Complexity and Emergence : Chaos and Complexity ( Part Two )


Michael is Research Associate at Henley Management College, U.K, and a Senior Advisor, Tripod Inc. He is also Director, Organization Science Related Programs, at the New England Complex Systems Institute.

Somehow Michael finds time to be Editor-in-Chief of "Emergence: A Journal of Complexity Issues In Organizations and Management".

Michael can be contacted at, and he has his own website at

BioTech and Internet - Complexity Theory In Action 

The two companies I have examined in depth for this paper are both in evolving high tech industries. One, which I will call BioTech, is a division of a many thousand employee industrial company. The other, which I will call Internet, is a young start-up Internet content provider with fewer than forty full-time employees.

Coming from very different stages in their corporate lives, these two companies share a few things in common. First their executives think about complexity on a day-to-day basis and their vocabulary and descriptions of their competitive environments reflect such thinking. Second, I was struck how a senior executive of each company told me a variant on an old J.P. Getty story, which states that success can be achieved if you: are in business for yourself, market products or services that are in demand, guarantee those products and services, give better services than your competitors, reward those who do the work, and build your success on the success of others.

BioTech's products provide cutting edge systems to the life science community with the most comprehensive line of integrated and automated systems for DNA research and analysis, protein and peptide characterization and analysis, carbohydrate analysis, separation and detection, as well as data research, management, and analysis. Application of excellence in microbial biotechnology ( bacteria, yeasts, fungi ), process design, scale-up and operation ( batch, fed-batch, and continuous ), and recombinant organisms is combined with a genuinely entrepreneurial culture and a receptivity towards external collaboration for business development. BioTech believes that it has brought together a scientific team capable of quickly and efficiently advancing its product development programs and identifying new proprietary products, technologies, and market opportunities.

In the biotech industry overall, diversification of revenue line has become the name of the game - - building businesses around such areas as research, manufacturing, contracting, shared development costs, co-marketing and co-promotion, desktop tools for scientists, bio-information systems, and, in some cases, actual product sales. "The challenge of my generation," said the president of one of BioTech's competitors, "is not just to have good technology and products in the pipeline. We also have to keep abreast of the ever-evolving changes in regulatory affairs, intellectual property, company alliances, and market niches from the start. Five or ten years ago, the biotechnology industry was a fraction of the size it is no w. Today, the complexity of the environment and intense competition push us from day one."

Internet, by contrast, is a start-up, a young business of mostly young people trying to make a "great thing" happen in a new medium. Internet provides its members with "tools for life," promoting an ethic of investment by offering six services: finance, career, travel, health, living, and community. Internet is pursuing a marketing strategy that builds upon its niche in the on-line market as a life skills development service aimed at the college market. Internet began as a school project for two of its founders and has developed into a small web community and a larger media company. Internet offers advertisers several venues for reaching people who need to know about their products or services: a web site, a 1,000,000 closed circulation print magazine, and some innovative-sponsored media opportunities. As a content provider ( not access ), Internet is attempting to compete in a value-added sector of a new medium.

As its founder puts it, "Internet will be demand-driven. Internet will be highly responsive to the expressed needs of young people. Ultimately, Internet's success depends on the demand from the students it intends to serve." As a demand driven company it may be greatly changed from the original vision -- "a computer network designed specifically for the interests of college students. This network could be accessed from your personal computer in your dorm room, or at home, or from computers generally available at school. The network would help you acquire important life skills, not currently included in college curricula, in such areas as personal finance, career planning and job search, and preventive health care. Additionally, there would be other services of interest to students, such as travel and research. To acquire these skills, students would participate in interactive simulations and play games. A low annual membership fee would provide unlimited access to all the network's programs and services."

Steve Case, the CEO of America OnLine defined the challenge, "the market is driven now by four components: 1 ) breadth and depth of content--we've got to have things that people see value in, such as major brands; 2 ) packaging and presentation of that content--having an engaging visual appeal, useful in terms of how services are integrated together, giving people the ability to personalize the information and so forth; 3 ) pricing--the pricing has to be simple and affordable and simple is actually as important as affordable; 4 ) providing a sense of community--a subtle component, but an important one."

Both Internet and BioTech have recognized a need to divide their company into two worlds. The old style manufacturing world where cost is king and a new style knowledge-based world where survival itself may be dependent upon new innovation and adaptation.

BioTech actively employs a technique which its executives ( and our discussion group at Santa Fe ) refer to as "flocking." Flocking is the ability of the organization to recognize good opportunities and to flock resources around those opportunities -- much as birds in an unselfconscious and reflexive way will rotate the leadership of their flying wing as one bird gets tired and another represents the "next hope." Having the ability to flock is having the ability to take advantage of opportunity. This is much easier said than done.

At BioTech there is a common pattern -- too many opportunities chasing too few resources. So how do the choices get made ? How is the search conducted over the opportunity frontier ( if I may use that set of words instead of fitness landscape ) ? BioTech has institutionalized two distinct processes to help this along -- first, researchers ( and this is a select group at BioTech ) are free to seek out their own resources among the company components, and second, BioTech has an institutionalized group ( which they label "operations" ) whose function is to satisfy the bureaucratic demands of the organization and keep the bureaucrats away from these select researchers. The freedom to fail has thus been institutionalized.

In some respects this is like an ad agency, where the account manager types shelter the "creatives" from the pressures of such things as budget overruns, and have the responsibility for acting as a liaison between the "numbers" part of the business and the "creatives." I note that the two parts of BioTech do not both speak of complexity -- this is a language used by the searchers ( those who get to decide flocking ) and managers, but not by manufacturing or operations. The ad agency liaison analogy seems apt.

Perhaps most important, BioTech has taken Kauffman's patches theory to heart. There is explicit recognition in the company of the differences between the manufacturing components ( which are run like traditional -- lowest cost margins win -- companies ), thinkers, researchers, and the operating divisions that bind them all together. Each part co-evolves with the others, yet is selfishly doing its own thing. And it seems to work.

At Internet, the processes are similar although the number of personnel involved is much smaller. As the company grew, its leaders recognized the need to patch and subdivide into co-evolving units. Tasks such as content and community development, web site architecture and design, advertising/sponsor relations, public relations, budgeting, and hiring and managing creative and technical staff were split out and made the province of a separate part of the firm. A select set of finders and thinkers are charged with the explicit task of seeking out the generative relationships of which Lane and Maxfield write. Internet's executives speak of opportunity frontiers and competition landscapes. They describe their ideal business partners and potential new services using words almost directly lifted from Lane and Maxfield. And indeed they must. In its brief corporate existence, the company has shifted from a "members who pay for access" model, to an advertising model, to a "we sell related services" model -- evolving as opportunities have presented themselves and as the net itself has evolved.

Even Internet has traditional production type aspects and these have been farmed off to more traditionally managed pieces of the organization. Tasks such as mail lists, marketing promotions, billing, membership updates, server maintenance are not part of the "creative side of the business." The complexity which Internet's officers see is in carving out an identity, an audience, and a revenue stream -- not in becoming the cheapest provider.

The challenge for Internet is managing an ever changing field. Esther Dyson, publisher of Release 1.0 said in a recent issue, "What new kinds of content-based value can be created on the net ? We believe the answers include services ( the transformation of bits rather than bits themselves ), the selection of content, the presence of other people, and assurance of authenticity -- reliable information about sources of bits and their future flows. The trick is to control not the copies but a relationship with the customer -- a subscription. And that's often what the customer wants, because he sees it as an assurance of a continuing supply of reliable, timely content. Much chargeable value will be in certification of authenticity and reliability, not in the content itself. Brand name identity and other marks of value will be important; so will security of supply. Customers will pay for a stream of information/content, from a trusted source . . . who add value -- everything from guarantees of authenticity to software support, selection, filtering, interpretation, and analysis. The redistributor's goal is to be the most convenient source of content and to put its own attitude or personality around the content; the underlying is unlikely to be exclusive, since the content provider wants to maximize distribution."

Brian Arthur refers to this split in tasks as the division between the world of traditional economics -- one defined by cost competition and diminishing returns, where business models are driven to commodity production -- and the knowledge-based world of "increasing returns." Other researchers make note of the network effects implicit in new technologies and suggest that to the consumer sunk costs are not sunk but represent emotional investments which must be accounted for in evaluating next steps. This seems to be what both Case and Dyson refer to when stressing the importance of "community." Internet's executives feel the need for community is a driving force if their firm is to prosper.

Let us go back to McMasters' definitions:

"Complex distinguishes those situations which can be understood by patterns and which can be maintained by paying attention to principles and patterns and leaving the details alone. Chaotic distinguishes those situations where no projections, plans or analysis will make sense and the way forward is to "clear an island of certainty" if that can be done or to make small efforts and observe or to get rid of the situation. Complicated distinguishes those situations which have grown beyond understanding by old patterns, can be continued in some state of operation by routines, but cannot be accessed by mapping. The action to bring complicated back to "simple" is to discover patterns that return some phenomena to the complex."

The key to McMasters' thought is the role of patterns -- patterns in information, patterns in growth, patterns in successful processes. Clearly the tools Kauffman suggests, patches and other forms of simulated annealing, can be thought of in pattern terms. The issue for managers is to get to that level of thought. Umanath and Vessey write of "cognitive fit" -- i.e., the task of getting the data presented in such a way that the brain will accept it most comfortably. Goleman writes that every thought "comes with a value already tacked onto it by the brain's processing system . . . and the evaluation added in the first microsecond of cognition stands if no further thought is given." Presenting issues in complexity terms can fall on deaf ears if there is no fit, and worse, complexity issues can be tagged with negative values which then must be overcome.

The vocabulary which both BioTech and Internet use is a vocabulary informed by the metaphors of complexity. By using that vocabulary at both the level of the opportunity seekers and higher management, there is created a cognitive fit which is important as ideas are presented and decisions are made regarding resource flocking and allocation. What both companies demonstrate most clearly is that if employees and management can think about such concepts as fitness landscapes and opportunity frontiers and local optima, they can strategize along similar lines. Highhouse and Hause's research would suggest the converse as well -- if there is no vocabulary used by management to perceive these metaphors, then the complexity issues will fall into the realm of "missing information" and be heavily and negatively discounted.

Cohen and Levinthal suggest that firms have differing capacities to absorb new information and to utilize the same. "Organizations with high levels of absorptive capacity will tend to be proactive, exploiting opportunities present in the environment. Conversely, organizations with modest absorptive capacity will tend to be reactive, looking for solutions to problems as they arise . . . A firm's absorptive capacity is not simply the sum of the absorptive capacity of its employees. It depends on the interface with the external environment, as well as the transfer of knowledge across and within sub-units."

Just as Biotech allows select employees to poke and prod ( with a goal of finding the right project to then be "flocked around" ), and Internet encourages a few employees to engage in net wandering, mind games, and imagination, Cohen and Levinthal conclude, "A firm's absorptive capacity depends on the individuals who stand at the interface of the firm and the external environment, or at the interface between the subunits within the firm. When the external actors have much different expertise than most individuals in the organization, some members may assume "gatekeeping" or "boundary spanning" roles. A gatekeeper both monitors the environment and translates the information to a form understandable to the research group. Under conditions of rapid and uncertain technical change, a centralized gatekeeper may not clearly see where the knowledge could be best applied. In this case, the firm should expose a broad range of 'receptors' to the environment."

In our discussion groups at Santa Fe we talked for hours about the role of vocabulary and metaphor in informing the very reality which the decision makers see. We outlined a process whereby that vocabulary is instilled into the organization's definition of itself. The process works something like this: Questions are asked, and patterns with suggested vocabulary are created. This is compared to perceived reality by independent review with other people. The patterns and vocabulary get better as they get refined. This is repeated both at levels of greater detail and up the corporate hierarchy. 

The effect of this process is :- 

  • The patterns created in the process become objects in a consensus reality, created by agreement. This "reality" did not exist before the discussion started.
  • The discussion alters the world for the participants, since the perception of the world is partially filtered through the words and metaphors in the new vocabulary.
  • The longer people work at developing a vocabulary to speak of the complexity they observe, the deeper the patterns, the words, and the metaphors are embedded in their minds. The new metaphors become intuitively "true" representations of reality for the people involved.
  • The new vocabulary becomes a shared linguistic domain, not just because of the language, but because the set of shared perceptions created about how the world works.

Thus, at both Biotech and Internet the "searchers" alter management's understanding of the fitness landscape on which competition is taking place by engaging management in a continual dialogue about the models and metaphors used to view the world. As that dialogue unfolds, the understanding changes, and more of the landscape is revealed. Data becomes information only as it is used to inform the dialogue and otherwise is discarded as a necessary component of the search process. The concept of Lane's generative relationships and opportunity mappings are kept to the forefront of the dialogue. Complexity is accepted as a management opportunity rather than disdained as a management constraint.

By accepting the vocabulary and metaphors of complexity, managers at BioTech and Internet have changed the way they view their competitive arena and the way they manage their businesses. Both firms have been careful, however, to also recognize that these metaphors do not apply to the more traditional -- i.e., manufacturing -- components of their organizations.

Overall Conclusion

Research into chaos and complexity has yielded a set of practical applications for business which are far more expansive than critics usually concede. The Economist writes of financial markets and crowd control. Management consultants speak of "learning organizations" and self organized groups. But the real value to-date lies in vocabulary and metaphor. One of my Santa Fe friends likes to quote Wittgenstein to wit, "The limits of my language mean the limits of my world." ( Proposition 5.6 )

Complexity research has developed a descriptive language which can help to shape the world around us. By use of the vocabulary of complexity, managers view the world in a different light. Meanings and metaphors matter. The meanings that we give to ourselves, our products, our competitors, our customers, and all the relevant others in our world determine the space of our possible actions -- and, to a large extent, how we act. Complexity metaphors -- fitness landscapes, simulated annealing, local maxima, patches, generative relationships, increasing returns -- when accepted within the vocabulary of an organization can change both the way mangers manage and the problems they choose to manage. Once these metaphors are accepted, it becomes reasonable to divide the organization by recognizing some pieces as "traditional" businesses and others as "knowledge based" businesses. By applying lowest cost margin approaches only to the traditional businesses, allowing simulated annealing approaches such as patches and the careful cultivation of "noise," and focusing on "search strategies" for the knowledge-based businesses, managers can have their cake and eat it too.

Additional Conclusion for Managers

An Arthur D. Little report ( 6 ) said it best:

"Remember that not all aspects of your operations are repetitive and machine like. Organic processes, such as the development and use of new knowledge can be just as important, yet by their nature, they are less predictable and replicable. Recognize that some changes have uncertain outcomes that may make people uneasy. Keep people moving on the road for change with an inspiring, tangible vision of the desired goal.

Trust your instincts. Organic process change may rely more on intuition and tacit knowledge than on traditional analytic methods. These processes need enablers, owners with a wide view and wide latitude to act. Revise your approach through: an appreciation of diversity, a willingness to experiment ( and safe places to fail ), a sense of community and constant communication and a tolerance for ambiguity."

Learn the metaphors of complexity and then apply what you have learned. 

Additional Conclusion for Complexity Researchers

I quote an expression I overheard in Paris, "It is easier to act our way into a new way of thinking, than it is to think our way into a new way of acting." There is great descriptive value in the research which is being done. It will not become prescriptive until the language is adapted into action. Success breeds success. We must cultivate increasing returns. Each time we use an idea, a language, or a skill we strengthen it, reinforce it, and make it more likely to be used again. Seek out practical applications. Teach decision makers the metaphors of complexity. Confidence builds confidence. When the managers accept the vocabulary, their actions will breed further successes.

We must learn to honor our errors. A trick will only work for a while, until everyone else is doing it. To advance from the ordinary requires a new game, or a new territory. But the process of going outside the conventional method, game, or territory is indistinguishable from error. Even the most brilliant act of human genius, in the final analysis, is an act of trial and error. "To be in error and to be cast out is a part of God's design," wrote William Blake. Error, whether random or deliberate, is an integral part of any process of creation and any process of changing the metaphors and language of our day. 

Exhibit 1


Arthur, B., 1994. Increasing Returns and Path Dependence in the Economy, University of Michigan Press.

Arthur, B., 1996. Increasing Returns and the Two Worlds of Business, draft of February 1996, prepared for the Harvard Business Review.

Cohen, M. et al, 1995. Routines and Other Recurring Action Patterns of Organizations: Contemporary Research Issues, Santa Fe Institute Working Papers, #95-11-101.

Cohen, J. and Stewart, I, 1994. The Collapse of Chaos, Viking Press.

Cohen, W. and. Levinthal, D., 1990. Absorptive capacity: A new perspective on learning and innovation, Administrative Science Quarterly, March, 128-152.

Coveney, P. and Highfield, R., 1995, Frontiers of Complexity, Fawcett Columbine.

Goertzel, B. 1995, Belief Systems as Attractors, in Chaos Theory in Psychology and the Life Sciences, Robertson, R. and Combs, A. ed., Lawrence Erlbaum Associates, 123-134.

Goleman, D., 1995. Brain May Tag All Perceptions With a Value, New York Times, August, 8, 1995, c-1.

Guastello, S., 1995. Chaos, Catastrophe and Human Affairs, Lawrence Erlbaum Associates.

Highhouse, S. and Hause, E. 1995. Missing Information in Selection: An Application of the Einhorn-Hogarth Ambiguity Model, Journal of Applied Psychology, Volume 80, No. 1, 86-93.

Horgan, J., 1995. From Complexity to Perplexity, Scientific American, June, 104-109.

Kauffman, S., and Macready, W. 1995. Technological Evolution and Adaptive Organizations, Complexity, Vol. 1, No. 2, 26-43.

Kauffman, S. 1995. At Home in the Universe, Oxford University Press.

Lakoff, G. and Johnson, M., 1995, Metaphors We Live By, University of Chicago Press.

Lane, D. and Maxfield, R.,1995. Foresight Complexity and Strategy, Santa Fe Institute Working Papers, #95-12-106

Levy, D., 1994. Chaos theory and strategy: Theory, application and managerial implications, Strategic Management Journal, 15 ( Summer ), 167-178.

Little, 1995. Arthur D. Little Inc., Best of the Best in Process Management, a promotional pamphlet.

March, J. and Simon, H., 1993. Organizations, 2nd edition, Blackwell Press

McMasters, M. 1995. Postings to the News Group: Learning Organizations.

Phelan, S., 1995. From Chaos to Complexity in Strategic Planning, paper presented to the Academy of Management 55th Annual Meeting, Vancouver, Canada

Stacey, R., 1991. The chaos frontier, Butterworth-Heinemann.

Timmermans, D., 1993. The Impact of Task Complexity on Information use in Multi-attribute Decision Making, Journal of Behavioral Decision Making, Vol. 6, 95-111.

Torre, C. 1995, Chaos, Creativity, and Innovation: Toward a Dynamical Model of Problem Solving, in Chaos Theory in Psychology and the Life Sciences, Robertson, R. and Combs, A. ed., Lawrence Erlbaum Associates, 123-134.

Tubbs, R. et al, 1993. Order Effects in Belief Updating with Consistent and Inconsistent Evidence, Journal of Behavioral Decision Making, Vol. 6, 257-269.

Umanath N. and Vessey I., 1995, Multi-Attribute Data Presentation and Human Judgment: A Cognitive Fit Perspective, Decision Sciences, Volume 25, Number 5/6, 795-823.

Wilson, D. and Comstock, E., 1995. Evolution of Usability at Digital Equipment Corporation, in Usability in Practice, Wilkund M. ed., American Institutes for Research.

Wittgenstein L., 1922 . Tractatus Logico-Philosophicus, trans. Ogden, C., Routledge & Kegan Paul.

The author thanks Patrick Walshe, Mark Rose, Merrill Bernstein, Erika Kelton, Jeffrey Cole, and Howard Sherman for their insights and suggestions as this project proceeded. Additional thanks are owed to the officers of the companies I have referred to as BioTech and Internet whose identity I have agreed to keep confidential.

Turn to Part 1 of this article

© 1996 Michael R. Lissack

Click here to view the first part of the article.

blog comments powered by Disqus