a proposed heuristic for a computer chess program...each chess piece 3 moves into the future, for...

A Proposed Heuristic for a Computer Chess Program

copyright (c) 2011John L. Jerz

[email protected], Virginia

Independent ScholarMS Systems Engineering, Virgina Tech, 2000

MS Electrical Engineering, Virgina Tech, 1995BS Electrical Engineering, Virginia Tech, 1988

December 2, 2011

Abstract

How might we manage the attention of a computer chess program in orderto play a stronger positional game of chess? A new heuristic for estimating thepositional pressure produced by chess pieces is proposed. We evaluate the ’health’of a game position from a Systems perspective, using a dynamic model of theinteraction of the pieces. The identification and management of stressors and theconstruction of resilient positions allow effective cut-offs for less-promising gamecontinuations due to the perceived presence of adaptive capacity and sustainabledevelopment. We calculate and maintain a database of potential mobility foreach chess piece 3 moves into the future, for each position we evaluate. Wedetermine the likely restrictions placed on the future mobility of the pieces basedon the attack paths of the lower-valued enemy pieces. Knowledge is derived fromFoucault’s and Znosko-Borovsky’s conceptions of dynamic power relations. Wedevelop strategic scenarios based on goals obtained from the lowest-scoring of thevital Vickers/Bossel/Max-Neef diagnostic indicators. Initial results are presented.

keywords: complexity, chess, game theory, constraints, heuristics, planning,measurement, diagnostic test, resilience, orientor

1

A Proposed Heuristic - copyright (c) 2011 John L. Jerz

Contents

1 Overview 4

2 Introduction 4

3 Principles of Positional Chess 5

4 Systems Engineering 6

5 Systems Thinking 7

6 Goldratt’s Theory of Constraints and Thinking Process 9

7 Soft Systems Methodology 9

8 Measurement 10

9 Vulnerability 11

10 Resilience 12

11 Inventive Problem Solving 14

12 Strategy and the Strategic Plan 15

13 Competitive Intelligence Leads to Competitive Advantage 23

14 From Orientors to Indicators to Goals 25

15 Shannon’s Evaluation Function 30

16 The Positional Evaluation Methodology 31

17 Observations from Cognitive Science 36

18 Serious Play Can Lead to Serious Strategy 37

19 John Boyd’s OODA Loop 42

2


20 Adaptive Campaigning model of Grisogono and Ryan 43

21 Endpoint Evaluation 44

22 Complexity 45

23 Results 46

24 Conclusions 49

25 Appendix A: Selective Search and Simulation 54

26 Appendix B: The Importance of Sustainability 55

27 Appendix C: Related Quotations 55

References 57

3


1 Overview

The complexity present in the game of chess of-ten hinders planning efforts and makes simplequestions like ”what’s going on?” and ”whichside has the better position?” difficult to answer.

Indeterminate and unexpected events in thenear future might make revisions necessary forthese plans, often after only a few moves havebeen played.

We theorize that dynamic planning modelsbased on perceptions of constraints, the man-agement of stress, the readiness of resourcesto support strategy, resiliency, sustainable de-velopment, and sensitivity to both incrementalprogress towards goals and the emergence of newopportunities can be used with greater success.We seek positions which can serve as a platformfor future success, in a future that is often un-certain.

A proposed heuristic for a machine playingthe game of chess, taking advantage of conceptsfrom multiple disciplines, can be used to bet-ter estimate the potential of resources to supportstrategy and to offer better insight for determin-ing whether progress is being made towards re-mote goals. In a future that is uncertain, thereis a benefit to develop a strategic position fullof resilience, flexibility, and structures with thepotential for seizing new opportunities as theyemerge.

As we evaluate each game position and ori-ent our search efforts, we now consider the poten-tial to exploit and respond to new opportunitiesas time passes and new situations emerge frombeyond our initial planning horizon. Our flexibil-ity ideally allows a smooth and resilient responseto concurrent events as they unfold. We theo-rize that our focus on the constraints, as well as

the development of a resilient position, is a moreuseful level of abstraction for our game-playingmachine.

We examine concepts and values useful forplaying a positional game of chess, we developa perception useful for measuring incrementalprogress towards goals, and then look at po-sitions in chess games where the heuristic of-fers insight not otherwise obtainable. We con-clude that our orientation/evaluation heuristicoffers promise for a machine playing a game ofchess, although our limited evidence (at present)consists of diagrams showing the strategic (dy-namic) potential of the game pieces and an ex-ample of how these ’building blocks’ can be com-bined into vital indicators.

We see the chess position as a complex adap-tive system, full of opportunities of emergencefrom interacting pieces. Our aim in this paperis to reengineer the work performed by our ma-chine, mindful of the values commonly adoptedby experts and the principles of Systems think-ing, so that it might be done in a far superiorway (Hammer and Stanton, 1995).

2 Introduction

This paper is concerned with heuristic algo-rithms. According to (Koen, 2003) a heuristicis anything that provides a plausible aid or di-rection in the solution of a problem but is inthe final analysis unjustified, incapable of justifi-cation, and potentially fallible. Heuristics helpsolve unsolvable problems or reduce the timeneeded to find a satisfactory solution.

A new heuristic is proposed which offersbetter insight on the positional placement ofthe pieces to a chess-playing computer program.

4


The heuristic will have usefulness in the orienta-tion/evaluation methodology of a computer pro-gram, or as part of a teaching tool which explainsto a human user the reasons that one side or theother has an advantage in a chess game.

The heuristic involves constructing a tableof the future mobility for each piece, taking intoaccount the other pieces on the board, as wellas the likely constraints that these other piecesplace on this future movement. The heuristicconcept is described, and then examples are pre-sented from a software application constructedto demonstrate this concept.

Computer chess programs have historicallybeen weak in understanding concepts relating topositional issues. The proposed heuristic offers amethod to potentially play a stronger positionalgame of chess.

3 Principles of Positional Chess

Understanding the principles of positional chessis a necessary starting point before designingconcepts useful for a machine implementation.We select the relevant concepts of positionalchess which have been addressed by multiple au-thors.

(Stean, 2002) declares that the most impor-tant single feature of a chess position is the ac-tivity of the pieces and that the primary con-straint on a piece’s activity is the pawn struc-ture. (Znosko-Borovsky, 1980) generalizes thisprinciple by declaring that if two opposing piecesmutually attack each other, it is not the weakerbut the stronger one which has to give way. (Re-shevsky, 2002) notes that a good or bad bishopdepends on placement of the pawns. Piecesshould be ”working” and engaged, delivering the

full force of their potential and avoiding influ-ences which constrain. (Levy, 1976) discusses agame where a computer program accepts a posi-tion with an extra piece out of play, making a windifficult, if at all possible. Our evaluation shouldtherefore consider the degree to which a piece isin play or is capable of forcefully contributing tothe game.

Stean defines a weak pawn as one which can-not be protected by another pawn, therefore re-quiring support from its own pieces. This isthe ability to be protected by another pawn, notnecessarily the present existence of such protec-tion. Stean declares that the pawn structurehas a certain capacity for efficiently accommo-dating pieces and that exceeding that capacityhurts their ability to work together.

(Aagaard, 2003) declares that all positionalchess is related to the existence of weakness ineither player’s position. This weakness becomesreal when it is possible for the weakness to beattacked. The pieces on the board and theirconstraining interactions define how attackablethese weaknesses are.

(Emms, 2001) declares that is an advantageif a piece is performing several important func-tions at once, while a disadvantage if a piece isnot participating effectively in the game. Emmsteaches that doubled pawns can be weak if theyare attackable or if they otherwise reduce themobility of the pawns. Doubled pawns can con-trol vital squares, which might also mean deny-ing mobility to enemy pieces. Isolated pawnsrequire the presence of pieces to defend them ifattacked.

(Dvoretsky and Yusupov, 1996) argue thatcreating multiple threats is a good starting pointfor forming a plan. Improving the performanceof the weakest piece is proposed as a good way

5


to improve your position as a whole.

(McDonald, 2006) gives an example of gooddoubled pawns which operate to restrict the mo-bility of the opponent’s pieces and are not easilyattackable. His view is that every position needsto be evaluated according to the unique featurespresent.

(Capablanca, 2002) and (Znosko-Borovsky,1980) speak of how the force of the chess piecesacts in space, over the chessboard, and throughtime, in sequential moves. Critical is the con-cept of position, which is valued by greater orlesser mobility plus the pressure exerted againstpoints on the board or against opponent’s pieces.Pre-eminence, according to Capablanca, shouldbe given to the element of position. We are alsoinstructed that the underlying principle of themiddle game is co-ordinating the action of ourpieces.

(Heisman, 1999) discusses the important el-ements of positional evaluation, including globalmobility of the pieces and flexibility.

(Albus and Meystel, 2001) have written thatthe key to building practical intelligent systemslies in our ability to focus attention on what isimportant and to ignore what is not. (Kaplan,1978) says that it is important to focus attentionon the few moves that are relevant and to spendlittle time on the rest.

The positional style is distinguished by po-sitional goals and an evaluation which rewardspieces for their future potential to accomplishobjectives. (Ulea, 2002) quotes Katsenelinboigenas saying that the goal of the positional style ofchess is the creation of a position which allowsfor development in the future. By selecting ap-propriate placement of pieces, combinations ide-ally will emerge. (Katsenelinboigen, 1992) fur-

ther describes the organizational strategy of cre-ating flexible structures and the need to createpotential in adaptive systems that face an un-predictable environment.

(Botvinnik, 1984) and (Botvinnik, 1970) at-tempt in general terms to describe a vision forimplementing long range planning, noting thatattacking the paths that pieces take towards ob-jectives is a viable positional strategy. Positionalplay aims at changing or constraining the attackpaths that pieces take when moving towards ob-jectives - in effect, creating or mitigating stressin the position.

(Hubbard, 2007) identifies procedures whichcan be helpful when attempting to measure in-tangible values, such as the positional pressureproduced by chess pieces. (Spitzer, 2007) de-clares that what gets measured get managed,that everything that should be measured, canbe measured, and that we should measure whatis most important.

4 Systems Engineering

A system (Kossiakoff and Sweet, 2003) is a setof interrelated components working together to-ward a common objective. A complex engineeredsystem is composed of a large number of intri-cately interrelated diverse elements. von Berta-lanffy is of the opinion (von Bertalanffy, 1968)that the concept of a system is not limited tomaterial entities but can be applied to any wholeconsisting of interacting components. This de-scription could also apply to the situation facedby an agent playing a game, where the piecesrepresent the interrelated diverse elements. vonBertalanffy further identifies dynamic interac-tion as the central problem in all fields of reality

6


(which would include playing a game), identi-fying system elements in mutual interaction asthe very core issue. Additionally, we are told tosuspect systems or certain systems conditions atwork whenever we come across something thatappears vitalistic or human-like in attribution.We therefore see an opportunity to apply prin-ciples of System Theory, and in particular, Sys-tems Engineering, to game theory.

How would we begin? We now apply basicprinciples of Systems Engineering from (Kossi-akoff and Sweet, 2003):

A needs analysis phase defines the need fora new system. We ask ”Is there a valid needfor a new system?” and ”Is there a practical ap-proach to satisfying such a need?” Critically, canwe modify existing designs, and is available tech-nology mature enough to support the desiredcapability? The valid need would be to playa stronger positional game of chess, and exist-ing technology has struggled with the conceptof positional chess, as reflected in recent corre-spondence games which use Shannon-based pro-grams. It would seem that we need a differentapproach, which might be as simple as attempt-ing to emulate the style of play performed bystrong human players.

The concept exploration phase examines po-tential system concepts in answering the ques-tions: ”What performance is required of the newsystem to meet the perceived need?” and ”Isthere at least one feasible approach to achiev-ing such performance at an affordable cost?” Wewould answer the first question as simply thatour software function as an adequate analysistool, capable of selecting high-quality positionalmoves (with quality of move proportional to theanalysis time spent) when left ”on” for indefi-nite periods of time. As far as the second ques-

tion, we might speculate that a new approachis needed, which feasibly we could model afterhumans playing the game.

The concept definition phase selects the pre-ferred concept. It answers the question: ”Whatare the key characteristics of a system conceptthat would achieve the most beneficial balancebetween capability, operational life, and cost?”To answer this question a number of alterna-tive concepts might be considered and their rel-ative performance, operational utility, develop-ment risk, and cost might be compared. The firstconcept we might consider would be the Shannonapproach, which has been the backbone of mostsoftware computer chess programs. We presentin this paper, defined in another section, anotherapproach. We therefore decide to explore theconcept definition phase in more detail, as welook for key system characteristics which con-ceptually could serve as the base of such a newsystem.

5 Systems Thinking

The heart of Systemsthinking, which is differentfrom analytical thinking,is the attempt to simplifycomplexity.

Systems think-ing is a disciplinefor observing wholes(Senge, 2006). Itis a framework forobserving interrela-tionships rather thanthings, for observing the effects of change ratherthan static snapshots. The heart of Systemsthinking, which is different from analyticalthinking, is the attempt to simplify complexity(Gharajedaghi, 2006). We see an opportunityto apply principles of Systems thinking to game

7


theory. (Gharajedaghi, 2006) discusses how in-dependent variables are the essence of analyticalthinking. We might find, on closer inspection,that our independent variables are not truly in-dependent - that the whole is more than a simplesum of the parts. Emergent properties of a sys-tem are a product of interactions and cannot(Gharajedaghi, 2006) be analyzed or manipu-lated by analytical tools, and do not have causalexplanations. We must instead attempt to un-derstand the processes that produce them bymanaging the critical interactions. One mightthink of emergent properties as being in the pro-cess of unfolding. What makes it possible to turnthe systems approach into a scientific approachis our belief that there is such a thing as approx-imate knowledge (Capra, 1988). Systems think-ing also shows that small, well-focused actionscan produce significant, enduring improvements,if they are in the right place (de Wit and Mayer,2010). Systems thinkers refer to this idea as theprinciple of leverage. Tackling a difficult prob-lem is often a matter of seeing where the leveragelies, where a change - with a minimum of effort- would lead to lasting, significant improvement(de Wit and Mayer, 2010).

(Gharajedaghi, 2006) informs us that un-derstanding consequences of actions (both short-and long-term, in their entirety), requires build-ing a dynamic model to simulate the multiple-loop, nonlinear nature of the system. Our modelshould aim to capture the important delays andrelevant interactions among the major variables,but need not be complicated.

We therefore attempt to approach the ori-entation/evaluation methodolgy from a Systemsperspective. We will look at the interactions ofthe pieces and their ability to create and miti-gate stress. We adopt constraints, vulnerability,

dynamic modeling, and resiliency as higher levelconcepts which will help cut through the com-plexity and steer search efforts along the linesof the most promising moves. The technique ofmodeling (Kossiakoff and Sweet, 2003) is one ofthe basic tools of systems engineering, especiallyin situations where complexity and emergenceobscure the basic facts in a situation.

From (Anderson and Johnson, 1997), we ap-ply Systems thinking to look at the web of in-terconnected, circular relationships present in achess position, confident that this is the propertool for doing so. Our reason for believing thisis that everything in a chess position is (An-derson and Johnson, 1997) dynamic, complex,and interdependent. Things are changing all thetime, analysis is messy, and the interactions ofthe pieces are all interconnected.

we apply Systems think-ing to look at the webof interconnected, circularrelationships present in achess position, confidentthat this is the proper toolfor doing so.

As we attemptto construct resilientgame positions, wefollow (Tierney andBruneau, 2007) andidentify 4 systemlevel components ofresiliency: Robust-ness - the ability of our game-playing agent towithstand our opponent’s forces without degra-dation or loss of performance; Redundancy -the extent to which pieces, structures or movesare substitutable, that is, capable of sustainingoperations, if degradation or a surprise move oc-curs; Resourcefulness - the ability of our agent todiagnose and prioritize candidate moves and toinitiate solutions by identifying and mobilizingappropriate amounts of search time and gameresources; and Rapidity - the capacity to restoreor sustain functionality in a timely way, contain-

8


ing losses by graceful failure and avoiding otherdisruptions.

6 Goldratt’s Theory of Con-straints and Thinking Pro-cess

Goldratt (Goldratt and Cox, 2004) has devel-oped a Theory of Constraints which postulatesthat organizations and complex systems are hin-dered from reaching their goals by the con-straints placed on that system. Identifyingthose constraints and removing them can speedprogress towards these goals. (Scheinkopf, 1999)describes how Golratt’s institute began to mod-ify the original concepts to serve the needs ofclients who wanted more generalized proceduresto solve a wider variety of problems outside of afactory production environment.

Goldratt’s ideas, while seemingly original,can be properly classified as a Systems thinkingmethodolgy which emphasizes raw human think-ing over the construction and implementationof computer models. Each approach is useful.Also emphasized is a vocabulary and terminol-ogy which allows groups to construct and discussanalytical diagrams of feedback loops and iden-tify root causes.

(Dettmer, 2007) explores Goldratt’s Think-ing Process and identifies procedures to logicallyidentify and eliminate undesirable effects fromsystems and organizations.

(Dechter, 2003) explains that a model of re-ality based on constraints helps us to achieve aneffective focus for search efforts, and is similar tothe heuristic process that humans use to searchfor effective solutions in complex situations. Re-

moving the constraints partially solves the prob-lem, and measured progress towards removingthese constraints can steer and prune our searchefforts when identifying positions and lines ofanalysis which are promising.

(Hollnagel et al., 2006) speak of identifyingand monitoring the ”barriers” which keep thesystem response within safe margins. Also, theuse of ”audit tools” is envisioned as a method tomeasure the effectiveness of the containment.

7 Soft Systems Methodology

(Checkland and Poulter, 2006) present a modi-fied Systems methodology where complexity andconfusion are tackled through organized explo-ration and learning. We envision the continuouschange present in the game of chess as a com-plex state that needs to be (at least partially)understood in order to make exploration efforts(of an exponentially growing search tree) moreefficient.

We conceptualize a learning agent whichgathers relevant information as it seeks to de-termine the cumulative stress present in the po-sition, in order to determine the paths of ex-ploration - the ones of promise and the onesof risk mitigation. Our Systems model (makingup our orientation/evaluation methodology) willideally suggest to us what moves are promisingor worth our time exploring, as well as to rec-ommend which paths can, justifiably, wait untillater. The heuristics which make up this learn-ing and decision making process will be discussedin a later section. Critical to these heuristics isthe concept that all dynamic behavior emergesfrom a combination of reinforcing and balancingfeedback loops (Anderson and Johnson, 1997).

9


what we really need is aninsightful and informed di-rection for exploration anda notion for how press-ing this direction becomesstrategically.

Curiously, ourorientation/ evalua-tion ’function’ willbecome a method-olgy rather than aformula. We shareBotvinnik’s puzzle-ment with an evaluation ”number” (Botvinnik,1970) when what we really need is an insightfuland informed direction for exploration (orienta-tion) and a notion for how pressing this directionbecomes strategically.

The insight we obtain by this method is usedas a spring for action (Checkland and Poulter,2006), as our software agent decides what to donext, after completing the current evaluation.Our ”evaluation” ideally produces candidate di-rections for exploration, as part of a carefullyconstructed strategic plan, and indicates whichpaths are critical and which can wait until later.For Checkland, our model is an intellectual de-vice we use to richly explore the future, usingstress transformation as our chosen strategy, orworldview. Simply put, our model tells us whichpaths to explore.

Our estimate of the winning chances of acandidate position critically depends on the iden-tification and exploration of the critical candi-date sequences of moves, and the correct classifi-cation of the worthiness (for timely exploration)of such candidate positions. A heuristic estimateof the cumulative stress present in the position,at the end of our principal variation, can be cor-related, if desired, with winning chances. How-ever, our operational use of this value is for (cy-bernetically) steering search efforts.

8 Measurement

Measurement plays a dual role (DiPiazza and Ec-cles, 2002): it focuses attention on what is impor-tant, as determined by strategy, and it monitorsthe level of performance along those dimensionsin the effort to turn strategy into results. Cer-tain measures can be predictive in nature, andwe aim for successful use of those measures as amanagement tool in steering search efforts.

Measurement systems create the basis foreffective management, since you get what youmeasure. Management therefore needs to focusits attention on the measures that really drivethe performance or success they seek (Spitzer,2007). Spitzer also speaks about the criticalneed to develop metrics which are predictive andwhich measure strategic potential. We seek tomeasure how ”ready” our pieces are (and thestructures they form) for supporting strategy(Kaplan and Norton, 2004), especially when thefuture positions we face are not entirely deter-minable. An asset (such as a game piece) thatcannot support strategy has limited value. Partof our orientation/evaluation of the promise of aposition should ideally include the readiness ofthe pieces and structures to support future de-velopments. We embrace the principle that whatyou look for is what you find.

For (Zeller and Carmines, 1980), measure-ment clarifies our theoretical thinking and linksthe conceptual with the observable. For mea-surement to be effective, we must first constructa valid sensor. In our attempts at measurement,we seek empirical indicators which are valid, op-erational indicators of our theoretical concepts.We desire to construct a diagnostic indicatorwhich gives, as a result, a useful predictive mea-sure of future promise and a direction for future

10


exploration.

Although it would seem that a perceptionbased on simplicity would yield the best all-around results, (Blalock, 1982) points out thedifficulties trying to simultaneously achieve sim-plicity, generality, and precision in our measure-ment. If we have to give up one of these three, itis Blalock’s opinion that parsimony, or the scien-tific idea that the simplest explanation of a phe-nomenon is the best one, would have to be sac-rificed in order to achieve the other two. Laszlo(Laszlo, 1996) suggests that science must bewareof rejecting the complexity of structure for thesake of simplicity. Therefore, our attempts to de-scribe a complex orientation/evaluation method-ology are grounded in the two-fold goals of gen-erality (it must be applied to all positions we en-counter) and precision (otherwise, search effortsare wasted on less promising lines). Essentially,oversimplifying complex problems is dangerousand can mislead an analyst to offer a detrimen-tal judgment (Fleisher and Bensoussan, 2007).

The alternative view is presented by (Gun-derson et al., 2010), who declare that experiencehas suggested to be as ruthlessly parsimoniousand economical as possible while still retainingresponsiveness to the management objectives andactions appropriate for the problem. Addition-ally, we are advised that the variables selectedfor system description must be the minimumthat will capture the system’s essential qualita-tive behavior in time and space. We are furthercautioned that the initial steps of bounding theproblem determine whether the abstract modelwill usefully represent that portion of reality rel-evant to policy design. We must therefore aimto simplify, but not so much as to impact theusefulness of the tool for predicting promisingpaths of exploration. We hypothesize that the

use of competition itself as an aid in constructingthe measurement model will allow complexity togrow as long as overall tournament performancedoes not decrease.

9 Vulnerability

Critical to the success of a computer chess pro-gram that attempts to play in the positionalstyle is the concept of vulnerability. The piecesand structures that are or have the potential tobecome vulnerable will become a focus of oursearch and exploration efforts and will serve astargets for our long-range planning.

We follow (McCarthy et al., 2001) and con-ceptualize vulnerability as a function of expo-sure, sensitivity, and adaptive capacity. Conse-quently, the sensor we develop should attemptto measure exposure to threats, the sensitivityto the effects of stimuli, and the ability to adaptand cope with the consequences of change. Weenvision a sensor that produces a forecast of po-tential vulnerability as an output. This forecastcan guide exploration efforts by identifying tar-gets for the useful application of stress and serveas one indicator of a promising position.

Additionally, we predict that any machine-based attempt to zero-in on vulnerability thatdoes not address this conceptual base runs therisk of missing opportunities in exploring theexponentially growing tree of possibilities thatexist for each game position. A missed oppor-tunity might equally prevent us from increas-ing positional pressure on our opponent, or in-stead, might dissipate the pressure that we havecarefully accumulated over time. Our orienta-tion/evaluation of the winning chances presentin the position might not be as accurate as it

11


could be unless we explore the promising posi-tions and consider the vulnerabilities that arepresent.

We conceptualize that the reduction of vul-nerability and the pursuit of sustainable develop-ment are interrelated aims (Smith et al., 2003).

10 Resilience

When something unexpected happens, it is re-silience we fall back on - resilience provides thecapacity to sustain strategy change (Valikangas,2010). Vulnerability is the condition that makesadaptation and resilience necessary as a mitiga-tion (Worldwatch, 2009). The scientific studyof resilience began in the 1970s when NormanGarmezy studied well-adapted children who hadovercome the stress of poverty (Lukey and Tepe,2008). Resilience is also an important researcharea in military science (Friedl, 2007) and in thestudy of ecosystems (Folke et al., 2002). We findthis concept useful in game theory.

We desire a generic, con-tinuous ability (both duringcrisis and non-crisis gamesituations) to cope with theuncertain positions that ar-rive from beyond our plan-ning horizon.

In our view,adapted from(Luthar, 2003), re-silience refers to anongoing, dynamic de-velopmental processof strategically po-sitioning resourcesthat enables the player in a game to negoti-ate current issues adaptively. It also providesa foundation for dealing with subsequent chal-lenges, as well as recovering from reversals offortune.

We desire a generic, continuous ability (bothduring crisis and non-crisis game situations) to

cope with the uncertain positions that arrivefrom beyond our planning horizon. Ideally, weseek to create a useful positional pressure to forcethese arriving positions to be in our favor, orminimally, to put a ”cage” of constraints aroundthe enemy pieces. Flexibility, adaptive capacity,and effective engagement of available resourceswill be our weapons against the dynamic changeswhich will unfold in our game (Hollnagel et al.,2008).

Ideally, we will look for and manage theheuristic early warning signs of a position ap-proaching a ”tipping point”, where a distinct,clear advantage for one side emerges from an un-clear array of concurrent piece interactions. Weagree with (Walsh, 2006) that resilience cannotbe captured as a snapshot at a moment in time,but rather is the result of an interactive processthat unfolds over time.

The failure to include resilience measure-ments like this in planning efforts might causea house-of-cards effect, as the weakest link inour plan might collapse, due to effects we cannotinitially perceive. This might create a situationfrom which we cannot recover, or from which wecannot continue to mount increasing positionalpressure on our opponent.

A central concept is the construction of aresilient position, one that ideally 1. possesses acapacity to bounce back from disruption in theevent of an unforeseen move by our opponent, 2.produces advantageous moves in light of smallmistakes by our opponent, or 3. permits us topostpone our search efforts at early points forless promising positions, with greater confidencethat we have sufficient resources to handle futureunforeseen developments if the actual game playproceeds down that route. In simplest form, wemight just measure the ability to self-organize.

12


When change occurs, the components thatmake up resilience provide the necessary capac-ity to (minimally) counter and (ideally) seize newopportunities that emerge (Folke et al., 2002).Resilience is (minimally) insurance against thecollapse of a position and (ideally) an investmentthat pays dividends in the form of better posi-tions in the future. With no pun intended, wesee the struggle to control the unknown, emerg-ing future positions as a ”Red Queen’s Race”,where in tough-fought games against a talentedopponent, it might take all the effort possible tomaintain equal chances. Extraordinary effortsinvolving hundreds of hours of analysis per move(such as in correspondence games) might be re-quired to maneuver to an advantage (Jerz, 2007).

For (Reivich and Shatte, 2002), resilienceis the basic strength. (Hollnagel et al., 2006)suggest that ”incidents”, which for us might bethe construction of short sequences of just thetop few promising moves (diagnostic probing),might reveal insight to boundary conditions inwhich resilience is either causing the system tostretch to adapt, or buckle and fail. Emergencyresponse teams use practice incidents to measureresilience as unforeseen events emerge during op-erations. Fire drills, random audits and securitysearches, even surprise tests are diagnostic toolsused to detect and correct situations lacking inresilient capabilities.

We speculate that the abil-ity to construct a resilientposition and the ability toperceive stress in a posi-tion are two primary con-ceptual differences betweena game-playing man andmachine.

We acknowledgethe reality that ourability to handle anunexpected move orcritical situation in agame depends on thestructures alreadyin place (Weick and

Sutcliffe, 2007). Wedesire (Weick and Sutcliffe, 2007) to pay closeattention to weak signals of failure that are di-agnostic indicators of potential problems in thesystem. We also perform diagnostic probing touncover and steer game play towards positionswhere there are multiple good moves - an addi-tional sign of resilience.

We speculate that the ability to constructa resilient position and the ability to perceivestress in a position are two primary conceptualdifferences between a game-playing man and ma-chine. We believe that these abilities can be emu-lated through the use of custom diagnostic tests.

Humans construct resilient positions (instrategic situations) almost by instinct and of-ten without conscious thought (Fritz, 2003), indiverse situations such as driving automobiles,playing sports games, conducting warfare, socialinteraction, and managing resources in businessor work situations. Humans have such refinedabilities (Laszlo, 1996) to make predictions, in-terpret clues and manipulate their environment,that using them is frequently effortless, espe-cially if performed daily or over extended periodsof time. (Aldwin, 2007) points out that humansappear to be hard-wired physiologically to re-spond to their perceptions of stress - so much sothat effective responses can be generated contin-uously with little conscious thought. We there-fore see the machine-based perception of stressas critical to successful performance in a game.

Additionally, much has been written (Fagreand Charles, 2009) (Folke et al., 2002) concern-ing ecosystems, resilience, and adaptive manage-ment that has direct application to game theory.

Conceptually, we desire the equivalent of a”mindset” that can successfully cope with prob-lems as they arise, as we attempt to 1. exam-

13


ine the promising positions, 2. evaluate the cor-responding winning potential and 3. steer oursearch efforts through an exponentially grow-ing ”tree” of strategically important move se-quences. This process is aided by the heuristicmeasurement of adaptive capacity, as the thou-sands of unexamined positions that lie just be-yond the point of our search cut-offs must beresilient enough to counter whatever unknownevents emerge. Before we cut-off our search ef-forts, we critically seek diagnostic evidence ofreadiness, which depends on the perceived abilityto quickly adopt, adjust, or abandon initiativesand investments once new conditions materialize(Beckham, 2002). Readiness describes an orga-nization that can be viable across a variety ofconditions (Beckham, 2002).

Rather than thinking about resilience as”bouncing back” from a shock or stress, it mightbe more useful to think about ”bouncing for-ward” to a position where shocks and stresseshave less effect on vulnerabilities (Walsh, 2006)(Worldwatch, 2009). Integral to the definitionof resilience are the interactions among risk andprotective factors (Verleye et al., prepub) at anagent and environmental level. Protective fac-tors operate to protect assets, such as pieces in agame, at risk from the effects of the risk factors.

We agree and conceptualize that, while riskfactors do not automatically lead to negativeoutcomes, their presence only exposes a game-playing agent to circumstances associated witha higher incidence of the outcome; protective ormitigating factors such as constraints can con-tribute to positive outcomes - perhaps regardlessof the risk status.

We accept as an operational concept of re-silience, the fourth proposal of Glantz and Slo-boda (Glantz and Johnson, 1999), which involves

the adoption of a systems approach. We con-sider both positive and negative circumstancesand both influencing and protecting characteris-tics and the ways in which they interact in therelevant situations. Additionally, this conceptu-alization considers the cumulation of factors andthe influences of both nearby and distant forces.In addition, (Elias et al., 2006) discuss a model ofresilience in which specific protective influences(which we see as constraints) moderate the ef-fect of risk processes over time, in order to fosteradaptive outcomes.

We propose (Gunderson et al., 2010) an ap-proach based on resilience, which would empha-size the need to keep options open, the need toview events in a larger context, and the need toemphasize a capacity for having a large numberof structural variations. From this we recognizeour ignorance of, and the unexpectedness of fu-ture events. The resilience framework does notrequire a precise ability to predict the future, butonly a capacity to devise systems that can absorband accommodate future events in whatever un-expected form they may take. If we could cramMacGyver into our software, we would certainlydo so.

11 Inventive Problem Solving

Our chess program attempts to be, like Mac-Gyver, an inventive problem solver. We see ef-fective problem solving as an adaptive processthat unfolds based on the nature of the problem,rather than as a series of specific steps (Albrecht,2007). We agree with (Browne, 2002) that know-ing the difference between what’s important andwhat isn’t is a basic starting point.

We attempt to navigate an exponentially

14


growing search tree, selecting those paths for ex-ploration that are promising, interesting, risk-mitigating, and resilient in the face of an un-known future. We are concerned at all timeswith the potential of a position to serve as an ad-vancing platform for future incremental progresstowards positional goals (Fritz, 2007). We willaccomplish this by knowing the outcomes wewant and looking tirelessly for them. (Savran-sky, 2000) lists three major requirements for aproblem-solving methodology, which we modifyslightly for the purposes of a machine playing agame:

1. It should focus on the most appropriateand strongest solutions

2. It should produce, as an output, the mostpromising strategies

3. It should acquire and use important, well-organized, and necessary information at all stepsof the process

(Savransky, 2000) additionally suggests thatwe should focus on gathering the important in-formation, information which characterizes theproblem and makes it clear, including contra-dictions. Any simplifications we perform shouldaim at reducing the problem to its essence andbe directed towards our conceptual, strategic so-lution.

As an example, typical American news re-ports each day announce the results of the DowJones index of stocks. This weighted index of30 representative companies serves as a good in-dicator of overall market performance and canhelp answer the question ”How did the marketsdo today?”. To obtain this numerical value, youjust sum the prices of each of the 30 specific com-panies and divide by a number which takes intoaccount stock splits and stock dividends.

We seek an equivalent summary numericalrepresentation of reality (March, 1994) which canserve as a guiding light and a measure of progresstowards our distant positional goals. We are notrestricted to the use of a single scoring metric,and can combine multiple, critical metrics in cre-ative ways, including the selection of the lowestscore from several indicators to provide a searchfocus. We should first form a concept of whatshould be measured, then create a sensor arraywhich allows us to measure and perform searchefforts (in an exponentially growing tree of pos-sible continuations) with reasonable efficiency.

12 Strategy and the StrategicPlan

I do not claim that strat-egy is or can be a ”sci-ence” in the sense of thephysical sciences. It canand should be an intellec-tual discipline of the high-est order, and the strate-gist should prepare himselfto manage ideas with preci-sion and clarity and imagi-nation... Thus, while strat-egy itself may not be ascience, strategic judgmentcan be scientific to the ex-tent that it is orderly, ra-tional, objective, inclusive,discriminatory, and percep-tive. -J.C. Wylie

We follow Beck-ham (Beckham,2000) and Wylie(Wylie and Wylie,1989) and definestrategy as a planfor using leverage toget from a point inthe present to somepoint in the futurein the face of uncer-tainty and resistance.We concur that with-out a future that in-volves some uncer-tainty and resistance,there is no need for astrategy. A strategyhas lasting power - its effects are sustained overa time horizon (Beckham, 2000). Strategy is akind of investment in that it aims to create or

15


sustain significant value. Strategy deals withthe important in a way that is deemed neces-sary for sustainable success. Leverage is criti-cal for Senge (Senge, 1990), as the leverage inmost management situations lies in understand-ing dynamic complexity, not detail complexity.Dynamic complexity arises when cause and ef-fect are distant in time and space, and whenthe consequences over time of interventions aresubtle and not obvious.

The only kind of strategythat makes sense in theface of unpredictablechange is a strategy tobecome adaptive.... whenchange becomes unpre-dictable, it follows thatthe appropriate responsewill be equally so. In thisenvironment, therefore,planned responses do notwork. -Stephan Haeckel

Anyone or any-thing lacking a strat-egy, a plan, a pro-gram - whatever wecall it - is undertak-ing a journey withouta map. Its actionswill be an incoherentseries of ad hoc andperhaps mutuallyconflicting responsesto new events (Mur-phy, 2005). A com-peting entity’s competitive capability dependsupon the resources at its disposal and how effi-ciently they are used. Winners need to combinea sound strategy with a fitting level of resources -they must also correctly identify the critical suc-cess factors for the environment in which theychoose to compete (Murphy, 2005). A strategydelivers significant improvements in the key in-dicators of success (Beckham, 2000). We needto get into a loop linking action, perception andthinking towards continual learning. An effec-tive strategy is one that triggers our successfullaunch into that learning loop (van der Heijden,2005).

We see a strategic plan (Bradford et al.,2000) as a simple statement of the few things we

really need to focus on to bring us success, as wedefine it. It will help us manage every detail ofthe game-playing process, but should not be ex-cessively detailed. It will encapsulate our visionand will help us make decisions as we criticallychoose, or choose not, to explore future positionsin our search tree. We see the formation and exe-cution of the strategic plan as the most effectiveway to get nearer to the goal state, especiallyin a competitive environment where our oppo-nent is also attempting to do the same. The sim-ple principles that govern strategy are not chainsbut flexible guides leaving free play, in situationsthat are themselves enormously variable (Cas-tex and Kiesling, 1994). Wylie’s general theoryof strategy, applicable in any conflict situation,is a worthwhile starting point and overall guide(Wylie and Wylie, 1989).

We see the role of themachine as merely that ofan executor of a strategicplan... we simply ask themachine to do what it istold.

We see the roleof the machine (inplaying a game suchas chess) as merelythat of an executorof a strategic plan,where we have previ-ously defined (through programmed software in-structions) the specific answers to the questions”Where do we want to be?” ”How will we knowwe have reached it?” ”What is changing in theenvironment that we need to consider?” ”Whereare we right now?” and ”How do we get from hereto our desired place?” (Haines, 1998). In our vi-sion, the intelligence is located in the strategicanswers to these questions and in the skill of theprogrammer in implementing them - we simplyask the machine to do what it is told.

16


computers... cannot un-derstand symbols (or in-deed anything else either),though they can manipu-late symbols according toformal rules with consum-mate speed and accuracy,far surpassing our own fum-bling efforts... they donot understand the ques-tions they are asked orthe answers they provide.- Richard Gregory, Minds,Machines and Meaning, In:Rose, Appignanesi, Scienceand Beyond, 1986.

If one game-playing computerprogram is betterthan another, asdemonstrated in atournament of manygames played, wespeculate that thereason is either a bet-ter strategic plan ora better software im-plementation of thatplan. Therefore, im-provements in com-puter chess programsideally should focuson these two areas, including answers to thequestions presented above. For Haines, all typesof problems and situations (which include se-lecting a move in a game) can benefit from astrategic approach.

A vision involves... ”antici-pative shaping” that seeksto discern not only thepowerful currents of the fu-ture but also how thosecurrents can be leveraged...it’s foolhardy to assumeyou can control the future.The future will consist ofpowerful flows that, like theweather, can be leveragedand ridden but can never becontrolled...

Before we de-velop our strategicplan, we ask our-selves and ponderthree critical ques-tions (Jorgensen andFath, 2007): 1. whatare the underlyingproperties that canexplain the responseswe see on the gameboard to perturba-tions and interven-tions, 2. are we able to formulate at least build-ing blocks of a management theory in the form ofuseful propositions about processes and proper-ties, and 3. can we form a theory to understand

the playing of chess that is sufficiently developedto be able to explain observations in a practicalway for choosing a move? We do not see the needto construct mathematical proofs - the conceptsof useful propositions and effective strategic prin-ciples allow us flexibility in choosing an approachand allow us to consider multiple options beforesettling on one with the most promise. We re-turn to these critical questions whenever we seekdirection or clarification in an approach, or con-sider starting over. We look to other disciplines -as suggested by (Boyd, 1987) - and to other pro-fessionals who have sought answers to the samequestions, which must be asked in a general wayto any management problem.

...Trips to the future be-gin with a struggle to seeand understand these pow-erful currents: their generaldirection, their power, andwhere they may collide andcoalesce. - J. Daniel Beck-ham

Central to ourstrategic plan are thefollowing concepts(Jorgensen and Fath,2007): system behav-ior frequently arisesout of indirect inter-actions that are dif-ficult to incorporateinto connected models, that we may not knowexactly what happens, but approximately whathappens, and that we can use holistic metricsto measure the growth and development of aposition in a game. We acknowledge that sys-tems have a complex response to disturbances,and that constraints play a major role in in-teractions. As a strategy we seek a method todetermine (and to resolve uncertainty concern-ing) 1. the promising candidate moves in a givenposition, and 2. the chances of sustainable de-velopment in a position, allowing us to postpone(if necessary) the exploration of future conse-quences.

17


In a building block for our strategic plan,we examine the position under inspection for thepresence of stressors (Glantz and Johnson, 1999)and determine their contribution to the cumula-tive stress in the position. A stressor is a realobject on the game board, such as a piece, oran object or property that might become realin the future, such as a Queen from a promotedpawn, a stone in the game of Go, or a King in thegame of draughts/checkers. Using our stressors,we seek to establish a structural tension (Fritz,1989) that, if resolved, leads to positions thatfavor us.

In a building block for ourstrategic plan, we exam-ine the position under in-spection for the presence ofstressors

The stress weseek to place on ouropponent (Glantzand Johnson, 1999)is the kind that inter-feres with or dimin-ishes the development of our opponent’s copingrepertoire, search and planning abilities, expec-tations and potential resilience. This stress isideally so effective that we create a platform fromwhich we can apply even more stress. We forceour opponent to divert additional resources tocontaining our threats, making fewer resourcesavailable for threats of his own.

We attempt to cope withthe stressors of our oppo-nent by weakening them orreducing their influence toa manageable level

We attempt tocope with the stres-sors of our opponentby weakening themor reducing their in-fluence to a manage-able level (Snyder,2001) - there is no compelling need to make theireffects go away completely. For (von Bertalanffy,1968), stress is a danger to be controlled and

neutralized by adaptive mechanisms. We gatherdiagnostic information that is used to determinethe readiness of the pieces to inflict stress on theopponent and lessen the stress imposed by theenemy pieces on our weak points. The creationof effective stress and the perceived mobilizationof forces to mitigate it will become a central con-cept in our orientation/evaluation. Our orienta-tion/evaluation looks not so much to goal seek-ing/optimizing a ”score” as to sustaining rela-tionships between/among the pieces and learn-ing what happens as stress is moved from onearea of the board to another. What is relevantcannot be known until later. The kinds of pre-dictions we most want to make, we feel, requireus to first determine which of all the things thatmight happen in the future will turn out to berelevant, in order that we can start paying atten-tion to them now (Watts, 2011). We acknowl-edge openly (Watts, 2011) that there are limitsto what can be predicted - we therefore seek todevelop methods for planning that respect thoselimits.

Figure 1: Simplified model (dynamic hypothesis) of posi-tional pressure for each piece

Figure 1 shows a simplification of the pro-posed model of positional pressure for each piece,based on principles of system dynamics. Thefuture mobility of each piece targets opponentpieces, the trajectories taken by these pieces, and

18


certain other weaknesses such as weak pawns,the opponent’s king, or undefended pieces. Thisthreat is mitigated (but not reduced completely)by the protective factor of constraints imposedby the lower-valued enemy pieces. The resid-ual ”Stock” is the effective stress that can befelt by our opponent, and which we seek to in-crease. For (Warren, 2008), the management ofcritical resources is part of an emerging theory ofperformance: performance depends on resourcecontribution, resource contribution accumulatesand depletes, and this depends on existing re-source contribution levels.

Figure 2 shows the plan for managing theperceived stress by incentivizing a coping strat-egy, such as the placement of constraints, in or-der to control the effects of the overall cumulativestress. We seek to maintain a resilient positionfull of adaptive capacity.

Figure 2: As perceived stress increases, we increase the in-centive to cope with the stress

Things start to get complicated when weremove stress (and the associated constraints)from one area of the board and apply it to

other areas. The short- and long-term effectsof these stress-exchanging maneuvers are exam-ined through prioritized search efforts, and in ouropinion represent the essence of playing a gamesuch as chess. This conceptual model will formthe basis of the machine’s perception. We relyon the simplifying principles of system dynam-ics to predict and anticipate the effects of suchstress transformation.

From (Friedl, 2007) we define a stressor asany challenge to a player in a game that evokesa response. Coping is the set of responses thatsustain performance in the presence of stressors.Resilience is the relative assessment of copingability. We desire to create in our opponent’sposition a condition similar to fatigue, definedby Friedl (and modified for game theory) as thestate of reduced performance capability due tothe inability to continue to cope with stressors.We follow Fontana (Fontana, 1989) and definestress as a demand made upon the adaptive ca-pacity of a player in a game by the other. Wetheorize a correlation between the state of stress-induced reduced performance capability and an”advantage”, or favorable chances for the morecapable player winning the game.

we are dealing with a pro-cess whose effects taketime in revealing them-selves

Strategically, weseek to identify thestress present in theposition by 1. exam-ining the demands ofeach stressor, 2. the capacity of each player torespond to those demands, and 3. the conse-quences of not responding to the demands.

19


we will predict the win-ning chances at some fu-ture point in time, afterthe present circumstancesprogress and the structuresin place are allowed to un-fold

We carefully de-fine weakness so thatthe stress and tensionwe create is focusedand effective. The in-formation we gatherfrom the interactingpieces should be pre-cise enough to get results - it does not need to beperfectly accurate. Information is power (Brad-ford et al., 2000), especially in strategic planning.Along the way, we will need to make assumptionsabout whether or not the stress we are inflictingon our opponent is increasing or decreasing, andwhether it is effective or not effective. We mightexplore promising paths in detail to confirm ourassumptions, or we might just rely on our mea-surements of resilience.

Critical is our ability to focus our search ef-forts on lines that are promising, with regardto the oriented application of stress and thepredicted effects on future lines of play. Inour opinion (Schumpeter, 2008), we are deal-ing with a process whose effects take time inrevealing themselves - we will predict the win-ning chances at some future point in time, af-ter the present circumstances progress and thestructures in place are allowed to unfold, includ-ing the newly emergent features which we are notcurrently able to perceive. We establish a port-folio of promising lines, and see where they go.We invest our time and processor resources inthe most promising, but only after investigatingthe promising via a swarm of lower-risk exper-iments (Hamel and Valikangas, 2003). We de-fine a concept of stress which lets us focus oursearch efforts on anticipated promising lines. Werely on the promise of adaptive capacity presentin resilient positions to sustain our efforts in

lines where the perception of weaker cumulativestress, time constraints, and our model of pur-poseful activity do not permit us to explore.

Figure 3: Conceptual Framework, from Chapin, 2009, p.21,(placeholder until new diagram is created)

We theorize that the dynamic forces ofchange during the playing of a game have anadaptation cost associated with them (Kelly andHoopes, 2004) (Zeidner and Endler, 1996). Thismight come from a shift in expectations, or froma required recovery from disruptions. We make”payments” for these adaptation costs from our”bank” of resilience. If we lose our positional re-silience, we lose our flexible ability to adapt tothe unknown requirements of change. Likewise,we can make ”deposits” to our resilience accountduring quiet periods of maneuver, if we choose,and if we value resilience as an element of our ori-entation/evaluation methodology. Friedl (Friedl,2007) refers to this concept as pre-habilitation.We seek to attack our opponent’s capacity to re-spond and to strengthen our own, so that thedynamic forces of change that drive the gamecontinuation will cause the unknown positionsarriving from beyond our planning horizon to bein our favor.

20


A strategic thinker never al-lows himself to lose sight ofthe key factors... he willshape his strategy - a strat-egy not for total war on allfronts but for a limited waron the fronts defined by thekey factors for success... itis this focus on key factorsthat gives the major direc-tion or orientation to theoperation we call strategicthinking -Kenichi Ohmae

We seek a re-silient mindset.Specifically, we fol-low Coutu (Coutu,2003) and aim forthree fundamentalcharacteristics: weidentify and facethe reality of thestresses and con-straints present inthe positions we eval-uate, we identify andreward the values ofpositional chess, and we develop an ability toimprovise solutions based on whatever resourcesare available to us. We seek to prepare for anunknown future that can be influenced by thestrategic placement of resources in the present.

In the generalized exchange of pieces,squares, and opportunities encountered in gameplaying (Botvinnik, 1970), we seek to establisha pressure that has a realistic chance to resolvein our favor, as determined by heuristic probingand the examination of promising future gamesequences. We desire to create and sustain aweb of stress which threatens to become real andtherefore has the property that (von Neumannand Morgenstern, 1953) have called ”virtual” ex-istence. Our opponent must ”spend” or dedi-cate resources to contain or adapt to the threats.Even if a particular threat is contained, it never-theless has participated in the dynamic shapingand influencing of the events that emerge andunfold in the game.

We will succeed at forming an effectivestrategic plan when we have identified our values,determined the key drivers to performance, de-veloped a sensor which is effective at measuring

them, and have focused on the lines of play thatare promising. At all times we wish to maintaina resilient position, which increases our ability toeffectively handle the unknown positions whichlie beyond the horizon of our explorations.

We will use two key strategies (Maddi andKhoshaba, 2005) to become and remain resilient:we will develop the vision to perceive changes inthe promise of a position (as they emerge fromour heuristic explorations), and we seek flexi-bility to act quickly, while remaining focusedon our goals of establishing and maintaining auseful structural tension. We seek (Kelly andHoopes, 2004) a balanced portfolio of resilienceskills, where ideally we are focused, flexible, or-ganized, and proactive in any given situation.We believe that resilient responses (Kelly andHoopes, 2004) are the result of resilience char-acteristics operating as a system, as we evaluateand predict the emergent results of change.

Following Jackson (Jackson, 2003), we avoidplacing a complete reliance on specific predic-tions of the future, concentrating on relation-ships, dynamism and unpredictability as muchas we do on determinism. In our plan, we willadapt as necessary and seize new opportunitiesas they emerge from the ”mess”. We seek to fo-cus on identifying and managing the structuresthat will drive the behavior of the game, andacknowledge the reality that large portions ofthe future possibilities will go unsearched andunexplored (until they emerge from beyond ourplanning horizon and into our perception). Aswe deepen our exploration and learning, we seenew opportunities emerging as much for us asfor our opponent, and requiring us to re-directour search (and planning) efforts. We see thewidest possible spectrum of adaptive responsescompeting for the fittest solution (Bossel, 1998).

21


Diversity is an important prerequisite for sus-tainability.

Where possible, we follow the advice ofFrench military strategist Pierre-Joseph Bourcet(Alexander, 2002) and spread out attackingforces over multiple objectives, forcing an ad-versary to divide his strength and prevent con-centration. Such divided forces - a ”plan withbranches”, can be concentrated at will, espe-cially if superior mobility is present, as recom-mended by French military strategist Guibert.As an end result of all this positional pressureand maneuver, we seek what Napoleon sought,that is (Alexander, 2002), the nature of strategyconsists of always having (even with a weakerarmy) more forces at the point of attack or atthe point where one is being attacked than theenemy. Such positions have the possibility of thewin of material, and are then approached froma more tactical perspective - one that currentheuristics handle well.

From a high level, we visualize the oppo-nent’s pieces in the game of chess as a network,and agree with Wilson (Wilson, 2006) that thebest way to confront a network is to create acounternetwork, a non-hierarchical organizationcapable of responding quickly to actionable in-telligence obtained from diagnostic efforts. Net-works are an essential ingredient in any complexadaptive system. Without interactions betweenagents, there can be no complexity (Beinhocker,2007).

We aim for control (Wylie and Wylie, 1989)(Kelly and Brennan, 2010), defined by Mc-Cormick (McCormick, 2005) as (1) the abilityto see everything in one’s area of operation thatmight pose a threat to security and (2) the abilityto influence what is seen. Our main efforts mustbe to establish dynamic control. Once control is

established, the opponent becomes an ineffectivefighting force - but only in the way a tiger be-comes contained within the cage. Direct actiondoes not provide control; control provides theability to conduct effective direct action (Canon-ico, 2004). More specifically, we seek to managethe leverage in dislocating the enemy (Wylie andWylie, 1989) (Palazzo et al., 2010) that leads tocontrol, and to face up to the questions surround-ing how influence and the threat of destructionlead (dynamically, now or later) to the controlwe seek.

We strategize with Schoemer (Schoemer,2009) that our success depends on changingquickly and effectively so that we can do whatneeds to be done in the future. Change is un-predictable - we can’t know which changes willoccur, so our most valuable skill is being able tomaster any changes that do. We need to learnhow to master the inevitable, yet unpredictable,change we will face in playing our game. Weseek to control the controllables - learning howto focus our time and energy on issues where wecan make a difference and learning how not towaste our time and energy on problems we can’tsolve. We theorize with Schoemer that misman-aged change leaves us worse off than before, andresults in even more change. We identify thosethings that we can control and then get busycontrolling them (Schoemer, 2009).

We see the indirect approach as the most di-rect path to victory (Wilson, 2006) (Hart, 1991).

We cannot improve on the centuries-old ob-servation that the secret of all victory lies in theorganization of the non-obvious (Marcus Aure-lius). To accomplish this, we follow (Maslow,1987) and critically focus our attention on theunusual, the unfamiliar, the dangerous and thethreatening, while seeking (from necessity, and

22


for exploration purposes) to separate the dan-gerous positions from the safe.

We desire to create, in the words of Vickers(Allison and Zelikow, 1999) (Vickers, 1995), anappreciative system, where our value judgmentsinfluence what aspects of reality we care to ob-serve, which in turn are influenced by instrumen-tal calculations, since what we want is affected bywhat we think we can get. We seek to establisha readiness to distinguish and respond to someaspects of a system rather than others, and tovalue certain conditions over others. Central tothis concept will be indicators which aim to mea-sure cause rather than effect, and the gatheringof early knowledge as the essence of preparation(Beckham, 2007). If our chess playing agent cansuccessfully act as a rational actor, it is throughthe mechanism of an appreciative system thatthis is accomplished.

13 Competitive IntelligenceLeads to Competitive Ad-vantage

We see one factor above all others as contribut-ing to the success (or failure) of the proposedheuristic: the gathering of useful competitive in-telligence. Very simply, competitive intelligenceis any information that tells us whether our posi-tion is still competitive, or how to make it morecompetitive (Gilad, 1994). The fundamental ob-jectives of competitive intelligence are to avoidsurprises and gain competitive advantage (West,2001). Knowledge has value, but intelligence haspower (Rothberg and Erickson, 2005). We followFuld and define intelligence as a time-sensitiveassessment that will direct someone to act (Fuld,2010). Gilad (Gilad, 1988) offers another useful

definition: processed information of interest tomanagement about the present and future envi-ronment in which [a competing entity] is operat-ing.

For Fuld, change will occur and the futurewill not be the same as today. To prepare our-selves for that future, we look to signs of earlywarning (the ability to see into the future) in theform of leading indicators. Early warning con-sists of four very simple and ”intelligent” steps,which we adapt for our purpose: (1) drawingthe road map of possible futures, (2) identifyingthe signals we need to watch for each of thesefutures, (3) constructing automated scripts towatch those signals in the course of a machine-played game analysis and exploration, and (4)making sure we create an approach to act quicklyonce one of the futures we have identified (aspromising) begins to emerge (Fuld, 2010). Weagree with Fuld that the signals are out there -we just need to construct a diagnostic indicatorsensitive enough (but not prone to false alarms)to guide our exploration efforts. We ask not,”Is this perfectly accurate?” But rather, ”Is thissufficient to make a good decision?” (Hooper andScott, 1996).

We use competitive intelligence to reducethe risk that our exploration efforts will not bepromising. We identify intelligence - not infor-mation - as helpful to us and our programmedmachine in choosing these paths (Fuld, 1995).By actively seeking intelligence and learning howto use it, we hope to turn information into apowerful weapon that will give us a competi-tive advantage (Fuld, 1995) - information bothvalid and timely becomes war’s most powerfulweapon (Luttwak, 2001). Each competitor play-ing a game has virtually the same access to infor-mation. We envision, with Fuld, that the player

23


that is more effective in converting available in-formation into actionable intelligence will endup winning the game. Without intelligence, youmay succeed in winning a battle or two, but youcan’t expect to win the war (Fuld, 1995).

Gilad (Gilad, 1988) explains how competi-tive intelligence translates into competitive ad-vantage, which we modify slightly for the pur-poses of a machine playing a game. The purposeof the data collected is to enable the machinegame-player to arrive at an assessment of thecurrent situation on the board (in terms of itsposition) based on the key success factors. Thebirth of a strategy follows logically and chrono-logically the assessment of the situation. This, inturn, is based on the environmental intelligencepicture provided by the competitive intelligenceprogram. For Gilad, and for us, the better thatinput, the better the resulting strategy.

For a business example, one of us (JLJ) re-cently filled out a multi-question employee satis-faction survey from his current employer - STGInc. This was necessary, we were told, ”to con-tinue to improve processes to help achieve ournumber 1 Critical Success Factor - to be recog-nized as one of the best places to work.” The sur-vey was actually conducted by another companyhired for that purpose, in order to allow employ-ees the ability to respond anonymously. We weretold that ”Once the survey closes, the data isanalyzed, charts and graphs are created and rec-ommendations are made by HR Innovative Solu-tions.” When the period of time allowed for em-ployees to complete the survey passed, we werethen told the results - ”The overall satisfactionrating (OSR) was X.XX out of 4.06;” (we wereasked to keep the results proprietary, but theywere very good) ”an overall Satisfaction Ratingof 3.71 is considered industry standard.” One

question we were asked was ”What would makeour company a better place to work?” We see thisexample as competitive intelligence in action,supporting the analysis of successful achievementof critical success factors, which drive correctiveactions. We can identify with Greene’s position(Greene, 1966) that management doesn’t careabout intelligence sources, nominal costs of col-lection, or clever filing techniques; they want (re-liable) answers to questions, and they want theanswers promptly. Intelligence is, in every sense,a control system - the intelligence system keepsthe competing entity on track with the externalenvironment - with reality (Page, 1996). STGcould have used an inexpensive method for thesurvey, such as e-mail or instructing managersto pass out forms and collect them. Employeesmight then be less than honest in their response,fearing that it could somehow be tracked backto them. STG management determined that anaccurate (although likely expensive) survey wasnecessary to make precise changes to companypolicies in pursuit of achieving good results inchosen critical success factors.

We proceed now with Kahaner’s first partof the intelligence cycle - planning and direction(Kahaner, 1997) - which involves a clear under-standing of the user’s needs (key success factors),and establishing a collection and analysis plan.What is essential here is knowing what needsto be known, at the moment it is needed foruse (Rothberg and Erickson, 2005), and turn-ing that knowledge into appropriate diagnosticaction. Rockart (Rockart, 1979) defined Criti-cal Success Factors as the limited number of ar-eas in which satisfactory results will ensure suc-cessful competitive performance for the individ-ual, department, or organization. Continuing,he felt that they are the few key areas where

24


things must go right for the competitor to flour-ish. If results in these areas are not adequate,the organization’s efforts for the period will beless than desired. We agree with Rockart thatthe critical success factors are areas of activitythat should receive constant and careful atten-tion from management. Specifically, the currentstatus of performance in each area should be con-tinually measured.

14 From Orientors to Indica-tors to Goals

We have found six basicsystem orientors (existenceand subsistence, effective-ness, freedom of action, se-curity, adaptability, coex-istence) that apply to allautonomous self-organizingsystems -Hartmut Bossel

We identify andadapt the frame-work independentlyarrived at by Vick-ers, Bossel andMax-Neef (Vick-ers, 1959) (Bossel,1976) (Bossel,1977) (Bossel, 1994)(Bossel, 1998) (Bossel, 1999) (Bossel, 2007)(Muller and Leupelt, 1998) and (Max-Neef,1991) to conceptualize the critical success fac-tors which guide diagnostic action, which in ourvision share much with that of an ecosystem.We seek indicators which realize Bossel’s six ba-sic high-level orienting properties of existenceand subsistence, effectiveness, freedom of action,security, adaptability, and coexistence. We the-orize with Bossel that these properties are eachvital diagnostic indicators of successful systemdevelopment, and we aim to steer our initialsearch efforts along paths which seek to improvethe weakest of these properties. Holistic indica-tors allow us to understand if the system understudy is globally following a path that takes

the system to a ”better” or to a ”worse” state(Jorgensen and Fath, 2007). These indicatorsmust give a fairly reliable and complete pictureof what really matters (Bossel, 1998).

If a system is to be viable in the long run, aminimum satisfaction of each of these basic ori-entors must be assured (Bossel, 1994). We the-orize with Bossel that the behavioral responseof the system is conditional on the chosen in-dicator set: problems not perceived cannot beattacked and solved (Bossel, 1977). Meaning-ful non-routine behavior can only occur by ref-erence to orientors, which are therefore key el-ements of non-routine behavior (Bossel, 1977)(Bossel, 2007). The possible successes of unori-ented non-routine behavior can be only chancesuccesses (Bossel, 1977) (Bossel, 2007). Bosseleven goes so far to declare (Bossel, 2007) thatorientor-guided decision-making will lead to sus-tainable development without requiring specifi-cation of intermediate or end states. A sustain-ability indicator should point the way to a courseof action (Bell and Morse, 2008). What is lack-ing is not data but an understanding of whatis important and the resolve to act (Lawrence,1997).

We directly follow Lockie (Lockie et al.,2005) in our conceptual foundation of indicators,in which we directly quote due to the importanceof the concept. Indicators are instruments to de-fine and monitor those aspects of a system thatprovide the most reliable clues as to its overallwell-being. They are used, in other words, toprovide cost and time-effective feedback on thehealth of a system without necessarily examiningall components of that system. According to pro-ponents, the validity of indicators is based on thedegree to which the wider network of componentsand relationships in which they are situated link

25


together in a relatively stable and self-regulatingmanner, and the degree to which the indicatorsthemselves represent the most salient or criticalaspects of the system that can be monitored overtime.

Needs provoke real im-pulses for action... whensufficiently gratified ceaseto exist as active determi-nants or organizers of be-havior

We also followLawrence (Lawrence,1997) and declarethat indicators areintended to answerthe question: ”Howmight I know objec-tively whether things are getting better or get-ting worse?”. What we are really interested inare directional indicators, which are less focusedupon numerical representations and are morefocused upon action, as in ”I should do some-thing about this.” For data to be useful to us,it must describe things which actually matter toour future. Objective and relevant data needsto be converted into information if it is to beuseful in the development of sustainability in-dicators (Lawrence, 1997). Information that ismeasured should evoke happiness when the sit-uation improves and unhappiness when it getsworse. If the change doesn’t matter, we are notmonitoring the right data (Lawrence, 1997).

Maslow (Maslow, 1987) notes that needs,along with their partial goals, when sufficientlygratified cease to exist as active determinants ororganizers of behavior. Bisogno (Bisogno, 1981)notes that the term need means a state of dis-satisfaction provoked by the lack of somethingfelt as being necessary. Needs provoke real im-pulses for action, which for Max-Neef, become(instead of a goal) the motor of development it-self (Max-Neef, 1991). Importantly for Bisogno,needs which would appear to be essential in a

particular moment, are no longer so when thesecircumstances - time, place, (or for Maslow astate of satisfaction), change. A need becomesa necessity when its satisfaction is absolutely in-dispensable to a given state of affairs (Bisogno,1981).

Health and fitness of a sys-tem require adequate satis-faction of each of the sys-tem’s basic orientors. Plan-ning, decisions, and actionsin societal systems musttherefore always reflect atleast the handful of basicorientors (or derived crite-ria) simultaneously. Com-prehensive assessments ofsystem behavior and devel-opment must also be multi-criteria assessments...

We see a valuein the two-phasedapproach of (Bossel,1976) and (Bossel,1994): first, a cer-tain minimum quali-fication must be ob-tained separately foreach of the basic ori-entors. A deficitin even one of theorientors potentiallythreatens our long-term survival fromour current position. Our computer software willhave to focus its attention on this deficit. Onlyif the required minimum satisfaction of all basicorientors is guaranteed is it permissible to try toraise system satisfaction by improving satisfac-tion of individual orientors further - if conditions,in particular our opponent, will allow this.

We see goal functions as operating to trans-late the fundamental system needs expressed inthe basic orientors into specific objectives linkingsystem response to properties observed on thechess board. We conceptualize that goal func-tions emerge as general properties in the coevo-lution of the chess position and dynamic, futuredevelopment. They can be viewed as specific re-sponses to the need to satisfy the basic orientors.For example, mobility is related to adaptabil-ity, constraints relate to coexistence, king safety

26


is related to the orientors of security and exis-tence, virtual existence and stress are related toeffectiveness, material is related to existence, se-curity and adaptability, etc. We can creativelycome up with new indicators for our search ori-entation, but we see them fitting within the pro-posed framework and ’dimension of concern’ asoutlined previously.

We see the vital orientors, which express ourvalues, as operating together to create a selectionmethod for our immediate goals. The goals weseek are not specific objects, but rather changesin our relations or in our opportunities for relat-ing (Vickers, 1995).

...a system’s developmentwill be constrained by theorientor that is currently’in the minimum’. Par-ticular attention will there-fore have to focus on thoseorientors that are currentlydeficient. -Hartmut Bossel

We see an inter-esting similarity withthe ”ABC” (airway,breathing, circula-tion) priority systemused in emergencyroom and rescue op-erations when decid-ing what to do nextwith an accident victim. The rescue team per-forms the set of vital diagnostic tests and thenfocuses their immediate attention on the criticalindicator that scores the lowest. The ”health”of the victim (and in fact the direction to takenext) would not be based on an average or sum-mation score of the vital indicators, but insteadon the vital indicator which scores the lowest.The goal, then would be to do something whichimproves the score returned by that indicator.If more than one indicator is below a certaincritical threshold (such as, the patient is notbreathing and there is no circulation), then Car-diopulmonary Resuscitation (CPR) would needto be performed - improvement of the airway,

breathing and circulation indicators are all si-multaneously attempted.

We also see a similarity to the commonyearly performance evaluation which is tradi-tionally performed by American management oneach company employee, or even a report cardgiven to a student. The ritual evaluation will liststrengths, weaknesses, and expectations, and itis also common to list improvements necessaryto reach the next performance level. The smartworker will examine his vital, multi-criteria di-agnostic assessment and orient his or her efforts(during the next year) towards improving theweakest scoring of these indicators, while con-tinuing to leverage strengths and meet the listedexpectations.

Our experience in computerchess over the past fewyears seems to indicate thatfuture chess programs willprobably benefit from eval-uation functions that alteras the general chess envi-ronment changes. -PeterFrey, Chess Skill in Manand Machine, 1977

We see similar-ities to Festinger’sprinciple of cogni-tive dissonance (Fes-tinger, 1957), wherea perception of disso-nance between an ob-served indicator anda desired value leadsto activity orientedtowards reduction ofthe perceived dissonance. Festinger believes thatreduction of dissonance is a basic process in hu-mans, preceded first by perception and identifi-cation.

We see the chess programs of the future asaddressing this conceptual foundation, in cre-ative ways and approaches that cannot yet beenvisioned by today’s developers. Our concep-tualization of stress management and the con-struction of resilient positions as indicators are,ideally, part of an operational realization of the

27


six orientors. If our concept fails as an orientor orfocus of search efforts, then it needs to be modi-fied or itself re-engineered. Perfectly usable indi-cators might overlap, or require too much proces-sor time to implement. Perhaps what is requiredis the art of a talented programmer/chess playerto select a set of indicators which also orient witheffective insight.

the set of basic orientorsderives from the question:”Given the global featuresof the system and of its en-vironment, what basic ori-enting dimensions must thesystem refer to in its non-routine behavior, and inparticular in fundamentalbehavioral decisions in or-der to fulfill the global in-struction of the supremeorientor?” -Hartmut Bossel

What we are say-ing is simply thatwe must pay at-tention to each ofthese orientatingqualities separately- we should not justroll them up intoa grand, universal”number” and ex-pect to effectivelyand efficiently driveour search effortsin that fashion. Aweakness in one of the six orientors criticallyimpacts sustainable development in the uncer-tain future and cannot be ”made up for” witha higher score from the others. A simple mech-anism for scoring our search efforts, such asaveraging the lower two indicators (of six total,one for each orientor), or using the lowest if it isfar beneath the others, will make sure that themachine pays attention to (and focuses attentionon) those orienting parameters that are in needof improvement.

We seek sustainability itself as a goal, whichmakes sense because our opponent can offer usanything else we would otherwise seek (and ini-tially appearing to move us closer to checkmate),but in a way that (for us) might not be sustain-

able in the long run, due to the hidden effects ofdynamic complexity. We can measure sustain-ability, in its simplest form, by the weakest ofour vital diagnostic indicators. A weighted sumof vital indicators would be used for endpointevaluation purposes, possibly including a limiteron each parameter.

Our present [evaluationfunction] is blind to thesimplest phenomena. Theevaluator gladly acceptsa position in which thecomputer is a knight aheadalthough its king is out inthe center of the boardsurrounded by hostileenemy queens and rooks.-David Slate and LawrenceAtkin, Chess Skill in Manand Machine, 1977

These orient-ing indicators, whichhelp us to constructa picture of the stateof our environmenton which we can baseintelligent decisions(Bossel, 1998), canall be based on acommon foundation,such as cumulativestress, but with aweighting that aimsto highlight the par-ticular dimension of concern. Our goal is simplyto determine, through appreciative indicators,”What matters most now?” (Vickers, 1995) andthen to (initially) focus attention on any movewhich we perceive to make progress in that areaor dimension of concern. We choose to behavelike an efficient business manager, besieged bynumerous concerns and pressed for time, de-ciding how to allocate attention in the face ofconstant demands, both known and unknown,in dynamically creating a response to the impor-tant and expensive (if wrong) question ”whatdo I do now?”. Curiously, how we allocate theattention of the machine becomes a decision ofprofound impact on the quality of the move wewill later decide to make.

What good is being a piece up if your King

28


is in the center of the board, surrounded by hos-tile enemy pieces? Better to see if we can returnthe King to a safe place, even at the price of ma-terial, so that we can continue the sustainabledevelopment of our position in the future. Wetherefore orient our attention and future search-ing in ways to improve King safety. Our imme-diate goals, therefore, emerge from the weakestindicators (results) of the vital diagnostic tests,and operate to focus the search efforts along linesthat allow sustainable development in the uncer-tain future.

test the [strategic] princi-ple for its ability to promoteand guide action. In partic-ular, assess whether it ex-hibits the three attributesof an effective strategicprinciple. Will it forcetradeoffs? Will it serve asa test for the wisdom ofa particular business move,especially one that mightpromote short-term profitsat the expense of long-termstrategy? ...

The orientorsrepresent our wantsor intentions - an in-tention doesn’t ex-actly require anydeep calculation orplan. We can ex-plore the moves that(partially) satisfy ourwants, and by sim-ple focused learn-ing, examine the con-sequences of whatemerges as we slideforward a few promising moves into the future.We need both the readings and the norms. Foronly if we know both where we are and wherewe want to go can we act purposefully in seeingabout getting there (Laszlo, 1996).

...Does it set boundarieswithin which people willnonetheless be free toexperiment? -Gadiesh,Gilbert, TransformingCorner-Office Strategy intoFrontline Action

We tentativelyenvision the follow-ing chess-based dy-namic leading indi-cators as orientorsand strategic guides

to action, based onBossel’s collection:existence and subsistence (short-range probingand quick exploration of promising moves indi-cates that position is sustainable and good movesare available), effectiveness (material, adjustedby positional engagement of each piece - level ofstress created by pieces makes sufficient short-and long-range threats to reduce resilience ofopponent while sufficiently avoiding opponent’sthreats), freedom of action (mobility - including2nd and 3rd order, penalty if pieces are trappedor pinned, multiple good moves available), se-curity (dynamic King safety), adaptability (po-sitional score is not decreasing with increasedsearch depth), coexistence (effective use of con-straints to weaken effect of enemy pieces, whileavoiding enemy constraints, pieces are not con-strained by friendly pieces). We simply ask,”What areas of competitor activity do we feelneed close attention?” (Fuld, 1995). With regardto the indicated direction for exploration, we asknot, ”Is this perfectly accurate?” But rather, ”Isthis sufficient to make a good initial decision?”

Hubert points out (Hubert, 2007) that whatis generally missing in sustainability programsis a set of leading indicators (such as thoseproposed above) that provide signals of systemchanges that will ultimately affect the system’soutput, and are timely enough to allow inter-vention that can change the outcomes. Whenproperly done, these leading indicators provideinsight into the state of a system’s health. ForHubert, an unbalanced dependence on laggingindicators (such as rewarding pieces for sittingon good squares) is to be fooled by early suc-cesses, or what is sometimes called the ”gettingbetter before it gets worse” - focusing on an out-come (maximum yield) rather than on leading

29


indicators of health. Without leading indicators,we cannot easily distinguish early successes fromthe early stages of looming failure. Additionally,he feels that the common cause behind many re-source management failures is this focus on man-aging for a single outcome, which first improvesperformance, but later leads to system collapse.Finally, Hubert declares his opinion that we cansustain systems that are evolving when we un-derstand that all we need to do is think in termsof sustaining a system’s health and functionalityrather than its specific form or condition (Hu-bert, 2007).

15 Shannon’s Evaluation Func-tion

when one is modeling somesituation... it is reasonableto use any assumptionsthat work, but it is notreasonable to make theseassumptions into ”laws,”or to forget that theseare assumptions that peo-ple made in the first place.-William Byers

Shannon pro-posed (Shannon,1950) a simple eval-uation to be per-formed in relativelyquiescent positions.While recent tour-naments have shownthat such evalua-tions (combined withalpha-beta pruningand the null-move heuristic) can be used to pro-duce world-class chess programs, we seek analternate approach with the capability of evenbetter performance. Programs that use Shan-non’s evaluation often have trouble figuring outwhat to do when there is no direct sequenceof moves leading to the placement of pieces onbetter squares (such as the center), or the acqui-sition of a ”material” gain.

We see a general correlation between theplacement of a piece on a ”good square” and theability of that piece to inflict stress on the oppo-nent, and to mitigate the effects of stress causedby well-placed opponent’s pieces. We even seethat the concept of mobility has value in a gen-eral sense. However, we see problems with thistechnique being used to build positional pres-sure, such as the kind needed to play an effec-tive game of correspondence chess. The long anddeep analysis produced by the machine is oftenfocused in the wrong areas, as determined by theactual course of the game.

Networks are comprised ofa set of objects with directtransaction (couplings)between these objects...these transactions viewedin total link direct andindirect parts togetherin an interconnectedweb, giving rise to thenetwork structure... Theconnectivity of nature hasimportant impacts on boththe objects within thenetwork and our attemptsto understand it. If weignore the web and lookat individual unconnectedorganisms... we missthe system-level effects.-Jorgensen, Fath, et al., ANew Ecology

The stress pro-duced by the Shan-non method is notof the type that re-duces the coping ca-pacity of the op-ponent, or increasesour own resilience,in certain game sit-uations where posi-tional play is re-quired. For exam-ple, in positions thatare empty of tacti-cal opportunities, themachine can be effec-tively challenged byopponents who knowhow to play a goodpositional game ofchess (Nickel, 2005).The terms of the Shannon evaluation functiondo not seem suitable metrics for guiding searchand planning efforts, in these cases.

Fontana (Fontana, 1989) advises us to ask:

30


what are the stressors, what needs to be doneabout them, and what is stopping us from doingit? There is little to be gained from generalizing,if our goal is to identify the stressors, accuratelyassess the levels of stress present, and mobilizeaccording to the results.

16 The Positional EvaluationMethodology

We propose that an approach which attempts toincrease the oriented positional pressure or cu-mulative stress on the opponent, even if unre-solved at the terminal positions in our searchtree, is a viable strategy and has the potentialto play a world-class game of chess. Our strate-gic intent is to form targeted positional pressure(aimed at weakpoints defined by chess theoryand at constraining the movement of the enemypieces) that will resolve at some future point intime into better positions, as events unfold andgameplay proceeds. At minimum, this pressurewill allow for sustainable development as onecomponent of a resilient position. We will notjudge pieces by the ”squares” they occupy, butinstead, by our heuristic estimate of the level offocused stress they can contribute (or mitigate)in the game.

So we need something likea map of the future. A mapdoes not tell us where wewill be going, or where weshould be going - it merelyinforms us about the pos-sibilities we have... Wetherefore need a descriptionof the possibilities ahead ofus...

We con-struct an orien-tation/evaluationmethodology withthe goal of makingour machine moreknowledgeable withregard to the posi-tional concepts dis-

cussed earlier. In de-signing our method-ology, we heed the advice of (Dombroski, 2000)that this methodology is our test of effects andconsequences and is our guiding light in oursearch for the consequences of our choices.

Our orientation/evaluation centers on aheuristic appraisal of the stress we inflict on theopponent’s position, and our mitigation of thestress created by the opponent. We aim to re-duce our opponent’s coping ability through care-ful targeting of stress. The dynamic forces ofchange, acting over time and in a future we of-ten cannot initially see, transform the reducedcoping ability of our opponent, our carefully tar-geted stress, and our resilient position full ofadaptive capacity, to future positions of advan-tage for us.

...Such a map would nothave to give us very de-tailed information... But itshould give us a useful im-age of what may be ahead,and allow us to comparethe relative merits of dif-ferent routes... before weembark on our journey. -Hartmut Bossel

Perhaps thisconcept is what in-spired Bobby Alli-son to race most ofthe 1982 Daytona500 without a backbumper - it fell offafter contacting an-other car early inthe event (NASCAR,2009). Some driversaccused Allison’s crew chief of rigging thebumper to intentionally fall off on impact. Al-lison’s car without the bumper had improvedaerodynamics, and the forces of dynamic changeoperating over the 500 mile race supplied thedriver with an advantage he used to win. Otherexamples (the winged keel of the Australia IIyacht and the new loop-keel design, hinged iceskates and performance enhancing swimsuits

31


come to mind) show how small changes, com-bined with other critical abilities and interactingwith a dynamic environment over time, can cre-ate a performance advantage.

We seek, in similar fashion, to favor certaininteracting arrangements of pieces, such that thedynamic forces of change (operating during theplaying of a game) cause favorable positions toemerge over time, from beyond our initial plan-ning horizon. We seek to re-conceptualize the”horizon effect” to our advantage. We cannotarrange for a bumper to fall off during a chessgame, but we can do the equivalent - we canactively manage the dynamics of change to im-prove the chances for persistence or transforma-tion (Chapin et al., 2009). This would includethe general approaches of reducing vulnerabil-ity, enhancing adaptive capacity, increasing re-silience, and enhancing transformability (Chapinet al., 2009). We manage the exposure to stress,in addition to the sensitivity to stress (Chapin etal., 2009).

We adopt the vision of (Katsenelinboigen,1992), that we define a ”potential” which mea-sures a structure aimed at forcing events in ourfavor. Ideally, one which also absorbs or reducesthe effects of unexpected events.

We follow the suggestion in (Pearl, 1984)to use as a strategy an orientation/evaluationbased on a relaxed constraint model, one thatideally provides (like human intuition), a streamof tentative, informative advice for managing thesteps that make up a problem-solving process,and use the insight from (Fritz, 1989) and (Ster-man, 2000) that structure influences behavior.

In order to more accurately estimate the dis-tant positional pressure produced by the chesspieces, as well as to predict the future capabilityof the pieces in a basic form of planning (Lakein,

1974) (Shoemaker, 2007) we create the softwareequivalent of a diagnostic probe which performsa heuristic estimate of the ability of each pieceto cause and mitigate stress. The objectiveswe select for this stress will be attacking enemypieces, constraining enemy pieces, and support-ing friendly pieces (especially those pieces thatare weak). To support this strategy, we calculateand maintain this database of potential mobilityfor each chess piece 3 moves into the future, foreach position we evaluate.

Sustainability... means, assaid before, that only theriverbed, not the exact lo-cation of the river in it, canand should be specified -Hartmut Bossel

We updatethis piece mobilitydatabase dynami-cally as we evaluateeach new leaf posi-tion in our searchtree. This databasehelps us determine the pieces that can be at-tacked or supported in the future (such as 2moves away from defending a piece or 3 movesaway from attacking a square next to the enemyking), as well as constrained from accomplishingthis same activity. Note that the piece mobil-ity we calculate is the means through which wedetermine the pressure the piece can exert ona distant objective. We can therefore see howmobility (as a general concept) can become avital holistic indicator of system health and onepredictor of sustainable development.

We reduce our bonus for each move that ittakes the piece to accomplish the desired objec-tive. We then consider restrictions which arelikely to constrain the piece as it attempts tomake moves on the board.

For example, let’s consider the pieces in thestarting position (Figure 4).

32


Figure 4: White and Black constraint map, pieces at thestarting position Legend: Red: pawn constraints, Yellow: Mi-nor piece constraints, Green: rook constraints, Blue-green:Queen constraints, Blue: King constraints

What squares can our knight on g1 influencein 3 moves, and which squares from this set arelikely off-limits due to potential constraints fromthe enemy pieces?

Figure 5: Influence Diagram and Simulation Diagram, Ng1at starting position Legend: Red - 1 move influence, Yellow- 2 move influence, Green - 3 move influence, Dark Red - 1move influence possibly constrained by opponent piece, DarkYellow - 2 move influence possibly constrained by opponentpiece, Dark Green - 3 move influence possibly constrained byopponent piece, Blue - no influence possible within 3 moves, X -presence of potential constraint ”Power... is diagrammatic... itpasses not so much through forms as through particular pointswhich on each occasion mark the application of a force, theaction or reaction of a force in relation to others” -Deleuze

We now construct the influence diagram(Shoemaker, 2007) and the simulation diagram(Bossel, 1994) (Figure 5), which are interpreted

in the following way. If a piece is on our influ-ence diagram for the knight, then it is possibleto attack it or defend it in 3 moves (this includeswaiting moves or moves which move a piece outof the way). We label this kind of map an influ-ence diagram because it shows the squares thatthe piece can influence in 3 moves, provided thatit is unconstrained in movement by the enemy.

Keep in mind that we need to take into ac-count the location of the other pieces on thechessboard when we generate these diagramsfor each piece. If we trace mobility through afriendly piece, we must consider whether or notwe can move this piece out of the way before wecan continue to trace mobility in that particu-lar direction. If we trace mobility through anenemy piece, we must first be able to spend 1move capturing that piece.

Foucault (Foucault, 1982) defines a relation-ship of power as a mode of action that does notact directly and immediately on others. Instead,it acts upon their actions: an action upon anaction, on possible or actual future or presentactions. A relationship of violence, in compari-son, acts upon a body or upon things; it forces,it destroys, or it closes off possibilities. We canthreaten the knight itself, or the potential ac-tions of the knight. The knight makes threatsof its own on the board - its power is exercisedrather than possessed (Foucault, 1995). An exer-cise of power shows up as an affect, since force de-fines itself by its very power to affect other forces(and in turn to be affected by them)(Deleuze andHand, 1988).

We focus on power relations because powerproduces knowledge (Foucault, 1995). Power andknowledge directly imply one another - there isno power relation without the correlative consti-tution of a field of knowledge, nor any knowl-

33


edge that does not presuppose and constitute atthe same time power relations (Foucault, 1995).We strategically apply Foucault’s conception ofpower to the pieces on the gameboard - we seethe maneuvers in terms of the network of rela-tions, constantly in tension, and as a perpetualbattle rather than a simplistic conquest of terri-tory (Foucault, 1995). The knowledge we acquireis meaningless if it is not derived from these ma-neuvers and the corresponding power relations.

Comparing this 3-move map with a diagramof the starting position, we can determine thatthe white knight on g1 can potentially attack 3enemy pieces in 3 moves (black pawns on d7, f7and h7). We can defend 8 of our own pieces in 3moves (the knight cannot defend itself).

The influence diagram cap-tures the behaviorally rele-vant structure of the sys-tem. It is therefore thebasis for any simulationmodel. Because of itsimportance for the suc-cess of model development,the influence diagram hasto be developed with careand precision... To cap-ture their dynamics cor-rectly, real systems can-not usually be representedby linear approximations -Hartmut Bossel

We decide toreward pieces fortheir potential abil-ity to accomplish cer-tain types of worth-while positional ob-jectives: attackingor constraining en-emy pieces, defend-ing friendly pieces,attacking squaresnear our opponentsking (especially in-volving collabora-tion), minimizing ouropponent’s ability toattack squares near our own king, attackingpieces that are not defended or pawns that can-not be defended by neighboring pawns, restrict-ing the mobility of enemy pieces (specifically,their ability to accomplish objectives), etc. Inthis way, we are getting real about what the

piece can do. The bonus we give the piece is1. a more precise estimate of the piece’s abilityto become strategically engaged with respect tocausing or mitigating stress and 2. operationallybased on real things present on the chessboard.In this way, our positional orientation/evaluationmethodology will obtain insight not usually ob-tained by a computer chess program, and allowour machine to take positive, constructive action(Browne, 2002). It is still an estimate, but thegoal here is to focus our search efforts on likelymoves in a positional style of play, and to eval-uate positions from a more positional point ofview.

What does the orientation/evaluationmethodology look like for the proposed heuris-tic? We model (and therefore estimate) thepositional pressure of our pieces, by following atwo-step process:

1. We determine the unrestricted future mo-bility of each chess piece 3 moves into the future,then

2. We estimate the operating range or levelof engagement of the pieces by determining thelimiting factors or constraints that bound the un-restricted mobility.

The concept of using limiting factors isbriefly mentioned (Blanchard and Fabrycky,2006) in the context of Systems Engineering.(Lukey and Tepe, 2008) argue that an impor-tant aspect of cognitive appraisal is the extent towhich stress-causing agents are perceived as con-trolled. Balancing processes such as constraints(Anderson and Johnson, 1997) seek to counterthe reinforcing loops created by a piece creatingstress, which, if unconstrained, can potentiallycreate even more stress (perhaps in combina-tion with other pieces). Once we have identifiedthe limiting factors, we can more easily exam-

34


ine them to discover which ones can be alteredto make progress possible - these then becomestrategic factors.

The consideration of constraints is a partof the decision protocol of Orasanu and Con-nolly (Orasanu and Connolly, 1993) and (Pless-ner et al., 2008) which also includes the iden-tification of resources and goals facing the de-cision maker. We therefore reduce the bonusfor accomplishing objectives (such as, attackingan enemy piece or defending a friendly piece) ifthe required moves can only be traced throughsquares that are likely to result in the piece be-ing captured before it can accomplish its objec-tive. We also reduce the engagement bonus formobility traced through squares where the pieceis attacked but not defended. We may use an-other scheme (such as probability) for determin-ing stress-application reduction for piece move-ment through squares attacked both by friendlyand enemy pieces where we cannot easily re-solve whether or not a piece can trace mobilitythrough the square in question (and thereforecreate stress). We think in terms of rewarding aself-organizing capacity to create stress out of thevaried locations of the pieces and the constraintsthey face (Costanza and Jorgensen, 2002).

We reward each piece for its predicted abil-ity to accomplish strategic objectives, exert po-sitional pressure, and restrict the mobility of en-emy pieces, based on the current set of pieces onthe chess board at the time we are calling ourorientation/evaluation methodology. Using an-ticipation as a strategy (van Wezel et al., 2006)can be costly and is limited by time constraints.It can hurt our performance if it is not done withcompetence. An efficient compromise betweenanticipative and reactive strategies would seemto maximize performance.

We give a piece an offensive score based onthe number and type of enemy pieces we can at-tack in 3 moves - more so if unconstrained. Wegive a piece a defensive score based on (1) howmany of our own pieces it can move to defend in 3moves and (2) the ability to mitigate or constrainthe attacking potential of enemy pieces. Again,this bonus is reduced for each move it takes toaccomplish the objective. This information is de-rived from the influence diagram and simulationdiagram we just calculated. Extra points can begiven for weak or undefended pieces that we canthreaten.

The proposed heuristic also determines kingsafety from these future mobility move maps.We penalize our king if our opponent can movepieces into the 9-square template around ourking within a 3 move window. The penalty islarger if the piece can make it there in 1 or 2moves, or if the piece is a queen or rook. We pe-nalize our king if multiple enemy pieces can at-tack the same square near our king. Our king isfree to move to the center of the board - as long asthe enemy cannot mount an attack. The incen-tive to castle our king will not be a fixed value,such as a quarter pawn for castling, but ratherthe reduction obtained in the enemy’s ability tomove pieces near our king (the rook involved inthe castling maneuver will likely see increasedmobility after castling is performed).

The king will come out of hiding naturallywhen the number of pieces on the board is re-duced and the enemy does not have the poten-tial to move these reduced number of pieces nearour king. We are likewise free to advance thepawns protecting our king, again as long as theenemy cannot mount an attack on the monarch.The potential ability of our opponent to mountan attack on our king is the heuristic we use as

35


the basis for king safety. Optionally, we will con-sider realistic restrictions that our own pieces canmake to our opponent’s ability to move piecesnear our king.

Pawns are rewarded based on their chance toreach the last rank, and what they can do (piecesattacked and defended in 3 moves, whether ornot they are blocked or movable). The piecemobility tables we generate should help us iden-tify pawns that cannot be defended by otherpawns, or other pieces - it is this weakness thatwe should penalize. Doubled or isolated pawnsthat cannot be potentially attacked blockaded orconstrained by our opponent should not be pe-nalized. Pawns can be awarded a bonus basedon the future mobility and offensive/ defensivepotential of a queen that would result if it madeit to the back rank, and of course this bonus isreduced by each move it would take the pawn toget there.

The information present in the future mo-bility maps (and the constraints that exist onthe board for the movement of these pieces) al-low us to better estimate the positional pres-sure produced by the chess pieces. From thesecalculations we can make a reasonably accu-rate estimate of the winning potential of a posi-tion, or estimate the presence of positional com-pensation from a piece sacrifice. This orienta-tion/evaluation score also helps steer the searchprocess, as the positional score is also a measureof how interesting the position is and helps usdetermine the positions we would like to searchfirst.

In summary, we have created an initialmodel of positional pressure which can be usedin the orientation/evaluation methodology ofa computer chess program, which can be re-fined in diagnostic tournaments of many short

games. (Michalewicz and Fogel, 2004) remindus that models leave something out, otherwisethey would be as complicated as the real world.Our models ideally provide insight and identifypromising paths through existing complexity.

(Starfield et al., 1994) emphasize that prob-lem solving and thinking revolve around themodel we have created of the process understudy. We can use the proposed model of po-sitional pressure to direct the machine to focusthe search efforts on moves which create the moststress in the position as a whole. For our searchefforts, we desire a proper balance between ananticipatory and a reactive planning strategy.We desire our forecast of each piece’s abilities tohelp us anticipate its effectiveness in the game(van Wezel et al., 2006), instead of just reactingto the consequences of the moves.

By identifying the elements and processes inour system (Voinov, 2008), identifying the limit-ing factors from the interactions of the elements,and by answering basic questions about space,time and structure, we describe and define theconceptual model of our system.

17 Observations from Cogni-tive Science

We make the following observations about ourapproach, from (Wood, 2009), which in our vi-sion also apply to the concept of a machine play-ing a game.

Our motives and needs, for whatever wechoose to do, affect what we see and don’t see.After carefully selecting what we choose to no-tice, we need somehow to make sense of theseperceptions and form strategic guides for our be-havior. Wood declares that the most useful the-

36


ory for explaining how we organize perceptionsis constructivism, which is the theory that we or-ganize and interpret experience by applying cog-nitive structures called schemata.

We use four types of cognitive schemata tomake sense of perceptions: prototypes, personalconstructs, stereotypes, and scripts. Scripts areguides to action based on experiences and obser-vations. A script consists of a sequence of ac-tivities that identify what we and others are ex-pected to do in certain specific situations. Manyof our daily activities are governed by scripts,although we’re often unaware of them. We the-orize, based on the interpretation of (Honeycuttand Cantrill, 2001) that scripts are a kind of au-topilot, that much subconscious activity whichtakes place in playing a game consists of follow-ing scripts, triggered by perceptions. In most ofthese activities, we use scripts to organize per-ceptions into lines of action. The script tells uswhat to do, in our case - how to gather and or-ganize information, when we find ourselves in ageneral or even a particular situation.

Scripts represent generalized knowledge(Lightfoot et al., 2009) and as such, can be usedto command a machine to take actions (or figureout what is likely to happen next) in a general-ized situation - such as addressing or determiningthe needs of a position in a board game.

For (de Wit and Mayer, 2010), Knowledgethat people have is stored in their minds in theform of ’cognitive maps’. These cognitive mapsare representations in a person’s mind of howthe world works. A cognitive map of a certainsituation reflects a person’s belief about the im-portance of the issues and about the cause andeffect relationships between them. A person’scognitive map will focus attention on particu-lar phenomena, while blocking out other data as

noise, and quickly make clear how a situationshould be perceived. Cognitive maps help to di-rect behavior, by providing an existing repertoireof ’problem-solving’ responses (also referred to as’scripts’) from which as appropriate action canbe derived.

Our machine will use scripts to, among otherthings, construct a map showing how fully en-gaged a piece is in the game. Maps are guides toaction (Hahlweg and Hooker, 1989) because theydepict genuine invariant relationships that exist,in this case, among the pieces on the game board.We will also use scripts to manage the stress in aposition, along particular dimensions of concern,and to manage search and exploration efforts.

18 Serious Play Can Lead toSerious Strategy

We follow (Brown, 2009) and (Sutton-Smith,2001) in a conceptualization of play that willform one foundation of our automated searchand exploration efforts.

Humans adopt play as a foundational be-havior that guides exploratory activity and insome cases becomes a basis for acquiring knowl-edge. Play is the basis of all art, games, books,sports, movies, fashion, fun, and wonder (Brown,2009). Play is the vital essence of life - it is whatmakes life lively (Brown, 2009). However, a ma-chine does not know how to play. It simply doeswhat we tell it to do, so we must tell it how toplay with the pieces on the board and the re-lationships among these pieces. Why must ourmachine play? because, Play is the answer tothe question, How does anything new ever comeabout? (Jean Piaget).

We further conceptualize play (Sutton-

37


Smith, 2001) as the extrusion of internal men-tal fantasy into the web of external constraints.Additionally, we adopt the practical aspect thatplay seems to be driven by the novelties, excite-ments, or anxieties that are most urgent to theperceptions of the players (Sutton-Smith, 2001).Finally, we note that the imagination makesunique models of the world, some of which leadus to anticipate useful changes - the strategicflexibility of the imagination, of play, and of theplayful is the ultimate guarantor of our game-based survival (Sutton-Smith, 2001). Play liesat the core of creativity and innovation (Brown,2009).

We desire our machine to always be busymaking up its own work assignments (Paley,1991). Specifically, the ”work assignments” in-volve choosing our courses of action and adjust-ing those courses based on the internal satisfac-tions we receive (Henricks, 2006). We desire fromour machine a behavior similar to ”playfulness”,and a set of creative, inquisitive, exploratory ori-entations centered on an object-based model ofthe game-world (Henricks, 2006). We desire anactivity of directed exploration, object manipu-lation and precise appraisal. We seek to managethe exploration of the new. This conceptuallyinvolves the creation of small-scale experimentsthat can be run outside the mainstream man-agement systems and learned from (Valikangas,2010). Whatever we do, we do not perform asimmutable policy, but as an experiment. We usethe action to learn. Learning means the willing-ness to go slowly, to try things out, and to collectinformation about the effects of actions, includ-ing information that the action is not working(Meadows et al., 2005). We resort to strate-gic experiments because more is unknown ratherthan known - the winner is often the one who

learns and adapts the quickest (Govindarajanand Trimble, 2005).

We agree with (Brown, 2009) that movementis primal and accompanies all the elements ofplay we are examining. Through movement play,we think in motion - movement structures ourknowledge of the world, space, time, and our re-lationship to others (Brown, 2009).

Our exploration of future game positionstherefore must take into account piece movementthat is likely, critical, interesting, stress induc-ing/relieving, or otherwise ”lively”. Children atplay engage other children or contemplate waysto engage their playmates. We therefore desireto create a heuristic which playfully examinesthe future consequences of the transformation ofstress on the board, as the pieces move about orare constrained by objects on the board, such asblocked pawns or lower-valued enemy pieces.

The creative person can beseen as embodying or act-ing as two characters, amuse and an editor... themuse proposes, the editordisposes. The editor crit-icizes, shapes, and orga-nizes the raw material thatthe free play of the musehas generated. -StephenNachmanovitch

We begin our ef-forts by conceptu-alizing the buildingof a principal varia-tion two moves intothe future, by firstexamining the sin-gle move that cre-ates the most per-ceived stress for ouropponent (or miti-gates the perceivedstress caused by his pieces). We then look atthe most likely response. For Roos and Vic-tor, the first thing we do with play is to activelyconstruct what we see in our imagination. Thisconstruction phase allows us to bring our intu-ition from all our experience and our analysesinto something concrete, something we can play

38


with (Roos and Victor, 1998). For de Geus (deGeus, 2002), we do not navigate to a predefineddestination. We take steps, one at a time, intoan unknowable future.

After ”sliding forward” in this fashion, wethen work backwards in our principal variation,examining the consequences of the next mostlikely move, and so on. In this ”serious play”we seek serious strategy (Roos and Victor, 1998)- we strive to retain control over the course ofthe imagined interaction by constantly reactingto its emerging results - what can and cannotbe done must depend on what the enemy can orcannot do (Luttwak, 2001). The ambiguity weface as we look into the future generates its ownpossible resolution (Byers, 2011). Any specificunderstanding of ambiguity must necessarily betentative - ambiguity is real but cannot be madeprecise. It is ambiguity and not certainty thatbest describes the way things are (Byers, 2011).

Play is experimenting witha toy that the player ac-cepts as representing his orher reality. This makesthe toy a representation ofthe real world with whichthe learner can experimentwithout having to fear theconsequences... Under-neath all the fun there is avery serious purpose: play-ing with one’s reality allowsone to understand more ofthe world we live in. Toplay is to learn. -Arie deGeus

Byers declaresthat the clarity of sci-ence has room withinit for the ambiguousand goes seriouslyastray when this am-biguity is unacknowl-edged (Byers, 2011).Further for Byers,the statement of thefundamental ambigu-ity (such as the prin-cipal variation wedevelop in attempt-ing to ”play” a gamesuch as chess) givesus an insight into what is going on; at everylevel, the same fundamental dynamic of ambi-

guity plays itself out (Byers, 2011). The resultsof science and the critical problems that we facedemand that we face up to uncertainty and am-biguity, no matter how stressful this is (Byers,2011).

We establish a few simple rules for our seri-ous play: we orient our search efforts (initially)along the lines of improving the score of theweakest, vital diagnostic test - the strategic prin-ciple which enables us to do something now byguiding our action and helping to allocate scarceresources (Gadiesh and Gilbert, 2001). We per-form a search ”cut-off” only after we confirmthat the position in question is resilient andthe moves left unexamined are not the mostpromising (and remain so), after performing ashallower search. We seek to focus our effortson uncovering the likely future consequences ofthe most promising extensions of managing thestress in the position. Unexpected discoveries inthe principal variation will cause the machine tore-focus its efforts on the next most promisinglines. We then begin to deepen our search ef-forts and spend more time exploring alternatemoves in our principal variation.

This is nothing more than Ashby’s model foradaptiveness (Bertalanffy, 1968), where the sys-tem tries different ways and means, and eventu-ally settles down in a field where it no longercomes into conflict with critical dynamic val-ues of the environment. Our leverage for deal-ing with ”driving forces” comes from recogniz-ing them, and understanding their effects. Lit-tle by little, our actions contribute to new driv-ing forces which in turn will change the worldof the gameboard once more (Schwartz, 1996).On some level, all three layers of intelligence -action, strategy, and prediction - need to occursimultaneously to create a seamless sustaining of

39


competitive advantage (Rothberg and Erickson,2005).

Intelligence for a system with limited pro-cessing resources consists in making wise choicesof what to do next (Simon and Newell, 1976).There is no easy solution for complex problems.What there is instead is an obvious direction (forexploration). The reason is that often there aretoo many (interacting) variables in a situation(Trout, 2008). This directed and flexibly persis-tent ”evolution” creates designs, or more appro-priately, discovers designs, through a process oftrial and error (Beinhocker, 2007).

Evolution is a general-purpose and highly power-ful recipe for finding inno-vative solutions to complexproblems. It is a learn-ing algorithm that adaptsto changing environmentsand accumulates knowl-edge over time. -Eric Bein-hocker

Evolution is apossibility generator(Beckham, 1998). Avariety of candidatedesigns are createdand tried out in theenvironment; designsthat are successfulare retained, repli-cated and built upon,while those that areunsuccessful are discarded (Beinhocker, 2007).Evolution is a method for searching enormous,almost infinitely large spaces of possible designsfor the almost infinitesimally small fraction ofdesigns that are ”fit” according to their particu-lar purpose and environment (Beinhocker, 2007).Evolution is a general-purpose and highly pow-erful recipe for finding innovative solutions tocomplex problems (Beinhocker, 2007). It is alearning algorithm that adapts to changing envi-ronments and accumulates knowledge over time(Beinhocker, 2007). The limits to this approachare seen to be the ability to manage complex-ity, and knowledge (Beinhocker, 2007). Beckham

agrees (Beckham, 2006), declaring that smart or-ganizations subject their most important deci-sions to a Darwinian environment in which thestrongest ideas survive and evolve to higher lev-els of fitness. The strategist looks at evolutionnot so much in terms of the survival of actual or-ganisms, but the survival of ideas (van der Heij-den, 2005).

Stephen Gould (Gould, 1996), speaking ofbiological evolution, notes that a species canevolve further only by using what physical prop-erties it has in new and interesting ways. Any bi-ological adaptation also produces a host of struc-tural by-products, initially irrelevant to the or-ganism’s functioning but available for later co-optation in fashioning novel evolutionary direc-tions. Evolution continually recycles, in differ-ent and creative ways, many structures built forradically different initial reasons (Gould, 2002).For Gould, much of biological evolution’s cre-ative power lies in the flexibility provided by thisstorehouse of latent functional potential. It isquirky shifts and latent potential, redundancy,and selected flexibility - three basic principleswhich define and permit the creativity of evolu-tion, the capacity to originate novel structuresand functions.

For Mitchell (Mitchell, 2009), the result ofevolution by natural selection, in our case simu-lated, is the appearance of ”design” but with nodesigner. We hypothesize with Mitchell that theappearance of computer-produced design comesfrom chance, the selection for exploration of thepromising moves which are fit for the game envi-ronment, and long periods of simulated time inorder to validate this fitness.

We see the search ”tree” formed in this fash-ion as an extended diagnostic test of how re-silient and adaptively controlling our position is

40


- the predisposed capacity to respond effectivelyto future situations that are beyond our abilityto predict. We see resilience as the basic strengthand adaptive control (with the flexible persis-tence of Beckham (Beckham, 2002) as a founda-tion) as the primary objective. These propertiesare more measurable and meaningful than esti-mates of winnability, especially in the case wherewe are deciding what to do next (and ignorantof what the future holds). The ”tree” is morea tool which is useful to plan what we want tolearn, rather than an expectation of where wewill be in the end (Cohn, 2006). We fully ex-pect that our opponent will (eventually) play amove which will take us outside of our currentlearning tree, and we fully expect, through themechanisms of resilience and adaptive control, tobe able to meet the challenges of the positionswhich newly emerge.

After ”evolving” a plan through a mecha-nism that ”proposes” and then ”disposes”, wecan test it using the principles of war gamingdeveloped by Gilad (Gilad, 2009). More specif-ically, we develop basic scenarios that illustratethe full range of potential strategic shifts (threatsor opportunities) (Page, 1996). A wargame de-velops scenarios (through the mechanism of sim-ulated competition) that otherwise might not oc-cur to us (Herman et al., 2009). The basic aimof a war game, which ideally captures the com-plexity of competitive dynamics, is to turn infor-mation into actionable intelligence (Fleisher andBensoussan, 2007). Gilad would have us envisionany and all plans that we develop as bets thatcome with risk, a risk originating from the com-petitive dynamics in our environment. We nowtest our plan and its assumption that the com-petitive response we will receive from our oppo-nent is containable. War gaming is nothing more

than role-playing in order to understand a thirdparty, with the goal of answering: What will theopponent do? What then is my best option? Gi-lad cautions that war gaming will not guaranteesuccess - nothing will - but states that it will in-crease the odds in our favor. Ideally, an effectivewar game produces a list of improvements for theexisting plan, or a list of options for a new plan.

Scenario-based planning attempts to makesense of the situation by looking at multiple fu-tures, which are treated as equally plausible, re-flecting not only the inherent uncertainty in thesituation, but also what is considered predictable(van der Heijden, 2005). The purpose of scenar-ios, wrote Pierre Wack, is to gather and trans-form information of strategic significance intofresh perceptions. When this works, it leadsto strategic insights beyond the mind’s previ-ous reach (Schwartz, 1996). For Schwartz, it isdriving forces, predetermined elements, and crit-ical uncertainties which give structure to our ex-ploration of the future (Schwartz, 1996). Theprocess of building scenarios starts with look-ing for driving forces, the forces that influencethe outcome of events. Driving forces are theelements that move the plot of a scenario, thatdetermine the story’s outcome. Without drivingforces, there is no way to begin thinking througha scenario. For Schwartz, they are a device forhoning an initial judgment, for helping to decidewhich factors will be significant and which fac-tors will not (Schwartz, 1996).

For Michael Howard, it is essential that weconstantly try to adapt ourselves to the unpre-dictable, and to the unknown. Our plans, what-ever they are, are likely wrong. This fact is, forHoward, amazingly irrelevant. What matters isthat we get them right when the critical mo-ment arrives (Howard, 1974). We affirmatively

41


answer Herman’s central question (Herman etal., 2009): if we had the opportunity to probethe future, make strategic choices, and view theconsequences of those choices in a risk-free en-vironment before making irrevocable decisions,that we would in fact take advantage of such anopportunity. For (Oriesek and Schwarz, 2008),wargaming is a form of accelerated learning.

This process is termed ”path analysis” byBossel (Bossel, 2007), who suggests that our firsttask consists of quickly finding the most relevantdevelopment paths despite a multitude of uncer-tain, time-dependent, or adjustable parameters.The efficiency of this search, in his and our opin-ion, depends on how cleverly possible parame-ter constellations are combined in consistent andplausible scenarios.

The second task of path analysis is the com-parative evaluation and assessment of differentdevelopment paths to clarify which path (orwhich group of paths) should be preferred. ForBossel, in this phase of the work, evaluation cri-teria have to be introduced that reflect the ex-istence and development interests of the system.We must make sure that the necessary minimumlevel of orientor fulfillment is achieved for eachindividual orientor, then we must determine thetotal quality of orientor satisfaction (for individ-ual orientors and some aggregated quality mea-sure).

We additionally note positions where imbal-ances are created (using our vital diagnostic in-dicators) and investigate the consequences, espe-cially when efforts to return to a resilient positionrequire extra efforts.

For software testing and configuration pur-poses, we envision the use of automated tour-naments of hundreds of games, each lasting per-haps three minutes long, to assess and fix the pa-

rameters of these orientation/evaluation/searchefforts so that we might succeed in the widestnumber of situations. We envision a tool whichidentifies and stores positions where faulty anal-ysis was generated. We see the program-mer/developer examining these saved positionsand identifying the reason for the failure to ori-ent/evaluate the indicated position.

We recognize certain positions as ”tacti-cal” in nature when responses become forced orwhen imbalances in vital indicators create fewbranches in our principal variation. We defer inthese cases to a search and evaluation method-ology designed for a more tactical situation.

We critically examine the trade-offs betweenexamining principal variations that are manymoves long, versus the exploration of the sec-ondary and tertiary lines that do not go as deep.We conceptualize our machine behaving like achild at play, creating novel combinations, andfinding or discovering what works and does notwork in an evolutionary fashion.

We can base our efforts on the ob-served behavior of large groups of Internet-connected humans examining a common chessposition, such as the daily chess puzzle featuredat http://www.chessgames.com/index.html (wehave no connection to the owners of this site -one of us (JLJ) pays a yearly fee to access cer-tain advanced site features).

19 John Boyd’s OODA Loop

The OODA Loop (Observe, Orient, Decide,Act) is a strategic methodology which wasoriginally applied by USAF Colonel JohnRichard Boyd to the combat operation pro-cess http://en.wikipedia.org/wiki/OODA loop.

42

http://en.wikipedia.org/wiki/OODA_loop


Boyd was of the opinion that without OODAloops, we will find it impossible to comprehend,shape, adapt to, and in turn be shaped by an un-folding, evolving reality that is uncertain, ever-changing, and unpredictable (Boyd, 1996). Boydadvocates an approach of pulling things apartand putting them back together until somethingnew and different is created (Boyd, 1987). Fur-ther, Boyd suggests we present our opponentswith ambiguous or novel situations in which theyare not capable of orienting their behavior orcoping with what’s going on (Boyd, 1987), whilewe maintain our fingerspitzengefuhl. For Boyd,orientation shapes the way we interact with theenvironment, and therefore the way we observe,decide, and act (Boyd, 2005). Boyd suggeststhat effective orientation demands that we createmental images, views, or impressions, hence pat-terns that match with the activity of our world(Boyd, 2005).

For Boyd, in a competitive encounter againsta talented opponent, our limited perceptionscause novelty to be produced continuously, andin an unpredictable manner. In order to main-tain a competitive position we must match ourthinking and doing, hence our orientation, withthat emerging novelty. Yet, any orientation weassume prior to this emerging novelty is perhapsmismatched after the fact, possibly causing con-fusion and disorientation. However, Boyd pointsout, the analytical/synthetic process permits usto address these mismatches so that we can com-petitively rematch ourselves and thereby reorientour thinking and action with that novelty (Boyd,1992).

We therefore envision our search and explo-ration process as an operational realization ofthis concept. We observe information, unfoldingcircumstances and interactions, orient our be-

havior according to Bossel’s concepts discussedearlier, decide which path to explore, and thenact by ”sliding forward” one move. We then re-peat the process, periodically ”backtracking” toexamine moves which were initially determinedto be the next best.

(Boyd, 1976) attempts to philosophically ar-rive at a theory useful for conducting warfareor other forms of competition, such as playinga game. Boyd concluded that to maintain acompetitively effective grasp of reality one mustoperationally follow a continuous cycle of inter-action with the environment oriented to assess-ing its constant changes. Boyd states that theOODA decision cycle is the central mechanism ofsuch adaptation, and that increasing one’s ownrate and accuracy of assessment (compared tothat of one’s opponent) provides a strategic foun-dation for acquiring an operational advantage ina dynamically changing environment.

Conceptually, we are exploring the presentand future consequences of the transformationof positional stress, with an emphasis on thesustainability of the intermediate positions, thesatisfaction of our operational needs, and (ul-timately) the perceived winnability of the finalposition. We are faced with a dynamic, novel,unstable world that we must constantly adaptto even as we try to shape it for our own ends(Hammond, 2001).

20 Adaptive Campaigningmodel of Grisogono andRyan

Grisogono and Ryan (Grisogono and Ryan,2007) propose the model of Adaptive Campaign-ing as a modified form of Boyd’s OODA loop

43


that presents a more relevant form for the chal-lenges of operating in an environment with highoperational uncertainty. Here we ’adapt’ theirapproach for game theory.

Adaptive Campaigning proposes a repeat-ing cycle of Act Sense Decide Adapt (ASDA).By placing ’Act’ first this model stresses theneed to act (make exploratory trial moves) withwhatever information is present, and by imme-diately following that with a ’Sense’ of what haschanged in our environment. The ’Decide’ func-tion follows to determine what is learned fromthe sensed feedback that results from the ac-tion, and what to do next - including possiblere-orientation based on results from the vital di-agnostic tests.

These first three elements of Adaptive Cam-paigning correspond closely to the four elementsof Boyd’s OODA loop, but with a different em-phasis on where the cycle starts, and with the’Orient’ function of OODA incorporated into the’Decide’ functions of ASDA. The object of the’decision’ is to choose the next trial move in ourforward exploration, or to begin backtracking byexploring alternative moves in our principal vari-ation. So ’Adapt’, the fourth element of ASDA,explicitly adds the need to invoke adaptation andconsider what, if anything, should be changed onevery cycle, before continuing to the next cyclewith another external ’Act’.

Ideally, successful application of the ’Adapt’element results in the machine improving its abil-ity to focus/orient its efforts on the right objec-tives at the right time and in the right place.Modern combat, including game playing, cantherefore be characterized as competitive learn-ing in which all sides are constantly in a pro-cess of creating, testing and refining hypothesesabout the nature of the reality of which they are

a part (Kelly and Brennan, 2009).

Recent criticism of Adapting Campaigning(Thomas, 2010) claims that Boyd’s work ade-quately addresses the issues in question, andshould be revisited. Time will tell whetherOODA or ASDA loops will prevail.

21 Endpoint Evaluation

Much as an employee receiving a yearly evalua-tion from his or her employer, we must come upwith a method to decide how desirable a positionis, at the point we stop diagnostic explorationand probing efforts. But the future is unknown,and unknowable. We seek therefore to estab-lish dynamic potential through a sum of laggingand leading indicators - the orientors discussedearlier, except that we are no longer interestedin guiding diagnostic action but instead in es-tablishing value, much as a Consumer Reportsmagazine evaluates, then ranks automobiles viaa score relating to their perceived value.

Where possible, our numerical score is basedon chances of winning. In certain cases, open-ing book databases can be consulted to establisha winning percentage, based on the number ofhigh-level games played and the win-loss-drawresults. We speculate that two computer chessprograms, each developed independently, mightconsistently arrive at nearly the same numer-ical endpoint evaluation, in the opening stageof the game, if calibrated to the winning per-centage expected from databases of recent high-level games, and where an identical strategy hasbeen selected of obtaining a resilient positionand adaptive control in the face of uncertaintyand resistance. Evidence for independent devel-opment might only be proven with longer time

44


controls, such as in correspondence chess, or inpositions obtained in middlegame or endgame,where the different selective search mechanismsare uniquely influenced and differentiated byfinding sustainable paths to advantage.

Alternatively, distance (in moves) fromcheckmate can be used, where we can directlyperceive the checkmate in our search tree.

We suspect that most positions faced by ourmachine will not fall into either category. Wepropose instead a method which uses orientedstress to establish the size of the advantage, mea-sured as the estimated size of the mistake whichcan be made by player with the advantage, whichwould lead to a sustainable, even game. This in-dicator is more directly measurable than winningchances, which are often shrouded in dynamiccomplexity, and can be used in a game strategywhich seeks to accumulate small positional ad-vantages over time. What we are saying is simplythat it is easier (in most cases) to measure andfavor ”increasing distance from draw” than ”de-creasing distance from checkmate” - this alignswith Lawrence (Lawrence, 1997), who declaresthat it may be more important to know whetherwe are making progress towards the goal than itis to know the size of the gap between the cur-rent situation and the (ultimate) goal we haveset. We hypothesize that this conceptual foun-dation is equivalent, in most cases.

This measurement philosophy will need tobe adjusted in certain well-known cases, such asRook and pawn endings, or Bishop of oppositecolor endings, where an additional pawn mightnot have direct leverage into winning potential.Such cases would need to be programmed in on acase by case basis, starting with the most likelyendgames, and consulting a reference such asFine (Fine and Benko, 2003).

The purpose of the endpoint evaluation (forthe principal variation) should be to establisha marker against which we compare competingmoves. We construct strategic challenge lines(with less effort in time) and see how close wecome to the marker score. Those challenge lineswhich come close in score to the marker will be-come strategic fallback positions, and will be-come elevated to the new principal variation - ora replacement branch of the principal variation -if discoveries are made which force us to changeour mind on which move is the most promising.The details need to be fixed through procedureswhich are developed and refined in diagnostictournaments of hundreds of 3-minute games.Obviously, we do not want to spend time (orattention) looking at unpromising moves whichhold little chance of becoming the principal vari-ation.

In comparing different paths of system de-velopment, we hypothesize with Bossel that themost favorable path will be the one for which (1)the minimum conditions are always satisfied forall orientors, and (2) the overall orientor satis-faction is better (Bossel, 2007).

22 Complexity

We insist that it is complexity itself which de-mands an approach much like the proposedheuristic in order to play a game like chess in atactically empty position. Complex systems arecontrolled by countless individual interactionsthat occur inside the system (Benyus, 2002).The complexity present when playing in the po-sitional style is due to connections - the moreconnected something is, the more complex it is(Beckham, 2001). A change in one connectedthing gives rise to changes in the various things

45


to which it is connected. More connections meanmore change (Beckham, 2001).

In a complex environment, the changes thatone action will generate are often beyond predic-tion because of all the other interactions theyset off (Beckham, 2001) (Byers, 2011). Smallchanges often amplify to become very largechanges - all we can do is watch for warning signs(Benyus, 2002). Complex conditions demandcontinuous adaptation. In a complex, highlyconnected system, things happen fast - or in away that involves a quick emergence into our per-ception. Maintaining a steady state of dynamicbalance requires continuous adjustment and ac-commodation. These shifts occur naturally asone change sets off another (Beckham, 2001).

In Beckham’s ”zone of complexity” muchdifferent approaches are needed to succeed.These approaches involve making short predic-tions, enabling self-organization, using simplematerials as building blocks, being continuouslyflexible and adaptive, all while looking for lessonsand metaphors in other complex systems, partic-ularly biological systems. Out there in the zoneof complexity, things are different. We agreewith Beckham that management that succeedswill be catalytic, facilitative, enabling, adaptive,incremental, and patient (Beckham, 2001).

Systems expert Russell Ackoff once empha-sized that success with a true system demandsthe effective management of interactions, notthe management of actions. Interaction is whathappens continuously at the various connectionsbetween things. It follows then that successfulmanagement in a densely connected system in-volves managing effectively in an environment ofcomplexity (Beckham, 2001).

23 Results

We have created software to demonstrate cer-tain features of the proposed heuristic and nowexamine four positions to see if we can obtaina better positional understanding of how wellthe pieces are performing. John Emms (Emms,2001), reached Figure 6 as white (black to move)with the idea of restricting the mobility of black’sknight on b7.

Figure 6: Emms-Miralles (Andorra, 1998) Constraint mapsLegend: The left diagram identifies the possible constraintsimposed by the white pieces, with red representing pawn con-straints, yellow minor piece constraints, green rook constraints,blue-green queen constraints, and blue king constraints. Theright diagram identifies possible constraints imposed by theblack pieces. The white and grey squares represent the stan-dard chessboard squares without constraints.

How fully engaged is this piece in the game?Let’s see what the influence diagram and simu-lation diagram from the proposed heuristic showus:

46


Figure 7: Emms-Miralles Tracing knight mobility from b7-a5-c4-b2 and b7-d8-e6-g5

Figure 8: Emms-Miralles Influence Diagram and SimulationDiagram for Nb7

We generate the constraint maps as in Fig-ure 6 in order to estimate the squares that theknight on b7 is likely to be denied access. Wethen apply the constraint maps to the individ-ual vectors which make up the influence diagramas in Figure 7 to create the simulation diagram.When a movement vector hits a constraint, fu-ture mobility through that square is constrained,and we use an ”X” to indicate constrained mo-bility. We can see from the X’s (denied potentialmobility) of Figure 8 that the movement of thepiece on b7 has been constrained. It is Emms’view that positional details like this one can bevitally important when assessing positions.

Figure 9: Constraint maps, white (left), black (right),Estrin-Berliner variation analysis (1965-68 corr.) after 12.Qe2Be6 13.Qf2, Black to move

Figure 10: Influence Diagram and Simulation Diagram forBc1

Figure 11: King safety heuristic maps: left - black kingsafety, right - white king safety. In the left diagram, darkersquares are safer squares for the black king, while lighter col-ored squares are more dangerous.

47


The organization and itsenvironment impinge oneach other in many ways.Strategy succeeds or failsby interacting with this en-vironment. It succeeds byavoiding, making use of, orovercoming, the impinge-ments. -Geoffrey Cham-berlain

Figure 9 ex-amines a sidelinefrom Estrin-Berliner(1965-68 corr.) af-ter the proposed im-provement 12.Qe2Be6 13.Qf2. Howfully engaged is thewhite Bishop on c1?We generate the con-straint maps and in-fluence diagram as before in order to constructthe simulation diagram. We see that the bishopon c1 can enter the game after moving a pawnout of the way, and become useful for creatingand mitigating stress in future positions.

Figure 11 displays an experimental kingsafety heuristic which is generated from all thepiece influence diagrams and a rule which awardspoints based on number of pieces which can at-tack a square and the distance/constrained effortrequired to do so.

Figure 12: Constraint maps, white (left), black (right),Umansky-World correspondence game (2009)

Figure 13: Influence Diagram and Simulation Diagram forQe8

Figures 12 and 13 examine a positionfrom the recent Umansky-World correspondencegame. The constraint map gives insight to thecontrolling influences present on the squares, andthe influence diagram/ simulation diagram forthe Queen on e8 gives insight to what this piececan threaten in 3 moves. Note that this piececan influence square c1 via the difficult to findmove sequence e8 to e6-h6-c1.

Figure 14: Influence Diagram and Simulation Diagramfor Rb8, Levy-Chess 4.4, simultaneous exhibition, 1975, after27.axb5

48


Figure 15: Influence Diagram and Simulation Diagramfor Rb8, Levy-Chess 4.4, simultaneous exhibition, 1975, after31.Bc8

Figures 14 and 15 show how a machine canpotentially recognize a trapped piece, with anexample first identified and discussed by (Levy,1976).

The computer can use the heuristic knowl-edge present in the influence diagram and simu-lation diagram to estimate the strategic potentialor how fully engaged each piece is in the game.The maps are a useful holistic measurement of acapacity to produce stress in a position, and canbe used as part of an oriented, vital system-levelindicator to predict and manage the sustainabledevelopment of a position in a chess game.

24 Conclusions

The elements of a systemand their interactions de-fine the system structure...

Alternative con-ceptual frameworksare important notonly for further in-sights into neglecteddimensions of the underlying phenomenon. Theyare essential as a reminder of the distortions andlimitations of whatever conceptual frameworkone employs (Allison and Zelikow, 1999). Only

by analyzing a phenomenon from an alternativeperspective (preferably multiple alternative per-spectives) can all the intricacies of a situation beunderstood (Canonico, 2004). Our alternativeconceptual framework for machine-based chesscan, at minimum, allow us deeper insight andbetter understanding of current methods. Par-ticularly in explaining and predicting actions,when one family of simplifications becomes con-venient and compelling, it is even more essentialto have at hand one or more simple but com-petitive conceptual frameworks to help remindus of what was omitted (Allison and Zelikow,1999). Allison and Zelikow believe this is a gen-eral methodological truth applicable in all areasof life, including, in our opinion, a strategy forplaying a game. One source suggests that weshould look at between four and six alternateconcepts for our design (Kossiakoff and Sweet,2003).

...By answering the basicquestions about space,time and structure, wedescribe the conceptualmodel of the system...Creating a conceptualmodel... very much re-sembles that of perception-Alexey Voinov

We agree withstrategist BernardBrodie that strategyis a field where truthis sought in the pur-suit of viable solu-tions, not at all likepure science, wherethe function of the-ory is to describe, or-ganize, and explain and not to prescribe. Thequestion that matters in strategy is: Will theidea work? (Steiner, 1991). Brodie believed thatstrategy was associated with problems involv-ing economy of means, i.e., the most efficientutilization of potential and available resources(Steiner, 1991).

A systemic (rather than analytic) approach,

49


focusing on interactions and feedback mecha-nisms rather than concentrating on agents, willoffer insights on where to apply leverage so asto contribute to the development of security andstability. The targeting derived from such an ap-proach will focus on building and fostering iden-tified sources of resilience and adaptive capacity,while mitigating or disrupting sources of stress.Complexity theory highlights the non-linearityof feedback mechanisms, implying a requirementfor the continuous monitoring of measures of ef-fectiveness in order to adapt effects-seeking op-erations (Calhoun and Hayward, 2010).

Ecosystems are working models of sustain-able complex systems, and it is reasonable tostudy them for clues to the sustainable manage-ment of the human enterprise (Jorgensen andMuller, 2000), including ’conflict ecosystems’mentioned by Kilcullen (Kilcullen, 2006). Weidentify systems thinking and the systems ap-proach as the theoretical basis for an orienta-tion/evaluation methodology, shifting our focusfrom the parts to the whole. The use of ap-proximate knowledge and the conceptualizationof a network of interacting components is real-ized through a system dynamics model of stress,or positional pressure.

The properties of the partscan be understood onlyfrom the dynamics of thewhole. In fact, ultimatelythere are no parts at all.What we call a part ismerely a pattern in an in-separable web of relation-ships. -Fritjof Capra, TheRole of Physics in the Cur-rent Change in Paradigms

The reality of theposition on the chess-board is seen as aninterconnected, dy-namic web of powerrelationships, withoriented, cumula-tive stress one driv-ing force of change.You can avoid re-ality, but you can-

not avoid the conse-quences of avoiding reality (Ayn Rand). We seekresilient positions and flexible, adaptive capacity(with the promise of sustainable development)to counter the effects of unknown positions thatlurk just beyond our planning horizon. Theconcepts of orientors and indicators, cumulativestress, constraints and virtual existence allowus to effectively simplify the dynamic reality ofeach game piece interacting with every othergame piece on the board - to the point where wecan predict promising directions of exploration(via the mechanism of stress transformation) andidentify the accessibility space (Bossel, 1998) offuture sustainable development.

In the final analysis, per-ception seems to be the keyto skill in chess... The dif-ference between two play-ers [when one defeats theother in a game] is usu-ally that one looks at thepromising moves, and theother spends his time go-ing down blind alleys. -Neil Charness, Chess Skillin Man and Machine, 1977

A model can beconsidered as a syn-thesis of elements ofknowledge about asystem (Jorgensenand Muller, 2000).Our model of dy-namic interactionpresented in this pa-per ideally capturesthe dominant vari-ables that controlthe transformation ofstress (Kossiakoff and Sweet, 2003), omittingthe higher order effects that have a cost/benefitdeemed to be overall not effective. No modelsare valid or verifiable in the sense of establish-ing their correctness (Sterman, 2000) (Voinov,2008). The question facing clients, academics,and modelers is not whether a model is truebut whether it is useful as a basis for some ac-tion, which in our case, is steering search efforts(through the critical lines) in an exponentially

50


growing tree of possibilities, in a way that ob-tains actionable intelligence and therefore allowsa strong positional game of chess to be played.

(Miller and Page, 2007) advise, with regardto computational modeling, that we judge thequality and simplicity of the model, the clev-erness of the experimental design, and examineany new insights gained by the effort. We shouldalso ask ourselves if our model has just enoughof the right elements, and no more. To be agood model, Miller is of the opinion that we havestripped phenomena down to their essentials, yethave retained enough of the details to producethe insights we require. For Nijhout et al., (Ni-jhout et al., 1997), the most important thingthat should be required of a model is that, withsmall quantitative changes in parameter values,it can produce the evolutionary diversity presentin that pattern, and the effects of perturbationexperiments and mutations on the pattern. Itmust also reproduce in its dynamics reasonableportions of the ontogenetic transformation thatthe real pattern undergoes (Nijhout et al., 1997).We conceptualize an equivalence with the gameposition, and the exploratory moves suggestedby our model.

Ideally, our responsibility would be to usethe best model available for the purpose at hand(Sterman, 2000) despite its limitations. Weview modeling (Sterman, 2000) as a process ofcommunication and persuasion among modelers,clients, and other stakeholders. Each party willjudge the quality and appropriateness of anymodel using criteria which reflect on their roleand perceived future benefits. This includes thetime and effort involved in the unending strug-gle to improve the model to the point where itsperformance reflects what theory would expectof the particular approach. Modeling team A

might not want to use a particular model due tosignificant time, money, belief, performance, andfamiliarity with their current approach. Team Amight not even be interested in discussing newapproaches. However, modeling team B mightbe looking for a new challenge, perhaps due todissatisfaction with the current model, a belief inpredicted performance, or perhaps due to a will-ingness to spend long hours and to engage withthe types of problems suggested by the new ap-proach. Team A might now become interested,seeing the preliminary success of team B.

Learning to handle a com-plex system means learningto recognize a specific setof indicators, and to assesswhat their current statemeans for the ’health’, orviability, of the system. Of-ten this learning of indica-tors is intuitive, informal,subconscious... - HartmutBossel

Our attemptsto reengineer theway machines playchess are, in thetrue spirit of reengi-neering (Hammerand Stanton, 1995),throwing away cur-rent methods andstarting over, butplacing at the fore-front of our designefforts the values and concepts of positional chessand Systems thinking. We acknowledge the dy-namic and static elements of a chess position,and construct a sensor array which responds toa perception of stress in the position in orderto orient our efforts to effectively navigate andexplore an exponentially growing search tree.We adopt a Soft Systems Methodology - that is,we see the game position as complex and con-fusing, and we seek to organize the explorationof future consequences through the means of alearning system (Checkland and Poulter, 2006).

The proposed heuristic offers insight on theability of the chess pieces to create and miti-

51


gate stress and aims for a rich awareness of dis-criminatory detail (Weick and Sutcliffe, 2007)between promising and less promising positions.We agree with Donohew, et al., (Donohew et al.,1978), that information seeking must be a pri-mary method for coping with our environment.Key components include the monitoring of struc-tural tension created by the pieces as they mu-tually constrain each other and seek to satisfyvital system-level needs, and the attempt to cre-ate positions which serve as a platform for fu-ture success, in a future that is uncertain. Allsustainable activities have to accept the naturalsystem of constraints in which the investigatedentity operates (Jorgensen and Muller, 2000).

Our orientation/evaluation centers on an ar-ray of vital diagnostic appraisals of the cumula-tive stress each side inflicts on the opponent’sposition, and the perceived mitigation of suchstress. (Selye, 1978) considers stress to be an es-sential element of all our actions, and the com-mon denominator of all adaptive reactions. Weaim to reduce our opponent’s coping ability andadaptive capacity through oriented targeting ofstress. The dynamic forces of change, acting overtime and in a future we often cannot initially see,ideally transform the reduced coping ability ofour opponent, our carefully targeted stress, andour resilient position full of adaptive capacity, tofuture positions of advantage for us. The entirepurpose of modeling stress is to aid search focus- that is, we orient or focus our search effortsin priorities based on the changing amounts ofstress in the position (and the results of vitaldiagnostic tests). We additionally monitor thestress that threatens to become real, having theproperty that (von Neumann and Morgenstern,1953) have called ”virtual” existence. Even if thethreat does not materialize, it nevertheless has

the capability to shape and influence the eventsthat do become real.

The invariance of basic ori-entors... as well as thechange in attention focusresulting from changes inorientor satisfaction, pro-vide the system with theability to cope flexibly andadaptively with a widelyand quickly changing stateof system and environment.-Hartmut Bossel

(Jorgensen,2009) and (Bossel,2007) discuss the ap-plication of Bossel’sorientor ideas to sim-ulated animals (an-imats) roaming insimulated environ-ments, where orien-tation rules are de-veloped over time todirect and controlthe behavior of the simulated animal and op-timize the acquisition of food and energy re-sources. These simulations involve the ’percep-tion’ by the simulated animal of clues in theenvironment to the presence of food as well asdanger. We ask ourselves what orientation ruleswould develop if the simulated environment wereinstead the board game of interest. Might wethen develop optimal rules (or minimally, a goodset of rules) for orienting our search behavior forplaying a board game?

It is the possibility of let-ting a great many relation-ships influence each otherunder precisely stated as-sumptions and of deter-mining the consequences,which gives the computermodel its enormous powerand advantage over con-ventional planning meth-ods...

We acknowl-edge that resilienceis a distinguishingcharacteristic of anysuccessful system(Sanderson, 2009)(Gunderson et al.,2010). The creationof resilient positionsfull of adaptive ca-pacity allows us tosharply and effec-tively postpone search efforts in less-promising

52


lines with the low-risk promise of sufficient re-sources to ’MacGyver’ the unknown future thatlies beyond. We determine the level of resiliencepresent in a position using a set of (heuristic) vi-tal diagnostic tests, such as the ones proposed byBossel. We desire a methodology which emulatesa productive thinking process, such as one envi-sioned by (Hurson, 2008), but where we playfullyconsider responses that reflect the changing, ur-gent stress in the position, and the resilience ofthe less urgent positions and analysis lines leftunexamined.

We configure our search/exploration activityusing the results from automated tournaments of3-minute games.

...The speed with which thegreat number of calcula-tions are accomplished en-ables one to experiment re-peatedly with different as-sumptions for the futurein different parts of themodel, i.e. with differ-ent ”scenarios”. - MichaelG. Strobel and HartmutBossel

From the high-est level, we desireto model the cumu-lative dynamic stresspresent in the po-sition so that wecan effectively ex-plore the possible di-rections of promisingdevelopment. Ourestimate of winningchances critically de-pends upon 1. exploring the promising and risk-mitigating paths and 2. correctly identifyingthose paths whose exploration of future conse-quences can justifiably wait until later. Inaccura-cies in these two areas of classification will createa limit to overall performance, as we strategicallyattempt to compete against other agents withdifferent and refined approaches to this sameproblem. We seek, as a strategy, to gain a sus-tainable edge over our opponent, and see thecareful formation and execution of the strategic

plan as the best and most productive way to ac-complish this.

The proposed heuristic offers promise as acomponent of an orientation/evaluation method-ology for a computer chess program, and shouldbe used to steer a search process (such as for-ward and backward chaining) to effectively re-duce search depth for lines deemed less promis-ing.

The concepts of competitive intelligence,critical success factors, serious play, evolution,wargaming, Boyd’s OODA Loop, Grisogono andRyan’s Adaptive Campaigning Model, and end-point evaluation critically complete the concep-tualization. We seek to ”play” the game of chessthrough a strategic orientation and explorationthat is guided by a playful-but-serious examina-tion of the future consequences of stress trans-formation, the tentative separation of positionsinto categories of uninteresting, not worthy of at-tention, probably sustainable (allowing a halt tofurther explorations) and interesting, worthy ofattention, possibly unsustainable (requiring ad-ditional searching), and vital diagnostic testswhich orient, summarize and simplify the com-plexity present on the game board. We gathercompetitive intelligence to measure our success-ful attainment of critical success factors - oursuccess or failure will serve as our guide to diag-nostic action.

We establish value through an endpoint eval-uation which sums critical parameters in orderto 1) critically perceive the size of the mistakewhich would need to be made to reach a sus-tainable, even game, 2) accumulate small, sus-tainable positional advantages and 3) establisha marker to develop challenge lines with strate-gic potential if or when problems develop withbranches in our principal variation, or with the

53


indicated move itself.Serious play can leverage the accumulated

strategic information and judgment gained overthe years. It can help develop original strategies(Roos and Victor, 1998). Serious play can enableus to explore, challenge, disagree, and come toagreement on how we will meet the future (Roosand Victor, 1998).

The presented results demonstrate the pos-sibilities of the proposed building blocks for fourtest positions. Perhaps chess is more than justcalculation (Aagaard, 2004), but the day maycome sooner than we think when computers useheuristics to play a positional game of chess atskill levels equal to their current strong tacticalplay. Correspondence chess would provide theideal testing ground for a positional heuristic.

We might borrow the words of economistJoseph Schumpeter (1883-1950) and theorizethat chess is a game of Creative Destruction.

Future work will involve the construction ofa prototype software application which imple-ments the concepts discussed in this paper.

We close with a quote from the Marine CorpsOperating Concepts - Third Edition (2010):

Old ideas can take on an entirely newlife when placed with a new context -and if there is one constant reflectedin our view of the future, it is thatthere is no longer a single contextbut many... Whether the ideas inthese pages are proven or disprovenis not the point - the act of thought-ful engagement in response to themis what matters. As steel sharpenssteel, ideas can - and should - do thesame. -G.J. Flynn, Lieutenant Gen-eral, U.S. Marine Corps

Note: colored diagrams were produced by a com-puter program in HTML format and rendered in a Fire-fox web browser in a method similar to that used by thesoftware program ChessDiagrams by Ambar Chatterjee.

Special thanks to all my friends at chessgames.com,

through whom I continue to learn about chess.

25 Appendix A: SelectiveSearch and Simulation

We recognize that the concept of selective searchis a critical concept in playing the game of chess,but we suggest an alternative way of thinkingabout the method where we choose to explorecertain future lines, and choose not to exploreothers.

The phrase dynamic simulation with strate-gic scenarios has certain advantages. First, werecognize that we are using a dynamic model.Second, the concept of a simulation permits usto think about the uncertain future where we en-counter resistance. Third, we are strategic in ourselection and rejection of lines. Last, we followcertain scenarios in our war gaming of the future- this allows us to learn what might lie ahead.

We suggest that this approach is more pre-cise and allows us to answer the question ”So,how are you doing selective search?” with theanswer, ”What we are doing is more than justsearching - it is more like conducting a compli-cated diagnostic test of how ’ready’ we are forthe uncertain future. As part of a strategy, weexplore the critical emerging results of stress in-teractions. We construct a dynamic model andcreate strategic scenarios. The process resemblesbiological evolution, as we first propose moveswhich satisfy orientors aimed at sustainability,and then dispose of lines judged by our compet-

54


itive intelligence to be not worth our attention.We aim for resilience, adaptive control and flexi-ble persistence in the face of complexity and theuncertain plans of our opponent. We developand expand scenarios which have strategic po-tential to become a replacement principal varia-tion (or replacement branches) when unexpecteddiscoveries are made. Through endpoint evalua-tion we establish a marker which is used to setthe threshold of our attention when constructingchallenge lines. We aim to create results nor-mally produced by a productive thinking pro-cess.”

26 Appendix B: The Impor-tance of Sustainability

We have placed much emphasis on the concept ofsustainability, and feel the need to explain whythis concept is such a critical strategy when play-ing a game in the positional style.

Whatever diagnostic test we use for explor-ing the future cannot prepare us for all possibil-ities. We instead must be ”ready” for whateveremerges from the ”mess” of interactions, someof which are foreseeable and are representativeof the types of situations we will later face. Sus-tainability allows us to continuously generate re-sponses - future positions which in turn are like-wise sustainable. Anyone who has played com-petitive sports learns to develop quick tests todetermine, on the fly, whether the current teamposition is sustainable, and if not, what needsto be done (personally, or calling instructions toothers) to correct it.

When facing a tactically empty position, wefeel that a strategy that develops, then selec-tively expands a portfolio of likely scenarios is

a good way to determine how ready we are toface an uncertain future. We prepare ourselvesto respond to the mistake of our opponent, orfor the situation where a scenario initially judgedto be not worth our attention, had unexpectedside effects which resulted a more favorable po-sition for our opponent. We seek to uncover un-expected situations ”down the road” which im-pact the ”health” and sustainability of the po-sition and cause us to shift our move played toone with a more favorable outlook. We are nowready in general, and will handle the specifics asthey come.

In short, we feel that the nature of the com-plexity which exists on the game board, of dy-namic and evolving systems in general and of’conflict ecosystems’ and the peculiarities of sys-tems in particular, must all be reflected in thesearch for general principles of sustainable de-velopment (Bossel, 2007).

27 Appendix C: Related Quo-tations

The analysis of general system principles shows thatmany concepts which have often been considered asanthropomorphic, metaphysical, or vitalistic are ac-cessible to exact formulation. They are consequencesof the definition of systems or of certain system con-ditions. - Ludwig von Bertalanffy

a good model enables prediction of the futurecourse of a dynamic system. - Bruce Hannon andMatthias Ruth

Perception, motivation, and values combine tocreate choice. - Joe Vitale

It’s your decisions about what to focus on, whatthings mean to you, and what you’re going to do aboutthem that will determine your ultimate destiny. - An-thony Robbins

55


We are successful because we use the right levelof abstraction. - Avi Wigderson

We can influence the future but not see it. -Stewart Brand

The mind will not focus until it has clear objec-tives. But the purpose of goals is to focus your at-tention and give you direction, not to identify a finaldestination. - John C. Maxwell

Of all the factors that contribute to adapting tochange, the single most important factor is the degreeto which individuals demonstrate resilience - the ca-pacity to absorb high levels of change and maintaintheir levels of performance. - Mark Kelly and LindaHoopes

Every piece of business strategy acquires its truesignificance only against the background of that pro-cess and within the situation created by it. It mustbe seen in its role in the perennial gale of creativedestruction; it cannot be understood irrespective of itor, in fact, on the hypothesis that there is a perenniallull. - Joseph Schumpeter

It is not the strongest of the species that survive,not the most intelligent, but the one most responsiveto change. - Charles Darwin

Resilience or some variation of this idea is aconcept that is explicitly if not tacitly implicit in al-most all explanatory models of behavior ranging fromthe biological to the social. It may be an inextricablepart of the ways in which we define and explain notonly human behavior but virtually all phenomena withvariable outcomes. - Meyer Glantz and Zili Sloboda

any approach able to deal with the changing com-plexity of real life will have to be flexible... It needsto be flexible enough to cope with the fact that ev-ery situation involving human beings is unique. Thehuman world is one in which nothing ever happenstwice, not in exactly the same way. This means thatan approach to problematical human situations hasto be a methodology rather than a method, or tech-nique... [Soft Systems Methodology] provides a set ofprinciples which can be both adopted and adapted foruse in any real situation in which people are intenton taking action to improve it. - Peter Checkland

and John Poulter

I think that resilience is manifest competence de-spite exposure to significant stressors. It seems to methat you can’t talk about resilience in the absence ofstress. The point I would make about stress is the crit-ical significance of cumulative stressors. I think thisis the most important element. - Norman Garmezy

No plan survives contact with the enemy. - FieldMarshal Helmuth von Moltke

In many ways, coping is like breathing, an auto-matic process requiring no apparent effort... Is cop-ing always a conscious process? ...we so often mayrepeatedly respond to a recurring stressor that we loseour awareness of doing so. - Charles Richard Snyder

What business strategy is all about; what dis-tinguishes it from all other kinds of business plan-ning - is, in a word, competitive advantage. Withoutcompetitors there would be no need for strategy, forthe sole purpose of strategic planning is to enable thecompany to gain, as effectively as possible, a sustain-able edge over its competitors - Keniche Ohnae

Rykiel (1996) defines model credibility as ”a suf-ficient degree of belief in the validity of a model tojustify its use for research and decision-making.”...there is no use talking about some overall universalmodel validity; the model is valid only with respect tothe goals that it is pursuing - Alexey Voinov

A principal deficiency in our mental models isour tendency to think of cause and effect as localand immediate. But in dynamically complex systems,cause and effect are distant in time and space. Mostof the unintended effects of decisions leading to pol-icy resistance involve feedbacks with long delays, farremoved from the point of decision or the problemsymptom. - John Sterman

everything in nature, everything in the universe,is composed of networks of two elements, or two partsin functional relationship to each other... The mostfundamental phenomenon in the universe is relation-ship. - Jonas Salk, Anatomy of Reality

What is the core of the matter? Why should amachine not be an excellent chess player? Is the taskinsoluble in principle? ... No. The problem seems to

56


be soluble... The machine may play chess badly, likea beginning amateur, but the machine is not guilty.Man is guilty. He has not yet succeeded in teach-ing the machine, in transferring his experience to it.What is involved in teaching a machine to play chess?- Mikhail Botvinnik

once you become aware of what means the mostto you, you’re less likely to put off something that’sreally valuable for something that matters much less...it’s knowing the difference between what’s importantand what isn’t that allows us to solve problems effec-tively. - Joy Browne

Intelligence is the ability to acquire knowledge,and not the knowledge itself. - George F. Luger

Where sustainability is not even a goal, it is un-likely that sustainability will be achieved by accident.And even if it is a declared goal, sustainability can-not be achieved where money, time, resources, and thecreative energies of individuals are wasted. - HartmutBossel

While a self-organizing system’s openness to newforms and new environments might seem to make ittoo fluid, spineless, and hard to define, this is not thecase. Though flexible, a self-organizing structure isno mere passive reactor to external fluctuations. Asit matures and stabilizes, it becomes more efficient inthe use of its resources and better able to exist withinits environment. It establishes a basic structure thatsupports the development of the system. This struc-ture then facilitates an insulation from the environ-ment that protects the system from constant, reactivechanges. - Margaret Wheatley, Leadership and theNew Science

If system behavior is guided by balanced referenceto basic orientors it will have the best chance for suc-cess in the long run... Systems which have evolvedunder evolutionary forces to be sustainable... can beviewed as having been designed in a way to achievebalanced satisfaction of basic orientors... To be ef-ficient and effective, path analysis, policy synthesis,and system design for sustainable development haveto take the orientor satisfaction of affected systemsinto account. - Hartmut Bossel

It should be obvious that in your workplace thereare some things you can control and some things thatyou can’t. The trick is being able to identify thosethings you can control and then to get busy control-ling them... My goal is to make you see that you havemore control over things than you think you do... Youcan regain a sense of control if you start to focus onissues where you can make a difference and stop wast-ing time on those where you can’t. - Karl Schoemer

Those who have to make the decisions should alsobe those who create the scenarios... We also recog-nize... that issues of power and influence are centralin determining how situations will unfold... power isa key determinant of... organizational... thinking...The key aim in writing scenarios is to grab the at-tention of the intended audience in order to conveyclear, concise and plausible stories about what typesof futures might unfold as a direct outcome of deci-sions made in the present and over time in relation tothe focal issue. - George Wright and George Cairns

Annotated bibliography:

http://mysite.verizon.net/vzesz4a6/current/sitemap.html

References

Aagaard, J. (2003). Excelling at Positional Chess. pp.35-36. Glouchester Publishers plc, London.

Aagaard, J. (2004). Inside the Chess Mind: How Playersof All Levels Think About the Game. p.157. EverymanChess, London.

Albrecht, K. (2007). Practical Intelligence: The Art andScience of Common Sense. p.325. Jossey-Bass, SanFrancisco, CA.

Albus, J., Meystel, A. (2001). Engineering of Mind. p.80.John Wiley and Sons, New York.

Aldwin, C. (2007). Stress, Coping, and Development, Sec-ond Edition. pp.2,28,37. The Guilford Press, New York.

57




Alexander, B. (2002). How Great Generals Win. pp.33-34,97-98,104. W.W. Norton and Company, New York.

Allison, G., Zelikow, P. (1999). Essence of Decision:Explaining the Cuban Missile Crisis, 2nd Edition.pp.8,52-53. Longman, New York.

Anderson, V., Johnson, L. (1997). Systems Thinking Ba-sics: From Concepts to Causal Loops. pp.17,19,57,127.Pegasus Communications, Waltham, MA.

Apgar, D. (2008). Relevance. p.26. John Wiley and Sons,Inc., San Francisco, CA.

Beckham, J.D. (1998). Organic Strategy.www.beckhamco.com/41articlescategory.html ac-cessed July 2011.

Beckham, J.D. (2000). Strategy: WhatIt Is, How It Works, Why It Fails.www.beckhamco.com/41articlescategory.html ac-cessed July 2011.

Beckham, J.D. (2001). Managing Complexity.www.beckhamco.com/40articlescategory.html ac-cessed August 2011.

Beckham, J.D. (2002). Planning for the Unknown.www.beckhamco.com/29articles.html accessed July2011.

Beckham, J.D. (2006). The Power of a StrategicPlan. www.beckhamco.com/27articles.html accessedJuly 2011.

Beckham, J.D. (2007). Data That Predict.www.beckhamco.com/27articles.html accessedJuly 2011.

Beinhocker, E. (2007). The Origin of Wealth: Evolu-tion, Complexity, and the Radical Remaking of Eco-nomics. pp.14,141,187,378. Harvard Business SchoolPress, Boston, MA.

Bell, S., Morse, S. (2008). Sustainability Indicators: Mea-suring the Immeasurable?. p.186. Earthscan, London.

Benyus, J. (2002). Biomimicry. pp.253,273. Harper Peren-nial, New York.

Bertalanffy, L. (1968). General System Theory. p.46.George Braziller, New York.

Bisogno, P. (1981). Scientific Research and Basic Needs.In: Research and Human Needs, A. Forti, P. Bisogno.pp.11,12,13-14. Pergamon Press, New York.

Blalock, H. (1982). Conceptualization and Measurementin the Social Sciences. pp.27-28. Sage Publications,Beverly Hills, California.

Blanchard, B., Fabrycky, W. (2006). Systems Engineer-ing and Analysis, Fourth Edition. pp.162-163. PrenticeHall, Upper Saddle River, New Jersey.

Bossel, H. (1976). Information Processing, Cognitive Dis-sonance, and Basic Needs: The Modelling of Behav-ior. In: Bossel, Klaczko, Muller, 1976, Systems Theoryin the Social Sciences. pp.423-472. Birkhauser VerlagBasel, Stuttgart, Germany.

Bossel, H. (1977). Orientors of Nonroutine Behav-ior. In: Bossel, H. (1977). Concepts and Tools ofComputer-assisted Policy Analysis: Vol. 1 Basic Con-cepts. pp.227,230,232-233,237,240,242. Birkhauser Ver-lag Basel, Stuttgart, Germany.

Bossel, H. (1994). Modeling and Simulation. pp.91,229-242,247. A K Peters, Ltd., United States of America.

Bossel, H. (1998). Earth at a Crossroads: Paths to a Sus-tainable Future. pp.7-8,17-18,20,70-108,283,284. Cam-bridge University Press, United Kingdom.

Bossel, H. (1999). Indicators for Sustainable De-velopment: Theory, Method, Applications. pp.44-45,57,64,73,84. International Institute for SustainableDevelopment, Canada.

Bossel, H. (2007). Systems and Models. pp.160,174,178-208,210,243. Books on Demand GmbH, Norderstedt,Germany.

Botvinnik, M. (1970). Computers, Chess, and Long-Range Planning. pp.2-3,5,7,9,10,13,14,22,23,24,26,63.Longman Group Ltd., London.

Botvinnik, M. (1984). Computers in Chess: SolvingInexact Search Problems. pp.16,38,39,64,67,105-106.Springer-Verlag, New York.

Boyd, J. (1976). Destruction and Creation, 3 Septem-ber 1976. http://www.goalsys.com/books/documents/DESTRUCTION AND CREATION.pdf accessed Jan-uary 2011.

58

http://www.goalsys.com/books/documents/DESTRUCTION_AND_CREATION.pdf

http://www.goalsys.com/books/documents/DESTRUCTION_AND_CREATION.pdf


Boyd, J. (1987). The Strategic Game of ? and?. pp.38,45,47. http://www.dnipogo.org/boyd/pdf/strategy.pdf accessed January 2011.

Boyd, J. (1992). Conceptual Spiral. p.28. http:// steven-shack.com/johnboyd/assets/discourse abstract.pdf ac-cessed January 2011.

Boyd, J. (1996). The Essence of Winning andLosing. p.3. http://stevenshack.com/johnboyd/assets/essence of winning losing.pdf accessed January 2011.

Boyd, J. (2005). Organic Design for Command and Con-trol. p.16. http://stevenshack.com/johnboyd/assets/organic design.pdf accessed January 2011.

Bradford, R., Duncan, P., Tarcy, B. (2000). Simpli-fied Strategic Planning. pp.10,24,51,71,92,136. Chan-dler House Press, Worchester, MA.

Brown, S. (2009). Play: How It Shapes the Brain, Opensthe Imagination, and Invigorates the Soul. pp.5,11-12,84-85,141,154,218. Penguin, New York.

Browne, J. (2002). Getting Unstuck: 8 Simple Stepsto Solving Any Problem. pp.20,81. Hay House, Inc.,Carlsbad, CA.

Byers, W. (2011). Blind Spot: Science and the Crisisof Uncertainty. pp.3,157,176-178. Princeton UniversityPress, Princeton, NJ.

Calhoun, R., Hayward, B. (2010). Stabilising ComplexAdaptive Systems. p.139. Australian Army Journal,Volume VII, Number 3.

Canonico, P. (2004). An Alternate Military Strategy forthe War on Terrorism. pp.1,64. Naval PostgraduateSchool, Monterey, CA.

Capablanca, J.R. (2002). A Primer of Chess. pp.115-117,120,139. Harcourt, New York.

Capra, F. (1988). The Role of Physics in the CurrentChange in Paradigms. In: Kitchener, R., The WorldView of Contemporary Physics: Does it Need a NewMetaphysics? p.146-152. State University of New YorkPress, Albany, New York.

Castex, R., Kiesling, E. (1994). Strategic Theories. p.21.Naval Institute Press, Annapolis, Maryland.

Chapin, S., Kofinas, G., Folke, C., (2009). Principles ofEcosystem Stewardship: Resilience-Based Natural Re-source Management in a Changing World. pp.20,22.Springer, New York.

Checkland, P., Poulter, J. (2006). Learning For Action:A Short Definitive Account of Soft Systems Methodol-ogy, and its use Practitioners, Teachers and Students.pp.9,10-11,15,20-21,168. John Wiley and Sons, Chich-ester, England.

Cohn, M. (2006). Agile Estimating and Planning. p.27.Pearson Education Inc., NJ.

Costanza, R., Jorgensen, S. (2002). Understanding andSolving Environmental Problems. p.58. Elsevier Ltd.,San Diego, CA.

Coutu, D. (2003). How Resilience Works. In: HarvardBusiness Review on Building Personal and Organi-zational Resilience. pp.1-18. Harvard Business SchoolPublishing Corporation.

Dechter, R. (2003). Constraint Processing. pp.v-vi,xvii.Morgan Kaufmann Publishers, San Francisco, Califor-nia.

de Geus, A. (2002). The Living Company. pp.64,155. Har-vard Business School Press, MA.

Deleuze, G., Hand, S. (1988). Foucault. p.60. UniversityOf Minnesota Press.

Dettmer, H. (2007). The Logical Thinking Process.pp.100,127-138. Quality Press, Milwaukee, WI.

de Wit, B., Meyer, R. (2010). Strategy: Process, Content,Context. pp.57,516. Cengage Learning, United King-dom.

DiPiazza, S., Eccles, R. (2002). Building Public Trust:The Future of Corporate Reporting. p.84. John Wileyand Sons, New York.

Dombroski, T. (2000). Creative Problem-Solving. p.119.toExcel Press, Lincoln, NE.

Donohew, L., Tipton, L., Haney, R. (1978). Analysis ofinformation-seeking strategies. Journalism Quarterly,55, 25-31.

Doz, Y., Kosonen, M. Fast Strategy. p.29. Pearson Edu-cation Limited, Edinburgh Gate.

59

http://www.dnipogo.org/boyd/pdf/strategy.pdf

http://www.dnipogo.org/boyd/pdf/strategy.pdf

http://stevenshack.com/johnboyd/assets/discourse_abstract.pdf

http://stevenshack.com/johnboyd/assets/discourse_abstract.pdf

http://stevenshack.com/johnboyd/assets/essence_of_winning_losing.pdf

http://stevenshack.com/johnboyd/assets/essence_of_winning_losing.pdf

http://stevenshack.com/johnboyd/assets/organic_design.pdf

http://stevenshack.com/johnboyd/assets/organic_design.pdf


Dvoretsky, M., Yusupov, A. (1996). Positional Play.pp.111,121. Henry Holt and Company, New York.

Elias, M., Parker, S., Rosenblatt, J. (2006). Building Ed-ucational Opportunity. In: Handbook of Resilience inChildren, Goldstein, Brooks. pp.317-318. Springer Sci-ence + Business Media, New York.

Emms, J. (2001). Simple Chess. pp.7,63,67,70. EverymanPublishers plc, London.

Fagre, D., Charles, C. (2009). Thresholds of ClimateChange in Ecosystems, Synthesis and AssessmentProduct 4.2. US Geological Survey.

Festinger, L. (1957). A Theory of Cognitive Dissonance.pp.3,4,5,18,24,55. Stanford University Press, Stanford,California.

Fine, R., Benko, P. (2003). Basic Chess Endings.pp.184,275. Random House, United States of America.

Fleisher, C., Bensoussan, B. (2007). Business and Com-petitive Analysis: Effective Application of New andClassic Methods. pp.44,378. FT Press, Upper SaddleRiver, NJ.

Fogel, A., Lyra, M., Valsinger, J. (1997). Dynamics andIndeterminism in Developmental and Social Processes.pp.3-4,32. Lawrence Erlbaum Associates, New Jersey.

Folke, C., Carpenter, C., Elmqvist, T., Gunderson, L.,Holling, CS, Walker, B., et al. (2002). Resilience andSustainable Development: Building Adaptive Capac-ity in a World of Transformations. pp.4,15-16,20,20-23.The Environmental Advisory Council to the SwedishGovernment.

Fontana, D. (1989). Managing Stress. pp.3,4,26-27. TheBritish Psychological Society, New York.

Foucault, M. (1982). Afterward: The subject and power.In H. L. Dreyfus and P. Rainbow (Eds.), MichelFoucault: Beyond structuralism and hermeneutics(pp.229-252). Chicago: University of Chicago Press.

Foucault, M. (1995). Discipline and Punish. pp.26-27.Random House, United States of America.

Friedl, K. (2007). What is Behind the Fatigue Concept?In: Operational Fatigue: RTO Meeting ProceedingsMP-HFM-151. North Atlantic Treaty Organisation,Research and Technology Organisation.

Fritz, R. (1989). The Path of Least Resistance: Learningto Become the Creative Force in Your Own Life. pp.5,6-7,8,9,11,12,31,38-39,49,51-54,116-118,133-137. FawcettColumbine.

Fritz, R. (2003). Your Life As Art. p.130. Newfane Press,Vermont.

Fritz, R. (2007). Elements. pp.81-82. Newfane Press, Ver-mont.

Fuld, L. (1995). The New Competitor Intelligence.pp.23,24,25,422. John Wiley and Sons, United Statesof America.

Fuld, L. (2010). The Secret Language of Competitive In-telligence. pp.143-144,272,281. Dog Ear Publishing.

Gadiesh, O., Gilbert, J. (2001). Transforming Corner-Office Strategy into Frontline Action. In: HBR’s 10Must Reads on Strategy, Harvard Business Review,2011, pp.191-208. Harvard Business Review, May 2001.

Gharajedaghi, J. (2006). Systems Thinking:Managing Chaos and Complexity. pp.13,15,32,33,46,47,50,107,108,110,112,126,137,140,141-142,315. Butterworth-Heinemann, Burlington, MA.

Gilad, B. (1994). Business Blindspots. pp.113,123-124.Probus Publishing Company, Chicago, IL.

Gilad, B., Gilad, T. (1988). The Business Intelligence Sys-tem: A New Tool for Competitive Advantage. pp.1,40.American Management Association.

Gilad, B., Herring, J. (1996). The Art and Science of Busi-ness Intelligence Analysis, Part B: Intelligence Analysisand its Applications. JAI Press Inc., Greenwich, Con-necticut.

Gilad, B. (2009). Business War Games. pp.16-19,97,184.The Career Press, Franklin Lakes, NJ.

Glantz, M., Johnson, J. (1999). Resilience and De-velopment: Positive Life Adaptations. pp.5-8,10-11,13,14,110,120,122. Kluwer Academic/Plenum Pub-lishers, New York.

Goldratt, E., Cox, J. (2004). The Goal. p.307. North RiverPress, Great Barrington, MA.

60


Gould, S. (1996). Creating the Creators. In: Discover, Oc-tober 1996. http://discovermagazine.com/1996/oct/creatingthecreat888 accessed April 2011.

Gould, S. (2002). The Structure of Evolutionary Theory.p.695. Harvard College.

Govindarajan, V., Trimble, C. (2005). 10 Rules for Strate-gic Innovators. p.89. Harvard Business School Publish-ing, MA.

Greene, R. (1966). Business Intelligence and Espionage.p.73. Dow Jones-Irwin, Homewood, IL.

Grisogono, A., Ryan, A. (2007). OperationalisingAdaptive Campaigning. pp.1-6. 2007 CCRTS. Aus-tralian Department of Defence, Defence Scienceand Technology Organisation, Edinburgh, Australia.http://www.dodccrp.org/events/12th ICCRTS/CD/html/papers/198.pdf

Gunderson, L., Allen, C., Holling, C. (2010). Founda-tions of Ecological Resilience. pp.46-47,303,339. IslandPress, United States of America.

Hahlweg, K., Hooker, C. (1989). Issues in EvolutionaryEpistemology. pp.70-71. State University of New York,Albany, New York.

Haines, S. (1998). The Managers Pocket Guide to Sys-tems Thinking and Learning. pp.31-32. HRD Press,Amherst, MA.

Hamel, G., Valikangas, L. (2003). The Quest For Re-silience. In: Harvard Business Review, September,2003. Harvard Business School Publishing Corpora-tion.

Hammer, M., Stanton, S. (1995). The Reengineering Rev-olution. pp.4,157,274. HarperCollins Publishers, Inc.,New York.

Hammond, G. (2001). The Mind of War: John Boyd andAmerican Security. p.191. Smithsonian Books, Wash-ington, D.C.

Hart, B.H. Liddell. (1991). Strategy. p.384. Meridian,New York.

Heisman, D. (1999). Elements of Positional Evaluation,Third Edition. pp.38,40,136. Chess Enterprises, MoonTownship, Pennsylvania.

Henricks, T. (2006). Play Reconsidered: Sociological Per-spectives on Human Expression. pp.12,95,130. Univer-sity of Urbana Press, Chicago, Illinois.

Herman, M., Frost, M., Kurz, R. (2009). Wargaming ForLeaders. pp.3,7,261. Booz Allen Hamilton, Inc., UnitedStates of America.

Hollnagel, E., Woods, D., Leveson, N. (2006). ResilienceEngineering. pp.71,303,314. Ashgate Publishing Com-pany, Burlington, Vermont.

Hollnagel, E., Nemeth, C., Dekker, S. (2008). ResilienceEngineering Perspectives, Volume 1: Remaining Sensi-tive to the Possibility of Failure. pp.5-6. Ashgate Pub-lishing Company, Burlington, Vermont.

Honeycutt, J., Cantrill, J. (2001). Cognition, Com-munication, and Romantic Relationships. pp.20-23.Lawrence Erlbaum Associates, New Jersey.

Hooper, T., Scott, L. (1996). Bulls Eye! Taking Aim WithTargeted Competitive Assessments. In: The Art andScience of Business Intelligence Analysis, Part B, Gi-lad, B., Herring, J. p.76. JAI Press Inc., Greenwich,Connecticut.

Howard, M. (1974). Military Science in the Age of Peace.In: RUSI Journal, March 1974. p.7.

Hubbard, D. (2007). How to Measure Anything.pp.23,39,43,115. John Wiley and Sons, Hoboken, NewJersey.

Hubert, R. (2007). Transforming Sustainabil-ity: Identifying the Critical Success Fac-tors for Sustainable Cities. pp.21,22,29.http://www.environment.nau.edu/research/Sustainabile%20Cities%20Project%20Report%20-%20Synopsis-1.pdf accessed November 2011.

Hurson, T. (2008). Think Better. pp.45,140,218. McGrawHill, New York.

Jackson, M. (2003). Systems Thinking: Creative Holismfor Managers. pp.131-132. John Wiley and Sons,Chichester, UK.

Jerz, J. (2007). The World vs. Arno Nickel 2006.http://mysite.verizon.net/vzesz4a6/current/id30.html accessed January 2011.

61

http://discovermagazine.com/1996/oct/creatingthecreat888

http://discovermagazine.com/1996/oct/creatingthecreat888

http://www.dodccrp.org/events/12th_ICCRTS/CD/html/papers/198.pdf

http://www.dodccrp.org/events/12th_ICCRTS/CD/html/papers/198.pdf

http://www.environment.nau.edu/research/Sustainabile%20Cities%20Project%20Report%20-%20Synopsis-1.pdf



http://mysite.verizon.net/vzesz4a6/current/id30.html

http://mysite.verizon.net/vzesz4a6/current/id30.html


Jorgensen, S. Muller, F. (2000). Handbook of EcosystemTheories and Management. pp.8,41,45. Lewis Publish-ers, New York.

Jorgensen, S., Fath, B. (2007). A New Ecology: Sys-tems Perspective. pp.3,4,51,79,84,155,199. Elsevier,New York.

Jorgensen, S. (2009). Ecosystem Ecology. pp.126-127. El-sevier, Oxford, UK.

Kahaner, L. (1997). Competitive Intelligence. pp.48-51.Touchstone, New York, NY.

Kahneman, D., Slovic, P., Tversky, A. (1982). Judgmentunder Uncertainty: Heuristics and Biases. p.69. Cam-bridge University Press, New York.

Kaplan, J. (1978). The ABCs of Chess. In: Hochberg, B.,Chess Life and Review, July. p.370. New Windsor, NY.

Kaplan, R., Norton, D. (2004). Measuring the StrategicReadiness of Intangible Assets. In: Harvard BusinessReview, February, 2004. Harvard Business School Pub-lishing Corporation.

Katsenelinboigen, A. (1992). Indeterministic Economics.pp.15,50,52-53,62,66,67,70,71,72-73,132,138,252-253.Praeger Publishers, New York.

Kelly, M., Hoopes, L. (2004). Managing Change withPersonal Resilience. pp.20-22,30,96. Mark Kelly Books,USA.

Kelly, J., Brennan, M. (2009). OODA Versus ASDA:Metaphors at War. p.47. Australian Army Journal,Volume VI Number 3, 2009.

Kelly, J., Brennan, M. (2010). Looking for the HedgehogIdea. pp.51-53 Australian Army Journal, Volume VIINumber 1, 2010.

Kilcullen, D. (2006). Three Pillars of Counterinsurgency.Remarks delivered at the U.S. Government Counterin-surgency Conference, Washington D.C., 28 September2006. http://www.au.af.mil/au/awc/awcgate/uscoin/3pillars of counterinsurgency.pdf accessed January2011.

Koen, B.V. (2003). Discussion of the Method. p.28. Ox-ford University Press, New York.

Kossiakoff, A., Sweet, W. (2003). Systems Engineering:Principles and Practice. pp.27,32,57,58,145,419. JohnWiley and Sons, New Jersey.

Lakein, A. (1974). How to Get Control of Your Time andYour Life. p.25. Signet, New York.

Laszlo, E. (1996). The Systems View of the World.pp.9,35,36,88. Hampton Press, New Jersey.

Lawrence, G. (1997). Indicators For Sustainable Devel-opment In: Dodds, F., The Way Forward: BeyondAgenda 21. pp.179-181,185-187,189. Earthscan Publi-cations Ltd.

Levy, D. (1976). 1975 U.S. Computer Chess Champi-onship. p.66. Computer Science Press, Woodland Hills,California.

Lightfoot, C., Cole, M., Cole, S. (2009). The Developmentof Children. p.295. Worth Publishers, New York.

Lockie, S., Rockloff, S., Helbers, D., Lawrence, K.,Gorospe-Lockie, Mahar. (2005). A Conceptual Frame-work for Selecting and Testing Potential Social andCommunity Health Indicators Linked to Changes inCoastal Resource Management or Condition: a Discus-sion Paper. Centre for Social Science Research, CentralQueensland University, Australia.

Lukey, B., Tepe, V. (2008). Biobehavioral Resilience toStress. pp.xiii,275. CRC Press, Boca Raton, Florida.

Luthar, S. (2003). Resilience and Vulnerability: Adapta-tion in the Context of Childhood Adversities. pp.249-250,258. Cambridge University Press, New York.

Luttwak, E. (2001). Strategy: The Logic of War andPeace. pp.110,127. Belknap Press, Cambridge, MA.

Maddi, S., Khoshaba, D. (2005). Resilience at Work:How to Succeed No Matter What Life Throws at You.pp.195-196. American Management Association, NewYork.

March, J. (1994). A Primer on Decision Making. pp.15,16-17. The Free Press, New York.

Maslow, A. (1987). Motivation and Personality.pp.18,30,32-33,196. Harper and Row, New York.

62

http://www.au.af.mil/au/awc/awcgate/uscoin/3pillars_of_counterinsurgency.pdf

http://www.au.af.mil/au/awc/awcgate/uscoin/3pillars_of_counterinsurgency.pdf


Max-Neef, M. (1991). Human Scale Development: Con-ception, Application and Further Reflections. pp.32-33,53. The Apex Press, New York.

McCarthy, J., Canziani, O., Leary, N., Dokken, D.,White, K. (2001). Climate Change 2001: Impacts,Adaption, and Vulnerability: Contribution of WorkingGroup II to the Third Assessment Report. pp.21,46,89.Cambridge University Press, New York.

McCormick, G. (2005). Wilson cites class notes from Dr.Gordon McCormick’s ”Seminar on Guerrilla Warfare,”Department of Defense Analysis, Naval PostgraduateSchool, October 2005.

McDonald, N. (2006). The Art of Planning in Chess.pp.35,86. Batsford, London.

Meadows, Do., Randers, J., Meadows, De. (2005). Toolsfor the Transition to Sustainability. In: The SystemsThinker, Volume 16, Number 9. pp.7-8. Pegasus Com-munications.

Michalewicz, Z., Fogel, D. (2004). How to Solve It: Mod-ern Heuristics. p.31. Springer, New York.

Mitchell, M. (2009). Complexity: A Guided Tour. p.79.Oxford University Press, New York.

Miller, J., Page, S. (2007). Complex Adaptive Systems:An Introduction to Computational Models of SocialLife. pp.246,254. Princeton University Press.

Muller, F., Leupelt. M. (1998). Eco Targets, Goal Func-tions, and Orientors. pp.4-22,190. Springer-Verlag,Heidelberg.

Murphy, C. (2005). Competitive Intelligence: Gathering,Analysing and Putting it to Work. pp.16,25-26. GowerPublishing Company, Burlington, VT.

Nascar official release. (2009). Best Daytona500 Debuts. http://www.nascar.com/2009/news/features/02/09/best.of.daytona.debuts/ in-dex.html accessed January 2011.

Nickel, A. (2005). http://www.amici.iccf.com/ is-sues/issue 05/issue 05 nickel hydra.html accessedJanuary 2011.

Nijout, H., Nadel, L., Stein, D. (1997). Pattern Forma-tion in the Physical and Biological Sciences. pp.10,291.Westview Press.

Orasanu, J., and Connolly, T. (1993). Reinvention of De-cision Making. In: Klein, G.A., Orasanu, J., Calder-wood, R. and Zsambok, C.E. (eds.) Decision Makingin Action: Models and Methods. pp.3-20. Ablex, Nor-wood, New Jersey.

Oriesek, D., Schwarz, J.O. (2008). Business Wargaming.p.xiii. Ashgate Publishing Company, Burlington, VT.

Page, A. (1996). Providing Effective Early Warning:Business Intelligence as a Strategic Control System. In:The Art and Science of Business Intelligence Analysis,Part B, Gilad, B., Herring, J. pp.6-7,10. JAI Press Inc.,Greenwich, Connecticut.

Palazzo, A., Trentini, A., Hawkins, J., Brailey, M. (2010).Projecting Force: The Australian Army and MaritimeStrategy. pp.20-21. Land Warfare Studies Centre, Can-berra.

Paley, V. (1991). A Child’s Work: The Importance ofFantasy Play. pp.1,7. The University of Chicago Press,Chicago, Illinois.

Pearl, J. (1984). Heuristics. pp.xi,115,131. Addison-Wesley, Reading, MA.

Plessner, H., Betsch, C., Betsch, T. (2008). Intuition inJudgment and Decision Making. pp.120-121. LawrenceErlbaum Associates, New York.

Reivich, R., Shatte, A. (2002). The Resilience Factor.p.59. Broadway Books, New York.

Reshevsky, S. (2002). The Art of Positional Play. pp.206,239. Random House, New York.

Rockart, J.F. (1979). Chief Executives Define Their OwnData Needs. In: Harvard Business Review, March-April. p.85. Harvard Business School Publishing Cor-poration.

Roos, J., Victor, B. (1998). In Search of Original Strate-gies: How About Some Serious Play? In: IMD Per-spectives for Managers, Vol 56, Issue No. 15, December1998.

63

http://www.nascar.com/2009/news/features/02/09/best.of.daytona.debuts/index.html



http://www.amici.iccf.com/issues/issue_05/issue_05_nickel_hydra.html

http://www.amici.iccf.com/issues/issue_05/issue_05_nickel_hydra.html


Rothberg, H., Erickson, G.S., (2005). From Knowl-edge to Intelligence: Creating Competitive Advan-tage in the Next Economy. pp.3,38,44,77. Butterworth-Heinemann, New York.

Sanderson, E. (2009). After the Storms, an Island ofCalm- and Resilience. In: The New York Times,September 11, 2009. The New York Times Company,New York.

Savransky, S. (2000). Engineering of Creativity. pp.17-18,165,166. CRC Press, Boca Raton, Florida.

Scheinkopf, L. (1999). Thinking for a Change. pp.2,3,4.St. Lucie Press, Boca Raton, Florida.

Schoemer, K. (2009). Change is Your Competitive Ad-vantage. pp.xxii,14,38,141,142. Adams Business, Avon,MA.

Schumpeter, J. (2008). Capitalism, Socialism and Democ-racy. p.83. Harperperenniel Modernthought, NewYork.

Schwartz, P. (1996). The Art of the Long View. p.37,101-102,107-108,117. Currency Doubleday, New York.

Selye, H. (1978). The Stress of Life. pp.64,407-408.McGraw-Hill, New York.

Senge, P. (1990). The Leader’s New Work: BuildingLearning Organizations. In: Sloan Management Re-view, Fall, 1990, V.32 No. 1. p.15. Massachusetts In-stitute of Technology.

Senge, P. (2006). The Fifth Discipline: The Art and Prac-tice of the Learning Organization. p.68. Doubleday,United States of America.

Shannon, C. (1950). Programming a Computer for Play-ing Chess. In: Philosophical Magazine, Ser.7, Vol. 41,No. 314 - March 1950.

Shoemaker, P. (2007). Forecasting and Scenario Planning:The Challenges of Uncertainty and Complexity. In:Koehler, D., Harvey, N. (eds.) Blackwell Handbook ofJudgment and Decision Making. pp.281,283. BlackwellPublishing, Malden MA.

Simon, H., Newell, A. (1976). Computer Science as Em-pirical Inquiry: Symbols and Search. p.121. Communi-cations of the ACM, March 1976, Volume 19, Number3.

Smith, J., Klein, R., Huq, S. (2003). Climate Change,Adaptive Capacity and Development. pp.61,177,326.Imperial College Press, London.

Snyder, C.R. (2001). Coping with Stress: Effective Peopleand Processes. pp.16,18. Oxford University Press, NewYork.

Spitzer, D. (2007). Transforming Performance Measure-ment. pp.13,42,68-69,79,84-85,209,226. American Man-agement Association, New York.

Starfield, A., Smith, K., Bleloch, A. (1994). How toModel It: Problem Solving for the Computer Age. p.21.Burgess International Group, New York.

Stean, M. (2002). Simple Chess. pp.7,39,42,132. DoverPublications, Mineola, New York.

Steiner, B. (1991). Bernard Brodie and the Foundations ofAmerican Nuclear Strategy. pp.8,207. University Pressof Kansas, Lawrence, Kansas.

Sterman, J. (2000). Business Dynamics: SystemsThinking and Modeling for a Complex World.pp.28,107,133,850,890. McGraw-Hill, United States ofAmerica.

Sutton-Smith, B. (2001). The Ambiguity of Play.pp.21,61,149. Harvard University Press, Cambridge,Massachusetts.

Thomas, J. (2010). Adaptive Campaigning: Is It Adap-tive Enough? In: Australian Army Journal, VolumeVII Number 1

Tierney, K., Bruneau, M. (2007). Conceptualizing andMeasuring Resilience: A Key to Disaster Loss Reduc-tion. In: TR News 250 May-June 2007. p.15. Trans-portation Research Board.

Trout, J. (2008). In Search of the Obvious. p.102. JohnWiley and Sons, NJ.

Ulea, V. (2002). The Concept of Dramatic Genre and TheComedy of A New Type: Chess, Literature, and Film.pp.17-18. Southern Illinois University Press, Carbon-dale and Edwardsville.

Valikangas, L. (2010). The Resilient Organization: HowAdaptive Cultures Thrive Even When Strategy Fails.p.13,14,103,109. McGraw Hill, New York.

64


van der Heijden, K. (2005). Scenarios: The Art of Strate-gic Conversations. pp.36,37,49. John Wiley and Sons,Hoboken, NJ.

van Wezel, W., Jorna, R., Meystel, A. (2006). Planning inIntelligent Systems. pp.23,65,96. John Wiley and Sons,Hoboken, New Jersey.

Verleye, G., Maeseele, P., Stevens, I., Speckhard, A. (pre-publication). Resilience in an Age of Terrorism: Psy-chology, Media and Communication. p.5.

Vickers, G. (1959). The Undirected Society. p.91. Univer-sity of Toronto Press, Canada.

Vickers, G. (1995). The Art of Judgment: A Study ofPolicy Making. pp.47-48,51-52,54,82-83,131. Sage Pub-lications, Thousand Oaks, CA.

Voinov, A. (2008). Systems Science and Modeling forEcological Economics. pp.12,17-18,20,25,29-30,135. El-sevier, New York.

von Bertalanffy, L. (1968). General Systems The-ory: Foundations, Development, Applications.pp.86,88,106,192. George Braziller, Inc., New York.

von Neumann, J., Morgenstern, O. (1953). Theory ofGames and Economic Behavior. pp.36,45. PrincetonUniversity Press, Princeton, NJ.

Walsh, F. (2006). Strengthening Family Resilience.pp.23,85. The Guilford Press, New York.

Warren, K. (2008). Strategic Management Dynamics.pp.124,184,229. John Wiley and Sons, Chichester, Eng-land.

Watts, D. (2011). Everything is Obvious Once You Knowthe Answer. pp.149,180. Crown Business, USA.

Weick, K., Sutcliffe, K. (2007). Managing the Unex-pected. pp.3,32,46. John Wiley and Sons, San Fran-cisco, CA.

West, C. (2001). Competitive Intelligence. p.49. Palgrave,New York, NY.

Wilson, G. (2006). Anatomy of a Successful COIN Opera-tion: OEF-Philippines and The Indirect Approach. In:Military Review, November-December 2006, pp.38-48. http://usacac.army.mil/CAC2/MilitaryReview/Archives/English/MilitaryReview 2008CRII0831art009.pdf

Wood, J. (2009). Communication in our Lives, 5th Edi-tion. pp.32-36. Wadsworth Cengage Learning, Boston,MA.

The Worldwatch Institute. (2009). 2009 State of theWorld: Into a Warming World. pp.152,153-154,168.W.W. Norton and Co., United States of America.

Wylie, J., Wylie, J.C. (1989). Military Strategy: A Gen-eral Theory of Power Control. pp.72,34,97-99,124,151-153. Naval Institute Press.

Zeidner, M., Endler, N. (1996) Handbook of Coping.p.362. John Wiley and Sons, United States of America.

Zeller, R., Carmines, E. (1980). Measurement in the So-cial Sciences. p.2. Cambridge University Press, NewYork.

Znosko-Borovsky, E. (1980). The Middle Game in Chess.

pp.11,12,29,34,47. Dover Publications, Inc., New York.

65

http://usacac.army.mil/CAC2/MilitaryReview/Archives/English/MilitaryReview_2008CRII0831_art009.pdf



a proposed heuristic for a computer chess program...each chess piece 3 moves into the future, for...

Documents