3 Emergence in emerging markets (the mechanism)
Exploring an interconnected world, as we suggested in earlier chapters, immediately makes the link to biology and the lessons we can learn from biology. Ecosystems, life itself and nature are interesting models of interconnectedness. The interactions that take place between different elements (be it amoebas, cel s, animals in a colony, the process of photosynthesis, etc.) on different levels of existence suggest some ideas to explore in order to improve our understanding of the interconnectedness between people in markets and companies. In this chapter, we consider some biological concepts that challenge the study of values-based leadership, and more broadly for human behavior within the context of groupings.
Does self-organization exist?
The answer to this question opens revolutionary possibilities for business. The relevant research for the answer comes from biology in general and neurobiology in particular. While Maturana and Varela are perhaps the best known representatives for the perceptual innovation known as self-organization or autopoiesis, the concept has been applied by several others in different domains and in different ways.
In their work, the two neurobiologists did not examine the systems from the perspective of genes or species, but from the simplest biological element, the amoeba. For them, the amoeba has a central role in each living being, and so Maturana and Varela studied how co-operation between these amoebas creates (complex) behavior. Each amoeba has individual autonomy at the center of a certain organism, and what appears to happen is that the living system basical y functions in a mechanical fashion. In effect, the total behavior of the system is generated by its elements and their interaction. As a result, observers find themselves entirely outside the system, and therefore, perceive the unit as well as the environment. The elements of a system each react uniquely in interaction with other components. Each declaration of a living system cannot therefore be based on either the idea of the goal, the direction or the final function. In this layout the systems seem to be autopoietic, they are circular, self-productive, self-conservative but also self-referring.
Here we have a few points of reference to see how people col aborate in a company or an organization. Accept for a minute that the purpose of a company is to create value for the market, value that is not yet available in that market. That could be the vision, longer term perspective and focus of the company. However, a company cannot be anything other than a col aboration of a number of individuals who are trying to attain their own individual goals and who employ a certain number of rules of interaction. If that was the case then it is remarkable, for such a system creates its own order and self-maintains in a good state (like our bodies) on the condition that no artificial order is imposed (something we could call an organization). By contrast, scientific management does precisely that. We impose an organization which we will then control if the requisite results are to be attained. This might be a reason for corporate failure.
Questioning organizing and observing theories
At this stage we must develop our understanding of the process by asking key questions: What does autopoiesis need to be true? What conditions must be satisfied to produce this self-production and self-organization? The answers lie in the theory. All perceptions, observations and experiences occur via our body (our senses) and our nervous system. The body then plays the role of the medium of transport. Once in the system, therefore, it is impossible for human beings to have a pure description of anything that is independent of themselves. Each experience is always a reflection of the observer. There is no object outside the field of the observer; instead this observation belongs only to them.
What is therefore true in an autopoietic system and how can we face up to knowledge and truth? What does truth signify? Who supports autopoiesis in maintaining a system in a good state? The survival of the system becomes a key criterion for measuring knowledge and success. Each approach that aims to be scientific can only clearly describe what the observer sees. In effect, the observer plays a crucial role. The comparison with the external world makes no sense. Therefore, the methodology and the manner of leading our investigations are specific. They cannot be detached from the view of the observer. In a company situation, each truth can be just as precious, and just as important, as another. It is not certain that the manager has more reason or a better understanding than someone who is closer to the company process, or to the customer.
Consequences follow from this. Autopoiesis real y says more about the observer than the subject (or should it be object) which is observed. In the case of autopoiesis, at least, that is clearly accepted but what about other scientific paradigms? An ulterior consequence is, of course, that any absolute claim of objectivity cannot be made, by whatever approach. All of this had already been confirmed by Gödel’s theory. Both belief and theory are pure human constructions, which then construct a reality instead of being a reflection of an existing reality. For this reason we sometimes speak of a paradigm of radical constructivism. Remember constructivism? How we make meaning of things to create order from randomness and chaos, in order to make sense. Reality is created and not perceived. Constructivism as research paradigm is gaining increasing support in social science research, but is also subject to a lot more discussion in the classical sciences.
Theory in action: computers, law and linguistic animals
Self-productive ideas have been successful y applied in the construction of self-generating computer applications. The application manipulates itself to be in an optimal situation at all times; in this case we are talking about genetic software. A telephone switchboard, for example, must at any moment of the day deal with a volume of swiftly changing traffic. We can easily imagine making a program which takes into account the multitude of possibilities but, in practice, that appears rather difficult. We can develop a software program which manipulates itself in relation to specific volumes of traffic that the switchboard intercepts. We do not produce a program that resolves the specific problem, but rather a program which uses ideas of self-reproduction.
Without expressing a value judgment, this chapter observes the same processes elsewhere in practice. The legal system, for example, organizes itself in the best manner to assure its survival and the survival of its practitioners, rather than an expressed aim such as enabling justice. For that the system is going to reproduce itself and establish its own frame of reference. In order to make the legal system more efficient, a common language (the law) and a number of procedures are created. The professionals of the legal system – judges, lawyers, etc. who work within it, know the language and procedures. Common understanding and efficiency advances inside the legal system. But the citizens that are only seeking justice (and have no a priori interest in the survival of the system itself), don’t understand the jargon and ceremonies around their – for them – very simple request. To outsiders, the insiders become extraterrestrials. To the insiders the outsiders don’t understand the importance of the survival of the system.
We have already mentioned the idea of self-reference as a strong, but potential y destructive, idea. By analogy with the legal system we can regard each human being (a society, an organization, a meeting, etc.) as an autopoietic system.
People seem to be “linguistic animals” (and that refers to communication and interaction in a network) who do nothing other than play the game that could be called the “practice of artificial living”. Human experience as an observer is not only crucial, but is more important than what real y happens in the world. The role of language and communication is core. All understanding happens through language and its representation but, in addition, all communication with others in a network takes place through language.
The number of misunderstandings in the world, within the same group of languages, is symbolic in relation to this central aspect of interaction. Even in the center of the same group of languages, the Dutch and the Flemish, for instance, use the same words differently. Sentence structure is different which leads to the gathering of the same ideas in a different way. The network of ideas that a speaker tries to transmit is a function of the construction of sentences, even the sequence of sentences. But also, from the listener’s side, how they make sense of the phrases determines the understanding of the message received. Communication is constituted by a network of agents (people) exchanging a network of thoughts with, perhaps, some hopes of being able to learn something, develop knowledge or get something done. In order to reach a real understanding of communication, we should not ignore common intention.
For facilitation of communication we create formal languages (a set of rules). These languages have the intention to standardize and therefore facilitate communication. However, they do not grasp the common intention and by formalizing the exchange of ideas, they sometimes obscure common intention. Children who communicate between themselves in different languages, without speaking those languages, may consequently make far fewer mistakes in the formal language and seem able to conceptualize more rapidly.
It seems evident that the messages children want to transmit are easier than those politicians wish to transmit. Here too, we immediately fall into the positivist trap: if we can measure, then we can know. But it would now seem that communication is not at all organized in the same way as thoughts are.
Is non-verbal communication not equal y efficient and less structured?
Does the extreme order help the process of communication, compared to a higher intensity of communication? If everyone in a Spanish café seems to be talking at the same time, and no-one seems to be listening, is the communication less significant and less precious? Do similar interlocutions lead to not-so-good decisions? Do we always have the intention of rational exchange, or do we use language for emotional connection? And would the latter be less value adding in communication? An example of disordered but effective communication is “open comms” in a live television broadcast which could, to an outsider, appear to be complete chaos. However, this communication is perfectly ordered to the insider where, being able to communicate to everyone from camera operators, sound technicians, commentators, and a host of technical operators, is vital to the success of the production. It is a highly ordered chaos focused on a very definite outcome – delivering the live sports event with most relevant sights and sound to the viewer.
Cognitive connections and artificial intelligence
Contemporary cognitive psychology seems to il ustrate a good number of these ideas. A great deal of research is done around language and interlocution, showing that language and action are tightly linked. Language is our “existence in the world”. Language is real y the entire thinking of humans. Even on the subject of language, we think about language, using language (a good example of self-referencing). Knowledge is not linguistic representation because it is possible to distinguish between different things beyond language. Language is, in fact, a social act. Organizations are therefore networks of recurrent interlocution constructed between individuals and groups of individuals. This thinking has parallels with a certain re-orientation in the developments in artificial intelligence.
The reigning paradigm of objective observation and the associated possibility of drawing up optimum rules, led to the research of “machines based on rules” in expert systems. The assumption is that all decision processes can be captured in rules. While these expert systems have known moderate success, the achievement of the objectivist orientation of artificial intelligence has certainly been over-estimated. Current developments lead in two rather different directions. One goes in the direction of research, which can be regarded as the search for self-learning behavior of systems (emergent behavior), and makes use of connective structures. This development is seen in artificial intelligence in particular. The connective structures are structures where many simple tightly-connected elements and communication make sense out of chaos. In practice, this involves such techniques as neural networks or networks of agents.
The second direction in current research also builds on the self-learning behavior of systems. In this case, it is perhaps better translated as constructing behavior, based on “enacted” technologies. “Enaction” here refers to what Varela cal s “enacted cognition” as an actor “enacts” a theatre play. An amateur asked to play the role of Hamlet might try but would be unlikely to achieve a satisfying theatrical outcome. If an actor is asked to play Hamlet, he does not ‘play’ Hamlet, he becomes Hamlet. Each evening he re-creates another Hamlet. Maybe the two find each other in the character of Shakespeare. A manager can no longer ‘play’ the role of manager. A manager can simply ‘enact’ his role. He ‘is’ his role and it is therefore very difficult to learn this ‘behavior’. You become a manager by experience. You cannot teach someone to become a manager and, as will be argued later, something which is based on skil s and competence cannot be taught anyway.
Developing knowledge: pathways, platforms, communities
For this reason, the personal development path of a manager is so crucial. Only a “learning” manager will be up to playing the role of the spider in the web, the inspirer and creator of good conditions for others. The manager must and should strive for continuously improving himself in this task. Operations or instruments only have a very limited utility and can never play a driving role in dynamic situations. Worse stil , they can therefore never have a general validity.
The vision is the glue that keeps the network together; a kind of weak force that gives basic stability to the network and prevents the network from disintegrating. The absence of this glue or vision will remove that basic centrifugal force. It is this vision, this glue should be values-oriented if we wish to move towards values-based leadership.
Knowledge is only knowledge if it offers a reasonable representation of action and creation. Otherwise it is only possible to speak of information. In this case therefore, the use and the analysis of communication (conversations), as well as a strong focus on the support of communication (by platforms, for example) only gives the context within which the actors create knowledge. To il ustrate: the “communities of practice” experiment carried out by numerous companies, seem to conform well to this preoccupation. That experiment was not uniquely led by artificial intelligence.
Nevertheless, the interest in dynamic re-creation in two directions appears in artificial intelligence research. Things are not fixed, but are produced afresh each time. If someone is asked their age, the information is not stored somewhere particular in the brain. Each time the question is asked, the person is going to produce the reply again. This seems inefficient from the point of view of a recurrent question, but it is, in fact, the case. On the other hand, a question is rarely truly repetitive, even if it is asked using the same words, since in the majority of cases it is looking for another signification. The intonation, for example, is often a clue to the question and the expected answer.
This dynamic reproduction as an approach therefore leaves open the possibility to reply very quickly (and differently) to similar or slightly changed questions. For an authentic, real-time conversation this principle is therefore a lot better and more efficient. Furthermore, if the brain would need to store all possible information in order to be able to answer all possible questions, it would need significantly more storage capacity (and hence a larger head to accommodate that). From a rational and positivist view of the conversation, this seems an aberration and an error of thinking. Language in general plays an important role in this research. The research methods themselves are more self-learning. The general validity of the observations is less declared; something which is done more quickly in classical science.
Contemporary theories of artificial intelligence regarding how to think about the subject of reasoning, offer a different attitude. They take more distance from earlier, rather positivist approaches. Possible interaction between reason and the soul are looked for, which was unthinkable before in cognitive psychology. Reason (intelligence) is considered a behavior: behavior is what counts. The link between the brain and reason (intelligence) is broken: reason is not only present in the brain but in the whole body (distributed intelligence). We are referring here to the concept of “embodied mind”. Reason, the brain and intelligence are considered less and less to be like a computer, which refers to the thinking of the machine behind reasoning. The interpretation, which is general y always vigorous becomes, in effect, more and more based on shifting sands.
Intelligence is strong in the organization of “the next step”. It is weaker in the planning of a number of next steps (multiple) and weaker still in the execution of subsequent multiple steps. Intelligence is the organ of control for an autonomous agent. The structure was formed in a descending manner or by a combination of various elements. A common example of a combination occurs when different people sit around a table. The result is in effect, a structure built up by a network of different elements (people).
In addition, intelligence is something continuous and not only a sort of metaphor for a machine which functions with zeros and ones. Intelligence does not function with numbers and symbols but with vague notions like tal , taller, smaller etc. Intelligence reacts to these sensations, which can be translated into sensorial perceptions. Then all of that is translated into information. It is not the action on the senses themselves which creates information, but the liaison of a specific perception with the existing network of perceptions, and information. Sensorial perception, action and knowledge go together and that is summarized in the notion of “enacted cognition”.
Biology lessons and enacted cognition
Self-organization and self-production are concepts which are strongly embedded in neurobiology; concepts which, when translated into social systems, receive more and more attention but which are radical y different from the tradition of western management. In this tradition, everything must be organized and controlled, based on whichever intellectual tradition it is inspired by. But, in practice, that does not work and can lead to frustration.
Through Varela’s concept of enacted cognition, it is possible to see how intelligence re-combines old information and experience to produce new actions. Intelligence is positioned in a diffuse way in the whole body, but it seems that, even if the modules are independent, they col aborate through connections between each other. These notions undermine what is used most of the time as a basis for our management, especial y specifical y concerning structure and control. In particular, it raises one crucial question: can a social system be organized from outside?
From enquiries into the behavior of autopoietic systems, the answer seems to be no. Instead, it appears that each system, even if examined from the simplest components, can only organize itself, replicate itself and assure its own survival. Although this could be taken as negative at first glance, it represents an undeniable strength. Systems do not necessarily need a tight direction with a lot of complicated rules, at least if we dare to give back the control and the direction to the system itself. Therefore, the shocking conclusion for traditional management is that intervening in such a system should probably be avoided.
Just as people are often not aware of the self-organization within themselves, management of large companies are largely not aware of the self-organization within the company. Many subsequent re-organizations directed from “outside” or “above” into traditional organizations actual y risk destroying the very organizational fabric that has allowed it to operate as an integrated whole. However, careful experiments in certain companies seem to indicate how more self-organized management could succeed.
In practice, how can this relate to knowledge management? Or to put it in more specific terms: how might knowledge be organized in a “self-searching” and preferably self-finding way (as opposed to the branch structure implemented in very large databases)? Groups responsible for the conception of new products seem to be more creative and efficient if they have more liberty. This is not to claim that nothing can be learned from the past with regard to successes and errors, on the contrary. However, the difference is that the lessons learned from previous experience are not translated into typical outcomes such as the “ten commandments of innovation”. Instead, they are used rather like stories from which, according to individual interests, advantages can be extracted.
The gift of learning and learning to walk
The potential of learning, in itself, in companies is an interesting gift. More and more companies try to provide a more flexible range of training and support better linked to the profiles of particular individual roles. They seek to provide “just-in-time, just-enough” learning, based on the development of necessary managerial competencies. Other research, including our own (Baets, 2005), has il ustrated that workplace learning is a very effective and efficient way of continuous learning. This is particularly marked if it is related to the development of managerial competencies. E-learning by itself (without speaking of classical e-teaching or of distance or correspondence learning), is a promising development through which a manager, while doing her or his job, is supported by the learning environment. To those who want to learn, it will offer potential y effective support, adapted to their needs. Such situations also abolish the classical teacher role, which is designed to transmit to classes of pupils how the real world is (or should be). Those who want to learn find what they want and need to learn, and then learn by doing (and therefore learning). All these developments are promising, on condition that the organization has the courage to relax control at all levels.
This does not translate into saying that everyone just does anything they want, and then see what happens. Let’s use Alice in Wonderland as an example. If, effectively, employees do not know where they or their organization wants to go, then each path is equal y relevant and good decision-making is difficult. Nor can organizations be too rigid with direction as in western management’s tendency to make the paths to follow fixed. That leads to problems when these paths change – as they quickly do in current conditions of rapid change – and organizations forget the need to adapt and change themselves, because they concentrate on the destination to reach instead of the path to follow.
Alice’s choice was made impossible because she didn’t have a final destination. And that is what values are in a company. Values-based leadership is based on the values that a company would like to reach. We will talk a bit more about values in a later chapter. But the role of values is not to replace profit, or even shareholder value in business management. The role of values is to be the lighthouse, the desired destination not just for the company, but for each and every employee of that company. Values as the lighthouse add a dimension to business management that we just don’t have today. And in the absence of those values as a longer term, sustainable purpose we are almost bound to rely on short term goals, like profit and shareholder value. If there is no longer term perspective, we can only manage the short term. If, however, we are able to agree about the values driving our business, they become the lighthouse for the mechanism of emergence, as described in this chapter.
From an autopoietic perspective, the path that leads towards the goal will be created by walking it, in the network of employees. As Antonio Machado says: Caminanto, no hay camino, se hace camino al andar (Wanderer, there is no path, the path is made by walking). In many cases it requires the hacking of bushes and branches to form the desired path. The path is also forged through the negotiation of obstacles such as large boulders, streams or cliffs. Management, or organizational strategy, must be concentrated on the goal to reach and then to share ideas with the network of employees. In practice, that does not often happen. Strategy is often considered a secret and so employees cannot even help management to attain the goal. Alternatively, the strategy often appears to be short term oriented. The agents in the network each have their own preferences and capabilities. In the interaction with other elements of the network they can walk on the path leading to the goal by different small steps. If the goal is clear and realistic, the path could be adapted each time it is necessary, and sometimes this could be very often. Knowing the goal, employees can take individual responsibility in contributing to this network of walking paths.
The search for an underlying theory (1): chaos and order
As already noted, certain contemporary disciplines challenge the tradition of western management and its propensity to organize and control. Another understanding of the functioning of social systems is possible and can give us other ideas for management if we are able to observe from another paradigm. Is there therefore a paradigm which is scientific (based on scientific discovery) and which provides a vision of social self-organization?
In seeking to ground this vision theoretical y, this book found inspiration in what is known as the theory of complexity and/or chaos. Scientists starting effectively from a positivist paradigm, but with an open-mindedness which lets them truly see, have made remarkable observations. A number of them, for example Prigogine, have received the Nobel prize for their research but very often their knowledge remains essential y within their own circle of researchers in hard sciences. Only during the last few decades have we gradual y seen what their theories, especial y their application, can mean for social systems. This aspect orients us towards “the search for an underlying theory”.
Look at the chaos of daily life, a point of common contact where writers and readers can supposedly agree. But how does it work? Chaos is clearly present at times, but fortunately we also regularly have order. Certain things often play out in a well-organized way. The administration is organized, the trains are well organized and planes leave more or less on time. Isn’t that order? And how does chaos play a role in that? Is there a theory of chaos? As we have already stated, positivism goes perfectly with the Cartesian attitude. It is, however, in the positivist sciences that the first doubts emerged. As early as 1903, Poincaré made a remarkable observation.
“Sometimes small differences in initial conditions produce very large differences in the
final observations. A minor change in the former can cause a tremendous error in the
latter. The phenomenon is becoming unpredictable; we have random phenomena.”
He could not prove it, but he could clearly see remarkable things happening. During the simulation of certain mathematical models, it was discovered that when very small differences in the initial values appear, big differences could be found in the final results. Poincaré could not explain the cause. Nor did he appear to have the least idea that he was making an observation which would later lead to what became known as the theory of chaos and complexity. It should also be remembered that Poincaré did not have computers at his disposal to experiment rapidly with all sorts of simulations.
It was not until 1964 that Lorenz, an American meteorologist, discovered and identified the problem with supporting data. In between these dates, in 1931, Gödel’s theory had sowed confusion by proving that any axiomatic system, let us say a mathematical system of variables and equations, would one day not be up to accepting or rejecting all possible statements. Therefore, there could be no unique perfect model of the world. This discovery was much more than a huge question mark, and represented a new turning point for mathematicians. Mandelbrot’s fractal algebra is just one of these new orientations. But Gödel’s theory, while adding to its significance, does not give a response to Poincaré’s problem.
Lorenz, however, did add clarity. As a meteorologist he worked with a simple system of three dynamic non-linear equations. Dynamic means, for example, that today’s temperature is a function of yesterday’s. Non-linear means that somewhere there is a variable with an exhibitor. With his system, Lorenz tried to predict the weather. He made a number of observations and simulations on this subject. Lorenz had a computer, which was not common in 1964, and thanks to this computer, Lorenz could clarify what Poincaré had suspected. The use of computers became indispensable to be able to do sufficiently large simulations.
During these simulations Lorenz had to interrupt his research. Since computers did not have screens at the time, they produced a mountain of paper which had to be analyzed. When he wanted to pick up the simulation later, he wanted to use the last value the computer had produced as the initial value of the rest of the simulation. He had, as a good scientist, certain doubt about the wisdom of such an approach. Therefore, instead of taking the last value, he took the result of the previous 100 observations and started the simulation with that value.