and Origin of Life Theory to Interactive Art on the Internet

Modeling Emergence of Complexity: the Application of Complex System

and Origin of Life Theory to Interactive Art on the Internet

Christa SOMMERER & Laurent MIGNONNEAU

ATR Media Integration and Communications Research Lab

2-2 Hikaridai, Seika-cho, Soraku-gun

61902 Kyoto, Japan

christa@mic.atr.co.jp, laurent@mic.atr.co.jp

Abstract

The origin of this paper lies in the fundamental question ofhow complexity arose in the development of life and howone could construct an artistic interactive system that canmodel and simulate this emergence of complexity. Based onthe idea that interaction and communication between entitiesof a system are the driving forces for the emergence ofhigher and more complex structures than its mere parts, wepropose to apply principles of Complex System Theory tothe creation of an interactive, computer generated andaudience participatory artwork on the Internet and to testwhether complexity within the system can emerge.

1. Introduction

The Internet seems to be especially capable of dealing withinteractions and transformations of data and information.Users on the Internet can be considered entities or particleswho transport information, such as written texts or images.As these data of information or entities are carried fromlocation to location they could, in principle, change theirstatus and value. We could imagine a system that canincrease its internal complexity as more and more usersinteract with its information. Just as a genetic string or“meme” (Blackmoore & Dawkins, 1999), these strings ofinformation would change and mutate as they aretransmitted by the users; they eventually could create aninterconnected system that features, similar to the modelspresented by Stewart Kauffman (1995), a phase transitiontoward more complex structures. Based on theseconsiderations, we propose a first prototype system formodeling a complex system for the Internet, introduce itsconstruction principles and translation mechanisms, andanalyze how the data of information have changed overtime.

creation of a computer generated and audienceparticipatory networked system on the Internet. ComplexSystems Theory is a field of research that allows simplersubsystems to increase in complexity by using phasetransitions. These phase transitions take place when anetwork of particles is given and these particles can switchone another on or off to catalyze or inhibit their production.The proposal of this paper is to test the principle of phasetransition for an interconnected web of people who cantransmit visual and written information over the Internet.As the information is transported from location to locationit will be transformed, creating an interconnected open-ended system that features phase transitions toward morecomplex structures. Before investigating how to actuallybuild the system, a short summary is given of the theoriesthat ground this research proposal.

3. Origin of Life Theories

The search for “laws of form” to explain the patterns oforder and complexity seen in nature has intriguedresearchers and philosophers since the Age ofEnlightenment. These searchers have included famousscholars such as William Bateson (1894), Richard Owen(1861), Hans Driesch (1914), D’Arcy WentworthThompson (1942), and Conrad Waddington (1966). Theirquest could generally be subsumed under the term RationalMorphology, a counterpart to the functionalistic approachof the Natural Theology promoted by Charles Darwin(1859, 1959) and Neo-Darwinist Richard Dawkins (1986).Whereas Natural Theology considers form mainly afunction of natural selection and adaptation, RationalMorphologists emphasize the creative principle ofemergence that accounts for the order of structures foundin nature. The quest for the “laws of form” is closely linkedto the question of the Emergence of Life. The discussionon how life emerged has a long tradition and basicallyinvolves two opposing views: the Artistotelian and thePlatonic. These two views of the natural world havedominated science over the past two millennia (Lewin,1993). Baltscheffsky (1997) notes that “Fundamental to adeeper understanding of complex biological functions are

2. Conceptual Objective

The aim of this research is to construct an Internet basedinteractive artwork that applies and tests principles ofComplex System Theory and Origin of Life theories to the

ideas about how life originated and evolved. They includequestions about how the first compounds, essential to life,appeared on Earth; how the first replicating moleculescame into being; how RNA and DNA were formed; howprokaryotes and the earliest eukaryotes emerged; howdifferent species, with traits like susceptibility, sentience,perception, cognition, and self-consciousness, and withvarious patterns of behaviors, evolved; and how with thesedevelopments, the environment and the ecological systemschanged. ”

Speculations of how life on earth might have originatedhave a long history, perhaps as long as the history ofhumanity. The widely accepted hypothesis that lifeoriginated from chemical processes largely derives fromthe work of Russian biochemist Alexander I. Oparin (1924,translated to English, 1938). In the 1930s, Alexander I.Oparin and J.B.S. Haldane (1932) suggested that life onearth could have emerged by natural means in an earlyatmosphere filled with different gases such as methane,ammonia, hydrogen and water vapor. Oparin and Haldanecalled this early atmosphere the Primordial Soup. In theirPrimordial Soup Theory, life would have originated in thesea as a reaction of these chemical gases triggered by theenergy of lightning, ultraviolet radiation, volcanic heat andnatural radioactivity.

In the early 1950s, Stanley Miller (1953) of the Universityof Chicago's Chemistry Department simulated such aprimordial atmosphere and was able to synthesizesignificant amounts of amino acids, main components ofall life forms, from methane, ammonia, water vapor andhydrogen. This experiment gave credence to the belief thatthe chemical building blocks of life could be created bynatural physical processes in the primordial environment.Modern proponents of the Primordial Soup Theory nowthink that the first living things were random replicatorsthat assembled themselves from components floatingaround in the primordial soup (Miller, 1953). Based onexperiments by Sol Spiegelman (1967), who was able tocreate self-replicating RNA strings in an environment filledwith a primitive “seed” virus and a constant supply ofreplicase enzymes, Manfred Eigen (1992) went a stepfurther by omitting the initial “seed” virus. Eigensucceeded in showing that self-replicating RNA strandscan assemble themselves from only replicase enzymes. InEigen’s theory of the origin of life, RNA molecules canevolve self-replicating patterns and finally develop aprimitive genetic code. As the molecules specify and takeon different functions, complex and cooperativeinteractions take place: Eigen calls these the “hypercycles”(Eigen, 1992). Mutation and competition among thesehypercycles finally create prototypes of modern cells andthe earlier chemical evolution is finally replaced bybiological evolution. A similar theory on the origin of lifewas also presented by Walter Gilbert (1986).

Even though the “RNA world” model seems veryconvincing, the question of where RNA came from in thefirst place remains open. L. Orgel (1987), C. Böhler(1995), and P. Nielsen (1991) found that a peptide nucleicacid, called PNA, could be a pre-form of RNA because itcan act to transcribe its detailed genetic informationdirectly to RNA; consequently, PNA could have initiatedthe RNA world. Another scientist, Hendrik Tiedemannsuggests that the nucleotide bases and sugars needed inRNA could have been built from hydrogen cyanide andformaldehyde, both available in the early atmosphere of theEarth.

Completely opposite to the “RNA world” theories on theorigin of life is the Dual-Origin Theory of A.G. Cairns-Smith (1982). According to Cairns-Smith, the startingpoint in early crystallization of life was not “high-tech”carbon but “low-tech” silicon, a component of clay. In histheory clay has the capacity to grow and re-assemble itselfby exchanging its ion components through mutation andmechanical imperfections. More recent proponents of themineral and early molecular based theories on themolecular evolution of metabolism subscribe to the “iron-sulphur world” theory of Wächtershäuser (1997), the“thioester world” theory of deDuve (1991), and the“inorganic pyrophosphate world” or “PPi world” theory ofBaltscheffsky (1991). Wächtershäuser (1994) proposes amodel where early evolution of life as a process beginswith chemical necessity and winds up in genetic exchange.Somewhat related to the question of how life occurred inthe first place, whether the first stages of life weremetabolic or genetic, is the question of how to draw theline between life and non-life. While generally it is agreedthat the RNA world (Gilbert, Eigen, Böhler, Nielsen,Orgel) is a first stage of life, Wächtershäuser (1997) andothers believe that rather primitive entities on mineralsurfaces can also be called alive; however he calls them“two-dimensional life. ” On the other hand, Maynard Smithand Szathmáry (1995) stress that a living organism needsto possess at minimum a reproduction mechanism, andGánti (1979) proposes that a minimum requirement for aliving organism is that it possesses three essentialsubsystems: a genetic system, a functioning unitsynthesizing the components, and a membrane part.Another big question in understanding life's origin is todetermine the origin of the translation apparatus and thegenetic code (Crick, 1968, Crick et al. 1976, Woese, 1967).Clas Blomberg (1994) claims that the only way to get astable translation mechanism is a feedback between thecode and the proteins that were synthesized by themechanisms they controlled. Furthermore, Maynard Smithand Szathmáry (1995) suggest that the relations betweenamino acids and nucleic acid sequences were establishedbefore the translation apparatus, serving as an improvedcatalyst in the RNA world.

It would exceed the scope of this report to describe all theother theories on the origin of life in detail; however someof them should be mentioned here briefly: the “MembraneFirst” theory of Harold Morowitz (1992) and the “Self-replicating protein” theory of Ghadiri et al. (1996).Theories that life was first introduced by meteorites thatcame from other planets or stars include the“Radiopanspermia” theory of Hoyle and Wickramasinghe(1979) and the “handedness of the solar system” theory andits influence on the origin of life of Carl Chyba (1997) aswell as the “Chirality” theories of Yoshihisa Inoue (1992).John Casti notes in the manuscript of his forthcoming book“Paradigms Regained” (Casti, 2000) (from which much ofthe above information is taken), that “when it comes todefining what it means to be alive, there are as manyanswers as there are biologists.” While the numeroustheories about the origin of life suggest that scientists todayare still in the dark about the details of life's beginnings andhave not been able to create it from scratch, RichardDawkins (1986) argues that this is rather to be expected.“If the spontaneous origin of life turned out to be aprobable enough event to have occurred during a few man-decades in which chemists have done their experiments,then life should have arisen many times on Earth and manytimes on planets within the radio range of Earth. ”

4. Complex System Theory

Closely related to the question of how life on earthoriginated is the question of how complexity arises.Complex System Theory, as a field of research, hasemerged in the past decade. It approaches the question ofhow life on earth could have appeared by searching forinherent structures in living systems and trying to definecommon patterns within these structures. Among others,researchers at the Santa Fe Institute in New Mexico, USAhave been looking at emergent structures in nature andhave called this approach the new science of ComplexSystem Theory. Stuart Kauffman is one of the mostprominent proponents of this new theory. According toKauffman (1995), the pure evolutionary view of nature inthe Darwinian sense fails to explain the vast structures oforder found in nature. By stressing only natural selection,patterns of spontaneous order cannot be sufficientlydescribed or predicted. In Kauffman’s view, this orderarises naturally as an “order for free.” As a consequence,life is an expected phenomenon deeply rooted in thepossibilities of the structures themselves. Kauffman arguesthat, considering how unlikely it is for life to have occurredby chance, there must be a simpler and more probableunderlying principle. He hypothesizes that life actually is anatural property of complex chemical systems and that ifthe number of different kinds of molecules in a chemicalsoup passes a certain threshold, a self-sustaining networkof reactions - an autocatalytic metabolism - will suddenlyappear. It is thus the interaction between these moleculesthat enables the system to become more complex than itsmere components taken by themselves.

4.1. Complexity through Phase Transition

Kauffman and other researchers at the Santa Fe Institutefor Complex Systems Research call the transition betweenthe areas of simple activity patterns and complex activitypatterns a phase transition. Kauffman (1995) has modeled ahypothetical circuitry of molecules that can switch eachother on or off to catalyze or inhibit one of theirproduction. As a consequence of this collective andinterconnected catalysis or closure, more complexmolecules are catalyzed, which again function as catalyzersfor even more complex molecules. Kaufmann argues that,given that a critical molecular diversity of molecules hasappeared, life can occur as catalytic closure itselfcrystallizes. A model built by Kauffman is the Booleannetwork model, which basically describes the connectionsand relations between three elements (Kauffman, 1995).The networks described by Kauffman in the Booleannetwork model show stablility, homeostatis, and the abilityto cope with minor modifications when mutated; they arestable as well as flexible. The poised state between stabilityand flexibility is commonly referred to as the “edge ofchaos.”

Other researchers have also analyzed this phase transitionbetween order and chaos. Brian Goodwin (1994) describesthis transition phase as a kind of biological attractor: “Forcomplex non-linear dynamic systems with rich networks ofinteracting elements, there is an attractor that lies betweena region of chaotic behaviour and one that is ‘frozen’ in theordered regime, with little spontaneous activity. Then anysuch system, be it a developing organism, a brain, an insectcolony, or an ecosystem will tend to settle dynamically atthe edge of chaos. If it moves into the chaotic regime it willcome out again of its own accord; and if it strays too farinto the ordered regime it will tend to “melt” back intodynamic fluidity where there is a rich but labile order, onethat is inherently unstable and open to change.”

4.2. Life at the Edge of Chaos

Two of the first scientists to describe the idea of complexpatterns and the ones who defined the term “life at the edgeof chaos” were Christopher Langton (1992) and NormanPackard. They discovered that in a simulation of cellularautomata there exists a transition region that separates thedomains of chaos and order. Cellular automata wereinvented in the 1950s by John Von Neumann (1966). Theyform a complex dynamical system of squares or cells thatcan change their inner states from black to white accordingto the general rules of the system and the states of theneighboring cells. When Langton and Packard observed thebehaviour of cellular automata, they found that althoughthe cellular automata obey simple rules of interaction of thetype described by Stephen Wolfram (1986), they candevelop complex patterns of activity. As these complexdynamic patterns develop and roam across the entiresystem, global structures emerge from local activity rules,which is a typical feature of complex systems. Langton and

Packard’s automata indeed show some kind of phasetransition between three states. Langton and Packardhypothesize that the third stage of high communication isalso the best place for adaptation and change and in factwould be the best place to provide maximum opportunitiesfor the system to evolve dynamic strategies of survival.They furthermore suggest that this stage is an attractor forevolving systems. Subsequently, they called the transitionphase of this third stage “life at the edge of chaos”(Langton, 1992).

Other researchers at the Santa Fe Institute have extendedthis idea of life found in this transition phase and applied itto chemistry. In 1992, Walter Fontana developed a logicalcalculus that can explore the emergence of catalytic closurein networks of polymers (Fontana, 1992). A relatedapproach is seen in the models of physicist Per Bak (1991),who sees a connection between the idea of phase transition,or “life at the edge of chaos,” and the physical world, inthis case a sand pile onto which sand is added at a constantrate (Bak, 1991).

To summarize, we can see that the various observationsand models of Kauffman, Langton, Packard, Fontana andBak describe complex adaptive systems, systems at the“edge of chaos” where internal changes can be describedby a power law distribution. These systems are at the pointof maximum computational ability, maximum fitness andmaximum evolvability. It is hypothesized that these modelscould indeed function to explain the emergence of life andcomplexity in nature. While Kauffman's concept of phasetransition is not the only model for creating complexity(many more approaches are currently being discussed on-line, see: www.comdig.org, http://necsi.org/ or published inrecent conference proceedings, see: Bar-Yam, 2000), itdoes however provide an advantageous starting point forcreating an artistic system that tries to incorporate some ofthe features of complex adaptive systems.

5. VERBARIUM - Modeling Emergence ofComplexity for Interactive Art on the Internet

Based on the above objective and the literature search inOrigin of Life and Complex System Theories, with specialfocus on the concept of phase transitions, we havedeveloped a first prototype to model a complex system forthe Internet (Sommerer and Mignonneau, 1999).

Artists have been working with the potential of userinteraction on the Internet over the past several years, andsome of the pioneering artworks include works by Anzai(1994), Fujihata (1996), Amerika (1997), and Goldberg(1998). A good overview of this work is also provided bythe on-line exhibition “Net-Condition” at the ZKM Centerin Karlsruhe, Germany (ZKM, 1999). While many of theabove works feature a significant amount of userinteraction, their main interest does not seem to be based

on the objective of modeling complexity as described inChapter 2 of this paper.

Our system, called VERBARIUM, is an interactive website where users can choose to write email messages thatare immediately translated into visual 3-D shapes. As theon-line users write various messages to theVERBARIUM's web site, these messages are translated byour in-house Text-to-Form editor into various 3D shapes.By accumulation, these collective shapes can create morecomplex image structures than the initial input elements. Itis anticipated that through the users increased interactionwith the system increasingly complex image structures willemerge over time.

5.1. VERBARIUM System Over View

VERBARIUM is available on-line at the following webpage: http://www.fondation.cartier.fr/verbarium.html.

The on-line user of VERBARIUM can create 3-D shapesin real-time by writing a text message within the interactivetext input editor in the lower-left window of the web site.Within seconds the server receives this message andtranslates it into a 3-D shape that appears on the upper-leftwindow of the web site. Additionally, this shape isintegrated into the upper-right window of the site, where allmessages transformed into shapes are stored in a collectiveimage. An example screenshot of the VERBARIUM website is shown in Fig. 1.

Fig. 1 VERBARIUM web page

VERBARIUM consists of the following elements:1. a JAVA based web site (Fig.1)2. an interactive text input editor

(lower-left window in Fig.1)

3. a graphical display window to display the 3-D forms

(upper-left window in Fig.1)

4. a collective display window to display the collective 3-D

forms (upper-right window in Fig.1))

5. a genetic Text-to-Form editor to translate text characters

into design functions

5.2. VERBARIUM´s Text-to-Form Editor

We have set up a system that uses the simplest possiblecomponent for a 3-D form that can subsequently model andassemble more complex structures. The simplest possibleform we constructed is a ring composed of 8 vertices. Thisring can be extruded in x, y and z axes, and during theextrusion process the rings’ vertices can be modified in x, yand z axes as well. Through addition and constantmodification of the ring parameters, the entire structure cangrow, branch and develop. Different possiblemanipulations, such as scaling, translating, stretching,rotating and branching of the ring and segment parameters,creates diverse and constantly growing structures, such asthose shown in Fig. 2.

Fig. 2 Example of VERBARIUMS’s growing structuresFigure 2a shows the basic ring with 8 vertices, and Fig. 2bshows the extruded ring that forms a segment. Figures 2cand 2d show branching possibilities, with branching takingplace on the same place (=internodium) (2c) or on differentinternodiums (2d). There can be several branches attachedto one internodium. Figure 2e shows an example ofsegment rotation, and Fig. 2h shows the combination ofrotation and branching. Figures 2f and 2g are different

examples of scaling. In total, there are about 50 differentdesign functions, which are organized into the designfunction look-up table (Fig. 3). These functions areresponsible for “sculpting” the default ring throughmodifications of its vertex parameters.

function1 translate ring for certain amount (a) in xfunction2 translate ring for certain amount (a) in yfunction3 translate ring for certain amount (a) in zfunction4 rotate ring for certain amount (b) in xfunction5 rotate ring for certain amount (b) in yfunction6 rotate ring for certain amount (b) in zfunction7 scale ring for certain amount (c) in xfunction8 scale ring for certain amount (c) in yfunction9 scale ring for certain amount (c) in zfunction10 copy whole segment(s)

function11 compose a new texture for segment(s)function12 copy texture of segment(s)

function13 change parameters of RED in segment(s)texturefunction14 change parameters of GREENinsegment(s)texture

function15 change parameters of BLUE insegment(s)texture

function16 change patterns of segment(s)texturefunction17 exchange positions of segmentsfunction18 add segment vertices

function19 divide segment in x to create branchfunction20 divide segment in y to create branchfunction21 divide segment in z to create branchfunction22 create new internodium(s) for branch(es)function23 add or replace some of the above functionsfunction24 randomize the next parameters

function25 copy parts of the previous operation

function26 add the new parameter to previous parameterfunction27 ignore the current parameterfunction28 ignore the next parameter

function29 replace the previous parameter by newparameter.............function50

Fig. 3 VERBARIUM’s design function table

The translation of the actual text characters of the user´semail message into design function values is done byassigning ASCII values to each text character according tothe standard ASCII table shown in Fig. 4.

33 ! 34 \" 35 # 36 $ 37 % 38 & 39 '

40 ( 41 ) 42 * 43 + 44 , 45 - 46 . 47 /48 0 49 1 50 2 51 3 52 4 53 5 54 6 55 756 8 57 9 58 : 59 ; 60 < 61 = 62 > 63 ?64@ 65 A 66 B 67 C 68 D 69 E 70 F 71 G72 H 73 I 74 J 75 K 76 L 77 M 78 N 79 O80 P 81 Q 82 R 83 S 84 T 85 U 86 V 87 W88 X 89 Y 90 Z 91 [ 92 \\ 93 ] 94 ^ 95 _96 ` 97 a 98 b 99 c 100 d 101 e 102 f 103 g104 h 105 i 106 j 107 k 108 l 109 m 110 n 111 o112 p 113 q 114 r 115 s 116 t 117 u 118 v 119 w

120 x 121 y 122 z 123 { 124 | 125 } 126 ~Fig. 4 ASCII tableEach text character refers to an integer. We can nowproceed by assigning this value to a random seed functionrseed. In our text example from Fig. 5, T of This has theASCII value 84, hence the assigned random seed functionfor T becomes rseed(84). This random seed function nowdefines an infinite sequence of linearly distributed randomnumbers with a floating point precision of 4 bytes (floatvalues are between 0.0 and 1.0). These random numbersfor the first character of the word This will become theactual values for the modification parameters in the designfunction table. Note that the random number we use is a so-called “pseudo random,” generated by an algorithm with48-bit precision, meaning that if the same rseed is calledonce more, the same sequence of linearly distributedrandom numbers will be called. Which of the designfunctions in the design function table are actually updatedis determined by the following characters of the text, i.e.,his; we then assign their ASCII values (104 for h, 105 for i,115 for s ...), which again provide us with random seedfunctions rseed(104), rseed(105), rseed(115). Theserandom seed functions are then used to update and modifythe corresponding design functions in the design functionlook-up table, between design function1 and function50.For example, by multiplying the first random number ofrseed(104) by 10, we get the integer that assigns theamount of functions that will be updated. Which of the 50functions are precisely updated is decided by the followingrandom numbers of rseed(104) (as there are 50 differentfunctions available, the following floats are multiplied by50 to create integers). Figure 5 shows in detail how theentire assignment of random numbers to design functionsoperates. As mentioned above, the actual float values forthe update parameters come from the random seed functionof the first character of the word, rseed(84).Example word: This

T => rseed(84) => {0.36784, 0.553688, 0.100701,...}

(actual values for the update parameters)h => rseed(104) => {0.52244, 0.67612, 0.90101,...} #0.52244 * 10 => get integer 5 => 5 differentfunctions are called within design function table #0.67612 * 50 => get integer 33 => function 33within design function table will be updated by value0.36784 from 1. value of rseed(84)

#0.90101 * 50 => get integer 45 => function 45within design function table will be updated by value0.553688 from 2. value rseed(84)........ until 5. value

Fig. 5 Example of assignment between random functionsand design functions

As explained earlier, the basic “module” is a ring that cangrow and assemble into segments that can then grow andbranch to create more complex structures as the incomingtext messages modify and “sculpt” the basic module by thedesign functions available in the design function table inFig. 3.

5.3. VERBARIUM´s Complexity Potential

Depending on the complexity of the incoming textmessages, the 3-D forms become increasingly shaped,modulated and varied. As there is usually great variationamong the texts, the forms themselves also vary greatly inappearance. As a result, each individual text messagecreates a very specific three-dimensional structure that canat times look like an organic tree or at other times lookmore like an abstract form. All forms together build acollective image displayed in the upper-right window ofthe web site: it is proposed that a complex image structurecould emerge that represents a new type of structure that isnot solely an accumulation of its parts but insteadrepresents the amount and type of interactions of the userswith the system. Another example of forms created by adifferent text message is shown in Fig. 6, this time the textwas written in French.

Fig. 6 VERBARIUM web page - example

6. Conclusions and Outlook

We have introduced an interactive system for the Internetthat enables on-line users to create 3-D shapes by sendingtext messages to the VERBARIUM web site. Using our

text-to-form editor, this system translates the textparameters into design parameters for the creation andmodulation of 3-D shapes. These shapes can becomeincreasingly complex as the users interact with the system.A collective image hosts and integrates all of the incomingmessages that have been transformed into 3-D images, andas users increasingly interact with the system it isanticipated that an increasingly complex structure willemerge. As it will no longer be possible to deconstruct thecollective image into its initial parts, some of the featuresof complex systems are thought to have emerged.However, it remains to be tested whether one can call thissystem a truly complex and emerging system. Futureversions of the system should address the currentshortcomings such as the limited amount of designfunctions as well as the somewhat nontransparenttranslation process. Furthermore, we plan to expand thecapacity of the system to simultaneously display allmessages in the browser's window; this should make itpossible for users to retrieve all messages ever sent.Finally, another crucial aspect will be the ability of theforms to start to interact with each other more actively; thiscould be done by using the genetic exchange ofinformation (text characters) between forms, creating off-spring forms through standard genetic cross-over, andmutation operations as we have used them in the past(Sommerer and Mignonneau, 1997).

References

Amerika, M. 1997. Gammatron,http://www.grammatron.com/

Anzai, T. and Nakamura, R. 1994. RENGA,http://www.renga.com/

Baas, N., Olesen, M. W. and Rasmussen, S. 1999.Generation of Higher Order Emergent Structures. PhysicaD. North Holland Publishing Company.

Bak, P. and Chen, K. 1991. Self-Organized Criticality.Scientific American. January 1991, pp. 46-54.

Baltschefsky, H. 1997. Major \"Anastrophes\" in the Originand Early Evolution of Biological Energy Conversion. J.Theoretical Biology, 187, pp.495-501.

Bar-Yam, Y (ed.), 2000. Unifying Themes in ComplexSystems: Proceedings of the First Necsi InternationalConference on Complex Systems, New York: PerseusBooks.

Bateson, W. 1894. Materials for the Study of Variation.Cambridge: University Press.

Blackmoore, S. and Dawkins, R. 1999. The MemeMachine. New York: Oxford University Press

Blomberg, C., Liljenström, H. Lindahl, B.I.B. and Århem,P. eds. 1994. Mind and matter: essays from biology,physics and philosophy. J. Theoretical Biology, 171, pp.1-122.

Böhler, C., Nielsen, P.E. and Orgel, L.E. 1995. NatureVol.376. pp. 578-581.

Cairns-Smith, A.G. 1982. Genetic Takeover. Cambridge:Cambridge University Press.

Cairns-Smith, A.G. 1986. Clay Minerals and the Origin ofLife. Cambridge: Cambridge University Press

Casti J. 2000. Paradigms Regained. New York: WilliamMorrow & Co. forthcoming.

Chyba, C. 1997. A Left-Handed Solar System? Nature,Vol.389, pp. 234-235.

Clancey, W. J. 1979b. Transfer of Rule-Based Expertisethrough a Tutorial Dialogue. Ph.D. diss., Dept. of Com-puter Science, Stanford Univ.

Crick, F.H.C. 1968. The origin of the genetic code. J.Molecular Biology, Vol. 38, pp. 367-379.

Crick, F.H.C., Brenner, S., Klug, A. and Pieczenik, G.1976. A speculation of the origin of protein synthesis.Origins Life, Vol. 7, pp.389-397.

Darwin, C. 1959 [1859]. The Origin of Species. Reprinted.London: Penguin.

Dawkins, R. 1986. The Blind Watchmaker: why theEvidence of Evolution Reveals a Universe without Design.London: Longman.

De Duve, C. 1991. The Beginnings of Life on Earth.American Scientist, Vol. 83, pp. 428-437.

Driesch H. 1914. The History and Theory of Vitalism.London: Macmillan.

Eigen, M. and Winler-Oswatitch, R. 1992. Steps TowardsLife: A Perspective on Evolution. New York: OxfordUniversity Press.

Fontana, W. 1992. Algorithmic Chemistry. In: C.G.Langton, J.D. Farmer, S. Rasmussen and C. Taylor (eds.)Artificial Life II: Santa Fe Institute Studies in the Sciencesof Complexity. Vol. 10. Reading, MA: Addison-Wesely,pp. 159-209.

Fujihata M. 1996. Light on the Net Project,http://www.flab.mag.keio.ac.jp/light/

Gánti, T. 1979. A Theory of Biochemical Supersystems andits Application to Problems of Natural and ArtificialBiogenesis. Baltimore, MD: University Park Press.

Gilbert, W. 1986. The RNA World. Nature, Vol. 319, p.618.

Goldberg, K. 1998. Memento Mori: an Interface to theEarth, http://memento.ieor.berkeley.edu/

Goodwin, B. 1994. How the Leopard Changed its Spots.London: Orion Books (Phoenix), p. 169.

Haldane, J.B.S. 1932. The Forces of Evolution. London:Longmans Green.

Hoyle, F. and Wickramasinghe, N.C. 1979. Diseases fromSpace. New York: Harper and Row.

Inoue, J. 1992. Asymmetric Photochemical Reactions inSolution. Chemical Reviews, Vol.92, pp.741-770.

Kauffman, S. 1995. At Home in the Universe. The Searchfor Laws of Complexity. New York: Oxford UniversityPress, pp. 23 - 112.

Langton, C. 1992. Life at the edge of chaos. In : C.G.Langton, J.D. Farmer, S. Rasmussen and C. Taylor (eds.)Artificial Life II: Santa Fe Institute Studies in the Sciencesof Complexity . Vol. 10. Reading, MA: Addison-Wesely,pp. 41-91.

Lee, D.H., Granja, J.R.; Martinez, J.A., Severin, K.,Ghadiri, M. R. A Self-Replicating Peptide. 1996. Nature ,Vol. 382, pp. 525-528.

Lewin, R. 1993. Complexity: Life at the Edge of Chaos.London: Macmillan Publishing, pp. 178.

Maynard Smith, J. and Szathmár, E. 1995. The MajorTransitions in Evolution. Oxford: W.E. Freeman.

Miller, S.L. 1953. Production of amino acids underpossible primitive earth conditions. Science. 117: 528.Morowitz, H. 1992. Beginnings of Cellular Life. NewHaven, CT: Yale University Press

Nielsen P. E., Egholm, M., Berg, R.H. and Buchardt, O.1991. Science. Vol. 254, pp. 1497-1500.

Orgel, L.E. 1987. The Origin of Self-ReplicatingMolecules. In: Self-Organizing Stystems: The Emergenceof Order. (F.E. Yates, ed.), New York: Plenum Press, pp.65-74.

Oparin, A. I. 1924 [translated to English 1938, new edition1957]. The Origin of Life on Earth. New York: AcademicPress.

Owen, R. 1980 [1861]. Paleontology or a SystematicSummary of Extinct Animals and Their GeologicalRelation. London: Ayer Co. Publisher

Ruse, M. 1997. The Origin of Life: PhilosophicalPerspectives. In: J. Theoretical Biology, 187, pp.473-482.Sommerer, C. and Mignonneau, L. 1997. Interacting withArtificial Life: A-Volve, In Complexity Journal. NewYork: Wiley, Vol. 2, No. 6, pp. 13-21.

Sommerer, C. and Mignonneau, L. 1999. VERBARIUMand LIFE SPACIES: Creating a Visual Language byTranscoding Text into Form on the Internet, In: 1999 IEEESymposium on Visual Languages, Tokyo, Japan, pp. 90-95.Spiegelman, S. 1967. An in Vitro Analysis of a ReplicatingMolecule. American Scientist, Vol. 55, pp: 3-68.

Thompson, D’A. W. 1942. On Growth and Form.Cambridge: University Press.

Von Neumann, J. 1966. Theory of Self-ReproducingAutomata.:Completed and edited by A.W. Burks,Champaign-Urbana: University of Illinois Press.

Waddington, C. H. 1966. Principles of Development andDifferentiation. New York: Macmillan.

Wächterhäuser, G. 1997. The Origin of Life and itsMethodological Challenge. In: J. Theoretical Biology, 187,pp.483-494.

Whitehead, A. N. 1971 [1920]. The Concept of Nature.Reprinted. Cambridge: Cambridge University Press.Woese, C.R. 1967. The Genetic Code - The MolecularBasis for Genetic Expression. New York: Harper and Row.Wolfram, S. 1986. Theory and Applications of CellularAutomata. Singapore: World Scientific Press

ZKM Center Karlsruhe, Germany, 2000. Net-Condition,http://on1.zkm.de/netCondition.root/netcondition