another exposition by the present writer, it is demonstrated that the chief design particulars for every major ancient Egyptian pyramid stem from one ingeniously contrived geometric configuration even the layout of the three main pyramids at Giza. The designs for the pyramids and the Giza layout are a spectacular example (perhaps the greatest) of Vitruviuss maxim on symmetry and proportion: The design of a temple depends on symmetry, the principles of which must be carefully observed by the architect. They are due to proportion, in Greek . Proportion is a correspondence among the measures of the members of an entire work, and of the whole to a certain part selected as a standard. From this result the principles of symmetry. Without symmetry and proportion, there can be no principles in the design of any temple, that is, if there is no precise relation between its members, as in the case of those of a well-shaped man. (Morgan, pp. 72/V3.1.1) The royal cubit is estimated to have measured around 20.62 inches, about 52.375 cm (Petrie, TPTG, p. 139 and Klein, p. 59). If the height of the Great Pyramid, 280 royal cubits, were divided by the time-related number 8.64 (refer 3.3. and 3.3.1.) the quotient 11 would be 32 /27 royal cubits, which is about 668.24 inches. W. M. F. Petrie recorded the height of the entrance to the structure as being about 668.2 inches above the level of pavement of the pyramid (Petrie, TPTG, p. 55). Note that 55 Greek feet of 12.15 inches is equal to 668.25 inches. There is a 297:175 proportional relationship between the royal cubit and the Greek/Parthenon foot (proved later). The two measures are 178

generated in a remarkable way in the geometric configuration mentioned above, along with the well-known 22:7 proportion. The number 8.64 is also a prominent mathematical feature in the Atlantis design and other designs that are discussed later. 3.5.2. The stade measure can be readily detected in ancient Egypt. In an article A Ground Plan at Giza, British researcher John Legon, using the measurements of the renowned archaeologist W. M. F. Petrie, observed the appearance of a design strategy in the rectangular layout of the three main pyramids at Giza. Legon demonstrated that the axial distance from the west side of the pyramid of Menkaure (Mycerinus) to the west side of the pyramid of Khafre (Khephren), the second largest of the pyramids, was 7289.5 inches (Legon, AGPG, p. 40). Compare this with the estimated measure of the stade above, 7290 inches; there is a near perfect fit. Note that a stade is equivalent to 500 ancient Egyptian remen, which can be expressed as 10,000 remen digits. Zupko and others have recorded the remen digit as measuring 0.729 inch and equal to the Roman digit (Zupko, p. 6 and Klein, p. 71). Petrie also noted the link between ancient Egyptian, Greek and Roman measures (Petrie, MW, p. 5). An examination is now made of the way Vitruvius, Plutarch and Herodotus referred in cryptic form to the real measure of the world. 3.6. Vitruvius. The following information is from The Ten Books on Architecture: a) In the Introduction to Book Five Vitruvius says: Then again, Pythagoras and those who came after him in his school thought it proper to employ the principles of the cube in composing books on their doctrines, and, having determined that the cube consisted of 216 lines, held that there should be no more than three cubes in any one treatise. The Pythagoreans appear to have drawn their analogy from the cube, because the number of lines mentioned will be fixed firmly and steadily in the memory when they have settled down, like a cube, upon a mans understanding. (Morgan, p. 130) It can be readily guessed where the Pythagoreans obtained the number of lines from nowhere else but from the measure of the world. (That said, there are astonishing geometric reasons as well. They are explicated in the larger version of this exposition and in Number and Divinity in Antiquity where more detailed accounts of the geometric knowledge of Plato and the Pythagoreans are given.) 179

b) The 600 Greek-foot stade was equivalent to 625 (25 squared) Roman feet. A Roman mile contained 5000 Roman feet, which was equivalent to 8 (2 cubed) stades (Zupko, p. 6). Consequently, the ancient world measure can be expressed as 27,000 (30 cubed) Roman miles. Vitruvius tells us in a discussion of aqueducts, wells and cisterns that: It is not ineffectual to build reservoirs at intervals of 24,000 [Roman] feet, (Morgan, p. 245/V8.6.7) Twenty-four thousand Roman feet was equal to 4.8 Roman miles. This was a mathematical formulation for those in the know, and possibly a conundrum for initiates. The quotient of the world measure 27,000 Roman miles (equal to 135,000,000 Roman feet) divided by 4.8 Roman miles is 5625, which is 75 squared. Other instances of such mathematical formulations can be found in The Ten Books on Architecture. Here is another; again related to the measure of time. c) In a discussion on hoisting machines, Vitruvius reports: In our own times, however, when the pedestal of the colossal Apollo in his temple had cracked with age, they were afraid that the statue would fall and be broken, and so they contracted for the cutting of a pedestal from the same quarries. The contract was taken by one Paconius. This pedestal was twelve feet long, eight feet wide and six feet high. (Morgan, p. 289/V10.2.13) The volume of the pedestal is 576 cubic feet: 576 is the time-number 24 squared. Apollo returns to the story later. 3.7. Plutarch. E. A. Wallis Budge, the renowned Keeper of the Egyptian and Assyrian Antiquities in the British Museum, wrote that the ancient Egyptian god Thoth, usually depicted as a man with the head of an ibis, was, amongst other things, he who reckons in heaven, the counter of the stars, the enumerator of the earth and of what is therein, and the measurer of the earth; (Budge, Vol. 1, p. 400) Plutarch, apart from being a teacher and a philosopher, was also a Delphic priest. (There was a famous Oracle at Delphi that was located in the temple of Apollo. The Oracle, seated on a tripod, went into a trance before she spoke.) Plutarch writes of the ibis in Isis and Osiris:

180

The strictest of the [Egyptian] priests take their lustral water for purification from a place where the ibis has drunk: for she does not drink water if it is unwholesome or tainted, nor will she approach it. By the spreading of her feet, in their relation to each other and to her bill, she makes an equilateral triangle. (Babbitt, p. 1757/S381) Here is an inventive reference to the ancient measure of the world. An equilateral triangle, which the feet of a tripod also form, has a particularly suitable mathematical property that connects to the ancient measure of the world: each internal angle is 60 and there are three of them.

Figure 13: an equilateral triangle And, as discussed many times already, 60 x 60 x 60 = 216,000. Why Thoth was the counter of stars and the enumerator and measurer of the earth is no longer a mystery, thanks to Plutarch. 3.8. Herodotus. Three striking examples from the writings of Herodotus provide more evidence. Book Two in The Histories, which deals with ancient Egypt, is the information source. 3.8.1. Herodotus tells us that the coastline of Egypt is 60 shoeni long, and that a schoenus (an Egyptian measure, he says) is equal to 60 stades. The coastline is thus 3600 stades long (Marincola, p. 88/H2.67). Clearly, the historian has designed the measure to be exactly one-sixtieth of the ancient measure of the world, 216,000 stades which is equivalent to 3600 (60 squared) schoeni. Everything is in proportion. 3.8.2. Herodotus also reports: From Heliopolis to Thebes is a nine days voyage up the Nile, a distance of eight-one schoeni or 4860 stades. (Marincola, p. 89/H2.9) The number nine is 3 squared and eighty-one is 9 squared. If the measure of the world 4 2 is divided by 4860 stades the quotient is 44 /9 (44.444), which is 6 /3 (6.666) 181

squared. Later, the geometric source of this formulation is unveiled. Note that nine days contains 216 (6 cubed) hours. This and the material in 3.8.1. provide compelling evidence, indeed, about the reality of the ancient world measure of 216,000 stades. 3.8.3. The famous historian tells us about the city of Buto at the Sebennytic mouth of the Nile: The city also contains two other temples, one of Apollo, the other of Artemis. The shrine of Leto [mother of the two gods], where the oracle is, is a building of great size with a gateway sixty feet high, but the most remarkable sight it has to offer is not the temple itself, but a small shrine within the enclosure made out of a single block of stone; it is cubical in shape, each side sixty feet long and sixty high. (Marincola, p. 144/H2.155) The purpose of the shrine is clear: the cubes volume is 216,000 cubic feet and thus it memorializes the number associated with the ancient measure of the world. Leto, as a matter of interest, suffered terrible birth pangs for nine days and nights, that is, 216 hours, before giving birth to Apollo (Grant, p. 118). 3.8.4. One further mention of Apollo. Robert Graves tells us that the name has two possible meanings: destroyer or appleman (Graves, Vol. 1, p. 57 also Vol. 2, p. 381). Apples are discussed shortly. 3.8.5. Ancient Egypt provides more proof. E. A. Wallis Budge notes in a translation from an ancient Egyptian text that: Soon after his marriage with Thi, Amen-hetep [Amenhotep] III. dug, in his wifes city of Tcharu, a lake, its length 3600 cubits, its breadth 600 cubits. (Budge, Vol. 2, p. 70) The surface area of the lake dug by Amenhotep III (c. 13911353 BC) is consequently 2,160,000 square cubits. The significance of the number is obvious and it provides further evidence that the measure of the world has been around for a very long time indeed. In this regard, recall the discussion in 3.5.2. of the manifestation of the stade measure in the Giza, pyramid layout.

182

4. In and around Atlantis

4.1. Time in Atlantis. A further examination of Platos design for Atlantis proves fruitful. A Greek cubit was one-and-a-half Greek feet and had 24 digit divisions (Dilke, p. 26). A stade therefore contained 400 Greek cubits making the measure of the world equivalent to 86,400,000 Greek cubits: cf. 86,400 seconds in a day. In the following calculations, real pi or /7 can be used. Plato used /7. This is proved in the larger version of this exposition and in Number and Divinity in Antiquity. 22 For ease and meaningful outcomes, the fraction /7 is used here. 4.1.1. The concentric ring arrangement of Atlantis was detailed in 2.1. The central 9 275 island, five stades in diameter, has an area of 19 /14 ( /14) square stades. The area 10 2376 of the largest ring island is 169 /14 ( /14) square stades. The quotient of 169 /14 divided by 19 /14 is 8.64. A day contains 86,400 seconds and the ancient measure of the world was 86,400,000 Greek cubits. Review the pentagram material in 3.3.1.
10 9 22 22

4.1.2. The area of the smaller ring island is exactly one-third that of the larger ring 8 792 island, 56 /14 ( /14) square stades. It is exactly equal to the area of a course for horseracing on the larger ring island. Plato writes: On the middle of the larger island, in particular, there was a special course for horseracing; its width was a stade and its length that of a complete circuit of the island, which was reserved for it. (Lee, Timaeus and Critias, pp. 1401/S117) Note how the philosopher has drawn attention to this feature and has provided mensural details. This setup is for one of the most unique mathematical feats ever to have come from antiquity. It is dealt with later. 4.1.3. The area of the largest water ring, the outer ring, is 226 /28 ( /28) square stades. It is equal to the sum of the areas of the two ring islands. This area is, 4 consequently, /3 times the area of the largest ring island. The 4:3 ratio is also found in a 3:4:5-proportion triangle. The significance of this triangle to the nature of Atlantis becomes apparent soon. 183
8 6336

4.1.4. The Atlantis ring arrangement, as already mentioned in 2.1. is 27 stades in diameter. The quotient of 216,000 stades, the ancient measure of the world, divided by 27 is 8000, which is 20 cubed. 4.2. Atlantis and the pentagon.

Figure 14: angles in a pentagon. As can be seen in the above illustration two angles are prominent in the pentagon component: 72 and 54. a) In the Atlantis concentric ring arrangement, the area of the largest water ring is 226 8 /28 square stades. This is equivalent to the area of a circle that has a radius of 72 stades. b) The area of the largest ring island is 169 area of a circle that has a radius of 54 stades.
10

/14 square stades.

This is equal to the

All the components of the Atlantis concentric ring arrangement are dealt with later. 4.3. Dodecahedron. In the Timaeus where Atlantis is first mentioned by Plato, there is a discussion of the five Platonic solids. Plato equates the dodecahedron with its twelve faces, each face a regular pentagon, with the whole heaven (Lee, Timaeus and Critias, p. 78/S55). In Phaedo, however, the philosopher tells us, through Socrates, that: the earths true surface, viewed from above, is supposed to look like one of those balls made of twelve pieces of skin, (Tredennick and Tarrant, p. 177/S110)

184

Figure 15: a dodecahedron. Here is Plato exercising his remarkable imaginative faculties. He chooses the dodecahedron to represent the earths true surface because information on the size of the planet and the nature of time is bound up in the geometric properties of this unique solid. 4.4. Plain facts. The other main physical feature of Atlantis, we are told by Plato, is of a great rectangular plain 3000 stades by 2000 stades (Lee, Timaeus and Critias, p. 141/S118). It was mentioned in 3.5.1. that a stade contained 600 Greek feet. Hence, the plain can be said to measure 1,800,000 by 1,200,000 Greek feet, an area of 2,160,000,000,000 square Greek feet. There are no prizes for guessing the number alludes to the ancient measure of the world. Recall the mention by Herodotus of the cubical block of stone with a volume of 216,000 cubic feet in the shrine of Leto at Buto (3.8.3.) and the area of the lake created by Amenhotep III in 3.8.5. 4.5. Finding time,
22

/7 and much else in 3:4:5 proportions.

The pentagram was not the only geometry involved in Platos design for Atlantis. The 3:4:5-proportion triangle had a key role too. (Much more is made of this triangle in How Plato designed Atlantis and where Vitruvius obtained his model for Man and in Number and Divinity in Antiquity.) The proportion clearly attracted interest in the ancient world: (i) At least eight ancient Egyptian pyramids, including the second largest, the pyramid of Khephren (Khafre) which is next to the Great Pyramid on the Giza 4 plateau, exhibit such proportions. The angle of slope (tangent /3) is about 53 7 48 (Baines and Malek, pp. 1401 and Petrie, TPTG, p. 202). 185

(ii) Plato discusses the 3:4:5 proportions in the Republic (S546) in relation to the divine creatures cycle; Vitruvius mentions them several times, particularly in the Introduction to Book Nine of The Ten Books on Architecture; in Isis and Osiris (Moralia V), Plutarch associates the three side with Osiris, the four side with Isis and the hypotenuse, the five side, with their son, Horus (Babbitt, 135/S3734). 4.5.1. The geometry in figure 16 below is of great historic interest. In this one diagram, solutions to important ancient mysteries can be found: Atlantis; the pyramids; key metrologies of antiquity. The following is a description of what has taken place in the illustrated geometry. Everything has been realized by the use of straight edge and a pair of compasses mostly straight edge. It is instructive to study this geometry carefully; there is a kind of rhythm to the process: i) ABC is a 3:4:5proportion triangle. Squares are drawn on the three sides. ii) K is the midpoint of BC. K is connected to D on the four square by a line. The line crosses AC at R. iii) A dotted line continues from K to (lower case) b on the base of the five square GF. From b a dotted line parallel to AC is drawn to (lower case) h on BG. The small triangle hGb has 3:4:5 proportions. iv) From K a perpendicular line is drawn to L on AC. v) From B a line is drawn through L to O on DE. vi) A perpendicular line is drawn from A to I on GF. AI crosses BC at M. AM is 2.4 units. MY is constructed to be also 2.4 units. YI is 2.6 units. vii) BO crosses AM at N. Through N, a line parallel to AC is drawn from P on AB to Q on BC. PQ crosses DK at U. viii) From U a line parallel to BC is drawn to V on AB. VU crosses AM at . ix) From M a line is drawn to S on AB; SM is parallel to AC. From S a perpendicular line is drawn to W on BC. SW crosses BO at X. x) From K, the middle of BC, a line parallel to AC is drawn to T, which is the midpoint of AB. From T a perpendicular line is drawn to Z on BC.

186

Figure 16.

4.5.2. Features and outcomes a) The perimeter of ABC measures 12 units (3 + 4 + 5): cf. the twelve greater Olympian gods and Platos many references to the number twelve in his writings: The Laws, for instance. A British imperial foot has twelve inches. Most analogue clocks express time in terms of hours numbered from one to twelve and this can be directly linked to ancient Egyptian water and shadow clock designs. A year contains 12 months, just like the ancient Egyptian year. The perimeter of the entire configuration measures 36 units (3 + 3 + 3 + 4 + 4 + 4 + 5 + 5 + 5). A British imperial yard contains 36 inches; a circle contains 360 degrees; an 187

ancient Egyptian sacred year contained 360 days (Gardiner, p. 203). Also see Adams remarks in 1.4.1. The product of 3 multiplied by 4 multiplied by 5 is 60: cf. the sixty-second minute and the sixty-minute hour. In 3.3.1. it is shown that the sum of the angles in a pentagram is 1440 and that a day contains 1440 minutes; SM measures 1.44 (1.2 squared) units. (The pentagon component of the pentagram contains five by 108; SB measures 1.08 units.) b) AM measures 2.4 units and SV 0.24 unit. A day contains 24 hours. The Greek cubit was divided into 24 parts called daktyloi (Dilke, p. 26). The Roman cubit and the ancient Egyptian common (short) cubit also had 24 digit divisions. (In Number and Divinity in Antiquity it is shown how the Greek cubit was created in the geometry presently under examination and how the geometry, in an astonishing natural process, automatically divides the measure into 24 digit divisions. c) MK is 0.7 unit. BD is ten times larger and is 7 units. A week contains seven days. QC is 1.4 units: a fortnight contains 14 days. The body of the ancient Egyptian deity Osiris is said, according to Plutarch, to have been cut into 14 parts by his brother Seth and those parts scattered throughout Egypt. d) Triangle BPQ has three measures: BP 2.16 units, PQ 2.88 units and BQ 3.6 units, a total of 8.64 units. A 24-hour day contains 86,400 seconds. It was shown in 4.1.1. that the area of the larger ring island in Atlantis was 8.64 times the area of the central circular island. In 3.5.1. the entrance to the Great Pyramid was described and it was shown that the original height of the structure, 280 royal cubits, divided by 8.64 accurately produced the height above pavement level of the entrance. SW measures 0.864 unit. All the main traditional numbers associated with the way time is measured have thus been located in full or miniature form: 12; 24; 60; 7; 14; and 360. e) Triangle ABC has an area of 6 square units. RKC, a scalene triangle, has an area of 1 10 /11 square units. The quotient of 6 divided by 1 10/11 is 22/7, the value often used in mathematical calculations to represent pi, a transcendental number. Its appearance in the design of the Great Pyramid has been much discussed: for example, see Petrie, TPTG, p. 183. KC is 2.5 units and RC is 2 /11 units. Although a perpendicular line from R to BC has 56 not been drawn to minimize clutter, RK slopes at tangent /17. This leads to a momentous finding.
6

188

f) This material is of major historic interest. RC is 2 /11 units and AR is 1 /11 units, a total of 4 units. The tangent of right-angled triangle DAR is 2.75. This is readily established. T is the middle of AB 3 units, as K is the middle of BC. AT is, therefore, 1.5 units and DT is 5.5 units (AD 4 + AT 1.5); TK is 2 units; DT 5.5 TK 2 = 2.75. Triangle DAR clearly has the same proportions as triangle DTK. RC can be expressed as /11 units and EC as /11 units. Instantly the numbers associated with the Great Pyramid arise: the height of the structure was originally 280 royal cubits and the base was 440 royal cubits square (Petrie, TPTG, p. 183). The first true pyramid constructed was at Meydum, about 40 miles south of the Great Pyramid. It originally had a base 275 royal cubits square and a height of 175 royal cubits (Legon, Meydum, pp. 189): AD 4 units AR 1
5 28 44

/11 units = 2.75.

RC 2

/11 units AR 1 5/11 units = 1.75.

Two pyramids in one design. It is possible to understand now why Khephrens pyramid, the largest 3:4:5-proportion structure ever built, has been located next to the Great Pyramid. There are other remarkable mathematical reasons as well and they are explicated in Number and Divinity in Antiquity. Something else of interest: AP is 0.84 unit and DP is 4.84 units; 4.84 is 2.2 squared. Since the tangent of right-angled triangle, DPU, like DTK, is 2.75 then PU must be 1.76 units. VU is 2.2 units and PV is 1.32 units. VPU is yet another 3:4:5-proportion triangle. The sum of its sides is 5.28 units. A mile contains 5280 feet or 1760 yards. Furthermore, the perimeter of the base of the Great Pyramid originally measured 1760 royal cubits, its half base 220 royal cubits. TP is 0.66 unit, so is TV: a chain in 66 feet and a furlong is 660 feet. The sum of VU 2.2 units and PU 1.76 units is 3.96 units which is half 7.92 units; a British imperial link measures 7.92 inches and there are 100 links in a 22-yard chain. The yard, as stated earlier, is 36 inches or 3 feet. It is possible to see quite clearly the birthplace of the so-called British imperial system of linear measures in this geometry. Even the number associated with the number of inches in a mile, 63,360, can be found; it is explicated in further developments that are at the far edge of the human imagination in Number and Divinity in Antiquity. The royal cubit is one of the most famous and controversial measures of history. The controversy usually concerns its origin. The matter can now be settled. It derives from the geometry presently under examination. W. M. F. Petrie thought the best value was one ascertained from the dimensions of the Kings Chamber in the Great Pyramid. He estimated the measure, which has 28 digit 189

divisions, to be 20.6200.005 inches (Petrie TPTG, p. 179). The acceptable range is, therefore, 20.615 to 20.625 inches. The construction of the small 3:4:5-proportion triangle hGb was described in 4.5.1. 165 (point iii). Gh measures /224 unit. Twenty-eight times 165/224 produces 20.625. As inches, this is the intended value for the royal cubit. The measure has a clear 22 geometric link to the design for the Great Pyramid as well as the /7 proportion. The quotient of AB 3 units divided by AR 1
5

/11 units is 2.0625.

LK is 1.875 units. If LK were extended to GF, the base of the five square, the line would measure 6.875 units; 6.875 is one-third of 20.625. In Number and Divinity in Antiquity, the measure as a whole, along with its digit division (not the one shown here), is created in this geometry in a more startling fashion. g) BP is 2.16 units. The real measure of the world in antiquity was 216,000 stades. The 3:4:5-proportion triangle PAN has sides that measure PN 0.63 unit, AP 0.84 unit, and AN 1.05 units, a total of 2.52 units. Eratosthenes fake measure of the world was 252,000 stades. Here we see an analogue of the 216/252 connection in the pentagram described in 3.5.1. WZ measures 0.252 unit. The ratio 216:252 can be expressed as 6:7. Compare this with the opening chapters of the Old Testament in the Bible, Genesis I and II: God created the world in six days and rested on the seventh. h) Right-angled triangle ODB has the proportions 7:24:25. OD measures 2 /24 units, 7 7 BD 7 units and BO 7 /24 units (7 /24 appears on a calculator display screen as 7.291666). BO, as inches, is exactly one-thousandth the measure of the stade (i.e. 7291 /3 inches), 7 7 as it was originally understood to be. Two times 7 /24 inches is 14 /12 inches (14.58333). The ancient Egyptian remen has long been estimated to measure around 14.58 inches (Zupko, p. 6 and Petrie, MW, p. 5). Recall the appearance of the stade in the Giza pyramid layout described in 3.5.2. The axial distance from the west side of the pyramid of Menkaure to the west side of the pyramid of Khephren is 7289.5 inches. The rate of accuracy in layout intention is around 99.97 percent. The ratio between the 20.625-inch royal cubit and the 14 /12inch remen is 99:70, a ratio that closely approximates the irrational number 2. (A4-size paper, which measures 297 mm by 210 mm, expresses this ratio.)
7 2 1

190

i) The stade of 7291 /3 inches when divided by 600 yields the Greek/Parthenon foot of 11 12 /72 (12.152777) inches: cf. the historic estimates of 12.15 inches (Zupko, p. 6 and Petrie, MW, p. 5). In Number and Divinity in Antiquity, it is shown how the 11 measure of 12 /72 inches is generated naturally in the geometry illustrated in figure 16, along with the royal and common cubits, the remen, the Greek cubit, and the Roman foot and cubit. The geometry even generates their digit divisions naturally. The socalled British imperial foot and yard are readily seen: refer to point (a). j) The ancient measure of the world can now be accurately fixed: 216,000 stades 2 21 multiplied by 7291 /3 inches produces 1,575,000,000 inches or 24,857 /22 (24,857.95454) miles: cf. the modern polar circumference measure of 24,859.82 miles. k) The Romans divided the stade into 625 (25 squared) Roman feet. Accordingly, the Roman foot measured 11 2/3 inches (296.333 mm). The measure of the world was therefore equal to 135,000,000 Roman feet. In figure 16, NM measures 1.35 units. l) BN measures 2.25 (1.5 squared) units: cf. the length of the Parthenon, 225 Greek feet: see 3.5.1. m) Triangle AMB has sides that measure BM 1.8 units, AM 2.4 units and AB 3 units, a total of 7.2 units. Inside AMB is triangle BMN whose sides measure NM 1.35 units, BM 1.8 units and BN 2.25 units, a total of 5.4 units. Compare this with the material on 72 and 54 in 4.2. n) The areas of the three water rings, two ring islands and the central circular island of Atlantis are as follows: Largest water ring: 226 /28 square stades, which is equal to the area of a circle that has a radius of 72 stades; the perimeter measure of triangle AMB is 7.2 units (also see 4.2.). Larger ring island: 169 /14 square stades, which is equal to the area of a circle that has a radius of 54 stades. The perimeter of triangle BMN measures 5.4 units (also see 4.2.). Middle water ring: 81 /7 square stades, which is equal to the area of a circle that has a radius of 26 stades; YI is 2.6 units (MY is 2.4 units, the same as AM; MI is 5 units.). Smaller ring island: 56 /7 square stades, which is equal to the area of a circle that has a radius of 18 stades; BM is 1.8 units. 191
4 5 10 8

Smallest water ring: 18 /7 square stades, which is equal to the area of a circle that has a radius 6 stades; the perimeter of triangle BTK is 6 units (BT 1.5 + TK 2 + BK 2.5). Central island: 19 /14 square stades, which is equal to the area of a circle that has a radius of 2.5 stades: BK is 2.5 units. o) The ring arrangement of Atlantis is 27 stades in diameter. The circumference of 6 Atlantis is therefore 84 /7 stades, which is 618,750 inches or 9.765625 miles; 9.765625 is 3.125 squared. In figure 16, LB and LC each measure 3.125 units. p) In 3.8.1. Herodotus description of the length of the coastline of ancient Egypt, 3600 stades, was examined. In figure 16 BQ is 3.6 units and the perimeter of the entire configuration measures 36 units. In 3.8.2. the historians report on the distance from Heliopolis to Thebes was discussed. From Heliopolis to Thebes is a nine days voyage up the Nile, a distance of eight-one schoeni or 4860 stades. (Marincola, p. 89/H2.9) 9 days voyage (216 hours): BZ measures 0.9 unit; 81 shoeni: BX measures 0.81 unit; 4860 stades: XW measures 0.486 unit; 216,000 stades, the measure of the world: BP measures 2.16, so does AV; 252,000 stades, Eratosthenes fake measure of the world: WZ measures 0.252 unit and the perimeter of triangle APN measures 2.52 units. The geometrical and numerical evidence that this is the source of Herodotus mathematical formulation for the voyage up the Nile is compelling indeed. q) Eratosthenes measure of the world, 252,000 stades, was not chosen for use because of numerical and geometric motives alone: there was a didactic reason as well. In antiquity, the world at the equator turned 216,000 stades every 86,400 seconds (24 11 hours). Every second, at the equator, the world moved 2 /12 (2.91666) stades; 2 11 2 11 /12 is one-quarter of 11 /3, the number of inches in a Roman foot; 2 /12 is one7 175 fifth of 14 /12 ( /12), the number of inches in an ancient Egyptian remen: see point (h) above. 192
9

Now 2 /12 stades is equal to (multiply by 7291 /3 inches) 21,267 13/36 inches. The 1750 latter number is /12 squared: cf. the number of inches in a remen in the preceding 1750 11 paragraph. Furthermore, / inches is equal to 12 /72 BI feet; a Greek/Parthenon 12 11 foot measured 12 /72 inches: see point (i) above. r) AV measures 2.16 (6 x 0.6 x 0.6) units. It is the hypotenuse of 3:4:5-proportion triangle AV. A measures 1.728 units; 1.728 is 1.2 cubed. V measures 1.296 units (3.6 x 0.36); the measure of 216,000 stades is equal to 129,600,000 Greek/Parthenon feet. The number 3600 squared (12,960,000) is discussed in Lees footnote to Platos reference to 3:4:5 proportions, the divine creatures cycle and the number of days in a Great Year (Lee, Republic, p. 299300). The perimeter of triangle AV measures 5.184 units; 5.184 is 7.2 multiplied by 0.72. 4.6. A racecourse to remember. The ancient measure of the world was 216,000 stades. It is equivalent to 24,857 /22 70,000,000 miles. The latter number can be written as /2816. If 22/7 is inverted and 7 dimidiated seven times, the outcome is /2816. All this is worthy of contemplation. 21 Plato has recorded the number 24,857 /22 in a memorable way. In 4.1.2. the following passage from Critias was cited: On the middle of the larger island, in particular, there was a special course for horseracing; its width was a stade and its length that of a complete circuit of the island, which was reserved for it. (Lee, Timaeus and Critias, pp. 1401/S117) Plato has provided an explicit mathematical description of the racecourse. It is one stade wide and it is in the middle of a ring island three stades wide. There is a reason for the formulation: he has created a formidable mathematical conundrum that only advanced students at his Academy or Pythagoreans, or privileged others could understand. (The calculations here use /7 to represent pi. It is easier to do all this on a calculator 22 that has memory facilities. It is also instructive to do it without the calculator. The /7 proportion is encrypted in the Timaeus; the location and method of encryption are explicated in the larger version of this exposition.) a) The area of the racecourse works out to be 56 /7 ( /7) square stades which is 2 equal to (multiply by 7291 /3 inches twice) 3,007,812,500 square inches. 193
4 396 22 21

11

b) Convert the area to square miles by dividing by 63,360 inches twice (63,360 inches is 24,609,375 /32,845,824 square the number of inches in a mile). The area works out to be mile. On a calculator display screen this will appear as 0.74923908 mile. c) Divide the measure of the world 24,857 /22 miles by quotient is 33,177.6, which can be expressed as 24 x 24 x 24 x 2.4.
21 24,609,375

/32,845,824 and the

That Plato or someone else in his Academy conceptualized all this is also worthy of contemplation. 4.6. The Pantheon and a Revelation. Other significant manifestations of the number 216 and thereby the measure of the world can be found in famous architecture and literature. (i) The first example is the Pantheon in Rome, one of the few great buildings of the Roman Empire to survive to modern times basically intact. It is the work of Hadrian, about 126 AD.

Figure 17: The Pantheon in Rome (illustration courtesy M. W. Jones) Sir Mortimer Wheeler writes: 194

It was dedicated to the seven planetary deities and was in effect an architectural simulacrum of the all-containing cosmos. (Wheeler, pp. 1045) Amongst the notable architectural features of this structure, normally the first to be seen by the visitor is the portico: see the illustration above. The width of the portico is 108 Roman feet, that is, about 31.99 metres that is about 1260 inches (Davies, Hemsoll, Wilson Jones, pp. 141 and 150). The number 108 is half 216 which is instantly recognizable for its reference to the measure of the world. Also, compare the measure with the angles of 108 in a pentagon as in figure 7. Furthermore, 108 Roman feet is equal to 86.4 ancient Egyptian remen: the ancient measure of the world was equivalent to 86,400,000 Greek cubits; there are 86,400 seconds in a day. One hundred and eight Roman feet is also equal to 1728 (12 cubed) Roman digits. (A Roman foot had sixteen digit divisions: see Dilke, p. 26.) In figure 16, A measures 1.728 units. (ii) In the Bible, the last book of the New Testament is titled The Revelation of St. John the Divine. It comes down to us in Greek and it is said it may have been composed in Alexandria, in Egypt. The work contains material from a range of sources, some of it ancient Egyptian. Chapter 21, the penultimate chapter, contains numerous mentions of the number twelve. The two most interesting mentions are (1) the visionary New Jerusalem, a giant cube whose length, width and breadth are 12,000 furlongs and (2) a wall 144 cubits high surrounding the city.

Figure 18: a cube. English translators long ago translated stade as furlong because of a vague similarity in length. It is correct to say that the New Jerusalem cube has sides that measure 12,000 stades. This super solid has a surrounding wall 144 (twelve squared) cubits high. The original text was Greek so it may properly be assumed the cubit is the Greek cubit. 195

A cube contains 24 right angles and 24 x 90 = 2160. According to the Pythagoreans, the cube consists of 216 lines: see 3.6. Each face of the New Jerusalem cube contains 144,000,000 square stades and since the object has six faces, the total surface area of the solid is 864,000,000 square stades. A day contains 86,400 seconds; the ancient measure of the world was equivalent to 86,400,000 Greek cubits. The number 864 has been shown to occur in the pentagram (3.3.1.) and the number 8.64 in the designs for the Great Pyramid (3.5.1.) and Atlantis (4.1.1.). A cube has twelve edges. An edge of the New Jerusalem cube measures 12,000 stades. The sum of the twelve edges is 144,000 stades, which is equal to 86,400,000 Greek feet. The wall around the city is 144 Greek cubits high. A cubit (both Greek and Roman) is equivalent to one and a half feet. Accordingly, the wall can be said to be 216 Greek feet high. There can be no doubt as to what the height refers to. More can be found in antiquity in relation to a square with sides that measure 12,000 stades. E. A. Wallis Budge compared ancient Hebrew notions of the Underworld with ancient Egyptian ideas of the Duat (or Tuat), the realm of the dead. The commonest of the names, which the Hebrews gave to the abode of the damned, is G HINNOM, or Gehenna, . According to the Rabbis Gehenna was created on the second day of creation, with the firmament and the angels, and just as there were an Upper and a Lower Paradise so there were also two Gehennas, one in the heavens and one on the earth. As to the size of Gehenna we read that Egypt was 400 parassangs long and 400 parassangs wide, ; that Nubia was sixty times as large as Egypt; that the world was sixty times as large as Nubia, and that it would require 500 years to travel across either its length or its breadth; that Gehenna was sixty times as large as the world; and that it would take a man 2,100 years to reach it. In Gehenna, as in Paradise, there were seven palaces , and the punishments, which were meted out to their inhabitants varied both in kind and in intensity. In each palace there are 6,000 houses, or chambers, and in each house are 6,000 boxes, and in each box are 6,000 vessels fitted with gall. (Budge, Vol. 1, pp. 2734) Gehenna is thus 216,000 times the size of Egypt. The number does not require comment; enough has been said about the measure of the world and 6 x 6 x 6 variants 196

already. The mythical area of Egypt, 400 parasangs square, is worthy of note. Herodotus informs us that a parasang (Budge spells it parassang) is equal to 30 stades (Marincola, p. 88/H2.6). Therefore, Egypt was 12,000 stades square, exactly the same size as each face of the New Jerusalem cube. Budge cites Eisenmenger, Entdecktes Judenthum, part ii, p. 328 as the source of the information on the size of Egypt. 4.7. Closing remarks. Atlantis is found. It rises from the depths of myth and speculation to be seen for what it really is: a powerful mnemonic and didactic device for those initiated into certain influential Mystery Schools in antiquity, notably those with Pythagorean and Neoplatonist interests. A large body of data drawn from authoritative ancient and modern sources, from the layout of the Giza pyramids, and from the properties of certain basic geometric shapes known to Plato, Vitruvius and others in antiquity, has revealed a distinctive approach to mathematical design. Unique and inventive mathematical feats previously unknown from the past have been unveiled. The design source for the Great Pyramid, too, has been located. So has the source for important metrologies of the ancient world, including the so-called British imperial system. Herodotus marvelous mathematical voyage up the Nile has not only been explicated, it has been placed in the geometry set out in figure 16. As has been repeatedly demonstrated, the ancient measure of the world was 216,000 stades. It has also been demonstrated that the measure of the world was linked to the nature of time. In addition, it was shown that the measure of the world was established at least as far back as the time of the construction of the pyramids at Giza. No information has come down from the past, though, as to who it was that realized the world was sphericalprobably some astute astronomer who witnessed a number of lunar eclipses. There is also no record of the survey method or of the survey work that established the necessary distances for the stade and other measures to be created. The survey work was likely to have been a religious duty as the world was considered a deity, just like the sun or the moon. How many times the task was undertaken until consistent results had been achieved cannot be estimated. One last surprise remains.

197

5. Golden apples
5.1. Storing knowledge. From Biblical times, particularly through Christian imagery, the apple has been seen as a symbol of knowledge (cf. the story of Adam and Eve in the garden of Eden). This fruit, a member of the rose family of plants,2 has also played an important role in Greek mythology. One myth is especially relevant to the matter of Atlantis and beyond. It concerns Heracles (Hercules) and one of his Twelve Labours. Using data obtained from ancient sources, Robert Graves, author of The Greek Myths, in the opening paragraph of chapter 133 entitled The Eleventh Labour: The Apples of the Hesperides, writes (bolded words are the present writers emphasis): Heracles had performed these Ten Labours in the space of eight years and one month; but Eurystheus, discounting the Second and the Fifth, set him two more. The Eleventh Labour was to fetch fruit from the golden apple-tree, Mother Earths wedding gift to Hera, with which she had been so delighted that she planted it in her own divine garden. This garden lay on the slopes of Mount Atlas, where the panting chariot-horses of the Sun complete their journey, and where Atlass sheep and cattle, one thousand herds of each, wander over their undisputed pastures. (Graves, Vol. 2, p. 145) What was so special about this fruit in ancient times? The photograph below says it all. When the apple is sliced crosswise rather than downwards from where the stem connects, the core is revealed as Natures pentagram, a geometric shape that encrypts the ancient measure of the world and the way we measure time.

198

Figure 19: Natures pentagram in the core of an apple. Thus, the key elements for the philosophers influential story can be found in an ancient legend that predates Platos lifetime. Plato has already told us that Atlantis is named after Atlas: refer 2.1.1. And we see that the apple-tree is a wedding gift from Mother Earth (Gaea). Who more appropriate? And the apples grow on Mount Atlas: a fitting location. Note, too, the mention of horses and recall Platos course for horse racing (4.1.2. and 4.6.). Recall also one of the possible meanings of Apollos name: appleman (3.8.4.). The apple truly is a symbol of knowledge. Notes 1. Zupko incorrectly calculates the Roman foot to be 11.65 inches and the stade as 625 times this value, that is, 606.9 BI feet. Zupko (correctly) rates the 625 Roman-foot stade as equal to 600 Greek feet like other historians of metrology. But he starts out with a slightly incorrect value for the Roman foot: 296 mm. The links between ancient Egyptian, Greek and Roman measures are well known and much discussed in works on the history of metrology (see Petrie, MW, p. 5, Klein, p. 71 and Zupko, p. 6). The Roman digit was the same as the ancient Egyptian remen digit, around 0.729 inch: see the references just mentioned. The Roman foot of 16 digits was therefore around 11.664 inches or about 296.27 mm. Zupko has simply rounded this number to 296 mm and obtained his value of 11.65 inches. Furthermore, when multiplied by 625 it led to his value for the stade measure being out by some seven inches. The well-known ratio (noted by Zupko) of 25:24 between the Greek foot and the Roman foot further 199

substantiates the corrected measure: the Greek (Parthenon) foot measured around 12.15 inches (Zupko, p. 6 and Petrie, MW, p. 5); twenty-four twenty-fifths of 12.15 inches is 11.664 inches. The so-called Roman foot has been observed in Greece under various guises, including the Nemean foot. In a critique of the book Athletics and Mathematics in Archaic Corinth: the Origins of the Greek Stadion by David Gilman Romano, British academic David W. J. Gill (see References) writes: This provides the information that the Nemea stadion must have been around 178 m long, judging by the location of the 100-foot marker at a distance of 29.63 m from the starting line; this gives a foot of 0.296 m. (Gill, Internet address in References) Note how an error in the measure of the Nemean/Roman foot arises: Gill has rounded 29.63 divided by 100 to 0.296. It is more accurate to say the Roman foot was about 0.2963 m (about 11.665 inches). 2. This should interest members of the Rosicrucian Society. 3. Historians of metrology have noted the use of a range of feet of varying values in ancient Greece. In all important cases, the present writer is able to demonstrate they are proportionally related to the Parthenon foot, the remen and the ancient Egyptian cubits. The matter is dealt with extensively in Number and Divinity in Antiquity, another paper by the present writer. The only Greek foot and cubit discussed in this paper are the so-called Parthenon foot and cubit. 4. A few remarks on several notable Greek feet measures: a) The so-called Doric foot of around 326 mm is simply the Parthenon foot of 12 11/72 (875/72) inches multiplied by 36/35 twice, that is, 36/35 squared. The Doric foots true value is 12 6/7 (90/7) inches (326.57 mm). It has an interesting mathematical function in relation to the measure of the world. As stated in the paper, 216,000 stades is equal to 1,575,000,000 inches (see 4.5.2. point (j). The quotient of the measure of the world, 1,575,000,000 inches, divided by 12 6/7 inches is 122,500,000, which can be expressed as 35 x 3,500,000. b) 1,575,000,000 inches divided by the Parthenon foot of 12 11/72 inches (about 308.6 mm) is 129,600,000, which can be expressed as 36 x 3,600,000.

200

c) The so-called common Greek foot of about 316 mm is more accurately expressed as 12 4/9 (112/9) inches or 316.0888 mm. The quotient of 1,575,000,000 inches divided by 12 4/9 inches is 126,562,500, which is 11,250 squared. d) One last example: the Halieis stade of 166.5 metres produces a Greek foot of around 278 mm. The correct value is 10.9375 inches (277.8125 mm) and is 9/10 of the Parthenon foot or three-quarters of an ancient Egyptian remen. The quotient of 1,575,000,000 inches divided by 10.9375 inches is 144,000,000, which is 12,000 squared. e) The following website may help http://ccat.sas.upenn.edu/bmcr/1995/95.09.19.html with these stade issues:

References
Adam, J. (with an introduction by D. A. Rees) The Republic of Plato. Vol. 2. Cambridge University Press, London, 1965. Babbitt, Frank Cole. Plutarch: Moralia. Volume V (Isis and Osiris). Harvard University Press, 1999 edition. Baines, J. & Malek, J. Atlas of Ancient Egypt. Time-Life Books, Amsterdam, 1996 edition. Budge, E. A. Wallis. The Gods of the Egyptians. (two volumes) Dover Publications, N.Y., 1969. Davies, P., Hemsoll, D., Wilson Jones, M. The Pantheon: Triumph of Rome or Triumph of Compromise? Art History. Vol. 10. No. 2, June 1987, ISSN 0141-6790 Dilke, O. A. W. Reading the Past: Mathematics and Measurement. British Museum Publications, London, 1991. Encyclopaedia Britannica (The New) Volume 9, Micropaedia, 15th edition. Encyclopaedia Britannica Inc, Chicago, 1992. Gardiner, A. Egyptian Grammar. Griffith Institute, Ashmolean Museum, Oxford. 1994 edition. Gill, D. W. J. (University of Wales, Swansea) Internet address: http://www.swan.ac.uk/classics/dghp.html . Grant, M. Myths of the Greeks and Romans. Mentor Books, New York, 1962. Graves, R. The Greek Myths. (Vols. 1 & 2) Penguin, England, 1978 edition. Klein, H. A. The World of Measurements. George Allen and Unwin Ltd., London, 1975. Lee, D. Plato: The Republic. Penguin, England, 1987 edition. Lee, D. Plato: Timaeus and Critias. Penguin, England, 1986 edition. Legon, J. A. R. A Ground Plan at Giza. Discussions in Egyptology 30. Oxford, 1988. ISSN 0268-3083. . (title abbreviation AGPG.) Also see the Internet: http://www.legon.demon.co.uk/index.htm . Legon, J. A. R. The 14/11 proportion at Meydum Discussions in Egyptology 17, Oxford, 1990. ISSN 0268-3083. (title abbreviation Meydum) Marincola, J. (revision of translation by Aubrey de Slincourt) Herodotus: The Histories. Penguin Books, 1996 edition. Morgan, M. H. Vitruvius: The Ten Books on Architecture. Harvard University Press, Cambridge, Mass., 1914. (Republished by Dover Publications in 1960.) Petrie, W. M. F. The Pyramids and Temples of Gizeh. Field and Tuer (Ye Leadenhalle Presse), London, 1883 (title abbreviation TPTG). Petrie, W. M. F. Measures and Weights. Methuen & Co. Ltd., London, 1934. (title abbreviation MW)

201

Saunders, T. J. Plato: The Laws. Penguin Books, London, 1975 edition. Tredennick, H. & Tarrant, H. The Last Days of Socrates. Penguin, England, 1993. Wheeler, Sir Mortimer. Roman Art and Architecture. Thames and Hudson, London, 1994 edition. Zupko, R. E. British Weights and Measures: A History from Antiquity to the Seventeenth Century. Madison University of Wisconsin Press, 1977.

202

THE MNEMONICS OF THE CRETAN LABYRINTH

TESSA MORRISON

Name: Tessa Morrison, Ph.D. candidate Address: The School of Fine Arts, The University of Newcastle, University Drive, Callaghan. NSW 2308. Australia. E-mail: c9520975@alinga.newcastle.edu.au Fields of interest: geometric and algebraic topology, group theory and history of ideas Awards: Certificate of Merit, Uiversitet im. Zhaksygarina, Aktobe, Zazakhstan, 1999 Diploma of Honour, Homage to the Poet, Ovidiu Petca. Cluj-Napoca, Romania, 2000 Best Graphic Print Bookplate, Australian Bookplate Design Award Exhibition, 2001 Publications and Exhibitions: The Geometry of History; 032147658. Vismath, http://www.mi.sanu.ac.yu/vismath Volume 3. No. 4. 2001, Het Stedelijk Museum, Sint-Niklaas, Belgium, 2001. Printmakers Perceptions of the Past One Hundred Years of Federation, Maitland City Council Art Gallery, Australia, 2001. Stadtmuseum Bruneck, Brunico, Italy, 2001. Roman Identity and the Roman Labyrinth, Journal of Inter-Cultural Studies, The University of Newcastle, Australia 2002 (forthcoming issue).

Abstract: The aim of this paper is to examine the various possible symbolic mnemonic devices that the Cretan labyrinth can be constructed from. The Cretan labyrinth has been used throughout history, sometimes as graffiti. Yet it has a complex and difficult structure, not a structure that you would expect to find in graffiti that would have been constructed in a hurry. The structure does suggest that it was drawn from the center expanded out form a very simple symbol. This paper concentrates on all the possible different ways of constructing it. These simple mnemonic symbols may explain the longevity and popularity of the complex structure of the Cretan labyrinth.

203

1.

HISTORY OF THE CRETAN LABYRINTH.

Symbols that consist of concentric levels appeared very early in history. Carved into a Neolithic passage grave in Ireland are symbols that comprise of concentric circles and are cut to the center by a line. These symbols are known as Cups and Rings marks. Despite the simplicity of the design these symbol would have been laboriously carved with no more sophisticated tool than a flint axe. This time-consuming process and precision of the carving does indicate that they have a ritual significant. The design of Cups and Rings continued to be carved into rocks well into the Late Bronze Age. Regardless of the painstaking carving of these symbols they can be easily remembered, once seen never forgotten. A ritual of significance would explain this symbols longevity while the simplicity of the symbol would guarantee its exact repetition. The next development of concentric level symbols was labyrinthine symbols. Although the Cups and Ring are not labyrinths, they are possibly an embryonic form of these later labyrinthine symbols. There are many examples of labyrinthine symbols that are dated to the Bronze and Iron Ages. They have been found carved onto rocks in Italy, Spain, England, Ireland and Sardinia (Kern 2000). A consistent structure began to emerge, which has become known as the Cretan labyrinth. The Cretan labyrinth has one and one only path from the outside to the center and it consists of eight concentric levels. Each level runs in an opposite direction to the one before and the level sequence is from the outside to the 3,1,2,4,7,6,5, levels and then into the center (Morrison 2001). Although there are rock carvings of the Cretan labyrinth have been dated to the early Bronze Age, the earliest dated Cretan labyrinth on the basis of archaeological and historical criterion is on a clay tablet from the palace of Nestor, Pylos (see Figure 1).

Figure 1. The Pylos Cretan Labyrinth thirteenth century BC. 204

This tablet was originally unbaked and only survived because of the fire that destroyed the palace. The writing on the other side of the tablet has no relevance to the labyrinth; this has lead to the assumption that this symbol was a doodle by an idle scribe (Heller 1961). However, the other tablets that have been found in the same archive indicate that the scribes had no time for doodles. The tablets show a city at war, fear and confusion reigned, inventories of weapons, troop movements and sacrifices, possibly human, to the gods were recorded (Chadwick 1976). Yet in this confusion a scribe draws this exact and difficult geometrical structure, perhaps a private prayer. This structure does suggest that it was drawn from the center, indicating that a mnemonic device was used. The palace of Pylos did not survive the war and it was leveled to the ground. Its very existence was forgotten until the excavations of Carl Blegen that begun in 1938. The Cretan labyrinth remerges again and again throughout history. In a tomb near the ancient city of Caera north of Rome, a seventh century BC Etruscan wine pitcher was found. On this wine pitcher is a Cretan labyrinth. The Etruscan labyrinth is round, and not square like the Pylos tablet, but they both have the same level sequence (see Fig. 2). The center of the labyrinth has a distinct cross and the placement of the turning points of rows 3 2, 2 1, 5 6, and 6 7 are equally placed in relation to the central cross. The cross and the equal placement of these points are in distinct contract to the rough execution of the rest of the labyrinth. This strongly suggests that here too a mnemonic device was used to draw this labyrinth.

Figure 2. The Etruscan Cretan Labyrinth seventh century BC For over six hundred years from the fifth century BC the labyrinth was used on coins from Crete. The symbol is used on the coins as a symbol of Crete (see Figure 3). Although the earliest literary reference to the term labyrinth is on a small clay tablet found at Knossos, dated 1400BC, no other labyrinth survives on Crete prior to these coins. The tablet has been translated to be One jar or honey to all the gods, one jar of 205

honey to the Mistress of the Labyrinth(Chadwick 1976). However, the tablet gives no clue as to what the labyrinth is or its meaning.

Figure 3. Cretan Coins (a) c190-100BC (b) c200-267BC Graffiti found on the surface of a crimson-painted pillar, in the peristyle of a villa at Pompeii, depicts a Cretan labyrinth (see Figure 4). It was preserved under the lava from Vesuvius in 79AD. The labyrinth is accompanied by an inscription Labyrinth. Here lives the Minotaur. The graffiti was scratched with a nail or stylus; one assumes that it was executed with some speed given the insulting nature of the inscription.

Figure 4. Graffiti for Pompeii 79AD Moreover, the Cretan labyrinth was used in Biblical manuscripts, as a symbol of the fall of the walls of Jericho. Figure 5 shows a page of a ninth century manuscript. Alongside the labyrinth is the inscription Uruem Gericho, a misspelling of Urbem Jercho, City of Jericho. The Cretan labyrinth remains in Biblical manuscripts for at least a thousand years (Kern 2000). This symbolic representation of the Cretan labyrinth has been used unchanged in its structure for thousands of years. The Cretan labyrinth is a complex structure, and it is difficult to draw freehand, yet it is used in graffiti that would have been executed with 206

some speed. The structure does suggest that it was drawn from the center with the assistance of a simple symbol that was expanded out to the more complex labyrinthine symbol. This simple symbol or symbols acted as a mnemonic and kept the Cretan labyrinth alive throughout the millennium.

Figure 5. Cretan labyrinth representing the walls of Jericho in a nine century Biblical manuscript.

2.1 MNEMONIC DEVICES.

Using mnemonic devices in training the memory was common in classical times. In a world devoid of printing and notepaper a highly trained memory was of paramount importance to an orator and rhetoric was an importance part of the classical education. Texts on training the memory for oratory have survived. The earliest treatise, was written c.86-5BC, was by an anonymous writer who came to be known as Ad Herennium, this name came from the dedication. Ad Herennium refers to Greek writings on the art of memory but these accounts have not survived. It is impossible to say how far back these mnemonic systems for an orator go. However, the texts such as Ad Herennium were written as if the knowledge of these mnemonic systems was common place (Yates 1966). In general a mnemonic is a device used to remember something that is otherwise hard to recall in detail. The classical orators used devices than involved memorizing a background and images. The background was like a wax tablet or papyrus and the arrangement of the images on the background was like the script. This type of mnemonic was short term and private. The orator would follow rules given in the rhetoric texts. However, each speech would comprise of different backgrounds and images according 207

to the orators own personal experience. A visual mnemonic for a symbolic depiction would be far more suitable than training the memory. The longevity of the Cretan labyrinth would depend upon its simplicity of the mnemonic and the clarity of the algorithm that explained that mnemonic.

2.2. POSSIBLE MNEMOMIC SYMBOLS FOR THE CONSTRUCTION OF THE CRETAN LABYRINTH.
One symbol and method of construct is continually pointed to as being the easiest way to draw a Cretan labyrinth (Attali 1999; Kern 2000) (see Figure 6). Denote symbol Figure 6.a as M(a). To expand M(a) beginning at the top vertical of the cross then inserting a right angle or arc between the vertical of the cross and vertical of the L-shape on the right hand side (see Figure 6b). Second, begin with the vertical on the left-hand side and follow the path made in the last step and terminate at the dot in the right-hand quadrant (see Figure 6c). The proceeding steps shown in Figure 6 continue to build up the labyrinth by beginning with the dot or line on the left-hand side, leaving the lines that have terminated at the dots. Then traversing the symbol in the same direction and terminating at the first dot or line on the right-hand side, again leaving the lines that have terminated at the dots. Denote this algorithm A(1). The nucleus M(a) to the expansion into the complete labyrinth has been reported to have been a game called Walls of Troy that was well known at the beginning of the twentieth century (Heller 1946) p.133. It is also claimed that this exact game continues to be played in India (Phillips 1992) p.322. However, is M(a) the mnemonic that the labyrinth on the Pylos tablet, the Etruscan wine pitcher or the Pompeii graffiti were drawn from?

208

Figure 6. (a) Is the mnemonic Symbol denote M(a), (b-i) are the expansion A(1) for M(a) The symbol M(a) does date back into history but this does not make it a mnemonic for the Cretan labyrinth. M(a) has been found on Babylonian seals that date back to 2000BC. A pottery shard, dated 604BC, found in Ashkelon in Israel has M(a) as a part of its design (see Figure 7a). The exact dating of the shard, can be established because of the destruction of the city by Nebuchadnezzar, King of Babylon was recorded. The pottery shard is of wild goat style and the wild goat style has lots of box type designs. Figure 7b-d show other wild goat shards found at the same archaeological excavation at Ashkelon. The pattern of M(a) may not have been used as a mnemonic in ancient times since it is not complicated and it is an attractive symmetrical symbol that is easily replicated accidentally. Moreover, M(a) gives the appearance that it was developed as a mnemonic well after the Cretan labyrinth was established.

209

Figure 7(a) the Ashkelon Shard containing M(a). 7(b-d) Examples of Wild Goat Pottery The first noticeable thing about M(a) is that the algorithm to expand it is very simple. The continuous joining of end points in the same direction is very straightforward. However, the L-shapes do appear to be redundant, by removing the L-shapes the steps of the algorithm are reduced. The Cretan labyrinth can be expanded from a simple cross and four dots in the quadrant of the cross, (see Figure 8a), denote this symbol M(b). First, begin with the top on the vertical of the cross moving right traverse the dot, following a path between the central point and the dot. Then return to the left side and terminate at the dot in that quadrant (see Figure 8b). Second, begin at the dot in the upper right quadrant, follow the path made by the last arc then traverse the last terminating point and proceed around to the horizontal of the cross on the right-hand side (see Figure 8c). Third, begin at the horizontal of the cross on the left-hand side follow the outside existing shape, traverse the dot following a path between the central point and the dot on the lower right quadrant. Then return and terminate at the dot on the left-hand quadrant (see Figure 8d). Finally, repeat the second step but begin at the dot on the lower right-hand side (see Figure 8d). M(b) and this simple algorithm complete the Cretan labyrinth in only a few simple steps. Denote the algorithm A(2).

210

Figure 8. (a) Is the mnemonic Symbol denote M(b), (b-e) are the expansion A(3) of M(b). A third possible mnemonic consists of the four L-shapes (see Figure 9a); denote this symbol M(c). The algorithm begins from the vertical line from the upper left-hand side and then arcs over to the right quadrant and terminates in the center of this quadrant (see Figure 9b). Second, begin at the horizontal line from the upper right-hand side return to the left-side traversing the path created in the last step and terminate in upper left quadrant level with the right terminating point (see Figure 9c). Third, begin at the horizontal line from the upper left side; follow the path created in the last step until the horizontal path between the quadrants is reached. Cross that path then continue around the outside of the upper half and terminating at the horizontal line of the lower righthand quadrant (see Figure 9d). Fourth, begin at the horizontal line on the lower left-hand side, traverse the path created in the last step and terminate in the center of the lower right-hand quadrant (see Figure 9e). Fifth, begin at the vertical line on the lower righthand side traverse the path created in the last step and terminate in the center of the lower left-hand quadrant (see Figure 9f). Finally, begin with the vertical of the lower left-hand side traverse the entire symbol until the vertical path between the quadrant is reached proceed through the center and terminate at the vertical of the right-hand side (see Figure 9g). This completes the symbol of the Cretan labyrinth. Denote this algorithm A(3).

211

Figure 9. (a) Is the mnemonic Symbol denote M(c), (b-e) are the expansion A(3) of M(c) The mnemonic symbols M(a), M(b) and M(c) are expanded by different types of algorithm. However, these symbols have variations and use algorithms A(1), A(2) and A(3). The mnemonic Figure 10a, is constructed with nine dots, denote this mnemonic M(d). M(d) is expanded by using A(1). The dots can be reduced to five dots, (see Figure 11a), denote this mnemonic M(e) and expand using A(2). Five dots and a vertical line through the middle also form a mnemonic, (see Figure 11b), denote this mnemonic M(f) and expand using A(3). A cross and eight dots, two in each quadrant (see Figure 11c), denote this mnemonic M(g) and use A(1) to expand it. Also M(c) can include four dots, denote M(h) and using the same algorithm, (see Figure 11d). Finally a vertical line a dot or broken at the center and eight dots four on each side (see Figure 11e), denote this M(i) and this is expanded using A(1). These nine suggested mnemonic devices are variations on the same central region of the Cretan labyrinth. However, the prime 212

purpose of this investigation is to consider what could be the mnemonic that the ancient labyrinths were drawn from and if there were any developments of the mnemonic itself.

Figure 10. (a) Is the mnemonic Symbol denote M(d), (b-e) are the expansion A(1) of M(d)

Figure 11. Further possible mnemonics

3. HOW DRAWN?

WERE

THE

ANCIENT

LABYRINTHS

It would be very satisfying to discover an explanation that would clearly link a mnemonic to the Minoan culture of ancient Crete. Assuming that the Cretan labyrinth was originally of ritual significant, the mnemonic would be more in keeping with this significance if it was based on a Minoan religious object or symbol. The swastika was an 213

ancient symbol well known to the Minoans, it has been found on early Minoan seals and sealings (Evans 1964 Vol, 4. p.570). A swastika can easily be constructed from the mnemonic symbol M(b), (see Figure 12).

Figure 12. Left M(b) and right M(b) expanded into a swastika On close examination of the seventh century Etruscan labyrinth, (see Figure 2) the center of the labyrinth has a distinct cross and the placement of the turning points of rows 3 2, 2 1, 5 6, and 6 7 these are equally placed in relation to the central cross. Looking at the way this labyrinth was drawn the cross and turning point were precisely placed then the walls of the labyrinth were drawn. The execution of the labyrinth is in stark contrast with the center where it was very roughly drawn. The cross and the equally placed turning point are M(b). The curve that traverses the turning point excludes the mnemonics that contains the L-shape and the distinct pre-drawn cross in the central exclude all those mnemonics without a cross in the center. The only other possibility is M(g), however it would be more suited to a square labyrinth, rather than the inaccurate curves that traverse the terminating points of the Etruscan labyrinth.

Figure 13. (a) The labyrinth in the Cathedral (b) the center of the labyrinth with the swastika highlighted. 214

The swastika does have a connection with a labyrinth in Algiers. A mosaic floor labyrinth now in the Cathedral of Algiers is the oldest surviving floor labyrinth in a church. It was originally at the Basilica of Reparata, which was founded in 324AD in Al-Asnam Orleansville and the labyrinth is dated approximately the same age. The labyrinth is a standard Roman layout of four quadrants, and in the center is a box that contains a matrix of letters, thirteen across and thirteen down (see Figure 13a). The center letter is S on close inspection there is an inscription. From the center letter S reads Sancta Eclesia, Holy Church, in all four directions forming a swastika, (see Figure 13b). The letters in the matrix are not well lined up so that the swastika is not visibly clear immediately. Moreover, the swastika was relatively unknown in North Africa and perhaps this is indicative of a strong Roman influence in the design and also in its meaning. The connection between the labyrinth and the swastika is obscure. Nevertheless, the inscription in the shape of the swastika in the center of the labyrinth in Algiers does have a sacred meaning. It is generally accepted that the origins of the word labyrinth is labyrinthos - laburifos and this has a direct association with an ancient cult symbol the double axe labrys labrud (Evans 1964 Vol. 3. p.283). The symbol of the double axe is carved into the walls of Knossos, it is on Minoan pottery and seals, in frescos, painted on sarcophaguses and it is a character in the writing of the Minoans and Mycenaeans. There were sacred knots in the shape of the double axe (Evans 1964 Vol. 1. p.430). Caves were used as religious sanctuaries; here votive objects were left including small minute double axes (Hogarth 1900). The double axe dates back to the borders of the Neolithic Age (Evans 1964 Vol. 1. p.57) and some of the mnemonics are in the shape of a double axe.

Figure 14. Mnemonics that are in the shape of the double axe In Figure 14 there are four variations of mnemonics that have the shape of the double axe, Figure 14a and 14b have the shape of the double blades while 14c and 14d have the added vertical, the shaft of the axe. While the vertical line is the only difference between the mnemonics 14a 14c and 14b 14d it does represent a visual difference in the flow of the finished drawing of the labyrinth.

215

Figures 14b and 14d are more suitable for a circular labyrinth, because the inner dots in the quadrants restrict the flow of the curve around the turning points. Nevertheless, it does not exclude them. Any of these four mnemonics in Figure 14 could have been the bases of the square Pylos labyrinth, which was drawn in an era when the symbol of the double axe appears to have been the most revered religious symbol. However, there is something more satisfying about the mnemonic 14c as it appears more complete as an image of the double axe.

4. CONCLUSION
The ancient rock carvings of the Cretan labyrinth are all circular. They are very precisely carved and this precision leaves no hint as to how they were laid out before carving. Furthermore, in most cases the dating has been inadequate. Many have been dated to the early Bronze Age, purely on the evidence of the symbol itself. An example of this inadequate dating was the Hollywood Stone. The Hollywood Stone was found in Ireland it is inscribed with the Cretan labyrinth. It was originally dated to the early Bronze Age, 2500-2000BC; however, it is now believed to be connected with the early Celtic Church. The Cretan coins with the Cretan labyrinth are also too accurately executed to reveal their pattern layout. In fact, the overall shape of the labyrinths, on the coins are a square, rectangle or a circle while the graffiti and the drawing of the Etruscan pitcher has a very defined raised entrance on one side. It is in the graffiti, of the Pylos tablet and on the column of Pompeii and the rough drawing on the Etruscan wine pitcher that hint at clues to their construction. The graffiti from Pompeii has a distinct cross in the center, the turning points are equally placed around this cross and vertical of the L-shapes are level with the opposite side. The center has an extra heel to it, which make the labyrinth more squat. Nevertheless the equally placed turning points and vertical of the L-shape around a precise cross suggest that M(a) was used in the center. This is in complete contrast with the Etruscan labyrinth where the orderly cross and the equally placed turning point are well defined but the rough drawing and placement of the curve that traverses the turning points indicates that the mnemonic M(b) was used and it was expanded by A(2). The swastika can also easily be constructed from M(b) and the swastika is a symbol that is found on Etruscan funeral urns dating to the same time that the labyrinth (Figure 2) and this wine pitcher was found in a tomb. Although the swastika was a religious symbol of the Minoans and Mycenaeans, the symbol of the double axe permeates both cultures. The Pylos labyrinth does have a distinct cross, the turning points are equally place around the central point yet the verticals that traverse the turning points are not at all level, this is particularly evident at 216

the verticals closest to the entrance. However, central endpoints of these verticals are equally placed around the central point or cross. This indicates three possible mnemonic M(d), M(g) and M(i) and using A(1) to expand. The cross would be formed by the application of the algorithm and on a square labyrinth it would be impossible to say which was used by visual evidence. Nevertheless, the deep religious significance of the double axe would indicate M(d) and M(i) are the stronger contenders as being the mnemonic device for the Pylos labyrinth. Also the general accepted etymology of the word labyrinth is directly related the cult symbol the double axe labrys and it would be difficult to believe that the words layrinthos and labrys are not cognate. Over thousands of years the Cretan labyrinth has remain unchanged in it structure, sometimes round, sometime square yet the level sequence remains the same. This static symbol has crossed time and cultural diversity. However, by examining the possible different mnemonics and algorithms it become evident that what appears to be static on the surface was undergoing continuous change. The continuous evolution of the mnemonics of the labyrinth emphasizes the constant change in its meaning and its religious significance as compared to the meaningless and rather pointless game it has now become.

References
Attali, J. (1999). The Labyrinth in Culture and Society. Berkeley, North Atlantic Books. Chadwick, J. (1976). The Mycenaean World. Cambridge, Cambridge University. Evans, A. (1964). The Palace of Minos at Knossos, Volume 3. New York, Biblo and Tannen. Evans, A. (1964). The Palace of Minos at Knossos, Volume 4. New York, Biblo and Tannen. Evans, A. (1964). The Palace of Minos at Knossos, Volume I. New York, Biblo and Tannen. Heller, J. L. (1946). Labyrinth or Troy Town? The Classical Journal Vol. 42: p.175-91. Heller, J. L. (1961). A Labyrinth from Pylos. American Journal of Archaeology Vol. 65: p.57-65. Hogarth, D. G. (1900). The Dictaean Cave. The Annual of the British School at Athens Vol. VI: p. 94-116. Kern, H. (2000). Through the Labyrinth. Munich, Prestel. Morrison, T. (2001). The Geometry of History: 03217658. Vismath Vol 3(4): http://www.mi.sanu.ac.yu/vismath/ Phillips, A. (1992). The Topology of Roman Mosaic Mazes. Leonardo Vol. 25(3/4): p.321-329. Yates, F. A. (1966). The Art of Memory. London, Routledge and Kegan Paul.

217

218

THE PARTHENON HEIGHT MEASUREMENTS: THE PARTHENON SCALE WITH ROOTS OF 2

A. M. BULCKENS

Name: Anne M. Bulckens, Ph.D. (Architecture), (b. Herentals, Belgium). Address: Bulckens, Jalan Intan Ujung 75, Jakarta Selatan, 12430, Indonesia E-mail: annepaul@cbn.net.id Fields of interest: Classical, Renaissance and Baroque Art and Architecture Publications: Bulckens, A.M. (1998) Did Plato Ever Refer to a Section in Extreme and Mean Ratio in his Writings?, Journal of Transfigural Mathematics, 3, no. 2, 27-31 and no. 3, 23-31. Bulckens, A.M. and Shakunle, L. (1998) Logic Numbers and the music of Greek Architecture, Journal of Transfigural Mathematics, 4, no.1, 21-47, 53-57. Bulckens, A.M. (1999) The Parthenons Main Design Proportion and its Meaning. [Ph.D. Dissertation], Geelong: Deakin University, 269 pp. Bulckens, A.M. (2001) The Parthenons Symmetry, in Symmetry: Art and Science, (Fifth Interdisciplinary Symmetry Congress and Exhibition of the ISIS-Symmetry, Sydney, July 8-14, 2001), no. 1-2, 38-41.

Abstract: The Parthenon (447 - 438 BC) epitomises the glory of Athens at the height of Classicism. Although enticingly forming the subject of numerous works of research, the understanding of several aspects of this temple to Athena still remains tentative. Many scholars adhere to the view that the Parthenon was designed in accordance with a simple 4:9 ratio. The dissertation, The Parthenons Main Design Proportion and its Meaning, offers new insights. Through an analysis of the specific dimensions of the temples building parts, this thesis establishes the measurement of a Parthenon Module and Parthenon foot (with 16 Dactyls). It then explains the main design proportion of the temple, expanding the idea that the Parthenon was built with a 4:9 ratio The measurements and design of the Parthenons building elements exalt a 4:6 = 6:9 geometric proportion. This proportion involves the ratio 2/3, which is the perfect fifth in the Pythagorean tetrachord. It is proposed here that the design of the Parthenon, in regard to its height measurements, was based on a Dorian musical scale, which the architects altered in order to incorporate simple ratios representing the first roots of 2. In the next part of the article, not presented here, it will be shown how the Parthenon design exemplifies these ratios (that closely approximate roots of 2 values) two219

dimensionally, as well as three-dimensionally. Since it will then be revealed how the subtle design of the Parthenon houses two cubes, with the large cube having double the volume of the small cube, it then becomes clear that the Parthenon offers an answer to 'the Delian riddle of the cube, which at the time was a hot mathematical issue concerning 3 2.

1. THE HEIGHT PARTHENON

MEASUREMENTS

OF

THE

In the faade and flank, a total of twelve different height measurements are identified as having importance. These measurements are listed below, and they are shown in figures 1 and 2. All measurements are based on the dimensions provided by Penrose [7], since it is well known that he was the leading archaeologist who measured the Parthenon in the most accurate and thorough manner. Penroses measurements were then translated into a system of Parthenon Feet consisting of 16 Dactyls (D), whereby each Dactyl measures 21.44 mm and each Module measures 40 Dactyls [See Bulckens, dissertation, 1 and article, 2]. 320 D = at the corner of the faade, acroterium + entire pediment (i.e. pediment face + raking cornice + rise in curvature of the horizontal cornice) 440 D = corner column minus the stylobate rise along the flank, minus the capital 480 D = column proper = central column height minus the stylobate rise along the flank 512 D = in the centre of the faade, upper step + column 560 D = in the center of the faade, 3 steps + column 623 1/3 D = at the corner of the faade, 3 steps + column + architrave 640 D = at the corner and also in the center of the faade, column + entablature 666 2/3 D = at the corner of the faade, 2 steps + column + architrave + frieze + its rise in curvature 720 D = in the center of the flank, 80 Dactyls of steps + column + entablature 800 D = in the center of the faade, column + entablature + pediment face 840 D = from the corner of the stylobate, column + entablature + pediment 960 D = from the corner of the stylobate, column + entablature + pediment + acroterium Only two measurements are no integers: measurements 623 1/3 D and 666 2/3 D. The fractions of these measurements indicate that the Dactyls are divided further in thirds of Dactyls. As will be explained in this article, these two measurements were incorporated to bring out root of 2 values. 220

Clearly, all other Dactyl measurements are integers, yet all these Dactyl measurements are very accurate conversions of Penroses dimensions into these Dactyl measurements, working well within a margin of 0.15 %. (It is noted that the 320 D measurement comprises the 200 Dactyl pediment (0.03 % off Penroses dimension) plus the height of the acroterium that does not survive. Likewise, the 960 D measurement constitutes the 840 D measurement (0.03 % off) plus the same acroterium. An acroterium height of 120 Dactyl is a valid supposition; it is based on drawings by Orlandos.) Surmising, then, that these accurate integers indicate that the Parthenon designers did their utmost to obtain ideal measurements, it will be explored which design intentions underlie these ideal measurements. It might well be that it is the Athenian conviction that they were especially blessed by the gods, hence superior of all others, which led the Athenians to demonstrate that this was indeed so. The astonishing level of perfection achieved by the architects and builders, as is evident in the Parthenons design and construction, puts the self-image of Periclean Athens on show for the whole world to admire.

221

Figure 1: Parthenon diagram of the north-east corner and center of the faade, with curvature and steps exaggerated; drawing not to scale; dimensions indicated to the right of their dimension line.

One of the key measurements is the 480 Dactyl height of the column proper, which is the height of the central column of the faade minus the stylobate rise in curvature along the flank. (The stylobate is the temple platform upon which the columns stand. As with all horozontal elements of the Parthenon, it is curved.) This measurement cannot be seen, but it will be explained how it plays a crucial role. Among the more tangible measurements at the center of the faade (shown in figure 1) are the 512 Dactyl measurement comprising the upper step plus the central column (25 2/3 D + 486 1/3 D), and the 560 Dactyl measurement consisting of the three steps and the central column (73 2/3 D + 486 1/3 D). Figure 1 shows also the 640 Dactyl measurement running from the stylobate center to the top of the horizontal cornice (central column + entablature, 486 1/3 222

D + 153 2/3 D), and the 800 Dactyl measurement running from the stylobate center to the top of the pediment face, excluding the perpendicular height of the raking cornice (central column + entablature + pediment face, 486 1/3 D + 153 2/3 D + 160 D).

Figure 2: Parthenon diagram of the north-east corner and center of the north flank, with curvature and steps exaggerated; drawing not to scale; measurements of the Parthenon scale in bold color.

In the center of the flank, shown in figure 2, is the 720 Dactyl measurement, which goes from the very bottom of the steps to the top of the horizontal cornice, including its rise in curvature along the flank (steps + stylobate curvature at the flank + central column + entablature, 73 2/3 D + 6 1/3 D + 486 1/3 D + 153 2/3 D). The 640 Dactyl measurement of the column and the entablature can also be taken at the corner (corner column + entablature, 486 2/3 D + 153 1/3 D). At this corner, the 623 1/3 Dactyl measurement (shown in figure 1), runs from the bottom of the steps to the top of the architrave (three 223

steps + corner column + architrave, 73 2/3 D + 486 2/3 D + 63 D). And the 666 2/3 Dactyl measurement (shown in figure 2) starts at the top of the lowest step and reaches the top of the frieze, including the friezes rise in curvature at the flank (second step + upper step + corner column + architrave + frieze + the friezes rise in curvature at the flank, 24 D + 25 2/3 D + 486 2/3 D + 63 D + 63 D + 4 1/3 D). Lastly, the height from the corner of the stylobate to the top of the pediment is 840 Dactyls (corner column + entablature + pediment, 486 2/3 D + 153 1/3 D + 200 D), and to crown the temple, the acroterium height is added to this measurement, resulting in an overall temple height of 960 D.

2. TWELVE LARGE MEASUREMENTS DETERMINING THE HEIGHT OF ALL THE BUILDING ELEMENTS.
In the extended article, it will be explained in detail how these twelve measurements determine the height of all the building elements of the faade and flank: the three steps, the column and its capital, the entablature with its frieze, architrave and horizontal cornice, the pediment with its pediment face and raking cornice, and the acroterium. It is proposed that these measurements determine the heights of the building elements in a peculiar way: when a large measurement is subtracted from a related larger measurement, the height of the building element in between these two measurements is obtained. And it is precisely the dimensions of the rise in curvature of the horizontal elements, such as the stylobate and the entablature, which facilitate ideal dimensions and simple ratios between large measurements. The reason for this method will become clear in later paragraphs, but the method itself is easy to understand when giving a few examples. Consider measurement 800 and measurement 640, which are both taken at the center of the faade, starting from the top of the stylobate (the curved platform upon which the columns stand). Take the largest measurement of 800 Dactyls, and notice that it goes to the top of the pediment face. Subtract from this the 640 Dactyl measurement, which only reaches the top of the entablature, including the rise in curvature of its highest element, the horizontal cornice. What is left is a 160 Dactyl height. This is the height that results from the subtraction (800 640), and it determines the height of the pediment face because this is the building element between the two large measurements. Equally important, the equation 800 x 4/5 = 640, establishes the relationship, i.e. the ratio, between the two large measurements. Indeed, with this method of subtraction between large measurements, the height of all the building elements will eventually be determined, while simple ratios underlie the relationship between the large measurements. Continue the same sequence and procedure with the 640 Dactyl height running from the center of the stylobate to the top of the entablature. Subtract from this the 486 1/3 Dactyl measurement comprising the central column. The result is the height of the 224

entablature: it measures 153 2/3 Dactyls. Interestingly, as explained on previous occasions [1], it is the temple refinements, such as the rise in curvature of building elements, which facilitate a system of simple ratios between measurements. This is also the case with the height measurements, of which 640 cannot be readily multiplied by a simple ratio in order to obtain 486 1/3. Instead, it is necessary to incorporate the stylobate rise, in this case along the flank, to understand the measurement. The rise in curvature along the flank is 6 1/3 Dactyls. The column proper, i.e. the height of the central column minus the rise in curvature along the flank, measures 480 Dactyls. When the 486 1/3 Dactyl height of the column is thus split up in the 6 1/3 Dactyl height of the stylobate rise along the flank and the 480 Dactyl height of the column proper, the result is that: (640 x 3/4 ) = 480. Thus, when 640 is multiplied by a simple 3/4 ratio, the column proper is the result. And to obtain this column proper, a stylobate rise along the flank of 6 1/3 Dactyls was established and then subtracted from the column height. In this fashion, the height of all the building elements will eventually be determined in the extended article. It will also be shown that there is a main sequence of measurements, which are related to each other by a continuous series of the ratios 1/2, 2/3, 3/4, 4/5, 5/6. Consider the next sequence of measurements, which starts with the 480 that originates from the first sequence, and compare it to measurement 512. The 512 Dactyl measurement, taken in the center of the faade, is the height of the upper step plus the central column. Take the 512 Dactyl measurement and subtract from this the 480 Dactyl column proper, which is the height of the central column minus the 6 1/3 Dactyl stylobate rise in curvature along the flank. The result is 32 Dactyls. Hence, the height of the upper step is (32 6 1/3) Dactyls, which is 25 2/3 Dactyls. Again, a simple ratio forms the basis, since 512 x 15/16 = 480. And by subtracting the stylobate rise along the flank, I was able to determine the height of the upper step. I propose that the temple was not only built with tremendous care, but that the architects already designed the temple adhering to this method, which allows working with a very small margin of error, always below 0.2 %. I propose that this is one of the reasons why the architects determined heights of building elements by taking simple ratios of larger measurements, while equating the left over, i.e. the subtraction result, with the height of the building element. This way, the architects could decide that, for example, 480 x 16/15 = 512, so that with this method employing 480, and a simple ratio, and 512, all works out correctly. On the other hand, if they would equate the height of the upper step, which in reality measures 25 3/4 Dactyls, with 25 2/3 Dactyls, the height of the step is off by a margin of error of about 0.3 %. This example shows the typical order of error of the building heights of the smaller elements, and what it makes clear is that: if it would solely be the height of each building element from which design ratios would be issued, the architects would not have enough leeway in order to form a coherent and logic design with simple ratios. When trying to find accurate simple ratios for the height of the upper step, the architrave, the frieze and the horizontal cornice, it becomes clear that there are no such accurate relationships within the design of the faade or flank. 225

Moreover, small measurements easily produce larger margins of error. On the other hand, if the architects worked with the large measurements as the ratio generators, and if they employed the increments of the rises in curvature in various ways, they achieved the necessary leeway to make an accurate, coherent and complete design from simple ratios. In the extended article, the heights of all the building elements of the faade are established by working with twelve large measurements, simple ratios and the rises in curvature. The only temple height not yet determined by these twelve large measurements is the height of the 2 steps of the cella. Beyond the faade, in the center of the cella, a 600 Dactyl measurement goes from the cella floor to the top of the entablature, excluding the entablatures rises in curvature. So, here is one more measurement that will be considered. Comparing the 640 Dactyl measurement taken at the corner, and the 600 Dactyl measurement taken in the center of the cella, the equation is 640 x 15/16 = 600, (This equation employs the same ratio as equation 480 x 16/15 = 512.) The difference is 1 Module, which comprises the height of the 2 cella steps and the rise of the upper floor in the 2 directions of the temple. (In fact, there is also a 600 Dactyl measurement, with 599 1/3 being 0.11 % off 600, at the corner of the faade going from the second step to the top of the architrave.)

3. THE PARTHENON SCALE WITH ROOTS OF 2

When the 600 Dactyl measurement is added to the height measurements, one gets a list of all the large measurements needed to obtain the heights of the main building elements of the temple. Measurement 512 can even be omitted, since it is possible, however cumbersome, to arrive at all building heights without this measurement. Thus measurement 512 is replaced by measurement 600, which also involves a ratio of 15/16 (the ratio of an enlarged half tone in music). (Concerning the ratios, the simplest route was taken to arrive at the heights of all building elements. Once many of the heights of building elements were determined, there were alternative routes on several occasions, which were not explained. These alternative routes provide the remaining ratios found between the heights in the Parthenon scale shown in figure 3.) When these twelve large height measurements are arranged in a linear scale as in figure 3, the temple heights reveal an elegant sequence of the ratios of the first ten integers, 10/9, 9/8, 8/7, 7/6, 6/5, 5/4, 4/3, 3/2, 2/1. Furthermore when 666 2/3 D is set as 100, it becomes clear that most of the heights are the same numbers as the tones of the basic HinduGreek musical scale [McClain, 5], later known as the reverse Dorian scale and currently the common scale: do re mi fa sol la ti do [Kappraff, 3]. This Dorian scale is: 72 80 90 96 108 120 135 144 C D E F G A B C 226

Basically, to arrive from the reverse Dorian musical scale to the Parthenon scale, 80 is enlarged to 84 and 135 is reduced to 126. The result is that the intervals from C to D and B to C are enlarged whole tones, 7/6 and 8/7. When next 93 1/2 and 100 are inserted, the Parthenon scale is obtained. Interestingly, according to Kappraff, a musician-mathematician, 100 falls on Fsharp [Kappraff, 3]. 72 84 90 93 C D E 96 (100) 108 120 126 144 F Fsharp G A B C

In figure 3, when 666 2/3 Dactyls are normalized to hundred, all Dactyl measurements are divided by 6 2/3, so that all the numbers that are the same as the ones from the Dorian scale result in integers, while their ratios become readily readable. As in the Pythagorean tetrachord, 72108 and 96144 form perfect fifths, while 7296 and 108144 form perfect fourths, and 640720 forms a whole tone. Surprisingly, 84126 is a 3/2 ratio also. When this ratio is split up to reach the 100 in between 84 and 126, the fourth root of 2 (which here is 100/84) and the cube root of two are obtained. In this case, the cube root of 2 is the old Babylonian 126/100. (According to McClain [6], the old Babylonian cube root of 2 was regarded as 125/100 + 1/100.) In all, the only numbers that differ from the Dorian scale, are 126 and 84, while 93 and 100 are inserted. These are the tones, which create ratios that closely approximate the roots of 2. So, by altering the Dorian musical scale in a unique manner, the Parthenon scale is created to provide the first roots of 2. The column height is crucial in the Parthenon scale. The 480 D height of the column proper forms the beginning of a musical octave. Its double is the overall temple height, and this is exactly where an octave ends. Also, the column height assists in providing the measurements 623 1/3 and 440. The ratio between these two measurements is 17/12, which is a Pythagorean approximation of 2. [Lasserre, 4, and Kappraff, 3].

227

Figure 3: TheParthenon scale; the 480 600 640 720 800 960 is the reverse Dorian scale, the ratio between 840 and 666 2/3 represents 32; the ratio between 666 2/3 and 560 represents 42; and the ratio between 623 1/3 and 440 represents 2.

In conclusion, by creating the Parthenon scale, simple ratios are revealed between numbers, of which 72 96 108 120 144 are exactly the same as the tones in the reverse Dorian scale. There are more similarities between the Parthenon scale and musical scales; In Pythagorean musical scales, the tetrachord is the framework and the ratios between other tones could vary, depending on the mode that was adhered to. With these other tones in between, there came Pythagorean commas, Pythagorean leimnas, etc, which in essence are left-overs that had to cover the distance to reach the octave. This is also the case in the height measurements of the Parthenon scale. The height of some building elements are literally what is left over after subtraction between large measurements. The difference between historic musical scales and the Parthenon scale is that the Parthenon scale incorporates ratios, which turn out to be excellent approximations for the roots of 2, whereas in the Pythagorean musical scales, 2 is the geometric mean of the octave, but it could never be reached by a coherent system of simple ratios. (The 126/100 ratio is less than 0.1 % off the real value of 3 2, while the 17/12 ratio is 0.17 % off the real value of 2.) 228

It needs to be stressed that the Parthenon scale was not made out of the blues! On the contrary, every single number of the scale accurately refers to a height measurement of the Parthenon. And as it is shown at length that the height of each building element could be determined by these twelve measurements and their interrelated ratios, I suggest that the architects had exactly the same measurements and ratios in mind in order to design the heights of all the building elements of the faade and flank. This article shows how the Parthenon design is rooted in Pythagorean music, whereby Kappraff and McClain contributed greatly to the understanding of ancient musical scales and theory. Concerning the Parthenon, the design issue was to marry a set of measurements exalting roots of 2 with a set of measurements, which were known to form part of the Dorian musical scale. When the 666 2/3 Dactyl measurement was normalized as 100, it provided the means to merge the two sets together in such a handsome way, that even the complete sequence of the ratios 10/9, 9/8, 8/7,, 2/1 was the result. In the next article, then, it will be explained why and how the architects incorporated the roots of 2, especially 32, in the Parthenons plan and elevations.

References
Bulckens, A..M. (1999) The Parthenons Main Design Proportion and its Meaning. [Ph.D. Dissertation], Geelong: Deakin University, 269 pp. Bulckens, A..M. (2001) The Parthenons Symmetry, in Symmetry: Art and Science, (Fifth Interdisciplinary Symmetry Congress and Exhibition of the ISIS-Symmetry, Sydney, July 8-14, 2001), no. 1-2, 38-41. Kappraff, J. (2002) Anne Bulckens System of Proportions of the Parthenon and its Meaning. A presentation at the Carleton University School of Architecture, Ottawa. Lasserre, F. (1964) The Birth of Mathematics in the Age of Plato. London: Hutchinson. According to Lasserre, Plato used the equation x 2y = 1, to find ratios between integers x and y that closely approximate 2. He further asserts that Plato, when using this equation, was Pythagorising. Since 17 x 17 2 x 12 x 12 = 1, it follows that 17/12 approximates 2, and thus according to Lasserre, the Pythagoreans knew this. According to Kappraff, the ratio 17/12 was also used in Roman architecture (Ostia), in order to approximate 2. McClain, E. (1976) The Myth of Invariance: The Origin of the Gods: Mathematics and Music from the Rg Veda to Plato. York Beach: Nicolas-Hays. McClain, E.(1997) The 'Star of David' as Jewish Harmonical Metaphor, International Journal of Musicology, no. 6, 43-47. Penrose, F.C. (1888) An investigation of the Principles of Athenian Architecture. 2nd ed., Washington, D.C: McGrath.

229

230

ARTISTIC DESIGNS BY MEANS OF ALGEBRAIC STRUCTURES

Ruiz, F. and Peas, M., Department of Didactic of Mathematics, University of Granada

Names: Francisco Ruiz and Mara Peas Address: Department of Didactic of Mathematics. University of Granada, Campus de Cartuja. 18071 Granada (Spain) tel. (34)958243951 fax: (34)958246359. Email: fcoruiz@ugr.es and mtroyano@ugr.es Fields of interest: Geometry, representations, numerical thinking, mathematics and art. Publications and/or Exhibitions: Ruiz, F.: Organizer of the exhibitions: - "M. C. Escher, entre la Geometra y el Arte" (M.C. Escher between Geometry and Art). University of Granada and Cordon Art. Granada. May 1990. - "M. C. Escher". Museum of Contemporary Art. University Complutense of Madrid. July 1990. Madrid. July 1990. Publications: - Ruiz, F. (1990). La simetra en la obra de M.C. Escher (Symmetry in the work of M.C. Escher). Catalogue of the Exhibition "M.C. Escher, entre la Geometra y el Arte". (M.C. Escher between Geometry and Art) University of Granada. May 1990. (pp. 1-3). Published by the University of Granada (Spain). - Ruiz, F. (1990). M.C. Escher. Catalogue of the Exhibition M.C. Escher. University Complutense of Madrid. July 1990. (pp. 9-11). Published by the University Complutense of Madrid (Spain). - Ruiz, F. & Rico, L. (2001). Visualization of Numerical Patterns by Means of Congruence Relation. Fifth Interdisciplinary Symmetry Congress and Exhibition of the International Society for the Interdisciplinary Study of Symmetry (ISIS-Symmetry). Intersections of Art and Science. Sydney, Australia. 2001.

Abstract

The study of algebraic structures usually presents difficulties because of its degree of abstraction. The visualization of elements of these structures can facilitate their recognition and the understanding of some properties. In order to visualize the additive and multiplicative groups of the residual classes Zn we substitute the numeric symbols for certain colored visual patterns in the corresponding Pythagorean table, in order to 231

highlight the regularities in the groups. This way, the resulting table maintains the original structure and provides new geometric elements to reduce the degree of abstraction and to obtain creative designs, which emphasize the symmetry. The students work, carried out manually or by means of computers, become motivational elements for the study of the algebraic structures.

1. Introduction
Visual mathematics has become an important reference point in researching mathematical education. Sensory perception is an important way to access knowledge. It is appropriate to consider the intervention of the senses in the transmission, acquisition and construction of any type of mathematical knowledge. Visual information as a generator of images and mental objects provides an important function in learning mathematics. The capability to visualize any mathematical concept or problem requires experimentation to interpret and understand figurative information about the concept and be able to manipulate it mentally, as well as to express it visually. The study of algebraic structures usually presents difficulties because of the degree of abstraction. Visualization of these structural elements can facilitate recognition and understanding of some of their properties. Since the Spanish curricula for Secondary Education have given up on the teaching of Modern Mathematics, algebraic structures are taught only to university students in some scientific careers, while Secondary students have never heard of the structure of group or field. In spite of the difficulties that the notion of algebraic structures has for many teaching students, we believe that teaching them contributes to enriching the view of future teachers about mathematics.

232

2. The aims of the experience

We aim to contribute to correcting the difficulties of this process of abstraction by means of: - a visual form of mathematics derived from the existing connections between algebra and geometry and the use of representations, - motivational elements for the students like the use of resources such as materials and computers. - strengthening the creativity of students by means of drawing artistic designs.

233

First of all we consider that the fields of knowledge that compose mathematics are related to one another, rather than being isolated subjects. In Mathematical Education we should make an effort to highlight these connections since they are not evident for our students. It is usual for students not to use their knowledge of algebra when they are studying statistics, or their knowledge of geometry when studying analysis, etc. For this reason we should promote experiences that relate the different fields of mathematics, and let future teachers make connections between algebra and geometry, by creating geometric patterns that highlight algebraic properties and relationships. A second objective of this experience is to take into account the elements of motivation in teaching. Mathematics is not the most popular subject for teaching students, which is why motivation becomes especially necessary in this case. The use of physical and technological methods can help to increase the pleasure of learning mathematics for students who have difficulties with abstract concepts. A third objective is creativity. Creativity has a decisive influence on the acquisition of certain learning skills that ought to be acquired gradually This creativity, in our experience, is linked to the perception of shapes and patterns, and it facilitates establishing a connection between numbers and geometric figures. To make compositions aesthetically combining shapes and colors involves an effort in the use of regularities, geometric figures, transformations, etc. until the students achieve a composition that satisfies their aesthetic expectations.

3. The study
The aim of our study was to answer questions such as the following: What do we achieve by replacing traditional numeric symbols with geometrical ones? Is visualization of the properties of algebraic structures facilitated in this way? Can we perform geometric transformations upon these tables? What is the role of computers in these activities? The experience involved two groups of primary student teachers at the University of Granada in two different sessions. The first one took place during the 1992/93 academic year. The lecturer made a short presentation with transparencies in black and white, and the work was done by hand, using a photocopier and several kinds of paper or transparencies. Only one student used color and another used a computer program. The second experience took place during the year 2001/02. The same short presentation was made by the lecturer but on this occasion he used a computer, with the same figures but 234

in color. In a first stage we aim for the students to come into contact with the mathematical knowledge that they will be working with. They are explained the mathematical structures they are going to use in order to generate patterns. We go into details of the properties of the elements of these structures and the operations of an algebraic system. In this case our purpose is that the students:

work to obtain the residual classes (Zn) determine the properties of an additive group check under what conditions the multiplicative structure provides the structure of a group be able to operate with these new elements familiarize themselves with the use of Pythagorean tables. use geometric figures and apply them in geometric transformations.

In order to introduce the residual classes module n to the students, we show some transparencies like the following, which is a sort of machine that schematizes Z4.

235

In a second phase the students receive information about the activities they have to develop, and they may be shown some examples made previously by the lecturer, and some materials such as transparencies, photocopies, grids, etc. We define the addition of classes as usual, and we display the addition and multiplication of classes on a Pythagorean table (fig. 2).

+ 0 1 2 3

0 0 1 2 3

1 1 2 3 0
Fig. 2

2 2 3 0 1

3 3 0 1 2

In order to represent numbers, mathematicians use symbols, which are special drawings with very well known meanings. Artists make different kinds of drawings without any restriction. What would happen if we substituted numerical symbols with geometrical drawings that have no previously agreed upon meaning? The students must assign a geometric form to each element of the residual class suggesting the idea that inverse elements should have complementary forms. Then, the students make the corresponding Pythagorean tables of these elements with the sum and the product.

0 1 8

2 3 4

7 6 5

+ 0 1 2 3 4 5 6 7 8

0 0 1 2 3 4 5 6 7 8

1 1 2 3 4 5 6 7 8 0

2 2 3 4 5 6 7 8 0 1

3 3 4 5 6 7 8 0 1 2

4 4 5 6 7 8 0 1 2 3

5 5 6 7 8 0 1 2 3 4

6 6 7 8 0 1 2 3 4 5

7 7 8 0 1 2 3 4 5 6

8 8 0 1 2 3 4 5 6 7

(Z9, +)
Fig. 3

236

An example shown to the students belongs to (Z9, +) in figure 3. In such a table, we can see numeric regularities, such as symmetry, along the main diagonal line (commutative property), or along parallel secondary diagonal lines (filled with the same number). After substituting each numeric symbol with its corresponding geometric figure, we obtain an isomorphic table (figure 4). (Forseth, S. & Troutman, A., 1974). This table can be modified by geometric transformations such as reflections or rotations (figure 5). Now, the students must go into in the algebraic structures working on their own, to find properties and regularities, identify structures visually and carry out artistic designs using different variables:

Additive or multiplicative structure Module Shapes of each element Color Shapes of the table (different grids) Geometric transformations (line symmetries, rotations, glide reflections, translations to make tessellations, etc.) Deformations using computers programs

Figure 4

237

Figure 5

4. Students work
Here are some examples from the first group of students, most of them done by hand, in black and white:

The shapes chosen for classes are squares, but the cells of the grid have different shapes. Then, this student uses mirror symmetries.

238

In this case the shapes for classes and the grid are triangles, but the cells of the grid are triangles and quadrilaterals. The resulting shapes using mirror symmetries (Fig. A) or rotations (Fig. B) are hexagons:

A 239

In this example the author uses the square and the arrow as a base for making his drawing. Most of the students did their work in black and white, because in this experience the introductory examples shown had this characteristic. The only student who used color started from these simple shapes. By applying line symmetries he got the following drawing:

240

The only student who used a computer program made a drawing based on (Z3, +) the pattern for both classes 1 and 2 being complementary.

The final result after mirror symmetries is rather simple:

He then created distortions with the computer, obtaining the following outputs:

241

Some examples from the second group of students: We found a great variety in the students work. A few of them made almost an exact copy of what they saw in the classroom, for example these two:

242

They both correspond to an additive group, but there are no relations between the symmetrical classes. The following used a greater variety of geometric shapes, including triangles, squares, parallelograms, etc. Since the work is framed with the neutral element, we can appreciate that its structure is a multiplicative one.

In the next output, the student used the square and the 3-D effect of MS Office drawing tools. She could not achieve a perfect symmetry with the elements. After trying repeatedly, she realized that the mirror reflection of these tools affects only the shape and not to the contents of the square that she made.

243

Some other examples follow:

Tessellation made with (Z10, x), framed in black. The following are three variations of the previous structure. 244

Another three variations of the same structure are:

245

In spite of the fact that the two following pieces of work look very different to each other, they have the same mathematical source. The first one reminds us of the mosque of Cordoba (Spain).

246

Next is an example of how to use colors to highlight the structures visually. Since the module increases, we can see how the colors and the shapes also increase. On the other hand we can appreciate the differences between these tessellations based upon the additive structure and the multiplicative one.

247

(Z2, +)

(Z3, +)

(Z4, +)
Additive structures

(Z5, +)

248

(Z2, x)

(Z3, x)

(Z4, x)
Multiplicative structures

(Z5, x)

5. Conclusions
The main conclusion of this experience is that it motivates students, and by means of these activities students are able to:

perceive and visualize algebraic structures and their properties 249

operate with: algebraic structures geometric figures geometric transformations

The use of computers:

Facilitates the design of geometric figures drawings, color, duplication, special effects Improves the application of geometric transformations Increases the students motivation and level of implication in the tasks Improves the quality of the students output.

We want to emphasize that our main purpose was not for the students to learn more mathematics, but to begin changing their attitude to mathematics. Many of them said that mathematics is not so awful as they thought previously, and they really enjoyed creating designs using mathematical concepts.

6. References
Forseth, S. & Troutman, A. (1974): Using mathematical structures to generate artistic designs. Mathematics Teacher, vol. 67, n 5. Ruiz, F. & Rico, L. (2001) Visualization of Numerical Patterns by Means of Congruence Relation. Fifth Interdisciplinary Symmetry Congress and Exhibition of the International Society for the Interdisciplinary Study of Symmetry (ISIS-Symmetry). Intersections of Art and Science. Sydney, Australia

250

ENDLESS FORM-WORLD GENERATED BY INTEGER PERMUTATION

GEORGE LUGOSI

Name: George Lugosi, director of R&D&I (b. Budapest, Hungary, 1936) Address: R&D&I, 2 Union Street, Kew 3101, Victoria, Australia E-mail: g.lugosi@hfi.unimelb.edu.au Field of interest: Patterns, forms, analogical thinking, crystallography, geometry; (sailing, gliding, sci-fi). Publications: Benapozs vizsglata, [Investigating the Sun-path in a room, in Hungarian] Mvszet, 1977, 3 Szabadg = semmibe vett szksgszersg? [Is freedom=ignored necessity? in Hungarian] Mvszet, 1978, 6 Patterns based on permutations, Symmetry: Culture and Science, Vol. 3 No. 2 (1992) More investigation on permutation-generated patterns, Symmetry: Culture and Science, Vol. 6 No. 2 (1995) Folk art and symmetries of permutation-generated signs, Symmetry: Art and Science, Vol. 1 No. 1-2 (2001)

Abstract: The permutation generated patterns, which are created in square matrices, consisting of n rows and n columns are naturally arranged in tables. Selecting an n, the number of basic forms (or signs) is n! x n!. This world offers really crystallike symmetries, and beyond this, several of its elements we can find in the folk art of several South American, Middle American, Asian and European countries. It would be a far reaching and fruitful task of etnomathematic to make serious research-work on this field.

1. Introduction
How to generate the above-mentioned signs? The simplest rule is: the M(c, r) n x n matrix has to be filled so that if the columns are c(1)=n, c(2)=n-1,c(n)=1 (from left to right), the rows are r(1)=1, r(2)=2,,r(n)=n (from up to down), and n(c) + n(r) > n, than M(c,r) = 1, otherwise it is 0. When we filled the first sign, than we take the next permutation of the rows and leave the columns intact, fill the next sign, etc. When we have no more possibility for the new permutation of the rows, keeping the r=1 at the 251

first place (which means that we filled (n-1)! signs), then we start a new row of the signs, starting with 2 at the first place. If we started with the n=4 and the sequence of the columns are 4,3,2,1, of the rows 1,2,3,4, then we permute the rows as 1,2,4,3; 1,3,2,4; 1,3,4,2; 1,4,2,3; and 1,4,3,2, all together 3! =6 signs. Then we cannot keep 1 at the first place, start a new row of the signs with 2,1,3,4 row-sequence, and continue with 2,1,4,3; 2,3,1,4; 2,3,4,1; 2,4,1,3; and 2,4,3,1. The next line of the signs will start with 3,1,2,4, and the last with 4,1,2,3. The 24 signs, arranged in four rows and six columns, will create the image of Fig. 1. It is the first table. We leave a gap, the size of a unit-square, between the signs, and the size of a sign between the tables.

Fig.1

Fig. 2

After this we have to change the column sequence from 4,3,2,1 to 4,3,1,2 and start again with the row 1,2,3,4. Similarly to the previous one, we can fill the second table, etc. When we finished the sixth table (the first row of the tables), we have to start the next column-sequence with 3,4,2,1. The third row of the tables will start with 2,4,3,1 column arrangements, the fourth (the last, if n=4) with 1,4,3,2. Finishing this, we can see that in the full block we will have n x (n-1)! number of tables with n x (n-1)! number of signs in each of them, which gives n! x n! number of signs. The number of signs in the tables (equally, the number of tables in a block) are growing so rapidly that we meet with a very interesting phenomenon: let us select n=5, and the size of the unit square=1 mm, which fits a sign in a 5 x 5 mm square. If we wish to print out the full block, then we will realize that the printout is more than 3.5 meters long. It means that if n > 5 then we cannot see the full block from one standing point. To be able to walk around and to see the generated form-world we need the help of a computer. We can imagine that we can move the screen around as a special window within our permutation-generated world.

252

2 The basic symmetries

Using the n=4 as an example, we can find within the block as well as within the tables several symmetry connections. Fig. 2 shows these connecting lines. The famous saying as above so below is true here also. The basic symmetries in the block are the same ones as in the tables. Regardless of the n, in one table there are two and only two diagonal-symmetric signs. When we investigate the signs in a table and the tables in the block, we can find symmetries, which are the same within a table and within the block. The lines are connecting those places where we can find the same diagonally symmetrical signs (in rotated positions) and in the same relative positions from these signs we will find the same surroundings in the connected tables also. These are important information about the so-called logical places.

3 Some important signs

It is a remarkable fact that in the folk art we can find certain signs, which are connected to each other. The stairs (the first and last signs of the first table) many times depicted together, in Hungarian cross stitch symbols as well as in North American Hopi Indian images or on Aztec shield and woven materials from Peru, with the spirals. We can find these forms together even as a clay vessel in Peru. Fig. 3 shows a Hopi-Indian sign and Fig. 4 are Hungarian cross-stitch elements from the region of the river Tisza. It is remarkable, that on this Hopi sign we can see two different types of spirals beside the stairs.

Fig. 3

Fig. 4

4 Symmetry of symmetries
In the folk art we can find Hanti shirts, which shows elements of the same logical places when n = 3, 4, and 5, and these are on the same item. (See Fig. 5). Naturally, it would be a rather silly idea to suppose that the maker of this item had any idea about 253

a sign system. The elements are coming from the folk soul. However, it is really interesting that on the stole we can see the n=4 system 11th table 3rd sign and under the V (which is at the end of the stole) we can find the n=5 version of it. Every country or area has some characteristic patterns, which they use as their formelements. It would be an interesting task of etnomathematic to investigate these patterns, and find out, why can we find certain sign-combinations (e.g. stairs and spirals together) in different areas of the world, thousands of kilometers from each other.

Fig. 5 The best way to see, understand and enjoy the symmetry-world of this endless and amazing system is to step into it and create the signs ourselves. It can strongly improve our form-recognition also. References
Ditfurt, H., (1973), A vilgegyetem gyermekei, [Children of the Universe, in Hungarian], Budapest McIntyre, Loren, Mystery of the Ancient Nazca Lines, National Geographic, May, 1975 Molnr, V. J., (1999), Vilg Virg, [World-Flower, in Hungarian], Budapest Purce, Jill, (1980), The Mystic Spiral, Thames and Hudson, Singapore Spinden, H. J., (1975), A Study of Maya Art,, Dover Publications, Inc., New York.

254

VISUALIZATION VS. VERBALIZATION, INSIGHT INTO THE MORPHOLOGY OF POLYHEDRA

IRIT WERTHEIM AND NITSA MOVSHOVITZ-HADAR

Name: Irit Wertheim. Address: Technion Israel Institute of Technology, Department of Education in Science and technology, Haifa 32000, ISRAEL. E-mail: weririt@techunix.technion.ac.il. Fields of interest: morphological approach to 3-D geometry. Name: Nitsa Movshovitz-Hadar. Address: Nitsa Movshovitz-Hadar, Ph.D., Head of Kesher Cham - National Center for Mathematics Education, Director of the Israel National Museum of Science, Planning, and Technology, Technion 32000, ISRAEL. E-mail: nitsa@techunix.technion.ac.il.

Abstract: The impact of visual representations on understanding, and even more so, on actively doing mathematics, has been intensively researched and is widely recognized. This is particularly true for the study of 3-d geometry. What about the role of verbal descriptions? Are they necessarily needed? Simply redundant? Or are they inappropriate, may be even disturbing? In this paper we attempt at demonstrating the crucial role verbalization plays as a complementary mode to visualization, for dealing with 3d geometry tasks, understandably and insightfully. Our claim is that neither visualization alone nor verbalization in itself suffice for meaningful conceptualisation. The cognitive processes involved will be demonstrated in the context of polyhedra.1 Introduction - One picture is worth a thousand words, or is it? In their struggle towards obtaining a sense of meaning for the hypercube (a 4-d cube), Davis & Hersh describe vividly, the important role played by manipulating a visual representation: "...I was impressed by ... the sheer visual pleasure of watching it. But I was disappointed; I didn't gain any intuitive feeling for the hypercube... I tried turning the 255

hypercube around, moving it away, bringing it up close, turning it around another way. Suddenly I could feel it! The hypercube had leaped into palpable reality, as I learned how to manipulate it..." (Davis & Hersh, 1981). "One picture is worth a thousand words". Underlying this well-known saying is the widespread experience that a relatively simple visual representation can replace a lot of, and save hours of talking. The impact of visual representations on understanding, and even more so, on actively doing mathematics, has been intensively researched in the past decades and is widely recognized nowadays. This is particularly true for the study of 3-d geometry (e.g., Parzysz, 1999; Barwise & Etchemendy, 1991; Zimmermann & Cunningham, 1991). Actually, the employment of visual representations in the study of spatial geometry comes very natural and handy. It is commonly agreed that 2-d drawings and 3-d concrete models are necessary tools, which provide comprehensive understanding of 3d configurations. What about the role of verbal descriptions? Are they necessarily needed? Simply redundant? Or are they inappropriate, may be even disturbing??? In this paper we attempt at demonstrating the crucial role verbalization plays as a complementary mode to visualization, for dealing with 3d geometry tasks, understandably and insightfully. Neither visualization alone nor verbalization in itself suffices for meaningful conceptualisation. This claim is supported by Kosslyn (1980) and by Stigler (1984) who suggested that visual and verbal representations of the same information have distinct attributes augmenting one another. To elaborate on the issue we focus our attention to the elementary processes of watching a solid or a 3-d configuration, followed by articulating its structure and properties. We'll show that the latter, promotes insight and understanding, which might not be gained through sheer observation nor by manipulating the visual representations. This we do in the context of convex Deltahedra - - Polyhedra with equilateral triangular faces. To conclude this introduction, here is a personal anecdote by the first author: The three Platonic polyhedra made out of equilateral triangular faces, were old "acquaintances" when I was first introduced to the set of convex Deltahedra (Note that if we do not restrict ourselves to convex deltahedra, there are infinitely many deltahedra, e. g., the stellated dodecahedron). The deltahedral dipyramids were quite easy to grasp (a dipyramid is a solid made of two congruent pyramids with a common base). However, there were three other deltahedra, D12, D14 and D16, which remained altogether "strangers". It was difficult for me to gain any intuitive feel for them - I was confused. I got a hold of them, only as I was able to construct a verbal description for 256

each (!). "Suddenly I could feel them! They had leaped into palpable reality", just like Davis' experience with manipulating the visual representation of the hypercube.

2 How many convex deltahedra possibly exist?

An n-face deltahedron is a polyhedron with n congruent equilateral triangular faces, conventionally denoted Dn. The name comes from the Greek capital letter delta, which has a triangular shape. An interesting problem is to find how many such convex polyhedra there are. In trying to exhaust the set of convex deltahedra, we note, first of all, that this set includes three of the five regular/Platonic solids, D4 - the tetrahedron; D8 - the octahedron; D20 - the icosahedron. Other convex deltahedra might be obtainable by attaching several triangles side to side in different spatial configurations. Our question is: How many convex deltahedra can thus be obtained?

2.1 How many faces can convex deltahedra possess?

To answer this question we first observe that in each vertex 3, 4, or 5 triangular faces only, can meet. These numbers define the valence of a vertex. Clearly, the valence of all vertices in any given deltahedron is not necessarily the same. Denote V3, V4, V5 - the number of vertices of valence 3, 4, 5 respectively. Next, note that (i) The minimum number of faces is four, as in the tetrahedron. (ii) Since no vertex can have more than five equilateral triangles meeting in it (as six give a flat 3600 angle sum around the vertex), the largest possible convex Deltahedron is a solid having 5 equilateral triangles meet in each of its vertices, i.e., the icosahedron. Thus twenty is the maximum number of faces for any Deltahedron. Therefore, The total number of faces of any existing convex Deltahedron is in the range between 4 and 20.

257

2.2 Candidates for existing deltahedra.

First observe that the number of faces in any Deltahedron must be even. To establish it note that each face has three edges and each edge is shared by two faces. Using Eulers formula for simple polyhedra, F(aces)+V(ertices)=E(dges)+2 , we get F=2(V-2) which implies the evenness of the total number of faces. Thus, we conclude that there are only 9 candidates for deltahedra those having 4, 6, 8, 10, 12, 14, 16, 18 and 20 (triangular) faces. To determine the various possible deltahedra, we employ Descartess formula for the total angular deficit of a convex polyhedron. First observe that: (i) The angular deficit of a vertex of valence 3 is 180o; (ii) The angular deficit of a vertex of valence 4 is 120o; (iii) The angular deficit of a vertex of valence 5 is 60o. Descartess formula for the total angular deficit of a convex polyhedron yields 180oV3 + 120oV4 + 60oV5 = 720o, or 3V3 + 2V4 + V5 = 12. Every Deltahedron must satisfy this (Diophantine) equation. This is a necessary but yet insufficient condition for the existence of a Deltahedron.

2.3 The solution

Every solution of the above mentioned (Diophantine) equation, which takes into account the evenness of the total number of faces, is a candidate for an existing Deltahedron. Altogether, there are nineteen possible solutions to the equation. They are listed in Table 1 (Lichtenberg, 1988). However, the only solutions that correspond to actually existing deltahedra are 1, 3-7, 16, and 19, total of eight convex deltahedra (a deltahedra is a polyhedra with equilateral triangular faces).

258

3. Visual and verbal description of the eight deltahedra.

As we have seen, altogether, there are exactly eight convex deltahedra. This set includes:

The Platonic solids: The tetrahedron D4, the octahedron D8, and the icosahedron D20; Two dipyramids - D6, and D10. (D8 which is also a dipyramid is not included here because it is mentioned above). D6 - triangular dipyramid - two attached tetrahedral. (D8 - square dipyramid - two attached square pyramids, actually the Octahedron). D10 - pentagonal dipyramid - two attached pentagonal pyramids.

And - three rather more complicated deltahedra: D12 - snub disphenoid. This solid can be thought of as a tetrahedron split into two wedges each made of two adjacent faces, which are then joined together by a band of eight triangles. D14 - triaugmented triangular prism - three square pyramids, each attached to a square side face of a triangular prism. D16 - gyro elongated square dipyramid - two square pyramids attached to a square antiprism.

Note that an antiprism has two congruent regular n-side polygons, as bases, in parallel planes. One base is twisted so that each of its vertices is midway between two vertices of the other, to each of which it is joined. The side faces are triangles. While twisting one of the prism's bases, the square side faces of the prism are folded into triangles, and the result is an antiprism.

3.1 D14 and D16 Can you tell the difference?

259

It is quite difficult to distinguish D14 and D16 from each other by their visual images. Holding their concrete real models and looking at them from various angles does not help much either. On the other hand verbal descriptions make it quite clear: D14 is a triangular prism with a square pyramid attached to each of its side faces; D16 is a square antiprism with a square pyramid attached to each of its bases.

3.2 Seeing and believing

We observed high school and college students attempting to describe verbally these two solids, after they saw their actual models and various drawings. The evidence collected suggest quite clearly, that only after one is able to give a coherent verbal description, one reaches an understanding of the structure of each solid, and is able to see the differences between the two solids. - - As if these solids leaped into palpable reality, in Daviss (ibid) language.

4. A Brief Discussion
Each of us tends to go through an initial "grasping" in which we understand key concepts but cannot converse about them fluently. As our exposure to the material increases, we are able to shape our comprehension through questions, tentative verbalizations, informal talks with others, reorganization of notes, and so forth. Through language, then, we gradually extend the meaning and gain understanding. We not only recognize the structure of the subject, but also verbally manipulate its ideas, expressing its orderliness in personalized and unique ways (Suhor, 1984). Mathematics educators have been busy studying mathematical discourse in the past two decades. It is commonly agreed within this community of researchers that intuition and operational experimentation serve successfully the development of higher level understanding, however, reaching this level necessitate reflective thinking which is anchored in verbal discourse. (E.g., Skemp 1973, Hiebert, and Carpenter 1992, Sfard 1994 and others) There is a tendency to communicate about visible entities by visual representations, and about formal mathematics notions, by symbols and verbal arguments. Although 3-d geometrical entities are real and visible, the examples given above illustrate that in order to develop a sense of meaning for their structures, their properties and the interrelationships among them, it is necessary to combine language and pictorial representation. "... thought unexpressed remains immature and eventually dies out. 260

Language therefore, is not just an expression of otherwise independent and fully formed thought, but rather is a necessary form of the thought's realization." (Vygotsky, 1986). While the visual representation provides an intuitive infrastructure, it is the language that guarantees the transition from the intuitive perception to an analytic-synthetic understanding, and even to innovative findings. For example, while describing in words the process of truncating (cutting off) the vertices of a tetrahedron through mid-edges, students observed that truncating the top vertex first, yielded a triangular face parallel to the basis of the original tetrahedron. This observation led to the insight that by truncating the other three vertices, a triangular antiprism is obtained. A second thought brought them to the conclusion that the resulted 8 triangular face polyhedron is actually an octahedron, and from here the way was short to the insightful finding that the triangular antiprism and the octahedron are nothing but two sides of the same coin. The property of the octahedron as an antiprism emerged as a result of the discourse among the group of students who had a tetrahedron in front of them and were challenged to study the results of various ways of truncating it. The students arrived at their discovery, totally on their own based upon the combination of the mental image they were able to operate on, and their linguistic ability to follow their imaginary action of truncating and connect it to a recently acquired concept of the antiprism. Visual representations and mental (pictorial) images, being wholesome and compact supported students overall structural conception, that can be grasped at one glance, while verbal encoding enabled a process of decomposition of the whole into its parts, analysing the inter-relationship among the parts and the complete new structure. Language played an important role in coming to grips with spatial relationships and structural properties. Using 'literary' forms in making connections, helped students in making sense of mathematical constructs, and in remembering it. Language is the medium within which the creation of new concepts takes place. We are not questioning the power of visualization. As Sfard (2000) noted: "Availability of visually manipulability means, either actual or only imagined, underlies our ability to communicate on the objects and operate them discoursively". On the other hand, we suggest that verbal representations for visual entities are crucial for the understanding of those entities. The concepts we are dealing with belong to the domain of concrete objects or processes, usually, accessible to us through perceptual experience. As these concepts are more complicated, the role of language and verbalization becomes more crucial. Students come to recognize the properties of a solid, by describing them verbally. While verbalizing, students organized their visual impressions by comparisons: distinguishing which aspects of the shape are to be noticed and which ones are to be ignored, which properties are similar to already known objects, and which are different. The immediate implication is that whoever strives to become a professional, must 261

develop the ability to express verbally, visual observations and to visualize verbal ideas mentally. Describing an object verbally is a process that involves the decomposition of the whole, followed by sequential reassembling of the parts. As early as 1969, Paivio observed that visual imagery has many of the properties of a spatially parallel system, whereas verbal processes are better suited for handling sequential, serial information (Paivio 1969). It takes cognitive processing to make sense of visual information, be it a 2d representation or a concrete 3-d model. The interplay between visualization and verbalization is the key to cognitive processing of that sort. The enhancement of this combination is therefore at the heart of a literate approach to any profession that has to do with spatial relations - - Architecture, Engineering, Mathematics and Arts. Finally, to balance the claim made so far, we leave you with a dilemma we started this paper with: Is verbal description always appropriate? Is it necessarily needed with no exception? Or is it sometimes simply redundant? Possibly disturbing???

References
Barwise, J. & Etchemendy, J. (1991). Visual Information and Valid Reasoning. In (Ed) Zimmermann, W. & Cunningham, S. Visualization in Teaching and Learning Mathematics. Mathematics of America. USA Davis, P.J. & Hersh, R. (1981). The Mathematical Experience. Birkhauser Boston. U.S.A. Hiebert, J., & Carpenter, T. P. (1992). Learning and teaching with understanding. In D. A. Grouws (Ed.), Handbook of research on mathematics teaching and learning (pp. 6597). New York: Macmillan. Kosslyn, S.M. (1980) Image and Mind. Harvard University Press. Lichtenberg, D.R. (1988). Pyramids, Antiprisms and Deltahedra. Mathematics Teacher. Vol. 81 Num. 1 pp. 261-265. Paivio, A. (1969). Mental imagery in associative learning and memory. Psychology Rev. 76 pp. 241-3. Parzysz, B. (1999). Visualization and Modelling in Problem Solving: From Algebra to Geometry and Back. Proceedings of the 23rd Conference of the PME. Sfard, A. (2000). Steering (Dis)Course Between Metaphors and Rigor: Using Focal Analysis to Investigate an Emergence of Mathematical Objects. Sfard, A. (1994). Reification as a birth of a metaphor. For the Learning of Mathematics, 14(1), 44-55. Skemp, R.R. (1973). Relational Understanding and Instrumental Understanding. Mathematics Teaching. Stigler, J. (1984). Mental Abacus: The effect of abacus training on Chinese children's mental arithmetic. Cognitive Psychology 16 145-176. Suhor, Charles (1984). Thinking Skills in English and across the Curriculum. ERIC Digest. Eric Product (071); Eric Digests (selected) (073). Vygotsky, L. (1986). Thought and Language. The MIT Press. Cambridge. Massachusetts.

262

Zimmermann, W. & Cunningham, S. (Eds). (1991). Editors' Introduction: What is Mathematical Visualization. In: Visualization in Teaching and Learning Mathematics. Mathematics of America. USA

263

264

EXPANDABLE POLYHEDRAL VIRUSES KINEMATICS IN BIOLOGY

FLRIN KOVCS

Name: Kovcs Flrin, Civil engineer, (b. Budapest, Hungary, 1973) Address: Research Group for Computational Structural Mechanics, Hungarian Academy of Sciences, Budapest, Megyetem rakpart 3, H-1521 Hungary. E-mail: kovacsf@ep-mech.me.bme.hu Fields of interest: Kinematics, deployable structures, geometry (also geography, maps and classical music). Publications and/or Exhibitions: Kovcs, F., Hegeds, I., Tarnai, T. (1997), Movable Pairs of Regular Polyhedra, in Proceedings of International Colloquium on Structural Morphology (eds. J. C. Chilton, B. S. Choo, W. J. Lewis and O.Popovic), Nottingham, 123-129. Kovcs, F., Tarnai, T. (1998), Foldable Bar Structures on a Sphere, in Proceedings of 2nd International PhD Symposium (ed. G. L. Balzs), Budapest, 305-311. Kovcs F. (2000), Foldable Bar Structures on a Sphere, in IUTAM-IASS Symposium on Deployable Structures: Theory and Applications (eds. S. Pellegrino, S. D. Guest), Solid Mechanics and its Applications, Vol. 80, Kluwer Academic Publishers, Dordrecht, Netherlands, 221-228. Kovcs F., Tarnai T. (2000), An expandable dodecahedron, in Bridge between Civil Engineering and Architecture Proceedings of International Colloquium on Structural Morphology (ed. J. M. Gerrits), Delft, Netherlands, 227-234. Kovcs F., Tarnai T. (2001), An expandable dodecahedron, HyperSpace, 10, no. 1, 13-20.

Abstract: The word mechanism is commonly associated with technical sciences. Altogether, there are some bright examples of them among living organisms such as in certain polyhedral viruses that are able to change their diameter by means of a symmetric rotational-translational motion. Mechanical modeling of this mechanism can provide deeper insight to natural kinematical processes and can give hints for structural engineering. First stage of modeling is the physical one that was built to simulate the observed motions, and then a numerical modeling follows done by computer analysis. Compound symmetric structures with multiple sets of independent infinitesimal displacements as occur in this analysis require efficient tools to interpret the motions in a physically plausible form: in our practice the group theory 265

was explored. Finally, some general conclusions are drawn about infinitesimal mobility of the physical model, extended to a family of slightly different polyhedral mechanisms.

Fig. 1: Cardboard model of an expandable dodecahedral virus, closed and expanded.

C B D F E

Fig. 2: Two pentagonal prisms and their connection

266

1 PHYSICAL MODEL
Biologists recently discovered some virus species that represent real mechanisms in nature. These structures are built of pentamers and hexamers (pentagonal and hexagonal faces) that are able to rotate and translate about their own axis if the pH of the environment changes. All this highly symmetric displacement is constrained by double protein links between two adjacent faces (Speir et al., 1995). In terms of structural mechanics, the question is how to model the protein links in these viruses in order to get the same type of mechanism like the original one? Our cardboard model is one of the possible answers where double links are substituted by single but planar ones, composed of triangular and rectangular elements. For the sake of simplicity, only twelve pentamers are used that produced a dodecahedral arrangement (Fig. 1). All members except for the hinges are supposed to be rigid, so all mechanisms must be inextensional. It is quite straightforward that linking structure between adjacent faces essentially determines the kinematical behavior of the structure. In this particular case, adjacent faces are connected to each other by means of a linkage composed of four (fold-type) hinges and a triangle (ABC), a rectangle and again a triangle (DEF) among them (Fig.2). Following just this one-parameter finite motion shown in Fig. 1, a direct relationship can be established between the amount of rotation of each face and the diameter of the circumscribed sphere of the structure: to the rotation of /5 (from the thoroughly closed configuration to the open state) belongs a 1.77 times larger diameter. Note that a fully expanded configuration means a truncated icosahedral form like a soccer ball and means a point of kinematical bifurcation, but the real range of expansion for the viruses does not include this extreme configuration. It can be verified on the physical model, however, that the structure has other probably not finite mechanisms: it is possible, for example, to lift infinitesimally just one face prism while the others remain fixed. The existence of this face mechanism can be easily proved through elementary geometrical arguments: it is evident that face mechanism means a pure translation in closed state, but pure rotation at the bifurcation point.

2 NUMERICAL ANALYSIS
Due to the design principles, the structure is able to imitate the expanding motion of viruses. At the same time, however, the structure might have other kind of mechanisms, coupled to the previous one. To discover these additional motions, a detailed cinematic 267

analysis was done in two steps (Kavas & Tania, 2001). First: just by counting cinematic constraints and unknowns it turned out that the structure must have at least six infinitesimal mechanisms (rigid body motions are not considered among them), second: the singular value decomposition of the compatibility matrix has shown nine more mechanisms. Regarding the possibility of coupling, it is quite complicated to interpret such a big amount of independent systems of displacements. To make this task easier, some considerations on symmetry were introduced by means of group theory (Kangwai, 1997). The result is that twelve of the fifteen mechanisms come from the individual infinitesimal translation-rotation of each face, while the remaining three ones belong to a certain symmetry group (T3) (Fowler et al., 2002) but their physical meaning is not clarified yet they are very likely to be also infinitesimal that would mean that only one of the mechanisms (the totally symmetric expansion) is finite, as experimented in nature. Note that in this case many of independent selfstress-states must occur in the structure that are not investigated till the moment. Interestingly enough, the analysis above can also be carried out for other expandable structures based on trivalent polyhedra, i. e. tetrahedron and cube (Fowler et al., 2002): according to the results, the six mechanisms by counting and face mechanisms plus three T3-mechanisms by matrix analysis are unchanged that supposes a more general cinematic property behind this kind of virus models. Acknowledgements Hungarian-British Intergovernmental Scientific and Technological Cooperation Programme (GB-15/98), OTKA grant no. T031931 and FKFP grant no. 0177/2001 are gratefully acknowledged.

References
Fowler, P. W., Tarnai, T., Guest, S. D., Kovcs, F. (2002) Expandable polyhedra (to be published). Kangwai, R. D. (1997) The Analysis of Symmetric Structures using Group Representation Theory [Ph.D. Dissertation], University of Cambridge, Department of Engineering, 162 pp. Kovcs, F., Tarnai, T. (2001) An expandable dodecahedron, HyperSpace, 10, no. 1, 13-20. Speir, J. A., Munshi, S., Wang, G., Baker, T. S. and Johnson, J. E. (1995) Structures of the native and swollen forms of cowpea chlorotic mottle virus determined by X-ray crystallography and cryo-electron microscopy. Structure, 3, 63-78.

268

Day 4

At the Horta van Eetvelde Hotel, 2002 April 12.

Location.
The location for the Friday talks was the Hotel van Eetvelde. It is described in the book Great Architecture of the World: "The Hotel van Eetvelde in Brussels was designed in 1898 by Victor Horta, undoubtedly the key European Art Nouveau architect. While most other architects flirted with the new style, Horta found it gave the best expression to his ideas. His skill is demonstrated in his ability to slip his domestic designs into narrow constricted sites. The interiors become of great importance as centers of light, which permeates through the filigree domes and skylightsusually in the center of the building. The Hotel van Eetvelde is a remarkable example of the way Horta handled the situation and used it to highlight the imposing staircase, which leads up to the first-floor reception rooms. "

The superb house can only very occasionally be visited. During the Monument Days in September, the house is open for the public, but there are lengthy cues. Owned by the federation of gas suppliers of Belgium, FIGAZ, the Mat mium organization could get access to the building thanks to colleague Arch. Jan Bruggemans, of the Brussels' SintLukas Archief.

269

Special Feature: mad math cartoons.

At the Hotel van Eetvelde Emiel De Bolle, a professional artist put his impressions about the Mat mium meeting and its participants on paper. It stressed the informal character of the meeting, at its final day. Note that these drawings are copy right protected, and by explicit wish of the author presented in reduced format to show they are not in their final finished and polished form.

Patrick Labarque printed computer outputs on old fashion slides first, to make sure the output would be fine.

Using a too ROUGH MathCad set-up, Labarque obtained FINE art work.

Virpi Kauko's talk was not a piece of pie, and the cartoonist was impressed.

Virpi's Mandelbrt sets were impressive.

270

Rebielak had a proof for a new type of roof, but it was not waterproof.

Rebielak used plenty of arcs in his roof constructions

Van Maldeghems talk was about spirals.

Van Maldeghems spiral could lead to an overdose, thought the cartoonist

Encarnacion Reyes talked about making polygons by folding paper.

The young architects from the 51N4E group talked about the restoration of the so-called Lamot building, a former brewery.

271

Tomasz Michniowski told the participants how to lie with mathematics.

Prof. Ogawa had a top-level talk about the true mathematical notion called ... frustration.

About Radmila Sazdanovic tilings

And again about Radmila Sazdanovic tilings

Were Schindler's dimensions too small?

Han Vandevyveres subject was about measuring Gothic town halls.

272

Dirk Frettloehs subject was tiling design.

Vera de Spinadel's topic was not about the lottery, but the cartoonist saw a lot of 'golden' numbers.

Some general cartoons

It was an interdisciplinary conference; mathematicians, architects and artists sometimes spoke different languages.

273

274

PATTERN DESIGN BY IMPROPER USE OF MATHCAD

PATRICK LABARQUE

Name: Labarque Patrick, architect, (b. Kortrijk, Belgium, 1945). Address: Hogeschool voor Wetenschap & Kunst, Sint Lucas Architectuur, Paleizenstraat 65-67 BRUSSELS E-mail: plabarque@archb.sintlucas.wenk.be Fields of interest: colour theory; the 4-colourproblem; geometry in general such as polyhedra, projection systems, geometric optics... Awards: Laureate of the urban design competition "Europa Kruispunt - Carrefour de l'Europe" Brussels 1970

Abstract: If one plots a function in Mathcad (or another math program) one has to define the x-values of the points to be plotted. These points are then successively joined by a line. When plotting a repetitive function, with the point spacing bigger than its repetition length, a nearly erratic plot is the result. The same can be done with contour plots. Here the contours are interpolated as well but they depend on the xy-values of the grid points. Some pattern designs extracted from the animation resulting from this improper way of plotting are shown.

INTRODUCTION
The pattern design started from a camera obscura or pinhole camera. One of the disadvantages of this camera is the very long exposure time it needs. A solution for this problem may be the use of a zone plate. Such a plate can be plotted with any mathprogram, printed, photographed and reduced on microfilm to the appropriate dimension. The plotted function is shown below as a surface plot (middle) and as contour plots. The contour plot with 2 levels (left) is a Fresnel zone plate. A Gabor zone plate (right) is the result when we plot it with an infinite number of levels from white to black, resulting in continuous greyscales from high to deep.

275

The illustration below explains the geometrical optics of the Fresnel zone plate, based on the interference of light. The ray in black has constructive, the one in white has destructive interference in the focal point with the central ray.

In order to define the pattern for the zone plate we start from the focal point. We divide the distance D by the wavelength and express it as a cosine function to obtain the amplitude in the z direction. 276

 x + y + f amplitude = cos 2

If the function is positive, this ray has constructive interference, if not destructive interference with the central ray in the focal point. If we eliminate now the rays with destructive interference by use of black rings we have an amplification of monochromatic light in nearly the same way as with an ordinary lens (in fact things are more complicated than explained here). A Fresnel zone plate with only black and transparent rings has not a sharp focal point for a given wavelength. A Gabor zone plate corrects this. However, a disadvantage of such a zone plate is the big chromatic aberration, as its focal distance is to a high degree depending on the wavelength. The use of a green filter gives a substantial amelioration, because green is one of the additive main colours (RGB), and these main colours give a more monochromatic light. Besides, as there is a lot of green light in normal white light, we lose less light with a green instead of a red or blue filter.

PATTERN GENERATION
The surprise at first was big when the result below, with 5 levels was obtained.

277

Afterwards the explanation was very simple; the contour plot above was made with a 9x9 matrix (for a "quick" result). The plot program then interpolates the contours (see the last figure in the paper). For a more correct plot a grid with much higher density was needed (and a lot more of rendering time). The first disillusion turned to curiosity; the lens design switched to pattern design. Experimenting with exactly the same formula, but with other grid values (11x11, 13x13 and 15x15 matrix), patterns as illustrated below are obtained.

Only by changing the grid density, a nearly infinite number of different patterns can be generated from one and the same formula.

The question arises whether the patterns can continuously be transformed into each other to make animations. This is impossible by changing the number of grids, as we need integers to build the matrix for the contour plot. But gradually changing the value can do the job (also "f" can do it). The next illustration shows such a continuous transformation of a pattern with the same grid density by changing the value step by step.

278

The patterns look too homogeneous. Therefore the amplitude is scaled from centre to edge. The figure below illustrates the principle of the erratic plot and the result of the scaling.

279

After careful experimentation with different step and scaling values, the final animation "MAT MIUM 2002" was made. The mini-animation on the Matomium website is made for internet use, as the original animation was 80Mb!

References
........,...Mathcad7, 1986-97 by MathSoft, Inc.; MathSoft Kernel Maple 1994 by Waterloo Maple; File Filters 1985-97 by Circle Systems, Inc. Feynman, Richard P. (1988) QED. De zonderlinge theorie van licht en materie. Amsterdam, Aramith Uitgeverij. ISBN 90 6834 037 9. Translated from: (1985) QED. The Strange Theory of Light and Matter. Princeton N.J., Princeton University Press Ernst, Bruno, (1986) HOLOGRAFIE: Toveren met licht (in Dutch). Amsterdam, Aramith Uitgeverij. ISBN 90 6834 011 5. Iovine, John (1990) Homemade holograms: the complete guide to inexpensive, do-it-yourself holography. TAB Books, a division of McGraw-Hill, Inc. ISBN 0 8306 7460 8.

280

GROWING SYMBOLIC TREES and BUILDING POLYTOPES

Virpi KAUKO

Name: Virpi Kauko, Mathematician, Licentiate in Philosophy Address: Department of Mathematics and Statistics, University of Jyv askyl a, P.O. Box 35, FIN40014 Jyv askyl a, Finland. e-mail: virpik@maths.jyu.fi Fields of interest: Complex dynamics, geometry (also evolutionary biology) Publications: Shadow Trees In Mandelbrot Sets (in preparation)

Abstract: Abstract mathematical concepts, fractals in particular, may carry somewhat unexpected similarities to the natural world. Generalized Mandelbrot set Md consists of those complex parameters c for which the orbit of zero under iteration of polynomial f (z ) = z d + c (of degree d 2) remains bounded. Combinatorial properties, like the tree structure, of these fractal sets can be studied by viewing them as subspaces of a larger, abstract symbol space. Not all such symbolic sequences refer to actual parameters in Md . This gives rise to a visual interpretation of the symbol space: the Mandelbrot set lying at on the complex plane and nonexistent component trees branching o it into another dimension, rather like peculiar shadows. The connection between symbolic sequences and tree structures also gives an analogy to studying evolutionary trees of living organisms by nding mutations in their DNA sequences. Visualizing objects with dimensions higher than three may be challenging, but possible. As an example, we construct 4-dimensional polytopes using the classical 3-dimensional Platonic solids, or regular polyhedra, as building blocks.

FRACTALS and TREES

Mandelbrot sets (dened above) are simply connected and compact, innitely complicated fractal sets on the complex plane with dihedral symmetry groups. Each Md contains hyperbolic components, connected by branching, thin threads [4], [8]. We will nd an abstract space which turns out to have a natural, similarly tree-like structure. 1
281

1.1

Abstract Mandelbrot set

The combinatorial properties of polynomials f : z z d + c (and hence also Mandelbrot sets) are based on the fact that to raise a complex number into power of d means to multiply its angle by d (adding c aects little for large |z |). Thus we study mappings of angles on the circle [1]; we are particularly interested in angles that are periodic under multiplication by d (modulo full turns). For example, 25 125 53 49 29 1 5 = , 72 72 72 72 72 72 72 so these angles are six-periodic under ve-tupling; note that 25/72 = 5425/(56 1).

Angles can be turned into symbolic sequences as follows: given an angle , divide the circle into d equal sectors at angles /d, ( + 1)/d, . . . and label them 0, 1, . . . , d 1 starting from the sector containing angle 0 = 1. The kneading sequence lists the labels of sectors where the iterated angles , d, d2 , . . . sit. For 1 1 1 example, K (25/72) = 244201 0 244200 244200 . . . = 244200 . Identifying angles with equal kneading sequences gives rise to a pinched-disk model [6]. It consists of pawprints with d 1 toes, connected by branching threads. For each degree d, this tree-like structure is actually same as the Mandelbrot set! 1.2 Symbolic sequence space

Each pawprint in the (abstract) Mandelbrot set has its own kneading sequence. A natural question to ask is, whether the Mandelbrot set contains a pawprint (or hyperbolic sector, actually) with any given sequence of appropriate form as its kneading sequence. The answer is no; there are sequences which are not realized. For example, the sequence 12112 non-exists in M3 because no angle of the form a/36 1 = a/728 has it as kneading sequence. Hence the abstract Mandelbrot set is a proper subset of a larger symbol space Ad consisting of all sequences a1 a2 a3 . . . with a1 = 0. 1.3 Growing trees

The denition of kneading sequence implies an important result [5], [7]: 2

282

When the angle moves counter-clockwise around the circle, the nth entry in its kneading sequence changes from j to j + 1 precisely when crosses a rational angle of the form (rd + j )/(dn 1). Given a pair of kneading sequences, A and B , one can thus nd the minimal period of angles separating them from each other and from the origin. This information can be used to gure out how the corresponding pawprints are arranged: either the paths leading to each from origin diverge, or one path is contained in the other. This algorithm [7] works for all sequences, realizable or not. Therefore the space Ad also has a natural tree structure, an extension of the abstract Mandelbrot set. One can also see from the sequence what other sequences there are ahead when looking away from the origin. Given a pawprint C c1 . . . c(nk ) , we nd the visible trees as follows (however, it may happen that some pawprints obtained this way are nonexistent even though the base C is not): for each q N and s {1, . . . , d 1}, B (c1 . . . c(nk ) )q1 c1 . . . (c(nk ) + s) = b1 . . . bn corresponds to a satellite of C for each B already in the tree, check if b1 . . . bl = b(nl+1) . . . bn for some l. If so, then A a1 . . . am = b1 . . . (bnl + r) is above B for all r {1, . . . , d 1}. 1.4 Evolutionary trees

A somewhat similar method of translating symbolic sequences into tree structures is used by molecular biologists when they reconstruct history of life by studying 3
283

stretches of DNA molecules or proteins coded by them. Suppose we have three extant species of animals X, Y, Z, such that the latest common ancestor of two of them has lived more recently than the latest common ancestor of all three of them. Then there are three possible ways they may have evolved. If they have (ctional) DNA sequences X: ...AAA AAC CCT GTG TGT GTT CGT CGC TCG GTC GTC ATA... Y: ...AAG AAC CCT GTG TGT GTC CGT CGC TCG GTC GTC ATA... Z: ...AAG ATC GTC ATA... . AAC CCT GTG TGT GTC . CGT CGC TCG . we see that two mutations separate Y and Z from X, whereas only one mutation separates Z from X and Y. Therefore the hypothesis that the lineage of X branched o earlier and thus Y and Z are more closely related to each other than either of them is to X seems more credible than the two alternavive hypotheses.

REGULAR POLYTOPES

The rst section of this paper dealt with visualizing an abstract mathematical object in two and three dimensions. Polytopes, on the other hand, may in general have any dimension by denition; because the human brain has evolved in a threedimensional world, visualizing higher dimensions is not easy. 2.1 Regular polytopes of dimensions 23

Regular polygon {p} can have any number p of vertices. Its corner angle is 2 (1 p ). Regular polyhedron {p, q } has q regular p-gons meeting at each vertex; the midpoints of their edges are the vertices of {q }. This is possible exactly when 2 q (1 p ) < 2 , or (p 2)(q 2) < 4, so we have the ve Platonic solids {3, 3} {4, 3} {3, 4} tetrahedron cube octahedron (4, 6, 4) (8, 12, 6) (6, 12, 8) {5, 3} {3, 5} dodekahedron icosahedron (20, 30, 12) (12, 30, 20)

where (V, E, F ) are the numbers of vertices, edges and faces, respectively. The dihedral angle (between the planes of adjacent faces) is 2 arcsin cos q /sin p . 2.2 Regular polytopes of dimension 4

Polytope {p, q, r} consists of a number (C ) of cells {p, q }, r 3 of them around each edge [2], [3]. Hence r times the dihedral angle must be less than 2 , so cos q < sin p sin r . It follows that there are six possibilities: {3, 3, 3} {4, 3, 3} {3, 3, 4} simplex hypercube co-cube (5, 10, 10, 5) (16, 32, 24, 8) (8, 24, 32, 16) {3, 4, 3} {5, 3, 3} {3, 3, 5} (24, 96, 96, 24) (600, 1200, 720, 120) (120, 720, 1200, 600) 4
284

The midpoints of all edges meeting at a vertex are the vertices of a regular polyhedron {q, r}, the vertex gure. For example, a vertex-tetrahedron can accomodate four cells with q = 3. We now construct the polytope {5, 3, 3}. Starting with one dodekahedron, we rst add twelve cells at each face. Then we add a second layer of cells, one into each dent left in between; there are twenty of them. Now the twelve outmost faces of the rst layer are still visible; we cover these with a third layer. The fourth layer of cells must be added one edge down, and there are thirty of them. Continuing this way, we see that we need 120 dodekahedra for this polytope; hence its other name, 120-cell.

References
[1] Atela, P. (1991) Bifurcations of Dynamic Rays in Complex Polynomials of Degree Two, Ergodic Theory and Dynamical Systems 12, pp. 401423 [2] Berger, M. (1977/1987) Geometry III (Translation of G eom etrie), Springer-Verlag [3] Coxeter, H.S.M. (1969) Introduction to Geometry; Wiley, New York [4] Carleson, L. and Gamelin, T.W. (1993) Complex Dynamics; Universitext, Springer-Verlag [5] Kauko, V. (2000) Trees of Visible Components in the Mandelbrot Set, Fundamenta Mathematicae 164, pp. 4160 [6] Keller, K. (2000) Invariant Factors, Julia Equivalences and the (Abstract) Mandelbrot Set; Lecture Notes in Mathematics #1732, Springer-Verlag [7] Lau, E. and Schleicher, D. (1994) Internal Addresses in the Mandelbrot Set and Irreducibility of Polynomials, Preprint #19, Institute for Mathematical Sciences, Stony Brook [8] Peitgen, H.-O. and Richter, P.H. (1986) The Beauty of Fractals, Springer-Verlag

5
285

286

VIRTUAL AND REAL STATES. INNER STRUCTURE OF THINGS AND OBJECTS

TOMASZ MICHNIOWSKI

Name: Tomasz A. Michniowski (b. Radomsko, Poland, 1963) Address: Division of Physics, Catholic University of Lublin, Raclawickie Ave 14, 20-038 Lublin, Poland, email: pool@kul.lublin.pl Fields of Interest: relativistic and Quantum Cosmology and Astrophysics, Philosophy of Science, Methodology of Physics Main publications (in reverse order): "Mathematical Representations of Physical Reality in Quantum and Relativistic Model", a part in a book: Philosophical and Scientific Elements of the Universe's Description. Part 2, ATK, Warszawa, Poland 1998, p.121-39 "Geometry of Space-time; Groups of Symmetry as the Representations of Physical Laws in Cosmological Models", ISIS-Symmetry, 1999 "Questions and Working Hypotheses in Recognising the Universe", The Man and the Nature 6, p.93-108, Lublin, Poland 1997 "Basic Research Schemes in Constructing the Models of Universe", Yearbook of Philosophy of Catholic University of Lublin, Part: Philosophy of Nature 44 , 3, p.75-87, Lublin, Poland 1996 "Inflationary Universes and Serial Model's Conception", KUL, Lublin, 1993

Abstract: Constructing the models of physical reality takes place in science with use of specific method. This method is based on general mathematical features of the Nature, called 'mathematization' and 'idealisation'. Thanks to it, there is possible to construct quite good emulators of reality, when having only limited knowledge about itself. The models are cognitive efficient, what means, that they are able to product verifiable predictions on Nature's behaviour. Symmetry is a tool for all formal verifications of these predictions and, as well, a tool for planning experiments for empirical verification. In the article there is shown, how in practice the method is being used in science. An example of advanced specific speculation with use of some features of main mathematical models of reality is being shown. It is pointed, what kind of strict scientific results may we expect when using the method. Cognitive importance of the procedure is being emphasised. Figures illustrating the content are included.

287

1. SCIENTIFIC RESULTS

METHODS

ANS

COGNITIVE

Scientific recognising the universe is possible thanks to a specific feature of the physical world called 'mathematization'. Well-known Einstein's joke about non-malicious way of creating the world by God identifies main idea of 'mathematization'. In simple words, the feature consists in being able to construct mathematical structures, which are able to emulate behaviour of physical objects up to the high level of conformity. These structures, called models of reality (or, shortly: models), allow calculating inside themselves. Results (often called 'predictions') derived in analytical way, are fully experimentally verifiable. Sets of predictions derived form models, until being checked experimentally, are usually called 'theories'. Sets of results verified positively in experimental way are known as 'scientific knowledge'. These relationships are obvious at all levels of scientific recognition. Both in the extremely simple, separate cases, as well as in very complicated physical situations, recognition with use of mathematical models goes always the same way. For example, simple, local model, created for emulating the process of falling down small balls (and being able to predict only point and/or moment of the fall of dropped ball), uses the same mathematics for emulating the phenomenon, as does it each one global cosmological model, constructed for finding behaviour of groups of galaxies in the far space. The only difference lies in the scale of mathematical complexity of both models and subtlety of accepted assumptions. Constructing models, of course, is not the easy task. It usually starts from very few separate information or only intuitions we have on the area of physical reality we intend to discover. The method of science (see Fig.1) allows us finally to reach the model and to derive its predictions (testable in the experimental way), which may lead us to new knowledge we hadn't got earlier.

Figure 1: Scientific method. Discovering unknown areas of physical reality in this way may be efficient enough thanks to special feature of mathematical models called sometimes 'idealisation'. It concerns in making models cognitive efficient even if their structures are not especially reliable. This means, that there may exist many models emulating the same set of

288

phenomena in a quite good way. All of them may be useful for discovering new physical facts in the same area of investigated reality. For better explanation see Fig. 2. Here are shown cosmological evolutions of the universe predicted by two different models: deSitter's and Opened Friedmanian ones. Both of them have been constructed in the same mathematical environment, but with use of different sets of dynamical assumptions. The model of deSitter (from 1917) assumes, that space-time of the universe is empty (no matter at all). Friedmanian model (from 1920) accepts the presence of matter of isotropic distribution inside the space-time. Both models predict expansion of the universe. It is easy to perceive, that both predictions unify for far future of cosmic evolution, when the average density and pressure of matter in Opened Friedmanian world vanishes and becomes close to assumed in deSitter's model.

Figure 2: Cosmological evolutions of Friedman and deSitter. Another non-trivial property of mathematical constructions and mathematics as a whole is impossibility of lying. In a well-known example [i. e. A = B 3A = 3B 3A B = 2B 2A B = 2B A 2A 2B = B A 2 (A B) = (A B) 2 = 1. Notice, that (A B) = 0, because A = B (first term of calculation)], there appears a trial of banned operation (dividing by zero). As a consequence of it, in last term there appears a hard contradiction, which is not allowing any further continuation of calculation. All these mentioned (and others) features of mathematical structures being an element of scientific methods, allow us to feel relatively safe in our trials of discovering unknown physical facts. The knowledge reached in this way (including final experimental verification of all predictions of models) is really reliable and unquestionable. Usually is yet much wider and richer than we expect. Except of having answers for what we wanted to know, we get much more. Large part of the knowledge obtained in this way is completely unexpected, mostly incomprehensible and at least at the very beginning seems to be incoherent with the rest of science. What is the reason?

289

2. MAPS AND SCIENTIFIC KNOWLEDGE

Mapping the land in geography is not easy operation. One must create the model (usually graphical one) of chosen area, respecting all details of its surface. Of course, it's possible only when we know exactly each one point of it. If not, on the map there exist 'blank spots' (for example see old maps of African interior from the beginning of 20-th century). In science, situation is quite different: 'blank spots' never occur. If we try to construct the 'map' of investigated area, we are not able to represent all details of reality by appropriate mathematical objects, because we don't know too much about them (and usually are even not conscious their existence). We rather try to create the model with use of only a few facts we suspect we know and are able to represent by mathematical objects. These objects, existing in their specific mathematical environment, thanks to the features of 'mathematization' and 'idealisation', do fix the main structure of the model (i. e. the 'map'; in mathematics, a map is each opened sub-set and a function, which represents the sub-set in Banach space. The set, together with all sets of maps being modelled in Banach space, is a manifold. Manifolds are natural environments for smooth functions, essential for emulating phenomena in all realistic models of physical spacetime). If we are really lucky, we sometimes can guess a set of assumptions and boundary conditions for this structure, appropriate enough to allow the model quite good emulation of reality. This means (see Fig.3), that 'map' we constructed, shows us investigated area of reality, but only approximately, not exactly (as geographical maps should do).

Figure 3: world and maps.

290

As luckier we were finding mathematical representations (for facts we had chosen as representative for investigated reality) and defining the set of conditions for the model, as better our 'map' emulates details of world it should do. This means, that the 'map' allows us to spot physical facts (mathematical details on the 'map') we couldn't find in any other way at all (note that these facts, being formally predictions of the model, must be verified in experiments until being accepted as elements of scientific knowledge). Of course, less luck or formal errors we could do in the process of constructing models imply that our 'map' emulates reality in a wrong way, up to the level of full fiction. Such situation means, that the 'map' shows reality, which doesn't exist within the space-time of our universe. Practical usefulness of such model disappears than.

3. SURPLUSAGE OF MATHEMATICS
The cognitive limitations for scientific model's domain are the consequence of assumptions we must accept in the process of constructing each one model. Arbitrary conditions we bring, don't deprive the mathematical structure of the model of possibility of emulating much wider areas than same model is intended to do. All models we use in physics and natural sciences have been constructed with use of the mathematics of very special kind, namely highly regular one. The general structure of manifolds we permanently use for emulating 4-dimensional space-time, is very comfortable. It assures the best possible environment for smooth enough functions (thanks to it analytical calculations are possible) and, as well, requires no more general environment for itself (in mathematics, manifolds often have been called self-contained structures). Very interesting question is - what do emulate mathematical structures of less regular kind than manifold is? Fig. 4 shows the situation. Manifold-like structure is good enough to emulate relativistic 4-dimensional world. But in whole body of mathematics, the area of regularity is of a very small measure. It would be strange, if the physical world would be created in such primitive way to have to be represented by such small part of mathematics only. Increase of scientific knowledge during the 20-th century shows, that physical phenomena have got much more complicated roots than classical only. Mathematical models of their mechanisms have got quite different nature. For example, model of quantum mechanics needs allowing, in general, the complex multidimensional spaces, instead of real ones.

291

Figure 4: Physical universe and its mathematical model.

4. QUANTUM AND REAL STATES

Scientific experiences of last century teach, that mechanisms of all physical phenomena we know are, in general, not explainable at the ground of classical paradigm only. This means, that everything what we can perceive and measure in physical universe, is only a part of much more complicated system of dependencies, not accessible to our senses. More simplified illustration of what we speak, is each quantum system; for example: quantum model of an atom. Fig. 5 shows measurable levels of energy in the atom. Understanding and possibility of making predictions of its behaviour is being possible thanks to Schroedinger equation only, solved for the specific set of boundary conditions. The mathematical environment of this model is not classical one. So, speaking clearly, we do use mathematics of complex spaces for predicting real behaviour of physically perceived system, called atom. The good question in this context is "how really the atom looks like"? It is hard to call "atom" the diagram shown at Fig. 5. But we really haven't got any other, different in quality, visualisation! It seems, visible world and phenomena in it, are only one of possible cross sections of much richer 'super reality' (see Fig. 6). Knowledge about it is highly desirable in context of our scientific and technological needs. Such statement, one hundred years ago was shocking, now presents itself as serious difficult technical and analytical scientific problem only. It is namely extremely hard to construct and to derive mathematical structures of unknown and (even potentially) not measurable areas of reality. It is difficult to solve them and to localise their separate predictions as well. But it isn't impossible!

292

Figure 5: levels of energy in the atom.

Figure 6: Quantum phenomenon.

5. PRECAUTIONS: EXPERIMENT AND SYMMETRY

There are two important problems connected with practical applying discussed method of science. The first one is: "how far may we be sure, mathematics is not showing us not-existing worlds?" The answer is simple: as far, as our physical experiments are able to verify predictions of models. Fig. 6 shows what part of 'super reality' may be checked by experiment. Until experimental methods are applicable (Nobody says it's easy task. Sometimes experiments seem to be more complicated than predictions of models, which should be verified), the danger of loosing scientific dimension of recognition doesn't exist. Up to this level, science is not being imminent by philosophy.

293

The second concern, connected with previous, is: "how can we differentiate predictions of physical meaning from other coincidences or interferences, without any cognitive dimension?" In this case, the solution lies in symmetries. Careful exploration of mathematical structures for identifying symmetrical structures leads us to tools, which allow us practically to find essential (in cognitive sense) objects among the formulas and equations of local importance (more about this was explained in a previous ISISSymmetry World Congress in Haifa 1998, see conference papers for details).

6. IDEAS
Cognitive efficiency of discussed scientific method manifest in practice. Historically, this efficiency had been tested many times with good result. It is interesting whether this method's opportunity of applying is of temporal or universal kind. For test, let's have now some speculations. For the moment we would try to apply our method for finding results of cognitive (or not) dimension, using the scientific facts from 'the edge' of present knowledge. In simple words, we would try to discover 'something' we don't know yet. If result is satisfactory, we will be able to accept main trains of our presentation. Let's take into account the problem of entropy. Equations of physics, including the most fundamental ones (for example, mentioned Schroedinger equation), don't distinguish the arrow of time. Therefore, physical phenomena ought to be reversible in time. Unfortunately, in real world we don't observe it: broken glass is not going to consolidate again and old houses don't transform themselves into new ones. For justifying the situation, we define the quantity called entropy, understood as the measure of sets of permissible configurations of the system. Second Law of thermodynamics says, that entropy of adiabatic system never decreases, what implies, that only some configurations of the system are permissible in physical processes (consequence of it is the arrow of time). This interpretation is a source of serious problems of modern physics. Steady rise of entropy forces the whole universe to be of zero-like entropy kind at the beginning, what is probabilistically negligent. What more, rise of entropy is not going to reverse after turning the arrow of time away (Exact explanations of this mechanism one can find in Penrose's books, for example "Emperor's new mind). All trials of justifying the effect by respecting expansion of space-time lead to other problems, for example to non-existence of 'white holes' (After turning away the arrow of time, black holes should spit the energy and matter away - to inverse themselves into 'white holes' -, what doesn't happen) or lack of the effect of reverse of time in very strong gravitational fields. This suggests, that rise of entropy is not connected with expansion of space-time, what was conceptual included into the definition of the term 'entropy'. It seems, such model produces contradictory predictions and mayn't be accepted for emulating evolution of the universe in a whole.

294

7. SPECULATIONS
Let's agree, that both models: relativistic and quantum ones, are simplifications of the more general structure (Finding it exactly, is a dream of probably all physicists in the world). Hoping, that 'super reality' in which our world does exist is physically and mathematically uniform, and, in general, it is able to create their models, we will try to 'guess' and to verify in formal way the mechanism of entropy. We'll do it using the scheme of discussed method, accepting separate elements of relativistic and quantum models as not contradictory. According to what we said before, we will now consider the 4-dimensional space-time of our universe as being the quasi-continuous (outside of singularities, of course) and smooth enough ensembles of quantum sub-space-times. Local and global curvatures of it will be interpreted as non-zero gravitational fields. The observer, travelling through this space-time, always perceives one real stratification (sub-world) from the quantum set of them. There are two quantum procedures of modelling systems. The first one describes the general evolution of quantum systems and is not breaking the symmetry of time. The second one, used for determining possible real manifestations of the system (levels of energy, by which the system is able to manifest their existence in physical world), is breaking symmetry of time permanently. It seems, the arrow of time comes out only in the act of reduction to one of real sub-worlds. Our observer, walking through the space-time, admires the reality of sub-worlds. Independently on the geodesic line he chooses, he always, with each step along the line, must perceive something. Each act of such perceiving is equal to observer's reduction to one of sub-worlds, what is connected with breaking the time-symmetry. Number of acts of breaking is proportional to the value of entropy measured by the traveller. Here we have another sense of the term 'entropy'. The quantity may be understood now as a parameter of reduction procedure. Let's check whether such interpretation helps in aborting mentioned cosmological problems connected with time-symmetry braking. Let's suppose that our observer starts his travel through the universe from the point very close to Big Bang. The space-time there is being described by extremely high values of Ricci tensor. This implies rapid expansion of the universe (Thanks to a specific shape of effective potential of vacuum and repulsive character of forces, existing in this symmetry of space-time). But each expansion ends very quickly in the Big Crunch singularity or in deSitter epoch, if only in the closest area of Big Bang there exist any, even the smallest, curvature of Weyl-type (Riemanian geometry distinguishes two types of curvature: Ricci- and Weyl-like). The second one is being understood as gravitational tide-force. To allow our observer travelling through the non-trivial world, we must accept the fact, that the whole area close to Big Bang is permanently flat in the Weyl sense (No world, which doesn't satisfy this condition, is able to exist much longer than the period

295

comparable to Planck time, i. e. 10-44s.) Any measurable Weyl-like curvatures are possible only quite far from Big Bang. Penrose suggests, that reduction procedure may happen when local Weyl-like curvature is more than so-called Planck mass (i.e. local gravitational gradients are strong enough). If he's right, our hypothetical observer will be able to notice growth of entropy as faster, as further from Big Bang he travels. At the beginning of his journey, in the almost completely flat universe, the traveller has nothing to observe, because reductions to real sub-worlds are not possible. Further from beginning singularity, where the universe comes not to be so flat, reductions are permissible and the observer is able to admire reality around, measuring in the same time the growth of entropy with his each next step along the geodesic line he had chosen. Let's notice, that if our observer is trying to reverse the line he came, he won't be able to visit again sub-worlds he admired before (reductions to the same sub-worlds are probabilistically impossible). Still the entropy he is permanently measuring, will decrease again when he approaches to Big Bang. This happens, because each reduction that observer notices, eliminates rich set of reductions, he potentially would be able to suffer a step earlier. Approaching to Big Bang by any possible geodesic line restores amount of sets of potentially possible reductions in different (not flat) areas of space-time. In this visualisation, 'further' to Big Bang means 'later' in the observer's time measure. This is equal to the growth of entropy. The effect of growth of entropy will happen as well each time when the observer approaches to local singularity (for example black hole) or other strong gravitational field source. Responsible for it is the fact, that strong fields reduce rapidly number of geodesic lines in the area (and even, after passing the Chandrasekhar border, cut off all the geodesic curves at all), what imposes strong limitations for set of potentially accessible reductions the observer could suffer.

8. RESULTS
As being announced, we tried to discuss importance of the term 'entropy', which is, in its classical concept, a source of serious problems of scientific and cognitive dimension. For clarifying the problem, we used traditional scientific method. We shaped the general outline of the new model, assuming only general conformity of well-known scientific models and mathematical character of investigated 'super reality'. We used as well some elements of presently accepted scientific knowledge and one hypothesis of quite nice likelihood of truthfulness. As expected results we had: another and more universal understanding of the term 'entropy'

296

elimination of conceptual and model contradictions, which were the consequence of traditional definition of entropy preservation of old models of thermodynamics, which work in the same way for both definitions of entropy. nothing discovered here is contradictory with what we know up to now Unexpected, but, as always, occurring results (scientific predictions) of physical importance, accessible for experimental verification, are: thermodynamical evolution of the universe satisfies symmetry of time (entropy increases after inverting the arrow of time), as closer to beginning singularity, as smaller is probability of virtualisation (reduction) the energy in real state, irreversible nature of reduction procedures is obvious, entropy increases near all sources of strong gravitational fields, Weyl-like flatness of space-time near the Big Bang singularity is necessary for existence of the universe, limitations for tracing time-like geodesic lines through the area of quantum stratification of the universe are of the Weyl-like curvatures shape.

9. COMMENTS
Let's finally collect everything what we discussed and discovered. The result is being shown at Fig.7. Phase space of vacuum curvatures, with diagrams drown for opened and closed Friedman-like evolutionary scenarios, is valuable and verifiable scientific result. Occurrence of it confirms correctness and cognitive efficiency of discussed scientific method. Explaining advanced problems of modern science seems to be, again, impossible with use of formalisms and complete, systematic knowledge only. This is a manifestation of Goedel statement (it is not possible to prove each one theorem, which can be formulated in specific system, with use of logical tools of this system only). Fascinating features of reality mathematization and idealisation give us opportunity of having deep insight into the nature of phenomena of our universe, even if we are not pretty sure what are we looking for.

297

Figure 7: phase space of curvatures of Friedman universes.

References
Penrose R., "Emperor's new mind", Oxford University Press 1989 Hawking S., Penrose R., "The Nature of Space and Time", Princeton University Press 1986 Michniowski T., "Cognitive function of physical models", in: Philosophical and Scientific Elements of the Universe's Description. Part 5, UKSW, Warszawa, Poland 1998, p.121-39.

298

TESSELATIONS OF EUCLIDEAN, RIEMANNIAN AND HYPERBOLIC PLANE

RADMILA SAZDANOVI, MIODRAG SREMEVI

Name: Sazdanovi, R., Mathematician, (b. Belgrade, Serbia, Yugoslavia,1977) Address: The Faculty of Mathematics, University of Belgrade, Studentski trg 16, 11000 Belgrade,Yugoslavia E-mail: seasmile@galeb.etf.bg.ac.yu Fields of interest: Geometry, Differential Geometry, Topology (Computer Science). Publications and/or Exhibitions: 1.Knezevi, I., Sazdanovi, R., Vukmirovi, S., (2002) L2Primitives, Userguide, Mathematica package, http://www.mathsource.com/Content/WhatsNew/0211-879 2. Knezevi, I., Sazdanovi, R., Vukmirovi, S., (2002), Visualization of the Lobachevskian Plane, Visual Mathematics, electronic journal, http://members.tripod.com/vismath7/sazdanovic/home.htm

Name: Sremevi, M., Physicist, (b. Kragujevac, Serbia, Yugoslavia, 1976) Address: Institute of Physics, University of Potsdam, Am Neuen Palais 10, D-14469 Potsdam, Germany E-mail: msremac@agnld.uni-potsdam.de Fields of interest: Polyhedra, Planetary sciences

Abstract: The very first human perception of this world was plane. Then one sphere among others in the universe. Or is it just a small part of the hyperbolic space? Different features of Euclidean, Elliptic and Hyperbolic geometry are presented through all types of planar edge-to-edge tessellations with regular polygons as tiles (not all congruent) and vertices of the same type, including regular, uniform, non-uniform and coloured tessellations. The most intriguing, hyperbolic geometry is the main topic of the algorithm implemented in Mathematica 4.0 that provides tools for creating and drawing tessellations and their animation under different isometries, in both Poincare models and Klein disk model. The package provides additional data such as: geometry and type of tessellation, number of all possible realizations, angles of tiles and transformation rules between vertices upon which the tessellation is constructed.

1 INTRODUCTION
According to the Oxford Dictionary the word tessellate means to form or arrange small squares in a checkered or mosaic pattern. Being familiar or not, with this or any other (more) formal definition, everyone can recognize tilings like brick walls and tile floors. 299

Tessellation as an art predates human history- Platonic polyhedra were well-known in antiquity and a toy regular dodecahedron was found in Padua in Etruscan ruins dating from 500 B.C. Excellent examples of tessellations can be found in Moorish architecture in Spain and Islamic architecture in the Middle East; not to mention Japanese and Chinese designs. Artists such as Albrecht Durer and Pierro della Francesca made drawings of many of the semi-regular polyhedra. Johannes Kepler was the first to give a complete description of such figures in his Harmonices Mundi which appeared in 1619. Except this initial study, some formal mathematical investigation took place before the end of 20th century. We must mention Grunbaum, Shephard, Robin, Sommerville, Andreini, M.S. Escher and that much was done also by chemist and crystallographers .

2 DEFINITIONS
Def. 1. A plane tiling is a countable family of polygons P1, P2,... called tiles such that: 1. Their union is the entire plane 2. Interiors of the tiles are pairwise disjoint Def. 2. An edge-to-edge tiling is plane tiling where each two tiles intersect along a common edge, common vertex or not at all. Def. 3. A vertex of a tiling is said to be of type n1, n2,,nN or that (n1, n2,,nN) is the vertex type (configuration) if the polygons about this vertex in cyclic order are an n1gon, an n2-gon, . . . and an nN-gon where N is the number of polygons at each vertex. Def. 4. An edge-to-edge tiling is called Archimedean if 1. Each tile is a regular polygon (not all congruent) 2. All vertices are of the same type. Def. 5. A regular tiling is an edge-to-edge tiling whose tiles are congruent to a single regular polygon. The other name is monohedral tiling with regular polygons as tiles. Def. 6. A semi-regular or uniform tiling is an Archimedean tiling whose symmetries are transitive on its vertices.

Figure 1. (4,4,4,6): two uniform (a, b) and non-uniform realization (c): colours make distinction between polygons that can not be mapped to each other by global symmetries.

300

The following array shows the relations between different types of tessellations: Regular Uniform Archimedean Edge-to-edge Plane Tessellations For the rest of this paper we restrict our attention to the tessellations whose tiles are regular polygons and vertices are of the same type. This gives us an opportunity to define tessellations with their vertex configuration, although this correspondence is not always one-to-one. The vertex symbol uniquely determines the type of a tiling in Euclidean and Elliptic plane (except (3,4,4,4) which has two realizations: rhombicuboctahedron and a non-uniform polyhedra, Fig. 3). However, non-congruent tessellations in the hyperbolic plane may have the same vertex symbol (Figs 1 and 2).

3 THEORY AND ALGORITHM

The general solution for constructing tessellations is based on a few rather simple ideas that we shall briefly explain. The algorithm can be divided in two logical parts: Step 1. Vertex configuration transformation rules for neighbouring vertices Step 2. Transformation rules for neighbouring vertices graph tessellation Step 1. The starting point of the algorithm and the only input is the vertex configuration. In order to find all possible realizations for a given vertex configuration we developed a very general algorithm. Our initial assumption is that each polygon belongs to a different equivalence class (each presented with different colours in our figures) under the symmetries of the whole tessellation (global symmetries) and then we, if necessary, decrease the number of classes. In each iteration we look for transformation rules i.e. mappings of the corresponding vertices and check if they really do form a tessellation. In this way we are able to find both uniform and non-uniform tessellations i.e. all Archimedean tessellations including enantiomorphic realizations (Fig 4) and all coloured realizations (Fig. 2.a, 2.b).

Figure 2. (3,3,3,3,3,4): uniform realization (a) and its coloured version (b), and one of the non-uniform realizations (c). Step 2. Obtained transformation rules posses two important properties: they are uniquely defined and sufficient for creation of the whole tessellation starting from any

301

vertex. So, once we have transformation rules we construct the graph and tessellation at the same time, using following ideas: Vertex configuration imposes geometry onto the graph. Hence, the graph can be realized only on a 2-dimensional, simply connected, unbounded, complete Riemannian surface of constant curvature, i.e. Euclidean, Elliptic or Hyperbolic plane [2]. In constructing tessellations problems arise on local and global level [5]. On the local level we consider only those images of the original tile obtained by a sequence of symmetries that move it in a way that the tile and its image are not totally disjoint. So the local tessellation problem is the question if these adjacent tiles cover a certain area completely enclosing the original one without gaps and overlapping. Global tessellation problem is the same question involving the whole 2-dimensional space. The main point is that in order to determine if the tessellation exists or not, we need to solve only the local problem.

Figure 3. (3,4,4,4): uniform tessellation (a, b) and non-uniform (c, d). This is the very core of the algorithm and additional calculations are necessary for drawing tessellations [1, 3, 7]. The algorithm contains methods for drawing Euclidean and Elliptic tessellations (wired and coloured model or a corresponding polyhedra, Figs 3, 4c) and for hyperbolic tessellations Mathematica [6, 8] package L2Primitives [4] is used (both Poincares and Klein disk model). Furthermore, it is possible to apply various isometries and obtain interesting animations. As a side effect, algorithm provides additional useful information such as geometry of tiling, type of tessellation, number of all possible realizations, angles of tiles [3] and complete listing of transformation rules. The detailed description of the algorithm and the theory behind will be published elsewhere.

4 GALLERY

302

Figure 4. Uniform enantiomorphic tessellations: (3,3,3,3,6) Euclidean (a, b), (3,3,3,3,5) Riemannian (c), and (3,3,3,3,7) hyperbolic in Klein disk model (d).

Figure 5. Tessellations with a single realization: uniform (4,8,10) and (3,6,4,6) (a, b) and non-uniform (3,4,3,4,4) (c). Tessellation with multiple realizations (3,3,3,4,3,4): two uniform (d, e) and a non-uniform (f).

References
1. Coxeter, H.S.M., (1988) Non-Euclidean Geometry, The Mathematical Association of America 2. Goodman-Strauss, C., Personal communication. 3. Har'El, Z., Uniform Solution for Uniform Polyhedra, Geometriae Dedicata 47 (1993), 57-110 4. Knezevi, I., Sazdanovi, R., Vukmirovi, S., (2002) Visualization of the Lobachevskian Plane, Visual Mathematics, e-journal, http://members.tripod.com/vismath7/sazdanovic/home.htm 5. Magnus, W., (1974) Non-Euclidean tessellations and their groups, Academic Press 6. Maeder, R., Programming in Mathematica, (1991) Addison-Wesley Publishing Company 7. Paraslovov, V.V., Tihomirov,V.M., (1997) Geometrija, MCNMO 8. Wolfram, R., The Mathematica Book, Mathematica Version 4, Wolfram Media & Cambridge University Press

303

304

A REMARKABLE SPIRAL
ANNIE VAN MALDEGHEM

Name: Annie Van Maldeghem, Mathematician, (b. Gent, Belgium, 1948). Address: Hogeschool voor Wetenschap en Kunst, Departement Architectuur, Sint-Lucas Gent, Belgium. E-mail: a.vanmaldeghem@archg.sintlucas.wenk.be Fields of interest: Ruled surfaces, minimal surfaces.

Abstract: This paper is the result of some mathematical amusement. In Dutch we call this with a German word: spielerei. The spiral that is subject of this talk is completely anonymous. At the occasion of Mat mium it makes sense to call it atom-spiral. This double spiral has a very interesting symbolic meaning and a lot of variations on its elegant shape are possible. All these spirals are related to the Archimedes and hyperbolic spirals. Moreover, they provide the possibility to design the character e surrounded by an appealing curl, comparable to the snail @. All the spiral drawings are computer generated by executing an Autolisp program (AutoLISP runs within the environment of AutoCAD). The subject is in harmony with the decorative composition in the superb Hotel Van Eetvelde. The elegant curved lines, the sweeping twists and twirls, the curls and the swirls that are characteristic of Art Nouveau are creating a perfect setting to put mathematical spirals into the scene.

1. THE ATOM SPIRAL

The atom-spiral is depicted in fig.1. This drawing is computer generated by executing an AutoLISP program. AutoLISP is a parametric programming language, the LISP variant, which is running within the AutoCAD environment. To write a program for this graph I used its polar equation: + r=k , k R0 (1). This polar equation is plotted with 1 respect to a pole o and a polar axis X. 305

Fig. 1

The program calculates the polar co-ordinates of a large number of points (r, ) on the spiral (r = directed radius and = polar angle; -values chosen between 0 and 1 result in negative r-values). Then the program draws line segment joining two consecutive points, joins all these line segments together to one polyline and finally changes the polyline into a spline. The range of is R \ 0 and the function defined by equation (1) is a monotone (decreasing) function. This means that equation (1) leads to a spiral that consists of two branches. It is easily verified that the spiral has two asymptotes: a straight line and a circle. One end of the interior branch tends to one half of the straight line, while the other end approaches the circle as limit from the inside; one end of the exterior branch tends to the other half of the straight line, while the other end approaches the circle as limit from the outside. I first noticed this spiral in 1966 in a mathematical magazine for youngsters (see references). I fell for this spiral because of its remarkable property to possess a circle as an asymptote: both branches of the spiral are winding themselves around the pole coming closer and closer to the asymptotic circle without ever touching it, from either side. And when we agree to consider a circle as a symbol for perfection we can see this spiral as a symbol for the never-ending search for perfection, which will never be achieved. In other words, the curve seems to tell that the hunt for perfection is to try for the impossible.

2. GENERAL SET AND ITS SUBSETS

Equation (1) has following form: r =
m + n , m, n, p, q R; (p, q) (0, 0) (2) . This p + q equation, in 4 homogenous parameters m, n, p and q describes a set of spirals, which falls apart in 4 subsets.

2.1 Subset 1 circles

If the parameters m, n, p, q fulfil

m n = 0 (p, q) (0, 0), then equation (2) changes p q into r = k (k R). This equation leads to a set of concentric circles (circle with radius = zero included), if k runs through R. So, a first subset of the set described by (2) consists of circles.

306

2.2 Subset 2 Archimedes spirals

0 p = 0 ,then p q equation (2) changes into: r = k ( a), k R0, a R (2.2). For any value of a (k is used to scale), if the range of is R, this equation represents a double Archimedes spiral which is directly similar to the one depicted in fig. 2.2. So, a second subset of the set described by (2) consists of (double) Archimedes spirals. The pole is a point of any spiral in this subset and the parameter a can be used to rotate.
If the parameters m, n, p, q fulfil

m n

Fig. 2.2

2.3 Subset 3 Hyperbolic spirals

If the parameters m, n, p, q fulfil r= m n p q 0 m = 0 , then equation (2) changes into:

k , k R0, b R (2.3). For any value of b b, (k is used to scale), if the range of is R \ b, this equation represents a double hyperbolic spiral which is directly similar to the one depicted in fig. 2.3. So, a third subset of the set described by (2) consists of (double) hyperbolic spirals. Any spiral in this subset has a straight line as an asymptote. Moreover, the pole is an asymptotic point, which Fig. 2.3 means that any curve of this subset spirals down to the pole without ever reaching it; the parameter b can be used to rotate.

k>0 b=0

2.4 Subset 4 variations on the atomspiral

m 0 0 , p q p 0 a then equation (2) passes into: r = k , k R0, a, b b If the parameters m, n, p, q fulfil m n
307

Fig. 2.4 a

 R, a b (2.4). For any value of a and b (k is used to scale), if the range of is R \ b, this equation represents a spiral which consists of two separate branches. All spirals described by equation (2.4) have two asymptotes: a straight line and a circle. So, a fourth subset of the set described by (2) consists of double spirals, which are all of the same type as the atom-spiral (a = 0, b = 1). The pole is a point on the interior branch of any spiral in this subset. Fig. 2.4 a shows one of the many variations on the atom-spiral (a = /2, b = + /2). Fig. 2.4 b shows spirals corresponding to random values for the parameters a and b: they are (in general) not at all similar.

r=k

+1 1.5

r=k

1 0.5

Fig. 2.4 b

k>0 r = k 0.618 + 0.382

0.618 r = k + 0.382

3. Variations on the atom-spiral: properties

But if a b = constant: all spirals are directly or indirectly similar. Fig. 3.1 a, fig. 3.1 b and fig. 3.1 c show 3 spirals all corresponding to the same k-value. If they were represented with respect to the same pole and pole axis the spirals in fig. 3.1 a and fig. 3.1 b would coincide up to a rotation; the spirals in fig. 3.1 a (or fig. 3.1 b) and fig. 3.1 c would coincide up to a rotation and a reflection.

ab=1 ab=1 ab=+1 Fig. 3.1 a Fig. 3.1 b 308 Fig. 3.1 c

The smaller a b, the larger the loop and the better the asymptotic circle is suggested (see fig. 3.2).

ab=8

a b = + 0.3

Fig. 3.2

In addition, there is the amusing fact that if a b > 0, the corresponding spirals described by (2.4) can be used to design the character e with an appealing curl just like in the snail @ (see fig. 3.3).

Fig. 3.3 Any variation on the atom-spiral, a (double) Archimedes spiral and a (double) hyperbolic spiral all belong to a set of spirals described by equation (2) while the m n 0 . This relationship is visualized in fig. parameters fulfill the same condition p q 3.4. Following spirals were chosen: * variation on the atom-spiral: r = k + 2 R0 (3.4 a) (3.4 b) (3.4 c)

* Archimedes spiral: r = k ( 2.77), R * hyperbolic spiral: r = k" , R0

k' k k" = = . For each diagram the range of 1 10 100 is restricted to subsets of the given domains in order to optimize the similarity between The scaling factors k, k and k fulfill: 309

the interior branch and the Archimedes spiral and between the exterior branch and the hyperbolic spiral.

Interior branch

Fig. 3.4 b Archimedes spiral

Exterior branch

Fig. 3.4 c Hyperbolic spiral

Fig. 3.4 a

Spiral (3.4 a) was chosen because one of the polar angles which corresponds to the

1+ 5 (golden number). This 2 angle corresponds to an angle 291. The line that joins the pole and this double point makes an angle 69 with the polar axis (see fig 3.5 a). One of the polar angles, which corresponds to the double point on the Archimedes spiral (3.4 b), is (2.77 /2) radians or (159 90). So, the line, which joins the pole and the double point on the Archimedes spiral (3.4 b), also makes an angle 69 with the polar axis (see fig 3.5 b).
double point of the loop is * radians where =

310

 291

69 159

69

Fig. 3.5 b Fig. 3.5 a

Remark
Equation (2), in polar co-ordinates, alludes of course to a similar equation in Cartesian mx + n co-ordinates: y = , m, n, p, q R; (p, q) (0, 0) (4) . This equation defines the px + q homographic function, which describes a straight line in case the parameters fulfil m n m n = 0 (p, q) (0,0) or: 0 p = 0 . If the parameters fulfil either: p q p q 0 p 0 , the homographic function p q describes a rectangular hyperbolar. The two congruent branches of a rectangular hyperbola (see fig. 4) explain the word homographic: homos (Greek) = the same + grapheini (Greek) = to write. m n

y=

x+6 x2

6 3

Fig. 4

Reference
Exlibris en Wiskunde, Pythagoras (Dutch mathematics journal for youngsters, Editor Bruno Ernst) (1966), 5, no.5, (pages 103-105).

311

312

RUDOLPH M. SCHINDLERS BRAXTON HOUSE: THE FIBONACCI AND LUCAS SEQUENCE.

JIN-HO PARK

Name: Jin-Ho Park, Ph.D., Professor of Architecture. Address: University of Hawaii at Manoa, USA.

Abstract: Most writings on the proportional study of Rudolph Michael Schindlers houses have been descriptive observations rather than in-depth analyses using computational techniques or making pedagogical inferences. At times, the significance of the architects proportional system has been blatantly ignored. Architectural historians, like David Gebhard (1971), considered Schindlers system merely one practical tool for wood frame structure and construction. It is true that Schindlers unit system was derived from practical needs for his space architecture, but it may be a mistake to consider the system without theoretic implications. Recently, March (1993) interpreted Schindlers proportional method as an analogy with classical proportion, much more than a mere construction tool. When he used a musical analogy to examine the proportional design of the How house, in his paper, Dr. Hows Magical Musical Box, he seems to portray architecture as frozen music. Schindler left behind numerous built and unbuilt projects, which demonstrate extraordinary spatial complexity as well as variety. To achieve such spatial sophistication, Schindler argues that the architect needs to not only improve his mental image of the space but also possess a system. Consequently, Schindler proposed a proportional system of space reference frame. With this simple technique, the forms of space are freely conceived and precisely measured in the architects mind. Forms of space are envisioned by being inside of it like musicians imagine and articulate their music with notes. By 1928, Schindler had sufficient command of his method in organizing his space and space forms to propose a design, called the Braxton House. This paper first introduces Schindlers proportional system, Reference Frames in Space, outlines the spatial organization of the house, and finally interprets Schindlers use of proportion in the house in relation to the Fibonacci series and Lucas sequence.

313

1. REFERENCE FRAMES IN SPACE

Although Schindler expressed his early interest in a proportional system as early as 1916, it was not clearly expressed until 1946 when Schindler published a comprehensive summary of his proportional system in an article, Reference Frames in Space. In the article, his idea of proportion is well defined as follows: Proportion is an alive and expressive tool in the hands of the modern architect who uses its variations freely to give each building its own individual feeling. The system for Schindler is indispensable to the creation of his notion of space architecture as a doctrine of spatial organization. Schindler argued that he started using the system as early as 1920. In fact, an analysis of Schindlers Free Public Library project of 1920 has demonstrated his use of the system. (Park, 1996) The reasons for his using this system are two fold. First, all locations and sizes of the parts with respect to the whole are precisely identified during the construction process. Thus, no obscure or arbitrarily unrelated measurements are involved in the unit system. Second, the unit grid system offers the means to visualize space forms in three dimensions. He argued, it must be a unit which he can carry palpably in his mind in order to be able to deal with space forms easily but accurately in his imagination. Although there are few exceptions, like the Schindler Shelter project where he used a 5foot unit module, Schindler recommended 48 inches (4-foot) as the basic unit, to be used with simple multiples and with 1/2, 1/3, and 1/4 subdivisions. Among the subdivisions, with only a few exceptions, 1/3 and 1/4 are used for vertical modules in his works. This single unit module with its multiples and subdivisions form the basis of all dimensions of rooms. This choice has two reasons. First, the unit must be related to the human figure to satisfy all the necessary sizes for rooms, doors, and ceiling heights; second, for practical reasons, the 48-inch module fits the standard dimensions of materials and common construction methods available in California at that time. Pueblo Ribera Court of 1923 is a typical example. This multiple housing project, built in San Diego, is superimposed on 4 foot by 4-foot grid lines. Its vertical module is based on a 16-inch dimensioning system, which controls not only the height of the room but also of all elements including built-in furniture, chair, table, windows, doors, and clerestory, providing, in Schindler words, a uniform scale. He utilized his unit system in a square grid pattern. Numbers and letters are laid out on the grid on the floor plans in sequence and the vertical module is identified with an elevation grade. This pattern was original to Schindler. The grid was presented on drawings and on the house in his earlier designs, yet they disappear from the house and, at times, from the drawings. However, this does not mean he abandoned his system; on the contrary, his system remains embedded in the designs as underlying principles. 314

Thus, every location of the buildings is identified accurately in the convenience of composition and construction. In addition, Schindler used the system to measure room sizes. In his preliminary sketches, his room sizes with the whole numbers on drawings are commonly presented. These numbers are increments of unit multiples with its subdivisions.

2. THE HOUSE
The Braxton house was designed for Shore brothers on the beach in Venice, CA by 1928-30 but never built. Gebhard (1980) argued Schindlers interlocking forms of building are indebted to the De Stijl mode, saying, The year 1928 marks Schindlers full commitment to de Stijl ... the use of intersecting rather than singular volumes to establish their forms. This feature is what most strongly differentiates Schindlers work from that of the closely knit Internationalists Gropius, Mies, Le Corbusier (Pre-1935) and Neutra. Then, he continued, the Braxton house shows the architect was slowly giving up his laissez-faire view of Southern California and its climate, and was demanding an increased control over nature. Giella (1985) points out some features of the house, the use of Wrightian windows, balanced asymmetry, cantilevered structure. There are two major elements of the house: one for the house and the other for a garage. They are connected by a horizontal structure. A courtyard covered with sand occupies the space between. The main building is lifted up from the ground floor. The rectilinear box form dominates its spatial outlook, in Gebhards words, [a] rectangular box into and out of which secondary volumes projected, and in Giellas words, a slightly topheavy composition is created by cantilevering out each successive story from bottom to top. Various box forms are projected, recessed, and interlocked along the longitude axis. It is Schindlers usual practice and signature of his designs. The ground level preserves an open playground and patio for the natural beach sand, forecourt, outdoor fireplace, the garage, a guestroom, maids quarters, and furnace. The ground level is an independent unit and there is no interior staircase connecting the ground to the upper level. The major portion of the house is raised above the lot, which might be indigenous to all beaches. The option of the lifted structure may be suitable for the beach house by the necessity of allowing sun and air to reach the entire floor, preserving some part of site beneath for a playground and providing an open outlook to the ocean. On the main level, the architect created a living room, two stories in height. It is open to the Pacific Ocean and enclosed by individual bedrooms on the balcony floor overlooking the living area, and below is the living room, kitchen facilities, dining room where porches are adjacent to dining room and kitchen. Inside the main floor, there is 315

little built-in furniture shown on the drawing compared to his other projects, so that the inner space remains visually unobstructed. And also, like in the Lovell house, finely proportioned window mullions are inset facing the courtyard, and at the same time, door mullions face the ocean side. Three bedrooms on the balcony floor are disposed equally overlooking the Pacific Ocean. The composite roof is used as a terrace; a portion of it is used for sunbathing, a sleeping porch and a small closet. The whole building is supported by means of five horizontal structures, different from the five reinforced vertical concrete frames in the Lovell house (1929). Rather than using concrete, these structures are made of wooden frames. All space volumes of the house are interlocked along the five major structures.

Figure 1. The Braxton house (1928): a quarter-inch scale model fabrication.

3. PROPORTIONAL ANALYSIS OF THE HOUSE

Despite its novel character, curiously enough, the Braxton house has remained to date unexplored in any depth, unlike its contemporaries the Wolfe house (1928) and the Lovell house (1929). In particular, whereas the proportional study of the Lovell house 316

was attempted earlier by August Sarnitz (1986, 1988), that of the Braxton house has never been investigated. Sarnitzs analysis of the Lovell house plans and elevations is based on his belief that simple square and double square determine the overall proportional system of the house. Sarnitz also applied the same method in the analysis of the Free Public Library project. There are various sets of drawings of the house in the Schindler Archive at the University of California at Santa Barbara, including preliminary sketches, structural and sash details, construction drawings, and well-organized presentation drawings. These sources provide valuable information for the proportional analysis of the house. On the drawings, dimensions and placements of various spatial forms and details of the house are controlled by Schindlers unit system. The 4-foot unit system is clearly identified in plan with numbers and alphabets, and the 16-inch vertical module in elevation with grades. All major space, details, and structure underlie its subdivisions and multiples where all parts are related to each other in terms of simple unit relations to produce a coherent unity. Most of the major rooms in the construction drawings are measured in whole numbers and written on the drawing as is usually in other designs. Although rooms are frequently not a simple rectangular form, Schindler approximated those to the whole numbers. It appears that simple whole numbers with respect to the space reference frame is easy to grasp its size, as Schindler implied in his space reference frame system. Given the evidence of the floor plan, the dimensions of the rooms include the following: a guestroom 9-foot x 15-foot, maid room 10-foot x 10-foot, a bathroom 5-foot x 8-foot, a furnace 6-foot x 8-foot on the ground level; the living room 24-foot x 27-foot, the kitchen 10-foot x 16-foot and a porch in front 6-foot x 10-foot, and the entrance 8-foot x 8-foot on the second level; his room 11-foot x 12-foot, her room 14-foot x 16-foot, another bedroom 10-foot x 16-foot on the third floor. The rafters of the horizontal wooden structures are regulated in a 2-foot distance half of the unit module. However, dimensions written on the construction drawings are found to be inconsistent with the real dimensions of rooms. The construction documents of Schindler provided actually consist of drawing which other architects might consider schematic like the instruction to the builder. Two reasons might be considered: First, this difference relies on their stages of design development, either conceptual or practical, that are common practice in the architectural field. Early schematic designs are primarily conceptual basis, and evolve depending on specific practical requirements in the design process and during fabrication as well as construction. Thus, room sizes shown on the drawings do not always accord with dimensions measured from the drawings, although the former might be what the architect had initially in his mind; rather, the latter is a result of some adjustments. For example, in the Braxton house, the dimension of the guestroom shown 317

on the drawing is 15-foot by 9-foot, yet the real measure of the drawing of the room is 14-foot by 11-foot. Before arriving at a final scheme of a project, exercises of various possible schematic layouts are an almost universal procedure employed by architects, including Schindler. For example, in the Gibling house (1925-26), Schindler provided four different stages of schematic sketches for the plan development until he arrived at the final scheme. Each scheme offers different room sizes from another. Second, various rooms are not rectangular in shape but interlocked, overlapped, and, at times, zigzagged. In this case it is not possible to present room ratios as a:b, other than approximation. In fact, Schindler approximately measured the room sizes in a rational manner with his unit system to determine its size of each room and probably their proportional relations. Although these inconsistencies may cause confusion for this analysis, the collection of room dimensions serves as an excellent basis for further explorations of his proportional designs. This proportional analysis solely relies on the architects room dimensions written on the drawings, since it maintained the architects original intent of room sizes and their relations to each other before the architects transfer of the ratios to the practical necessities. Here is a list of all rooms and their relations:
Floor Room Names Room Dimensions Space Reference Frame Ratios Decimal Value of the Ratios

Guestroom Ground Floor Maid room Bathroom Furnace Living room Second Floor Kitchen Porch Entrance

15 x 9 10 x 10 8x5 8x6 27 x 24 16 x 10 10 x 6 8x8

3x2 2x2 2x1 2x 6x6 4x2 2x1 2x2

5:3 1:1 8:5 4:3 9:8 8:5 5:3 1:1

1.67 1.00 1.60 1.33 1.125 1.60 1.67 1.00

318

His room Third Floor Her room Bedroom Bathroom

12 x 11 16 x 14 16 x 10 10 x 6

3x2 4x3 4x2 2x1

12 : 11 8:7 8:5 5:3

1.09 1.14 1.60 1.67

There are some interesting results from the analysis as shown in the table. First, among three fractions that Schindler used which include1/2, 1/3, and 1/4, 1/3 of 48 inches (16inch) is not used in room dimensions of the plan. Instead, the fraction is used only for the vertical module. The16-inch vertical module measures the room height of the house. It also governs window mullions, door mullions, and the thickness of a series of horizontal structures. Accordingly, it is inferred that Schindler used those fractions separately: 12 increments for plans and those of 16 for elevations. The door height is his typical 6 feet 8 inches. The room heights vary, but they are subdivisions and multiples of 16 inches vertical module. The room height of the ground floor including the maid, guestroom, furnace, and garage is 8 feet. But the height of the open playground is 10 feet. The height of the kitchen and the dining room in the second floor is 8 feet. The height of the two story high open living room is 14 feet 16 inches. The height of His room and another bedroom is 8 foot 16 inches in the third floor. The height of Her room is 9 foot 8 inches, but that of the bathroom and the dressing room next to Her room is exceptionally 8 feet 10 inches. A series of five horizontal structures are apart 8 feet, except for the height of the first floor, which is 8 feet 9 inches. Second, these surprisingly few room ratios are worthy of the profound study of their interrelationship. In the house seven different ratios: 1:1, 4:3, 5:3, 8:5, 8:7, 9:8, and 12:11 are collected. There might be a certain relationship between these ratios. In order to deduce relationship of ratios, a particular method as an effective tool is introduced to derive all these ratios. Subsequently, these ratios are identified with the extreme and mean ratio geometrically incorporated in the regular pentagon. The ratios suggest a possible relationship with the Fibonacci sequence, in which each successive number is equal to the sum of the two preceding numbers, 1, 1, 2, 3, 5, 8, 13, 21 and the Lucas sequence, 2, 1, 3, 4, 7, 11, 18, 29 The ratios of successive numbers of both series converge to approximately 1.618, which is known as the extreme and mean ratio. As goes on to the right in these sequences, the ratios get close to the Golden Ratio, which is 1.618.

319

Figure 2. A graph showing its successive Fibonacci sequence closes to 1.618. Its proportion with a regular geometrical pentagon can be identified in the following diagram.

320

Figure 3. A double square and pentagon with root 5. Ratios including 1:1, 4:3, 5:3, and 8:5 are certainly among these. Geometric constructions using existing ratios on the drawings derive new proportional ratios. By computing other ratios of the two series together, the remaining three ratios can be constructed: 7/4 2/1 = 8/7, 4/3 3/2 = 9/8, and 18/11 3/2 = 12/11. All ratios are derived directly from other ratios previously used. The implication of these newly constructed ratios with old ones is that all room ratios become associated each other, thus a ratio consistency is achieved. Schindler might deliberately incorporate sizes of each room in his projects. This computation is delineated in the following diagrams. 321

Figure 4. Geometric construction of the room ratios. Thus, all room ratios of the project are suggestive of methods of computation coming out of the geometric and proportional construction. By this method, all other room ratios are constructed with regard to each other. Schindler might be aware of such ideas as he wrote, We must realize that proportion is not any more a simple mathematical relationship (Golden Rule, etc.) which can be applied universally in all buildings as it was in classical times. However, it is incorrect to suppose Schindler derived his room ratios this way or mathematically. Perhaps, some sort of computing procedure may have been involved in the architects mind. The architect could choose the whole number of room dimensions in the project with reasons to reference proportions derived from simple geometry or to practical solutions to the problems of a particular design. 322

4. SUMMARY
The proportional design of Schindlers Braxton house has been analyzed with regard to the Fibonacci or Lucas sequence. It is unlikely that while designing the project, Schindler calculated those numbers and their ratios with deliberate intent relying on the particular computational method. As discussed, the architect had no particular belief of a simple mathematical relation or Golden Rule. Thus, the analysis has based on the hypothetic speculation rather than documented evidence of the architect. Also it is hard to believe Schindler favored these individual room proportions as a whole or their relation to one another. That does not mean, however, that the analysis is useless. Even if the architect never claimed that his theory of proportion is based on a particular mathematical ground, there are some implications inherent in his designs where each room associates with others in terms of their ratios. Schindler wrote, The house of the future is a symphony of space forms - each room a necessary and unavoidable part of the whole. In the design, the architects approach was extremely simple. The composition of various rooms did not have to follow mere mathematical play of rules but have to express its relations in size between rooms so as to form an organic unity: a whole composed of related parts in orderly arrangement.

323

Figure 5. 3-D representation of rooms with their ratios hung on the horizontal structures. Although speculative, the proportional relations of those ratios examined in this article represent a valid method for analyzing proportional relationship of the rooms in the house, their harmonic use in the spatial organization of the design. Consequently, it could be said that the house is composed of simple room ratios and their relations in its entirety. Their relations in 3-D space are constructed to illustrate its integrity. Therefore, the observation leads to a conclusion that proportional relations of these ratios tie together in a cohesive whole in a single design, thus creating such a complex design. Whether the design was intentional or arrived at by chance, R. M. Schindler must have a splendid sense of proportional eyes to project such a simple and rational system to a very complex spatial composition.

5. ACKNOWLEDGEMENT
The author is indebted to professor Lionel March for his advice in carrying out this proportional analysis.

REFERENCES
Gebhard, D. (1971). Schindler, Thames and Hudson, London. Giella, B. (1985). R.M. Schindler's Thirties Style: Its Character (1931-1937) and International Sources (19061937), Ph.D. Dissertation, New York University, New York, NY. March, L. (1993). Dr. How 's Magical Music Box, RM Schindler: composition and construction, Eds. L. March and J. Sheine, Academy Edition, London. March, L. (1993). Proportion is an Alive and Expressive Tool RM Schindler: composition and construction, Eds. L. March and J. Sheine, Academy Edition, London. March, L. (1998). Architectonics of Humanism, Academy Editions, London. Park, J-H. (1996). Schindler, Symmetry and the Free Public Library, 1920, Architectural Research Quarterly, 2(2) pp. 72-83. Sarnitz, A. Proportion and Beauty the Lovell Beach House by Rudolph Michael Schindler, Newport Beach, 1922-26 JSAH, December 1986, Vol. XLV, No. 4. August Sarnitz (1988) R. M. Schindler-Architect, 1887-1953, New York: Rizzoli. Schindler, R.M., (1946). Reference Frames in Space, Architect and Engineer, San Francisco 165: pp10, 40, 44-45.

324

IN SEARCH OF ELWIN BRUNO CHRISTOFFEL; IS THIS FAMOUS MATHEMATICIAN NOT SO WELL-KNOWN?

ALEXA and ROBERT WILLEM VAN DER WAALL

Name: Dr. H. A. van der Waall (*1970, Nijmegen), Dr. R. W. van der Waall (*1941, Den Haag), mathematicians. Address: (H. A.): Simon Fraser University, Dept. of Mathematics and Statistics, 8888 University Drive, Burnaby, BC, Canada V5A1S6; (R. W.): KdV-Institute for Mathematics, University of Amsterdam, Plantage Muidergracht 24, 1018 TV Amsterdam, The Netherlands. E-mail: awaall@cecm.sfu.ca; waallr@science.uva.nl Fields of interest: (H. A.): algebraic number theory, computer algebra, differential Galois theory (architecture, art, design, skiing, traveling). (R. W.): number theory, group theory, geometry, coding theory, history of science (chess, genealogy, architecture, old scripts like hieroglyphs, linear B, maya) Awards: (R. W.): 1954, first prize in school education at the Gymnasium Haganum, The Netherlands; 1989: vita published on invitation in: Who is who of intellectuals, Melrose Press, Cambridge, U.K., ISBN 0948875305; 1991: award for Dutch paleography at Utrecht, The Netherlands. Publications and/or Exhibitions: (H.A.) Lam equations with finite monodromy; dissertation Utrecht; 2002 145 pages, ISBN 90-393-2927-3. (R.W.) 1) (with E. B. Kuisch) Homogeneous character induction II; Journal of Algebra 170 (1994), 584-595. 2) (with E. B. Kuisch) Modular Frobenius Groups; Manuscripta Mathematica 90 (1996), 403-427. 3) (with R. C. Lindenbergh) Ergebnisse ber Dedekind-Zeta-Funktionen, monomiale Charaktere und Konjugationsklassen endlicher Gruppen, unter Benutzung von GAP; Bayreuster Mathematische Schriften 56 (1999), 79-148. 4) (met L. de Clerck) Kijk op kegelsneden (Dutch) (View on conics); volume 21 in the Doe-Boek reeks (dobook series) of the society "Vierkant voor Wiskunde", for development of mathematics skills for high-schoolstudents; 36 pages; publisher CWI-Amsterdam; 2000. 5) On iterated group actions and direct products; Journal of Algebra 223 (2000), 57-65.

Abstract: While on vacation in Monschau (Germany) we suddenly stood in front of a commemorative plaque on behalf of the 150est birthday of the famous mathematician Elwin Bruno Christoffel (1829-1900). The talk at the Mat mium 2002 conference

325

presents an overview of his work and his influence and of the color locale of places where he used work and to teach. It will be embedded in the historical surroundings of the cities of Aachen, Monschau, Berlin, Zrich, and Straburg. A survey of our investigations is due to appear in the column The Mathematical Tourist of the journal The Mathematical Intelligencer. The talk will be accompanied with lots of slides, photographs, and journal pages.

326

1 INTRODUCTION
The subtitle of this contribution seems to contain a contradiction. Indeed, is it possible that someone is famous whereas it looks as if that very same person is not so well known? Well, here we present a way to get rid of this riddle. It appears that the mathematics of Elwin Bruno Christoffel (born 1829, Montjoie deceased 1900, Straburg) is well known, no doubt about that! Many of you, dear mathematically inclined readers, will be aware of the so-called Christoffel-DarbouxEinstein summation formulae, Christoffel symbols, or of the Christoffel-Schwarz mapping theorem. There is more, of course, much more. Among other things Christoffel's thirty-five printed papers deal with subjects like function theory including conformal mapping theory, propagation of electricity, Gaussian quadrature, continued fractions, dispersion of light, movements of points in periods, continuity conditions as to differential equations, minimal surfaces, theory of invariants, geodetic triangles, geometry and tensor analysis, orthogonal polynomials, shock waves, potential theory, Riemann's integrals, Jacobi's theta-sequences, irrational numbers, all yielding a beautiful lot of 19th century mathematics; here you should consult his Collected Papers (8). There also exist contemplative contributions on the mathematical works of Christoffel. To start with, view (9) and (12), being about a century old. On the other hand, Christoffel as a person acting in his society and habitat, his social, mathematical and intellectual acquaintances in Berlin, Zrich and Straburg (the places where he used to lecture) happened to be another story at our initial sight, i.e. without making use of the so-called electronic search-machines of today. But around the year 1981 it turned out already, that three elucidatory sources, other than (9) and (12) were in existence, as we discovered in 1999, in a rather surprising way.

2 A MEMORIAL
During a short holiday in the summer of that year 1999 in the neighborhood of the socalled "drielandenpunt" (meaning: the meeting point of the three countries The Netherlands, Belgium, Germany) both of us took a stroll through the colorful and picturesque city of Monschau (until 1918: Montjoie), where suddenly our attention was fixed upon a plaque on one of the walls of the building at the Rurstrae 1. The inscription on it reads as follows: Dr. Elwin Bruno CHRISTOFFEL 327

Professor der Mathematik in Zrich, Berlin, Straburg * 10.11.1829 in Monschau + 5. 3.1900 in Straburg GEBURTSHAUS

Figure 1. The memorial. So here it is. It concerns the building in which Christoffel was born. More precisely, as we found out, the original home in which he was born, was burned down in 1835. The plaque contains a sculpture of his face en face too; it is derived from the only known photograph for which one is absolutely sure that it represents Christoffel (he is then about 40 years of age).

328

Figures 2-3. Left: the only known photograph representing Christoffel. Right: the (German) subscript in (6) to the second photograph indicates that it might be another picture of Christoffel, at older age, as physiognomy suggests. Nowadays the building at the Rurstrae 1 comprises the Parfumerie-Foto-Drogerie "Servaes" at the bottom floor. Its proprietor Mrs. Ingrid Hermanns did tell us that the plaque has been unveiled in the year 1979 in commemoration of the 150th birthday of Christoffel. She also showed an invaluable item in connection to all aspects of life and work of Christoffel. It turned out to be a three-issues-in-one volume of a journal dealing with the local history of the city of Aachen and its surroundings; view (6). That source (not so well-known to mathematicians we presume) did trigger our investigations on Christoffel. We daresay, that the eighty pages of (6) are thoroughly filled will all sorts of profound and important information; as such it is an indispensable toolkit. It provides a lot of photographs (Lejeune-Dirichlet, Geiser, Prym, Rost, Hilbert, Riemann, Einstein, Weierstrass, Klein, von Laue, Reuleaux, Ricci-Curbastro to mention a few) besides a description of the social environment of Monschau and a genealogy of Christoffel, a history on his life and work, his school years in Cologne, mathematics and politics combined with intrigues in Zrich, Berlin and Straburg. Of course, one finds there the list of all of his printed publications.

Figure 4. View from dormer window of the building at the Rurstrae 1, in 1889, at the place where Christoffel was born.

3. ADDITIONAL INFORMATION AND CONCLUSION

Be aware that in (6) all of the beautiful and clarifying material, presented over three columns per page and supplied with an enormous amount of (pragmatic) references, is brought to you in the German language. The reader not in command of the German language should not be discouraged, however. Our investigations revealed, indirectly from source (6), that a conference took place in Aachen and Monschau (both in Germany) in November 8-11 of the year 1979, on the influences and aspects of the work of Christoffel into our era. Much is to be learned from the twelve lectures given at the conference and from forty-five papers written on invitation; see (7). The beginning of (7) yields an English version of some of the papers presented in (6), be it in a form slightly more condensed in comparison to the ones in (6). It is nevertheless clear that the starting part of (7) provides a good impression of time and circumstances in which Christoffel used to work. An accurate, short but complete, overview of the contents of the conference is also to be found in the very handy non-mathematical source (1). Above we have told our story how we did unveil Christoffel's life and work up to the year 1981. An electronic search-machine investigation provides relevant and additional material of later date (read up Christoffel's geometry in (11) for instance). The so-called "Riemann example" of a continuous non-differentiable function is scrutinized in dept in the remarkable paper (4). Furthermore one finds that mathematics used to be a vivid affair in the region of the "drielandenpunt" during the last 1200 years (view (3) and (5)), 330

whereas mathematics in Zrich and Berlin over the ages has been described in (10) (see the contributions of G. Frei and E. Knobloch therein, respectively). In summary, it has come to light that an almost conclusive overview of life and work of Christoffel in his days and its impact into our days, has been described in (1), (3), (5), (6), (7) and (11), while its relevant mathematics is presented in the greater part of (7) and (of course) in (8). Besides that, modern electronic data comprise source (14) and above all, the unsurpassed source (13). Hundreds of papers have been published since 1980 in which Christoffel's mathematics plays the leading part. All this being said, we hope to have convinced you, that the question brought up in the subtitle of this mathematical tourist contribution, has to be answered in the negative, i.e., Christoffel is famous and well known!

Figures 5-6. View on the house and the memorial plate, shown by the authors. 331

Acknowledgement
We are indebted to Mrs.Ingrid Hermanns and her husband in providing an original issue of (6); without it, this note would never have been written.

References
1. Bericht ber das Internationale Christoffel-Symposium in Aachen und Monschau, 8.-11.November 1979; von P. L. Butzer and F. Fehr; in: Berichte zur Wissenschaftgeschichte 3 (1980),193-201. 2. P. L. Butzer- An outline of the life and work of E. B. Christoffel (1829-1900), in: Historia Mathematica 8 (1981), 243-276. 3. P. L. Butzer- Mathematics in the region Aachen-Lige-Maastricht from Carolingian times to the 19th century, in: Bull. Soc. Roy. Lige 51 (1982), 5-30. 4. P. L. Butzer and E. L. Stark- "Riemann's example" of a continuous non differentiable function in the light of two letters (1865) of Christoffel to Prym; in: Bull. Soc. Math. Belge, Sr. A, 38 (1986), 45-73. 5. P. L. Butzer- Scholars of the Mathematical Sciences in the Aachen -Lige-Maastricht region during the past 1200 years, an overview (with the assistance of Helga Butzer Felleisen), pages 43-90 in: P. L. Butzer (ed) et all., Karl der Groe und sein Nachwirken: 1200 Jahre Kultur und Wissenschaft in Europa. Band 2: Mathematisches Wissen, Turnhout Brepols (1998); ISBN 2-503-50674-7. 6. Elwin Bruno Christoffel (10.November 1829, Monschau - 5.Mrz 1900, Straburg), Professor der Mathematik in Zrich, Berlin und Straburg - Gedenkschrift zur 150.Wiederkehr des Geburtstages, in: Heimatbltter des Kreises Aachen, Jahrgnge 34/35 (1978,3/4 und 1979,1) - 80 Seiten (in einem Band). 7. E. B. Christoffel - The influence of his Work on Mathematics and the Physical Sciences; edited by Paul Leo Butzer und Franziska Fehr (International Christoffel Symposium in honor of E.B. Christoffel on the 150 Anniversary of his Birth, 1979, Aachen and Monschau (Germany)); 1981 - Birkhuser Verlag, Basel-Boston- Stuttgart; ISBN 3-7643-1162-2; XXV + 761 pages. 8. E. B. Christoffel- Gesammelte Mathematische Abhandlungen; unter Mitwirkung von A. Krazer und G. Faber herausgegeben von L. Maurer; zwei Bnde, 1910 - Teubner Verlag, Leipzig und Berlin. 9. C. F. Geiser und L. Maurer- Elwin Bruno Christoffel, in: Mathematische Annalen, 54 (1901), 329-341. 10. Jahrbuch Ueberblicke der Mathematik 1994 (S. D. Chatterji (ed) et all.); Vieweg - Braunschweig; ISBN 3526-06578-8; VII + 265 pages. 11.K.E.B.Leichtwei- Christoffels Einflu auf die Geometrie, Seiten 93-103 in: Sitzungsberichte der Berliner Mathematischen Gesellschaft, Jahrgnge 1972-1987 (1.10.1971-31.10.1987); Berlin 1987, 185 Seiten. 12.W. Windelband- Zur Gedchtnis E. B. Christoffel; in: Mathematische Annalen 54 (1901), 341-344. 13. http://www-groups.dcs.st-and.ac.uk/~history/Mathematicians/Christoffel.html 14. http://www.stetson.edu/~efriedma/periodictable/html/CF.html

332

SPACE PRODUCTION.
51N4E

Name: 51N4E, Brussels (Belgium).

Abstract: 51N4E is a Brussels based office working on matters of space production, and visual aspects of buildings. Space production implies the creation of an adapted reality. 51N4E does not consider reality as a mere given or context, instead, it provides a new uncanny context a necessary luxury for the production of successful space. Advanced domestics, European space, scaling and outsmarting building regulations are but a few tools in the quest of mastering space production. Today, 51N4Es main activities focus on: E.D.E.N. (public park Luxembourg Brussels), CELL (a mass producible minidwelling), LAMOT (production centre for an urban culture) and -space.

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

FRUSTRATION: SOURCE OF COMPLEXITY

TOHRU OGAWA

E-mail: Ogawa-t@koalanet.ne.jp

An important role of fundamental science is to prepare some views and concepts, which are available in extending the scope of science. The attempt in this paper is based on this philosophy. Cellular automata are often studied from a similar point of view, in which simple basic laws governing local processes give rise to highly complex phenomena. These studies are very important in the extension of physical science in the future even though the correspondence with the real physical world may be at most metaphorical or suggestive. For further development along this line, it is desirable to try to include some conservation laws and frustration phenomena in models of cellular automata. A frustrated system is very interesting and important from this point of view because of abundance of states and phenomena appearing in it. We may bear in mind that, in living organisms, a wide variety of phenomena are contained within a very narrow energy range. We may expect the study of a frustrated system to provide an insight into more complex systems than treated hitherto. The cellular automaton models are usually deterministic. However it is very difficult to construct a purely deterministic model with conservation and frustration. Therefore a kinetic model containing stochastic (probabilistic) terms is studied in this paper. The antiferromagnetic Ising model on a triangular lattice is typical of such systems. The frustration produces a degeneracy of the ground-state. The number of degrees of freedom of the ground-states is exp[0.3231] = 1.381 per site. The residual entropy of the system, which was rigorously calculated by Wannier, is proportional to the logarithm of this value. Almost nothing is known of the details of the degenerate ground-states. Some years ago, the author studied their cluster structure in relation to some maze-like patterns observed in the monodispersive latex system in a (2+e)-dimensional space. Though the present investigation is on similar lines, its purpose is more abstract. It is the so-called "glass transition" that is the closest physical effect to the present research. It is supposed that there is no essential difference between a liquid and a glass so far as their

351

static structures are concerned. Or, at least, no abrupt change in the static structure can be expected. The difference in structure between a liquid and a glass may be expected to lie in the dynamics of their phase spaces. If so, the present model is the simplest one that can be investigated in detail.

Reference
T. Ogawa and Y. Nakajima; Frustration, Degeneracy, and Forms: A View of the Antiferromagnetic Ising Model on a Triangular Lattice Progress of Theoretical Physics, Supplement No. 87 (1986), pp. 90101. It is reprinted at http://members.tripod.com/vismath5/ogawa1/forms.htm (VisMath)

352

DIGITAL SHAPING OF SPATIAL STRUCTURES

JANUSZ REBIELAK

Name: Janusz. Rebielak, architect, (b. Bierutow, woj.wroclawskie., Poland, 1955). Address: Department of Architecture, Wroclaw University of Technology, ul. B. Prusa 53/55, 50-317 Wroclaw, Poland. E-mail: j.rebielak@wp.pl Fields of interest: Architecture, morphology of space structures, roof covers, high-rise buildings, formex algebra ( history of science, art and culture). Awards: Distinguished Leadership Award of the American Biographical Institute (2001). Publications and/or Exhibitions: 1. Rebielak Janusz: Bar Space Structures - Rules of Shaping, in: Symmetry: Natural and Artificial, Proceedings of the Third Interdisciplinary Symmetry Symposium and Exhibition, Washington, USA, August 14-20, 1995, Symmetry: Science & Culture, Quarterly of the ISIS-SYMMETRY, Vol. 6, No 3, pp. 442-445. 2. Rebielak Janusz: Space structures shaping and visualisation of their digital models by means of Formian, in: Proceedings of the Fifth Interdisciplinary Symmetry Congress and Exhibition, Sydney, Australia, July 814, 2001, Symmetry: Art and Science, Quarterly of the ISIS-SYMMETRY, pp. 158-161. 3. Rebielak Janusz: Structural systems of cable domes composed of concentric spatial hoops, in: ed. H. Kunieda, IASS 2001 Extended Abstracts, International Symposium on Theory, Design and Realization of Shell and Spatial Structures, October 9-13, 2001, Nagoya, Japan, pp. 328-329. 4. Rebielak Janusz: General morphology of new types of tension-strut systems, Lightweight Structures in Civil Engineering, Local Seminar of IASS Polish Chapter, Micro-Publisher Jan B. Obrebski Wydawnictwo Naukowe, Warsaw- Wroclaw, 7th December, 2001, pp. 50-55. 5. Rebielak Janusz: Prismatic space frames as the main support structures for high-rise buildings, Lightweight Structures in Civil Engineering, Local Seminar of IASS Polish Chapter, Micro-Publisher Jan B. Obrebski Wydawnictwo Naukowe, Warsaw-Wroclaw, 7th December, 2001, pp. 81-89.

Abstract: The paper will present the proposals of the space structure shaping as the structural systems for lightweight roofs of large spans and as the main support for the high-rise building. It will be presented the concept of the special type of the retractable dome. The author recently invented some new types of these systems are by and for the main types of them he has prepared several programmes written in the programming language Formian. These programmes determine the digital models of the proposed forms of the space structures systems. Owing to the suitable application of the systems the buildings designed by means of them may obtain the interesting and the individual architectonic views.

353

1 INTRODUCTION
The aesthetic perception of the form of many types of buildings has been radically changed after the wide application of the space structures in architecture and in the civil engineering during the 20th century. The specific way of the arrangement of the component parts in spaces of these structures causes, that they have their own homogeneous architectonic views having an extremely great imprint to the architecture of the objects, in which they are applied. It is one of the most important factors justified the highest interest of many architects in processes of their shaping. Many evidences show that the develop potential of the space structures is very great and makes possible to discover their new and numerous applications in the future. Because of the basic rules of the theory of structures the general aspects of symmetry or sometimes asymmetry have to be taken into consideration during the design processes of the space structures. The are many propositions given by the author in the area of shaping of the structural systems devoted for various types of the roof covers and for the high-rise buildings. Many of them are of very complex shape therefore the design process of them is complicated and it needs special tools. One of the very useful and powerful tool in this field is the programming language Formian, Nooshin & others (1993). It is invented by Professor Hoshyar Nooshin and developed by his team in the Space Structures Research Centre at the University of Surrey in England on the basis of the r a bra , also developed by Prof. H. Nooshin, Nooshin (1984). It was intended as a tool for improving these processes easier and faster and in order to make the better co-operation between the architects, the civil engineers and the other designers involved in the project. Formian makes possible the fluently changes of the geometry of the structure at any stage of the design. The visualisation of any, even the very complex form of a space structure, is by means of Formian relatively simple.

2 SHAPING OF TENSION-STRUT SYSTEMS

The tension-strut structures are in the last few decades used in the design and realisations of various types of the lightweight roof covers. Some types of these covers, particularly the cable domes developed by D. Geiger, are of the significant importance in the developing of the structural solutions in this domain of the engineering activity. The structural systems, proposed by the author for these purposes, were developed as a result of suitable transformations of the bar arrangement of the dome, the structural form of which was proposed earlier by the author, Rebielak (1996). The tetrahedron and the octahedron were assumed as the bases in these processes, Rebielak (1999, 2001). One on the results of the shaping processes presents Figure 1. The digital model of this structure was defined by means of Formian and than it was transported to Corel, the software designed for preparing the visualization processes. The proposed structural system may be applied for the designing of very lightweight covers s of large spans.

354

Figure 1: Perspective view of the one of the tension-strut systems proposed by the author for the lightweight structures of large span covers. The appropriate usage of the symmetry formulas make the processes of preparation of the digital models of these structure much easier and shorter than before. The structure as itself and its digital model have to be built according to the basic rules of symmetry.

3 RETRACTABLE AND FOLDABLE ROOF STRUCTURE

Figure 2 shows the basic schemes of the structural system proposed by the author for the foldable and retractable structure of e.g. the dome cover. The system consists of the triangular bar sets, see Figure 2a. The chosen bars, marked there by grey colour, build the main girders located onto the perpendicular surface marked by the dash lines. These

Figure 2: General schemes of structural system proposed by the author for the foldable and retractable structures of e.g. the dome cover.

355

girders are the main ribs of e.g. the dome cover and they are focused in its crown-node. This basic configuration of the structural module shows Figure 2b. Almost all of these bars are of the fixed lengths, only the vertical, marked in Figure 2a by the letter symbol V, have to be able to change their lengths. These bars can be made as the struts of equal lengths with the appropriate screws, what enables the necessary changing of the bar set shapes. The mechanisms of the retractable and foldable covers are of the high technical requirements, Ishii (2000). These types of structures and the assembly processes of them have to be designed according to the rules of symmetry. The special bars, marked by symbol V in Figure 2a, are perpendicular to the middle surface of the planned roof. Owing to the suitable length changes the entire space structure can be folded and refolded. By means of the length controlling of these bars the structure is retractable and it can obtain the planned shape. The proposed system can be used in the design and in erection of the structure being at the same time the foldable and the retractable structure. The digital models of these systems, prepared in Formian, will be used to carry on many comprehensive analyses in order to estimate their suitability to the proposed purposes.

References
Nooshin, H. (1984) Formex Configuration Processing in Structural Engineering, Elsevier Applied Science Publisher, London and New York. Nooshin, H., Disney, P., and Yamamoto, C. (1993), Formian, Multi-Science Publishing Co. Ltd, Brentwood, England. Ishii, K., (2000) Structural Design of Retractable Structures, WITPress, Southampton, Boston. Rebielak, J. (1996) Examples of Shaping for Large Span Roofs and for High-Rise Buildings, International Journal of Space Structures, special issue on Morphology and Architecture, Volume 11, Nos 1&2, pp. 241-250. Rebielak, J. (1999) Cable Dome Shaped on the Ground of the {T T} Double-Layer Space Structure. Example of Formians Application in Creation of Numerical Model of a Structure, Lightweight Structures in Civil Engineering, Local Seminar of IASS Polish Chapter, Micro-Publisher Jan B. Obrebski Wydawnictwo Naukowe, Warsaw, pp. 86-87. Rebielak, J. (2001) Examples o Space Structures Shaping and Visualisation of Their Digital Models by Means of Formian, Proceedings of the Fifth Interdisciplinary Symmetry Congress and Exhibition, Sydney, Australia, July 8-14, 2001, Symmetry: Art and Science, Quarterly of the ISIS-SYMMETRY, pp. 158-161. Rebielak, J. (2001) Structural systems of cable domes composed of concentric spatial hoops, ed. Kunieda, H., IASS 2001, Extended Abstracts, International Symposium on Theory, Design and Realization of Shell and Spatial Structures, October 9-13, 2001, Nagoya, Japan, pp. 328-329. Rebielak, J. (2001) General morphology of new types of tension-strut systems, Lightweight Structures in Civil Engineering, Local Seminar of IASS Polish Chapter, Micro-Publisher Jan B. Obrebski Wydawnictwo Naukowe, Warsaw- Wroclaw, 7th December, 2001, pp. 50-55. Rebielak, J. (2001) Tension-strut systems built by means of modules with V-shaped bar sets, Lightweight Structures in Civil Engineering, Local Seminar of IASS Polish Chapter, Micro-Publisher Jan B. Obrebski Wydawnictwo Naukowe, Warsaw-Wroclaw, 7th December, 2001, pp. 78-80.

356

PROPORTIONS AND DISSECTIONS IN POLYGONS

ENCARNACION REYES IGLESIAS

Name: M Encarnacin Reyes Iglesias, mathematician, ( Burgos, SPAIN., 1954). Address: Departamento de Matemtica Aplicada Fundamental, E.T.S. Arquitectura. University of Valladolid SPAIN. E-mail: ereyes@maf.uva.es. Fields of interest: Geometry, proportion, symmetry (Art and Architecture, Mathematics with Paper Folding, Phyllotaxis, and Mathematics in Daily Life). Publications and/or Exhibitions: - Proporciones y Arquitectura. ICME, (Congreso Internacional de Educacin Matemtica) Sevilla 1996. - Proportions mathmatiques et leurs applications l' architecture. Bulletin IREM (Institut de recherche pour l'enseignement des mathmatiques de Toulouse) Toulouse. France, 1998, 39 pp. - Papiroflexia y proporciones dinmicas en rectngulos. JAEM (Jornadas Nacionales para la Enseanza y Aprendizaje de las Matemticas.) 1999 - Mathematics: a determinant strategic for the development of an architectonic project. Symposium on Symmetry. Alhambra 2000, Granada (Spain) -Paper folding and proportions in polygons. ISIS Symmetry Congress & Exhibition. Sydney 2001.

Abstract: In this work I will focus mainly on some proportions and dissections in polygons, particularly hexagons and octagons. The classical method of the decomposition of polygons into smaller parts that can be rearranged to form other polygons is a pedagogical tool to understand the geometry of polygons and their properties. An interesting problem to be solved en each case is to find an equivalent polygon by using the minimum number of pieces. I will show some dissections of the hexagon and the octagon with their equivalent figures. In the other hand, teaching mathematics through paper folding is an alternative way to understand them. I will present in this exposition some constructions of the hexagon and the octagon by using these techniques. My interest focuses on researching the underlying mathematics involved in the paper folding process. In each case (hexagonal and octagonal) it will be calculated the relationships of proportion between diagonals and sides the same in convex polygons as in their stars. I will establish also the concept of polygonal proportion (PP) using it to calculate the hexagonal and octagonal global proportion.

357

We define the proportion of the regular polygon Pn, written p(Pn), as the quotient between the area A (Pn ) of the polygon and the square of the length of its side L: p(Pn) =

A(Pn ) L2

If H6 is a regular hexagon with side L, this definition gives us

3L2 3 A(H 6 ) 3 3 2 = = . p( H 6 ) = 2 2 2 L L

In the hexagon we also have the following proportion: D = 3 . We can construct a hexagon and its star 6/2 by using L

paper folding. We get a strip of paper with the same proportion than the rectangle

D around its center and repeat this operation once = 3 . Rotate the strip an angle of L 3 more.

358

In case of the regular octagon O8 with side L we have: p(O 8 ) = A (O 8 ) L2 = 2 = 2( 2 + 1) where

denotes the silver number. In the figure, we have L2 = 2x 2 ; 2 x + L = D , L ; solving this system, we obtain: x = 2

therefore

L D 2 + L = D L = 2 +1 = . 2 around 4

By rotating the silver rectangle (a rectangle with proportion ) an angle of its center, we can obtain an octagon and its star 8/3.

RATIOS OF LENGTH OF SIDES IN HEXAGONS AND HEXAGRAMS 6/2

Let

{Li }be

the decreasing sequence of the

length of the sides of the convex hexagons (see figure). This sequence is a geometric series of ratio
L n +1 1 = and whose sum is Ln 3

359

S=

1 1 1 3

3+ 3 If we consider the sequence {A i } of lengths of the hexagrams, we 2

also obtain the same relation.

A n +1 1 = . An 3

RATIOS OF LENGHTS OF SIDES IN OCTAGONS AND THEIR STARS 8/3

Let {L i } be the decreasing sequence of the length of sides of convex octagons (see figure). This sequence L n +1 1 = Ln is a

geometric series of ratio

The is:

sum S= 1 1 1

of =

the

series

{Li }

1 =1+ . If we 1 1

consider the decreasing sequence of lengths of sides

{A i }

in the star

polygon 8/3, we also obtain the same relation. A n +1 1 = . An

360

In the following dissections of the hexagon and the octagon we can rearrange the pieces to form equivalent rectangles.

This interesting dissected octagon appears in CEAC (Center for the Study of Contemporary Art) in Barcelona (Spain); it is conceived by J. L. Sert. This kind of division of the octagon into nine parts is also useful in pavements, because it can be constructed using only two types of tiles: a square and its half. However, such an octagon is irregular.

361

References
Boltianski, V.G. Book. Equivalent and Equidecomposable Figures. D.C. Heath, Boston, 1963 Fernndez, I, Reyes E. Construcciones y disecciones del octgono, SUMA, 38, (2001) 69-72 Kappraff, J. Connections.Book. The geometric bridge between art and science. Mc Graw Hill, New York, 1991. Kappraff, J. Systems of Proportion in Design and Architecture and their Relationship to Dynamical Systems Theory. http://members.tripod.com/vismath/kappraff/kap4.html Reyes, E. Papiroflexia y proporciones dinmicas en rectngulos. JAEM (Jornadas Nacionales para l Enseanza y Aprendizaje de las Matemticas.) 1999. Reyes, E. Paper folding and proportions in polygons .Symmetry: Art and Science. Congress. Sydney, 2001 Vera W. De Spinadel The metallic means family and multifractal spectra. Non Linear Analysis, 36 (1999) 721-745. Vera W. De Spinadel Book. Title: From the Golden Mean to Chaos, Ed. Nueva Librera, Buenos Aires, Argentina,1998.

ADDENDUM: QUESTIONS ABOUT THE EXPOSITION

It is possible the construction of the regular pentagon by using a DIN A4 sheet of paper? The answer was: A Din A 4 sheet of paper is not suitable for the construction of a regular pentagon

because the proportion of a DIN A4 is the irrational number the regular pentagon involves the golden number 1+ 5 . 2

2 , and the proportion of

1

The right construction of the regular pentagon starting with a DIN A 4 sheet of paper requires to cut a small strip 8 mm (7,96 mm), making the long side of the sheet of paper shorter. The proof is the following: 540 = 108. The internal angle of a pentagon is 5 Starting from a DIN A4 tg = 2 = actg 2 54,74. If we fold through the transversal line we obtain the angle 2 109,47 , that which is different from 108. 108 We need another rectangle with = = 54 , tg54 = 1,376, 2 therefore if we denote by y the long side of the rectangle A4, y tg = y = 210. tg54 = 289,04 mm. 210

2
210 y

362

And since the length of the long side of a DIN A4 is 297 mm, we must cut: 297 - 289,04 =7,96 mm. 8mm.

Therefore, -Cut 8 mm to the long side of a DIN A4 format. (Then, cut along the short side, see figure) -Join the two opposite vertices of paper and fold it. -Fold the figure along its axes of symmetry -Fold the smallest sides once more until they coincide with the axis of symmetry.

THE REGULAR PENTAGON IS OBTAINED

Another question was: What about the heptagon? Has it been constructed? The answer was: So far no heptagon construction using paper folding has been executed. The only regular polygons whose construction was studied by the author were those that can be drawn by ruler and compasses.

363

364

THE GEOMETRY OF FLEMISH GOTHIC TOWN HALLS

HAN VANDEVYVERE

Name: Han Vandevyvere, engineer-architect, scient. coll. KU.Leuven, Belgium (b. Brugge, Belgium, 1966). Address: Departement Architectuur, Stedenbouw en Ruimtelijke Ordening, K.U.Leuven, Kasteelpark Arenberg 1, B 3001 Leuven (Belgium). E-mail: han.vandevyvere@asro.kuleuven.ac.be Fields of interest: Architecture, building & engineering, geometry (also arts, philosophy). Publications: Vandevyvere, H., Het stadhuis van Leuven: een geometrische analyse; in Jaarboek van de geschied- en oudheidkundige kring voor Leuven en omgeving, 39 (2000), pp 171-192; and Vandevyvere, H., "Gothic Town Halls in and around Flanders, 1350-1550: A Geometric Analysis", Nexus Network Journal, vol. 3, no. 3 (Summer 2001), http://www.nexusjournal.com/Vandevyvere.html.

Abstract: Can we find evidence for the application of design rules in the plans of the gothic town halls in Flanders? This question was dealt with in a recent investigation into the layout of some medieval town halls of the southern low countries. The research was initiated following a series of accidental discoveries. These indicated that not only medieval churches were built according to a symbolic geometry, but certain civic buildings as well. Starting from a graphical analysis of the building plans of the town halls under investigation, an attempt was made to reconstruct the scenario of their design. As a result, we may put forward a number of conclusions. A particular system of applied mathematics, characteristic of the gothic tradition, was used to set out the building plans. The geometry is based on manipulations of the compass and the carpenters square. A taste for symbolic series of numbers and figures appears. Acting together with the local measurement systems that were used to set out the plan, we often obtain a subtle game of numbers in the entire composition, expressed in local feet or rods. If the medieval building master is known to have applied symbolic geometries in religious buildings, thereby referring to the metaphysic views of his time, then we may now state that he did this also in the civic architecture of the town halls. The medieval city had at that time become a powerful entity, challenging the established orders, and the town hall was in its design the very expression of this new reality. As such it

365

confirmed the consciousness of the citizen in way proper of the Middle Ages: by symbolic representation.

Recent research on the design of a number of gothic town halls in the southern low countries indicates that not only medieval churches were built according to a symbolic geometry, but certain civic buildings as well. Starting from a graphical analysis of the building plans of a number of representative town halls in and around Flanders, an attempt was made to reconstruct the scenario of their design.

Methodology
The research was started from a selected set of premises, after a building module in the design of the Leuven town hall was discovered rather accidentally. A first assumption deals with the validity of a research on hidden design and proportion systems in a building. Do we have reasons to look for such a system, and what is the probability rate of the findings? As there is a good deal of evidence about the existence of medieval design systems in general, a closer look to the town halls under consideration was at least defendable. As a further constraint, only those schemes would be accepted that are relatively simple, evident, and robust. The last condition means that conclusions should have a tolerance margin, accounting of e.g. survey errors in the plans, and building particularities. Carrying on, a set of design principles has been selected for consideration. We can summarize them as follows: Buildings are set out in the local measurement units, in use at the time of construction. Documentation about the measures in use in Flanders during the middle ages is available, with a high degree of reliability; Following some medieval writings, we should consider that what is to be found in plan, is reflected in elevation [1]: we have to find a three-dimensional approach based on pulling up the plan; Symbolic, simple integer number series, such as to obtain simple ratios between dimensions, are typical of medieval architecture and sculpture. So they may well occur in the current research subject; An applied geometry, based on manipulations of the compass and the carpenters square, is likely to be a major design canon for similar reasons. Emphasis was put on a graphical analysis of the buildings as they stand; research on literary sources was limited to building history information necessary to reconstitute, as much as possible, the medieval situation. It should be noted that in the 18th and 19th centuries a lot of correcting restoration was performed on medieval buildings, often creating an idealized reconstruction of something that never existed before. Doors were moved, balustrades added, and so on. Finally, used building plans should have a reasonable degree of accuracy for the purposes of the research. In the case of Leuven, a new survey of some cardinal points of

366

the building was made, confirming the satisfactory precision of the survey plans at disposition. In what follows, we bring forward some results of the research that point towards the existence of a number of generally applied design principles. The buildings that were investigated are situated in Brugge, Oudenaarde, Brussel, Leuven (B), Veere (NL), Arras and Saint-Quentin (F).

Issue 1: the cubic grid

For the buildings situated in present Flanders (B), it could be verified that they are set out in the local, medieval measurement units of the city where they stand. All of the four cities had a different basic foot, and a rod composed of a different number of these local feet. Moreover, for the buildings outside actual Flanders, it was possible to find a similar modularity in the overall dimensioning of the composition. We can conclude that the contouring volumes of the buildings are simple rectangular parallelepipeds, with the ribs made up of a simple number of rods. For example, in Brugge we find a contouring volume of 3 x 5 x 7 Brugge rods. As an illustration, we give two graphical examples of buildings where this cubic grid is a major characteristic of the design. Figure 1 shows Leuven, figure 2 shows Veere.

367

Figure 1: Leuven grid

Figure 2: Veere grid For Leuven, the basic cube of the grid has a side of two Leuven rods, or 5.71 meter. See further on for comments on this dimensional module. For the buildings outside the actual Belgian borders, it should still be checked if the apparent module coincides in a similar way with a precise number of local rods or feet.

368

Issue 2: the contouring square

In connection with the first finding, the front elevation of the town halls, including the projection of the roof, is dictated by a contouring square. This square may be put on a base in some cases. As an illustration, we show the contouring square for Leuven. Further on the ones for the other town halls will appear in connection with other design elements.

Figure 3: the principle of the contouring square (Leuven)

Issue 3: circle, triangle and square; or the numbers 1, 3 and 4

Not only do the front elevations of the town halls display the contouring square, but often, this square is also embedded in a symbolic construction composed furthermore of a circle and its inscribed equilateral triangle. So we have a composite figure created from a 1-sided, 3-sided and 4-sided polygon. At this point it is interesting to notice that the construction of this figure also gives good approximations for the division of a side of the square into 3rd and 7th parts.

369

Figure 4: symbolic circle, triangle and square. The fat lines give the exact proportions of 1 to 3 and 1 to 7. The intersections of the square and the triangle are very good approximations of these positions. The symbolic meaning of this figure could probably be explained by tracing compagnonesque knowledge, but by now we may assume already that a medieval verse is stating this kind of construction as follows: A point in a circle And that can be situated in the square and the triangle Do you know the point? All is for the better Dont you know it? All is in vain. [2] Moreover we find, corresponding with this figure, a number series that is especially confirmed in the design of the town hall at Oudenaarde. The series is as follows: 1 3 4 7 (i.e. 4 + 3) 12 (i.e. 4 x 3). One Oudenaarde rod is 21 (i.e. 3 x 7) Oudenaarde feet, hence a number game in all of the town halls composition. Together with the scheme applied for Oudenaarde, we give a few other significant examples of this design canon in the other buildings. Note that entrance doors often reach until the 1:7 level of the square side, and roof balustrades until the 4:7 or 5:7 level. In two cases, Oudenaarde and Saint-Quentin, the scheme can be repeated half of the radius of the circle upwards, i.e. putting the base of the second triangle through the center point of the first circle. Doing this we find the position of other cardinal points of the front elevation. In Brugge, by contrast, the scheme seems to have been repeated half a Brugge rod, or seven Brugge feet, upwards.

370

Figure 5: facade composition for Oudenaarde with shifted scheme (R/2), relations between plan & elevation.

371

Figure 6: facade composition for Brussel; relations between plan & elevation. Note that for Brussel only the oldest front wing of the building is considered. The more recent half of the building, right of the central gate and tower (not on plan), does not seem to obey to the design system of the first building concept.

372

Figure 7: facade composition for Brugge; relations between plan & elevation, shifted scheme (7 feet)

373

Figure 8: facade composition for Arras; relations between plan & elevation.

374

Figure 9: facade composition for Saint-Quentin, shifted scheme (R/2).

Some considerations about symmetry.

In our times, we tend to interpret a composition as being symmetrical when it can be perfectly mirrored around an axis or a point. Through the research, it can be illustrated that the concept of compositional symmetry had a different interpretation in the Middle Ages. Often, the design of a building and of its facades is (slightly) asymmetric, due to the program of use, the situation of the building site, or for other reasons. It was not considered a problem that, e.g. the main entrance door did not sit centrally in the front facade. As such, it was at first found strange that a notorious historian of the 16th century, Justus Lipsius, praised the beautiful symmetry of the building: Magnitudo iusta est, symmetria bellissima [3]. Denes Nagy has pointed out that we indeed should refer in this case to a different meaning of the word symmetry, which lasted until the 17th-18th century. It goes back to the ancient Greek concept of symmetria, which refers to commensurability and stands

375

for the correspondence between the part and the whole, for the common measure between the different parts of the composition and their relation to the whole. It is moreover very significant that the main facades of some town halls were modified in the 18th century, to obtain the new kind of symmetry. In Leuven, one front door was added beside the existing one, and in Bruges one of the two existing doors was moved, in both cases to obtain a strictly axial symmetry. But as mentioned higher, we should consider a different approach for the medieval building master. A look back to the ground plans of Bruges, Brussels and Veere indicates already the ease with which these plans were adapted to the particularities of the building site. In the facades at Leuven, we find a particular game of asymmetry as well, for example in the series of balustrade motives (figure 10), or in the east gable facade. In the last case, a stair tower imposes a subtle shift of the axial positions of different building elements. However, these translations are combined in such way that a very balanced overall composition is obtained (figure 11).

Figure 10: roof balustrades of the town hall in Leuven

376

Figure 11: east elevation of the town hall in Leuven. Note the asymmetry, caused by the lower left tower drum, provoking eccentric positions of the gable and the corner towers as well.

A final note on medieval measurement units

Considering the different measurement systems that were in use in the towns (each Flemish town had its own privileges, jurisdiction, and even measurement units), Leslie Greenhill has suggested that those units would probably be re-combinations or reinterpretations of ancient measurements. For the Leuven rod, which is known to be 5.7102 meters, he considers that this length corresponds with 225 inches in the British Imperial system. 5.7102 meters at 39.37 inches a meter brings us indeed to 224.8121 inches, which gives a deviation of only 0.1% from the symbolic 225 inches. That measure should be seen as 15 x 15 inches, or as 6.25 (2.5 squared) yards. Square or cube numbers tend indeed to indicate

377

symbolic design. The British system is on its turn based on older, antique measurement units, but this last issue is still under investigation. The appearing of the number 225 is especially interesting in relation to antique design, as e.g. the Parthenon in Athens is known to exhibit the 225:100 proportion (length to breadth of the top of the stylobate). This ratio is often expressed as 9:4. But the use of the number 225 occurs also in other places, and would further go back to Egyptian times. Here we probably come on the track of the building lodges of the free masons, which have their roots in antiquity as well. We may mention the Greek Pythagorian brotherhoods, with their three degrees of initiation, as forefathers of the medieval companions. A consistent body of knowledge that was transferred through the centuries is the most probable assumption to start from when we want to further investigate this subject.

References
[1] Shelby, L.R., Gothic design techniques the fifteenth-century design booklets of Mathes Roriczer and Hanns Schuttermayer, Southern Illinois University Press, Carbondale and Edwardsville, 1977, p. 168, citation from "Die Satzungen des Regensburger Steinmetzentages nach dem Tiroler Huttenbuche von 1460", in Zeitschrift fur Bauwesen, 46 (1896). [2] Ghyka, M., Le nombre dor: rites et rythmes pythagoriciens dans le dveloppement de la civilisation occidentale, Gallimard, Paris, 1969, p. 72; and also Ballegeer, J., Stadhuis Brugge, De Gulden Engel, Wommelgem, 1987, p. 20 [3] Justus Lipsius, Lovanium: sive opidi et academiae eius descriptio. Libri tres, Antwerp, 1605, p. 88 (2nd edition, Antwerp, 1610, p. 90).

378

"

&

&

"

"

"

"

u d d w  "  h i j "  ! l o x   i r  t t & & u x   i r  w "  i l

"

"

"

| u } d | u

u d w x !  " " w " "  l  i "  r # i !  r " # " !

"

"

"

"

"

&

&

u u r

u u

| d u } d | t | } r

| } | d | | u u

d r u t r | t

| u d

u r t } u u | r

u | d u

u |

r } t d u |

| | u r d | u

| | } | d u u r t d u

d u | } | d

r d | u

| | } | d r } r d r u r |

r | r } d | t | } r

| r

d u } d | t | } r | } | u u

u d t t u

| | u } r d d | r } d u d u

| u | u u

r u r r | r }

t |

d t t u

d u } d | t | } r

| | d u } r u | u u

r d r

| d t t u

u r

u r

| | r } u d d t t u

| u d

| r u d u

| | } | u

r u } r

u u u d t t u

} | u

| r | z u r | d t t u

} | t | u

| | u u | } } u } r | u d u u d

r | u d t u

u } r | d u u u u t

| } | !  ! i "  # r " "  # # " i o | } u d  !  " i "   ! o "

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

&

"

"

"

"

"

379

   " : ;  o ! " # o ! #  i i "  r # i l # o " i ! = ? A B C

"

"

"

"

"

"

# i R ! # # " r d u } d | t | } r

| r

T  o ;

"

"

"

"

"

"

"

"

L M

Z T ! " \  #  i r

&

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

&

"

"

"

"

"

"

"

&

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

T o   "  i i o " r " ! o " ! # " r # i

T R ! !  ! " o "  i i o "

"

"

"

"

&

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

# i R ! # # " r u

|  o R r R v w o  " ! w r ( " i !  o " j "

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

&

"

"

"

"

"

"

"

380

 ! #  l # " " o i i   " :  ! i i " # l i l !  i !

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

&

"

"

i " i  "  o # " !  # # " r | } u d  " "  o 

T i r   # " 

T r ! ! " " # " ! ! # " r l o  i # " ! i   " i  # " !

"

"

"

"

"

"

"

"

"

T o  " i " r ! ! " " ! # " r # " !  r i " j " i   # " "

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

d ~  u u u r t | } r } } | } | d u u r u

r | r | z

r | d r d u } d | t | } r | } | u

d u } d | t | } r

| } | | u d

| d u r | r |

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

&

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

(a) HalfHex


(b) an inflation rule by Eric Lord

 " & ;   i i   # " ! o

(c)

(d) a tangram tiling

"

381

 i i i   # " i " # " r # " !  i " ! " ! o # " !  i r # ! # i !

"

"

"

"

"

"

"

"

"

"

"

"

"

&

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

d ~  u u B | r | | } | R | | u } t r } r } z

r | d

r d } r | | d u

| r

} | | r d r

u r |

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

&

"

"

"

"

"

"

"

"

"

# "  # l "  j # ! \   " # i ! & "  r

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

i i  "  l o i i "  r ! " # o ! #  # i ! i " "  o  ! 

"

"

&

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

&

"

"

"

"

"

"

"

"

"

&

"

"

"

"

"

"

&

"

&

"

"

"

"

"

"

"

"

"

"

"

"

&

"

"

"

"

i &

"

"

"

"

"

"

"

"

"

"

"

"

"

"

o # " ! !  "  l ! # # " r } | t | u |

i  # "  r l ! !  i # l

"

"

"

"

"

"

"

"

"

"

i "  "  o  # # " i "  " i i "  r # i ! g " i " i i g ! "

"

"

"

"

"

"

"

"

"

382

"

"

"

"

i r " i ! i # ! " i & r " i ! i # ! " o  # r " !  i ! ;

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

&

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

"

&

383

 r " " # \  !  l ! # !  =  " ;

r d | d r } d  " r ! 

"

"

"

"

"

"

"

"

"

! " ! ! \  !  l ! # ! z r i "  i # ! " !  d w r u

"

"

"

"

"

"

"

&

&

= ? A B w r i z   ! ; i  # " ! i r  !  i # i ! 

r } d

&

&

"

"

"

"

"

"

"

| d u u t r | r } u t u

& ( ( : : u & ( :

"

"

"

"

"

! #   i ! # i # # l i " # i !  | d u u t r | r } u t u

&

&

384

SYMMETRY GROUPS IN MATHEMATICS, ARCHITECTURE AND ART

VERA W. DE SPINADEL

Name: Vera W. de Spinadel, Mathematician, (born Buenos Aires, Argentina, 1929). Address: Centre of Mathematics & Design, Faculty of Architecture, Design and Urban Planning, University of Buenos Aires, Jos M. Paz 1131 Florida (1602) Buenos Aires, Argentina. E-mail: vspinade@fibertel.com.ar; vwinit@fadu.uba.ar; maydi@cvtci.com.ar Fields of interest: Fractal Morphology, Onset to Chaos, Number Theory (Continued Fraction Expansions of Real Numbers). Awards: Several national Research and Development prizes. Publications and/or Exhibitions: 1) The Metallic Means family and forbidden symmetries, International Mathematical Journal, vol. 2, Nr. 3, pp. 279-288, 2002; 2) Exhibition Fractal Art, Fifth Interdisciplinary Symmetry Congress and Exhibition of the ISIS-Symmetry, July 14, 2001, Sydney, Australia; 3) Continued fraction expansions and Design, Proc. of Mathematics & Design 2001 The third International Conference, July 3-5 2001, Deakin University, Geelong, Australia; 4) The family of Metallic Means, Visual Mathematics, vol. I, Nr. 3, 1999. http://members.tripod.com/vismath/; 5) From the Golden Mean to Chaos, book edited by Nueva Librera, ISBN 950-43-9329-1, 1998.

Abstract: The word symmetry has two meanings. A symmetric object is well proportioned but the concept is not restricted to concrete objects; the synonym harmony refers to its use in Acoustics and Music. The second meaning is that of the geometric bilateral symmetry, the symmetry so evident in superior animals, especially in men. From the mathematical point of view, a whole symmetry theory can be considered for applications. From this theory, we have chosen the symmetry group of the square to present interesting uses in Architecture and art.

385

1. Symmetry in Culture
The word symmetry comes from the Greek symmetria, meaning the right proportion. From the historical point of view, the term symmetry has denoted many meanings, depending on the field of human knowledge where it was used. Notwithstanding, symmetry is a unifying concept, as Hargittais, Magdolna and Istvn, have proved in their beautiful and unique book Symmetry. Indeed, the concept of symmetry can provide a connecting link among many different fields of endeavor, perhaps the best and more appropriate link to protect human studies from the increasing and separating compartmentalization within our scientific world. Going back to the year 27 B.C., we found a monumental work: the 10 books written by the roman architecture Vitruvio (probably Marco Vitruvio Pollione) and dedicated to the Emperor August. Architecture, says Vitruvio, depends from order, disposition, eurhythmy, property, symmetry and economy. These terms have today, completely different meanings. E.g., order, says Vitruvio, confers the appropriate measure to the elements of a certain building, when considered separately and symmetry, gives concordance to the proportions of the different parts of the construction. This approach to the meaning of symmetry is quite similar to the mutually corresponding arrangement of the various parts of a human body around a central axis, producing a proportioned balanced form. Vitruvio dedicated much time to the study of the proportions in the human body in his considerations on symmetry. Symmetry, says Vitruvio, comes from proportion, that is, from a correspondence between the dimensions of the parts of a whole and of the whole with respect to a certain part selected as a model, the module. Such a selection of parts of the human body as a module, initiated probably by Vitruvio, was the very beginning of a historical ergonomic chain linking Vitruvio, Albrecht Drer, Leonardo da Vinci and many, many other artists, including the modern contemporary architect Le Corbusier.

2. Symmetry in Mathematics
In Euclidean geometry, a symmetry of a figure is a rigid motion that leaves the figure unchanged. And what is a rigid motion? A rigid motion of the plane is any way of moving all the points in the plane such that

The distance between points stays the same The relative position of the points stays the same.

386

This concept of symmetry in the plane is easily generalized to symmetry in threedimensional space. The simplest rigid motion is translation. In a translation, everything is moved by the same amount and in the same direction. We specify a translation by drawing an arrow. Another rigid motion of the plane is rotation. A rotation fixes one point and everything rotates by the same angle around that point. A third rigid motion is reflection. A mirror line determines a reflection and it is easy to prove that

points on the mirror line are unchanged by reflection; the distance from a point to the mirror is the same as the distance from the image of that point to the mirror.

Finally, combining reflection with translation, we get a glide reflection, that is, a mirror reflection followed by a translation parallel to the mirror. To make the list complete, we add the do-nothing operation as another rigid motion of the plane. All these rigid motions of the plane generate different types of symmetries: Translation symmetry or repetition: it means shifting and repeating the same motif, producing a periodic pattern. In ornamental art, this type of symmetry is called infinite ratio, in which case it is a glide reflection, like in the frieze of Persian archers at Dares Palace, in Susa (at present Khuzistan, Iran), depicted in Fig. 2.1 or the Palace of the Doje in Venice, Italy (Fig. 2.2).

Fig. 2.1

Fig. 2.2

Rotational symmetry: it is an operation that iterated, brings the configuration again to its original position. To avoid ambiguity, it is normally assumed that counterclockwise rotations are positive. On the plane, the simplest figures that exhibit rotational symmetry are regular polygons. In three-dimensional space, a body has rotational symmetry with respect to an axis r if every rotation around r, takes the configuration back to its original 387

position. E.g., a four-blade pinwheel has rotational symmetry and if all its petals are of different colors, it is easy to prove that after a whole turn, we are back at the starting position. This is called a 4-fold rotational symmetry. Mirror symmetry or bilateral symmetry: it occurs when two halves of a whole are each others mirror images. This symmetry was also called heraldic symmetry, because the old inhabitants of Sumer (southern part of Mesopotamia) were the first ancient people using heraldic drawings, having its use being extended later on to Persia, Syria and Byzance. To introduce a mathematical language and redefine symmetry, let us consider a spatial configuration F. We shall say that the motions that leave F invariant -- or unchanged --, form an algebraic structure called a group of transformations G, defined in the following way: Let G be a set together with a composition law which associates to each pair g,h G another element g h G, called the product of g and h. Suppose that this product satisfies the properties: a) I G (the identity is in G) b) if s G then s-1 G (the inverse transformation is in G) c) if s and t G then the composition s t G. Then the set of transformations G forms a group. This group of transformations describes exactly the symmetries of the figure F. To describe three-dimensional space, the notion of congruence is very useful. Two spatial regions are congruent if a rigid body in two different positions can occupy them. Congruent transformations form a group and, obviously, the simplest types of congruencies are translations, reflections and rotations. The symmetry of any figure in three-dimensional space is described by a subgroup of the group G. Example 1: Let us consider the symmetries of a square. The eight distinguishable spatial transformations, which comprise this group, are four quarter-turns and four reflections (see Fig. 2.3): d1 : Do-nothing d2 : Rotation by 1/4-turn d3 : Rotation by 1/2-turn d4 : Rotation by 3/4-turn 388 d5 : Reflection in mirror m1 d6 : Reflection in mirror m2 d7 : Reflection in mirror m3 d8 : Reflection in mirror m4

Fig. 2.3 By direct inspection of the figure, it is evident that a square has four rotation symmetries and four reflection symmetries. This is a property that can be extended to all regular polygons, and it is easy to prove the following general result: The regular polygon with n sides, has exactly n rotation symmetries and n reflection symmetries. Adopting the usual composition of transformations as the fundamental operation, we may write the composition table called Cayley diagram, that gives the results of the composition dj dk , for j = 1,...,8; k = 1,...,8 and describes the group of symmetries of a square.

389

j 1 1 2 3 k 4 5 6 7 8 1 2 3 4 5 6 7 8 2 2 3 4 1 7 8 6 5 3 3 4 1 2 6 5 8 7 4 4 1 2 3 8 7 5 6 5 5 8 6 7 1 3 4 2 6 6 7 5 8 3 1 2 4 7 7 5 8 6 2 4 1 3 8 8 6 7 5 4 2 3 1

The symmetries of a square form a group (non-commutative, as is easily verified by the non-symmetrical table) and this result is generalized in the following sense: The collection of symmetries of any figure will always be a group. Furthermore, it is easy to prove that d3 = d22 ; d4 = d23 ; d6 = d22 d5 ; d7 = d2 d5 ; d8 = d23 d5 . Notice that the elements d6 , d7 and d8 can be expressed by the compositions of d2 and d5 alone. Example 2: Another interesting example is the famous pentagonal star with which Faust exorcised Mephistopheles (Fig. 2.4). It coincides with itself by performing 5 rotations of angles 72, 144, 216, 288, 360 and the 5 reflections with respect to the lines linking the center of the figure with the 5 vertices of the pentagon. These 10 operations form a group and this group indicates what sort of symmetry this star has.

390

Fig. 2.4 Note: The most fascinating and comprehensive analysis of all sort of symmetries, can be found in the book written by the mathematician D. Schattschneider (Visions of Symmetry) who has devoted much of her research to studying the work of the famous graphic M. C. Escher. In this beautifully illustrated volume, she presents a detailed symmetry inventory of Eschers periodic drawings with one motif, two or more motifs and non-interlocking patterns.

3. Symmetries in Architecture
An interesting application of the above mentioned point group symmetry of the square, has been recently presented by Jin-Ho Park. Two housing plans have been analyzed by him to show how the mathematical methods of symmetry operations are employed as thematic elements in the unit design as well as the variations in the planning of the site. The first example was Frank Lloyd Wrights social housing project, called the Quadruple Building Block. Wright (1867-1959) used a standard unit plan for the project, based on his earlier design of his own Home & Studio, Oak Park, Illinois, 1889. A version of this project was originally published as a vignette in the Ladies Home Journal article of 1901 on his Small House with Lots of Rooms in It. Whereas the unit itself was asymmetric, various local symmetries were involved in the unit. The assembly in their site layout included two types: one is the pinwheel type and the other is the mirrored reflection type. Each house was set on four equally subdivided lots, sharing a common backyard in the center for all four houses. The whole complex of the housings was laid out in a way that their elevation could be varied according to their arrangement. The Quadruple Block plan was a scheme with which Wright was actively 391

concerned from 1902 till the next 10 or 15 years. We cannot be sure that his purpose was a practical one at all because they are not inexpensive row houses designed for mass production; they are substantial dwellings on half-acre plots. Unfortunately, he never found either four families who wanted identical dwellings or a real state investor who could believe that such blocks would constitute a salable commodity. But he never abandoned this revolutionary idea of urban design. Wrights next housing designs share a common compositional theme: the pinwheel type of symmetry, characteristic of the point group symmetry of the square (Figs. 3.1 and 3.2).

Fig. 3.1

Fig. 3.2

The second example was Rudolf Michael Schindler(1887-1953) unexecuted Shelter project, developed from 1933 to 1942. In this project, the overall floor plan was based on a 5-foot square grid. Along with the square grid, the symmetry governed the internal structure of the spatial composition in each shelter unit as well as the unit variations. The internal organization was subdivided by the removable closet partitions for spatial flexibility and set along the pinwheel type of rotational symmetry. Then the garage was added to any side of the house as a separate unit. But instead of providing only a standard unit with fixed layouts, a sequence of variations were provided, variations that shared the same underlying organization system. Using the symmetry operations of the square, new eight different shelter designs were produced. Taking advantage of the possibilities of combining units into groups, Jin-Ho Park presented an experimental grouping exercise using the shelter unit corresponding to the eight transformations of the symmetry of the square in an abstract level, quite similar to what Wright envisioned in his Quadruple project.

392

4. Forbidden symmetries
Quasicrystals, which belong to a class of quasiperiodic systems, gave diffraction patterns that show local n-fold rotational symmetry forbidden in Crystallography: n = 5, 8, 10, 12. In this context, quasilattices can be thought as mathematical discrete sets supporting Bragg peaks or atomic sites. They play the same role as lattices do for crystals. Quite recently, discrete sets of numbers, the -integers Z, have been proposed as numbering tools for coordinating quasicrystalline nodes in 1, 2 or 3 dimensions, and also the Bragg peaks in diffraction patterns. In the observed cases: Golden Mean (penta- or decagonal quasilattices):

==

1+ 5 = 2 cos 2 5

Silver Mean (octagonal quasilattices)

= Ag = 1 + 2 = 1 + 2 cos

Subtle Mean (dodecagonal quasilattices)

= 3 = 2 + 3 = 2 + 2 cos

12

An important mathematical common characteristic is that the continued fraction expansions of these irrational numbers are the following:

= [ 1 ] ; Ag = [2 ] ; 3 = [ 4

where the usual notation [] for continued fractions is used. All of them are purely periodic continued fraction expansions. The relevant scale factor is a quadratic Pisot-Vijayaraghavan number or more simply, a PV number, i.e. an algebraic integer > 1 which is solution to equations of the type 393

x 2 = ax 1

( a { 1, 2, 4} )

such that all respective second roots (called Galois conjugates of ) have modulus strictly smaller than 1. These PV numbers are a subset of the Metallic Means Family (MMF), introduced by the author. The more outstanding member of this family is the well known Golden Mean, then comes the Silver Mean, the Bronze Mean, the Copper Mean, the Nickel Mean and many others. These positive quadratic irrational numbers intervene in the determination of the quasi-periodical behavior of non-linear dynamical systems, being therefore an invaluable key in the search of universal roads to chaos. Besides, the members of the MMF satisfy simultaneously many additive and geometric properties, having been in consequence advantageously adopted as bases of many architectonic systems of proportions. The role played in lattice theory by the ring of integers Z and the planar rotational compatibility condition = 2cos(2/n) Z is in quasilattices replaced by Z. In terms of the members of the MMF, quadratic Pisot numbers have the following equivalences

394

n
5 8

2 cos 2 / 5 = 1 /

1 + cos 2 / 5 =

2 cos 2 / 8 = Ag 1
2 cos 2 / 12 = 3 2 cos 2 / 12 = 3

1 + 2 cos 2 / 8 = Ag
1+2 cos 2 / 12 = 1 + 3 = [2, 1,2] 2 + 2 cos 2 / 12 = 2 + 3 = [3, 1,2]

1 2 = 2 A 1 3 = [ 1,2] 2 3 =[ 1,1,2]

12 12

The discovery of quasi-crystals with crystallographically forbidden symmetries is one of the most striking examples where a pure symmetry analysis determines mathematically forbidden symmetries appearing in a new solid state of matter. Interesting to mention, Silver films with a close-packed structure modulated by a Silver Mean quasi-periodic sequence have been experimentally obtained. This represents a new type of quasicrystal that is fundamentally different from the commonly known quasicrystals, which possess at least two-dimensional rotational order.

5. Symmetry and fractals

Fractals are geometric configurations with a built-in self-similarity, that is, configurations that remain invariant in the presence of scale changes. More simply, they possess borders, surfaces or internal structures with patterns that zoomed and zoomed until infinity, are invariant, exactly or statistically. Fractal structures are important to study because they are intrinsically related to the important notion of chaos. A physical process is said to be chaotic in relation to its dynamics that means, when it is impossible to make any type of prognosis about its future evolution, since it is verified that very similar initial conditions give rise to system behaviors that differ enormously among them. To be able to find the connection between fractal structures and chaotic processes, it is necessary to geometrize the system dynamics. Using computerized graphics, it is feasible to detect fractal patterns that dynamically are considered as universal scenarios of the roads to chaos.
395

Fig. 5.1 Let us consider some variations of a classical geometric fractal: the so-called Sierpinski gasket, depicted in Fig. 5.1. Beginning with a square, it is possible to set three contractions that reduce the initial square by a factor of 1/2 and place the resulting square appropriately

x y v1 ( x, y ) = , 2 2 x +1 y v 2 ( x, y ) = , 2 2 x y +1 v3 ( x, y ) = , 2 2 Returning to the 8 symmetry transformations of a square, we may define a family of patterns specifying a triplet w1, w2, w3, where each wi is given by the product of transformations wi = vi dk for k = 1,2,...,8 and i = 1,2,3. This makes altogether 83 = 512 different triplets w1, w2, w3, and each one describing a specific pattern. To illustrate this amazing variety of fractal patterns, all of which are close relatives of the Sierpinski gasket, let us consider 8 different sets of transformations, which are symmetric with respect to the diagonal and the corresponding fractal patterns (see Fig. 5.2).
396

Fig 5.2 This theoretical description is based on the so-called Hutchinson operator, introduced in 1981 by the mathematician J. Hutchinson. Using this operator, M. F. Barnsley was able in 1985, to devise an IFS (Iterated Function System) with which he succeeded in getting fractal images that were so near from a natural image as desired.

6. Symmetry and visual design

Ljubisa Kocic, a Serbian mathematician, has developed a simple technique for converting real numbers into linear ornaments and vice versa. To achieve this visualization of real numbers, he has recently presented two algorithms A and B, based on the continued fraction expansion of a real number. Algorithm A 1. 2. Take the continued fraction expansion of a real number x = [a 0 , a1 , a 2 , ] . Cut the sequence {a k } up to the (m + 1)-th term to obtain a 0 , a1 ,  , a m . Replace ai by ai = ai (mod 8) . The number x will be represented by a 0 , a1 , m , a m , where a i {0,1,  ,7} and it is denoted as ( x) = a , a , m , a ..

397

3.

Starting

from (x0 , y0 ) = (0,0) create {(x k , y k ), k = 1,2,  , m + 1} following the rule

x k +1 = x k + p

sequence

of

points

y k +1 = y k + q where p and q are given in the following table ai P Q

1 1 0

2 1 1

3 0 1

4 -1 1

5 -1 0

6 -1 -1

7 0 -1

0 1 -1

4.

Draw the polygonal line ( x) = {(x 0 , y 0 ), (x1 , y1 ), m , (x m+1 , y m +1 )}.

Note: This table encodes the direction of the path of a moving particle according to the compass rose: E, NE, N, NW, W, SW, S, SE. Curiously, this pseudo-random Brownian walk was used by Berthelssen et al. to investigate the quantity of information hidden into the DNA molecule chain. It is easy to see that Algorithm A, applied to the Golden Mean

1+ 5 =[ 1,1, m] , 2

will produce a horizontal line. The same will happen with all the Metallic Means that have a purely periodic continued fraction expansion, of the form [n]. But there is a lot of very interesting Metallic Means that originate very nice Brownian friezes, so called because the algorithm produces a kind of Brownian motion trajectory. In fact, since the members of the Metallic Means family are defined as the positive solutions of quadratic equations of the form x2 - px - p = 0 (p, q natural numbers), i.e.

( p, q ) =

p+

p 2 + 4q
2

it is possible to apply Algorithm A to produce beautiful Brownian friezes, like the shown at Fig. 6.1 that correspond to the following Metallic Means

398

 (3,3) =

3 + 21 1 + 21 =1+ = 1 + [2, 1,3] 2 2 1 + 33 = [3, 2,1,2,5] (1,8) = 2 1 + 129 = [6, 5,1,1,2,3,2,1,1,5,11] (1,32 ) = 2 1 + 2313 = [24, 1,1,4,1,5,5,5,1,4,1,1,47 ] (1,578) = 2

Notice that all these Metallic Means are periodic of the form [m, n1 , n 2 , ] and these periods are palindromic i.e. the periods are symmetric about their centers, except for the last number of the period, which equals 2m 1 .

Fig. 6.1 Algorithm B 1. 2. 3. Take the continued fraction expansion of x.. Cut the sequence {a k } up to the ; k = 0,1, m , m . Create a sequence of pairs {a k , a mk } Rotate each point (m+1)-th term to get a 0 , a1 , m , a m .

draw this part of the polygonal line ( x, m) = A0 , A1 , m , Am .

Ak = (a k , a mk ) counterclockwise about the origin and

The graph ( x, m) is a symmetric figure with respect to the y-axis and is called the symmetrogram of x. The reason of introducing this second algorithm is that the Brownian frieze method produces friezes that are not univocal, that is, two different numbers may have the same graph. Let us apply Algorithm B to irrational numbers containing e = 2.71828, the basis of natural logarithms.
399

For example: (e,50)

(e 2 ,65) (e 3 ,30) (e1 / 4 ,70) produce the symmetrograms of Fig. 6.2.

Fig. 6.2

400

Summarizing, the application of these two algorithms for visualization of real numbers produce patterns and ornaments that range from straight lines over periodic and cyclic to chaotic ones.

7. A visual notation for rational numbers

The Australian mathematician Julie Tolmie has presented at the ISIS-Symmetry Fifth Interdisciplinary Symmetry Congress and Exhibition held at Sydney, Australia, July 814, 2001, an interesting methodology to construct a visual notation for rational numbers. Rational numbers mod 1 are represented as equivalence classes of pairs (p,q). The torus, with its two directions of rotational symmetry is used as a phase space; its longitudinal cycle for the denominator and its meridian cycle for the numerator. Using cyclic motion and a discrete map of {1,2,3,,37} into the RedGreenBlue RGB-color space, the numerator is encoded as cyclic permutations of color coded dots. These configurations are placed in the appropriate meridian slices (in radial vertical planes), being the result a three dimensional navigable object contained in a torus (a visual abstract is provided at the site http://www.ozemail.com.au/~jatolmie/isis0.html). As is well known, the rational numbers mod 1 have a Farey tree, structure, which is called Stern-Brocot tree because it was independently, discovered by the German mathematician Moritz Stern (1858) and the French clockmaker Achille Brocot (1860). To introduce rotational symmetry, the Farey tree is made circular and embedded in the longitudinal plane of the torus. Rational numbers are then defined as curve segments, which span the longitudinal region, bounded by its Farey parents. These curve segments are cut from curves, which wind around the torus at a constant integer velocity and in this way, a three-dimensional navigable object is made.

References
Barache D., Champagne B. and Gazeau J. P. (1998), Pisot-Cyclotomic Quasilattices and their Symmetry Semi-groups, ed. J. Patera, Fields Institute Monograph Series, volume 10, Amer. Math. Soc. Barnsley M. F. (1988), Fractal modelling of real world images in The Science of Fractal Images, H.-O. Peitgen and D. Saupe (eds.), Springer-Verlag, New York. Berthelsen C. L., Glazier J. A. and Skolnick M.H. (1992), Global fractal dimension of human DNA sequence treated as pseudorandom walks, Phys. Rev. A45, 8902-8913. Bertin M. J., Decomps-Guilloux A., Grandet-Hugot M., Pathiaux-Delefosse M. and Schreiber J. P. (1992), Pisot and Salem numbers, Birkhuser Verlag. Bogomolny A., Farey Series, http://www.cut-the-knot.com/blue/Farey.html 1996-2001.

401

Burdik C., Frougny C., Gazeau J. P. and Krejcar R. (1998), Beta-Integers as Natural Counting Systems for Quasicrystals, J. Phys. A: Math. Gen., 31, 6449-6472. Elser V. (1985), Indexing problems in Quasicrystal Diffraction, Phys. Rev. B32, 4892-4898. Farmer David W. (1996), Groups and Symmetry, American Mathematical Society. Gazeau J. P. (1997), Pisot-cyclotomic Integers for Quasicrystals, The Mathematics of Aperiodic Long Range Order (ed. R. V. Moody) NATO-ASI Proceedings, Waterloo 1995, Kluwer Academic Publishers. Gazeau J. P. (2000), Counting Systems with Irrational Basis for Quasicrystals, editors: F. Axel, F. Dnoyer, J. P. Gazeau. In From Quasicrystals to more Complex Systems, No. 13. Les Houches School Proceedings, Springer-Verlag, Heidelberg, Berlin. Gazeau J. P. and Lipinski D. (1997), Quasicrystals and their Symmetries, Symmetries and structural properties of condensed matter, Zajaczkowo 1996. Ed. T. Lulek. World Scientific: Singapore. Hargittai Istvn & H. Magdolna (1960), Symmetry: a unifying concept, Shelter Publications Inc., Bolinas, California, 1994. Hutchinson J. (1981), Fractals and self-similarity, Indiana University Journal of Mathematics, 30, pp. 713747. Ikezawa K. and Kohmoto M. (1994), Energy spectrum and the critical wavefunctions of the quasiperiodic Harper equation the Silver Mean case , J. of the Phys. Soc. of Japan, 63, Nr. 6, pp. 2261-2268. Kocic Ljubisa (1999), Portraits of Numbers, http://www.mi.sanu.ac.yu/vismath/pap.html Visual Mathematics, vol. 1, No. 4,

Kocic Ljubisa (2001), Numbers and Ornaments, Third International Conference on Mathematics & Design M&D-2001, 3-5 July 2001, Deakin University, Australia. Pedoe Dan (1976), Geometry and the liberal arts, Penguin Books Ltd. Schattschneider Doris (1990), M. C. Escher: Visions of symmetry, W. H. Freeman & Co., New York. Shih Chih-Kang, Growing atomically flat metal http://www.aps.org/BAPSMAR98/abs./S278002.html films on semiconductor substrates,

Smith A. R., Chao Kuo-Jen, Niu Qian, Shih Chih-Kang (1996), formation of atomically flat Silver films on GaAs with a Silver Mean quasi periodictiy, Science, 273, pp. 226-228. Spinadel Vera W. de (1998), From the Golden Mean to Chaos, Nueva Librera, Buenos Aires. Vitruvio (1960), Los 10 libros de Arquitectura, Editorial Iberia S.A. Weyl Hermann (1952), Symmetry, Princeton University Press.

402

ADVANTAGES OF DECENTRALIZED ELECTRICITY AND HEAT SUPPLY FOR BUILDINGS, USING FUEL CELLS.
ERICO SPINADEL

Name: Erico Spinadel, Industrial Engineer ( born Vienna, Austria, 1929). Address: Argentine Wind Energy Association AAEE, Universities UBA, UNLu, EST-IESE, UFASTA. Jos Mara Paz 1131, RA1602 Florida, Buenos Aires, Argentina. E-mail: espinadel@fibertel.com.ar ; gencoeol@cvtci.com.ar ; homepage www.ecopuerto.com/organizaciones.asp Fields of interest: Wind Energy, Renewable Energies, Electric Energy, Hydrogen Technologies. Awards: Several national & international R&D prizes Publications (2001): 1) Mathematical Model for Optimizing Sizes of PEM Fuel Cells in Combined Natural Gas and Electricity Energy Supply. Proc. M&D2001, The Third International Conference, Deakin University, Geelong, Australia. ISBN 0 7300 2526 8, pp. 166 173. 2) Stand Alone Energy Islands. Proc. HYPOTHESIS IV, The 4th. International Symposium on Hydrogen Power, University of Applied Sciences, Stralsund, Germany. ISBN 3-9807963-0-2, Vol I, pp. 40 47. 3) Patagonian Wind Exported as Liquid Hydrogen. Proc. HYPOTHESIS IV, The 4th. International Symposium on Hydrogen Power, University of Applied Sciences, Stralsund, Germany. ISBN 3-9807963-0-2, Vol I, pp. 88 92. 4) Feasibility Study for the Hacienda Project (Wind-H2 Stand Alone Energy System). Proc. HYPOTHESIS IV, The 4th. International Symposium on Hydrogen Power, University of Applied Sciences, Stralsund, Germany. ISBN 3-9807963-0-2, Vol III, pp. 446 450. 5) Definicin de las Caractersticas de un Generador Eoloelctrico para Alimentacin de Electrolizadores Alcalinos. Revista Ingeniera Militar, Argentina. ISSN 0326-5560, year 18, No.43, JanJun.2001, pp. 36-37. 6) Hay conciencia de que la energa elica existe, pero la legislacin an no es suficiente. Revista Electrogremio, Argentina. ISSN 0329-3009, year 15, No.l45, Jan. 2001, pp. 74-75. 7) Celdas de combustible para uso domiciliario. Revista Megavatios, Argentina. ISSN 0325-352X, year 24, No.242, Jun 2001, pp. 102-106. 8) Condicionantes socio-econmicos para un mayor aprovechamiento de la fuente energtica primaria elica en la Argentina. Revista Megavatios, Argentina. ISSN 0325-352X, year 24, No.242, Jun 2001, pp. 120-142. 9) Energa elica, una inversin a futuro. Revista Tecnoil, Argentina. RPI 324-856, year 23, No.232, Oct. 2001, ps. 54-60.

Abstract. The fast development of fuel cells technologies helps to consider their use in decentralized energy supplies. They have an advantage over solar power: In addition to investments by a handful of start-ups, several corporations are spending more than $1 billion annually on fuel cell research. This technology is tout as a way to wean nations

403

off foreign oil, as fuel cell systems extract their hydrogen from natural gas, and to rid the countries of coal-fired and nuclear power plants. Fuel cells are also money-savers. Given the cost benefit of natural gas over gasoline and the fuel cell's more efficient extraction technology, experts predict that fuel cell systems will become attractive when they reach $1,000 a kilowatt in the next seven to 10 years. Fuel cells will then save money in places where utilities are charging more than 10 cents per kilowatt-hour, offering energy security, energy quality, and environmental quality.

1. General description of the proposal in countries with access to Natural Gas NG resources.
For several reasons, in certain countries natural gas is commercialized at very low prices. Argentina is one of those countries. For long years, electricity had been generated in Argentina by hydro energy (some 50%), by thermal energy (some 40%) and by nuclear energy (some 10%). The resulting mix was commercialized at the gross electricity market at about 32 mils per MWh (US$ 0,032/kWh). Nowadays, using the combined cycle generation (gas turbines plus steam turbines using the former wasted process heat from the gas turbine cycle) in about 33% of the total argentine active generation park of some 17 GW, the resulting mix can be commercialized at only 28 mils. Normally, small households as well as big office or apartment houses, are connected as well to the electric grid as to the natural gas distribution system, using simultaneously both energy services, for lightning and powering electrical devices the former and for heating and cooking purposes the later. Consequently, as analyzed in some papers by different authors, a typical middle class household pays for both energy services together, as a yearly average, some US$ 120 per month. The expected presence of PEM fuel cells on the market (to be used in combination with reformers producing hydrogen from natural gas, in small 10kW units with an output capacity of some 7 kW thermal and 3 kW electricity and at prices in the order of about US$5.000 per unit) will substantially modify the commercial energy demand. As analyzed by the same authors, the mentioned typical middle class household will be able to reduce the energy services bills to only some US$ 70 per month, using only the gas connection and the combination of PEM fuel cell and reformer described later. The same criteria may be applied to the energy consumption prognosis for large buildings. In this case, a very careful study must be done in each case, analyzing the required balance between thermal and electricity needs, in order to optimize the size of 404

the fuel cells to be installed, and, consequently, the overall costs of the energy provision to the building. Some guidelines for developing a mathematical model to be used in each case are also proposed, contemplating the specific thermal and electricity demand and assuming current costs for the kWh of electricity and for the cubic meter NG, in order to obtain the price-optimal system (fuel cell - commercial energy provision).

2. Strategic advantages.
A flexible, decentralized energy system allows the simultaneous generation of both electricity and heat, exactly covering the users demand. A service based on fuel cells fed by NG is the first step to an integrated Hydrogen Energy System. The second step would be the use of renewable primary energy sources, for instance, wind energy, to produce pure H2 by electrolysis. This alternative is of particular interest for countries with the possibility of implementing large wind farms (onshore or offshore). The strategic advantage becomes evident once the no dependence of fossil fuels for electricity generation is considered. Besides that advantage, hydrogen may be used as an energy accumulator as well as an energy vector. If hydrogen is produced using a renewable primary energy source and stored as pressurized gas, or in the liquid phase or by means of metallic hydrides, the entire prime energy offer may be used time independent and electricity and heat may be generated only when really needed. And last but not least, the possibility of a decentralized electricity generation system eliminates the necessity of transmission lines. Resuming, no more dependence of fossil fuels coming from abroad, as well as a more rational use of electricity, as it is generated in accordance with the real demand and without losses due to transmission lines.

3. Increased anti-terrorist security

Doubtless, high pressure NG-ducts are less vulnerable to sabotage as power lines are, or as any nuclear or thermal powerhouse. It is relatively easy to produce severe damage to a power line and black out even several big cities for a considerable time extension. In the case of an underground high-pressure gas duct, not only the sabotage would not be so easy, but also the rehabilitation of the service would demand less time. The same could be said for a distribution network.

405

4. Schematic lay out of a fuel cell.

In the next section, different fuel cells operating nowadays will be presented. All of them act according to the same principle, shown in the following diagram (Fig.1). Cathode and anode are separated by an electrolyte and are covered on the inner face by a catalyst. On the anode, molecular hydrogen is atomized, adsorbed by the catalyst and split into a proton H+ and an electron e-. On the cathode, oxygen, in most cases coming from the air input, is also atomized and adsorbed by the catalyst. It recombines then with the proton H+ that traveled through the electrolyte and the electron e- that traveled through the electric load. Surplus air or oxygen as well as water vapor is liberated at the cathode and excess hydrogen is liberated at the anode. Also a given amount of heat is liberated and may be used.

Figure 1.

5. Different types of fuel cells.

The different types are presented in (Fig.2). Values and composition of the input gases are given, as well as the average operating temperature, type of electrolyte used in each case and their average efficiency. Historically, the alkaline cell, AFC, is the oldest 406

developed and has been used and is still widely used in quite different applications. The first space missions had been equipped with this type of cells. The Proton Exchange Membrane Cell, PEM, (Fig.3) is the other known low temperature cell. Several automobile producers are experimenting with this type of cells and their cars will most probably be in series production in the very near future. This type of cells is also of great interest for its use in the theme of this presentation. The BEWAG utility of the City of Berlin in Germany has a 250 kW unit of this type currently in operation.

Figure 2.

407

Figure 3. In the range of medium temperatures, the PEM cell may also be used fueled with CNG in cars, provided that they have a reformer unit on board, for splitting the methanol into H2 and CO2. It should be noted that the Proton Exchange Membrane is made of very special materials, with different names and covered by patents by their manufacturers, characterized by its property of being only permeable to protons. In this type of cells, the catalyst is Pt, very sensitive to CO and CO2 , gases that should be avoided in order to improve the lifetime of the cell. The membrane, covered by patents by their manufacturers, are, as shown in Fig. 4, molecules containing only one hydrogen atom, loosely linked to the molecule and therefore permitting an easy hydrogen transport through the electrolyte.

408

Figure 4. The Phosphoric Acid (PAFC) is another medium temperature cell, already extensively used for combined electricity and heat production. The trouble is the high aggressivity of the phosphoric acid at the working temperature; the assumed lifespan of this type of cells is therefore of no more than some four years. In the high temperature range, lot of research is done nowadays with the Molten Carbonate (MCFC) and the Solid Oxide Fuel Cells (SOFC). As at these high temperatures both the emitted CO and theCO2 are cracked, coal gas as well as biogas may also be employed as fuel, with no reasonable fear of contaminating the catalyst with these gases.

6. Cell and System Costs.

First of all (Fig.5), the definition of efficiency in fuel cells has to be considered. It is also necessary to keep in mind the impossibility of generating electricity or heat separately, as both of them are produced simultaneously.

409

Figure 5. At present, costs are relatively high as fuel cells are produced one by one. The goals to be attained in the next years are:

A power-to-mass ratio of some 1 kg/kW. A specific cost around 100 US$/kW for the cell stack. A service life of at least 2.500 operating hours (in the case of mobile use, 5 years with an average of 30.000km/year).

The current cost of the membrane foil (in the case of a PEM cell stack of only 1kW electric power, about 1 m2) is around 500 US$. If the mentioned goals were attained, the decentralized use of those devices for simultaneous electricity and heat supply will be able to compete favorably with the current commercial electricity and heat prices. It is not necessary to mention the 410

important reduction of the environmental impact for any given amount of commercial energy used, specially the reduction of the green house effect due to the less intensive CO2 emissions. For each case, the demands of electricity and heat must be carefully analyzed, in order to optimize the design of the system, reducing the number of fuel cells required for each particular application. The peak seasonal demands of electricity and of heat (respectively during summer and winter) must be carefully studied. A mathematic model should be developed in order to perform this optimization.

7. Examples of uses in buildings in a decentralized energy supply.

The typical complete fuel cell power system is shown in Fig.6. In certain applications, some of the components may loose significance and could not to taken into account. Also, not all of the emissions should necessarily be evaluated in a first approach.

Figure 6.

411

8. Steps to be followed for an adequate system proposal.

Normally, one should analyze the possible utilization of fuel cells for the decentralized co-generation of electricity and heat, connected to the natural gas net. After studying different extreme summer and winter conditions, the requirements for a mathematical modeling to optimize the quantity of fuel cells should be stated. This optimization is of an economical nature, since its main aim is to obtain the price-optimal system (fuel cell commercial energy provision) for different values of the variables. The following items must be taken into consideration for establishing this mathematical model. First of all, it is to point out that this is really one of the first historical analyses of a combined decentralized electricity and heat supply for buildings. No significant previous experiences are therefore available. A careful balance of the electric and thermal demand as a function of time must be established. Proper use of the eventual excess of electricity or heat must also be carefully studied. As previously pointed out, both energy forms, electricity and heat, will always be produced simultaneously by the fuel cell, and this is why the ENERGY demand must be optimized. But, at the same time, peaks in the electricity and heat demands must be diminished, in order to optimize also the POWER demand. The designer must also take into account different possible technological alternatives related to each energy demand (for instance, refrigeration, lightning, etc.). Lots of variables appear simultaneously in the problem; all of them are easy to analyze one at a time, but sometimes very difficult to evaluate while acting together, due to their interactions. Optimizing means that both, energy offer and energy demand, need to be simultaneously dimensioned, in accordance one with the other.

References.
Barlow R. Residential Fuel Cells: Hope or Hype? Published by Homepower Magazine N 72, August 1999. Available under http://www.humboldt1.com/~michael.welch/extras/homefuelcells.pdf. Hart David L. Hydrogen Power The commercial future of the ultimate fuel. Published by Financial Time Energy Publishing, London, UK, 1997. Lehmann J. Costs an niches hydrogen prices today. Proc. HYPOTHESIS IV, The 4th International Symposium on Hydrogen Power, University of Applied Sciences, Stralsund, Germany September 2001.

412

Lehmann J. Fuel cells and hydrogen. Proc. Fuel Cell Seminar, Universidad Tecnolgica Nacional, Unidad Acadmica Concordia, Argentina, March 2002. Ogden Joan M. and Nitsch J. Solar Hydrogen, Published as Chapter 22 of Renewable Energies Sources for fuel and electricity by Island Press, Washington DC, U.S.A., 1993. Ogden Joan M. and Williams R.H. Solar Hydrogen Moving beyond fossil fuels, Published by the World Resources Institute, Washington DC, U.S.A., October 1989. Reidpath M.M. Fuel Cell Technology Overview. Federal Energy Technology Center, U.S. Dept. of Energy, U.S.A., August 1999. Schnurnberger W. Technologies for a sustainable energy economy: renewable energy resources and hydrogen. Proc. HYPOTHESIS IV, The 4th. International Symposium on Hydrogen Power, University of Applied Sciences, Stralsund, Germany, September 2001. Spinadel E, Gracia Nuez S.L., Maislin J., Wurster R. and Gamallo F. Domestic Electricity Generation by means of hydrogen and fuel cells, Proc. Conference of Technological Innovations in Electricity Distribution Systems, School of Electrical Engineering, Catholic University of Valparaiso, Valparaiso, Chile, October 1999. Spinadel E., Spinadel V., Gamallo, F. Mathematical Model for Optimizing Sizes of PEM Fuel Cells in Combined Natural Gas and Electricity Energy Supply. Proc. M&D2001, The Third International Conference, Deakin University, Geelong, Australia, July 2001. Sun Media GmbH H2 Tec The Magazine for H2 and Fuel Cells. Published by Sun Media GmbH, Hannover, Germany, April 2001. Tetzlaff K.H. The better way to an energy economy without emissions. Proc. HYPOTHESIS IV, The 4th. International Symposium on Hydrogen Power, University of Applied Sciences, Stralsund, Germany, September 2001.

10. Acknowledgment.

To Florencio Gamallo, my first assistant in the GENCo R&D Group at the University of Buenos Aires, board member of the AAEE and co-author of several of my publications.

IN MEMORIAM OF PROF. SABAS LUIS GRACIA NUEZ

413

414

Day 5

Study visit to Bruges, 2002 April 13.

The final day led the participants to Bruges, where Ann Ratinckx organized a guided walk (the charming girl on the right of the picture on the left). It led to the statues of Simon Stevin (below), the Flemish da Vinci, and Van Eyck, the master of Flemish art. An exclusive visit to the new concert hall was equally part of the program (left, second photo from top). One of the architects of the building, Hilde Daem, was the guide. This remarkable concert hall was an exponent of the Bruges 2002 art festival, when the city of Bruges was the Cultural Capital of Europe. The brand new concert hall is hoped to play a role similar to the Guggenheim museum in Bilbao. More information can be found on the site http://www.concertgebouw.be/browser_ok.asp Finally, the visit led to the Memling museum, located at the "St. Jans Hospitaal" (picture down left; see http://www.trabel.com/brugge-m-memling.htm). Hans Memling (1435/40(?) - 1494) is one of the most important Flemish primitives. One of the masterpieces is the Shrine of St.Ursula, and another the Altarpiece of Saint-John the Baptist and SaintJohn the Evangelist (painting down right).

415