SYSTEMS COMMUNITY SHOULD FOCUS ON EMERGENCE IN COMPLEX COGNITIVE AND BIOLOGICAL SYSTEMS
Eliano Pessa
Dipartimento di Filosofia - Sezione Psicologia
Università degli Studi di Pavia
Piazza Botta 6 (Palazzo S.Felice), 27100 Pavia, Italy
Tel.: +39-(0)382-986276 eliano.pessa@unipv.it
The goal of this short contribution is to propose to the systems community an important challenge to be dealt with in the immediate future: the study and characterization of general features of what is commonly qualified as ‘emergence’, chiefly in complex systems such as biological and cognitive ones. Such a topic was a fundamental one at the very beginning of Systemic movement, and to it the founding fathers, such as Von Bertalanffy, Ashby and Von Foerster, devoted most efforts. In more recent times, however, the interests shifted towards an empirical study of systemic properties characterizing human organizations, and the subject of emergence was partly abandoned. Notwithstanding, the understanding of what is emergence, and of the circumstances which allow for its occurrence within a complex system, is of crucial importance for Systemics. Namely all systemic properties – the ones which allow a system to behave as a whole and not as an aggregate of constituents – are just emergent properties.
In this regard, we are reminded that, starting from the Sixties, many scientific disciplines, sometimes inspired by systems thinking, but most often completely unaware of it, produced a number of fundamental contributions to the topic of emergence. We can quote Physics (the quantum theory of collective phenomena, the theory of symmetry-breaking phase transitions, the reformulation of Quantum Field Theory, the theory of nonlinear phenomena, the study of classical and quantum chaos, the theory of dissipative structures, the birth of Synergetics), Biology (the birth of structuralism), Neuroscience (the discovery of long range correlations in the brain, the discovery of role of chaotic processes in olfactory bulb, the birth of psychoneuroimmunology), Cognitive Science (the introduction of connectionist models of cognitive processing), Artificial Intelligence (the introduction of neural and neuro-fuzzy networks, of soft computing, of evolutionary algorithms, of Artificial Life), Engineering (the birth of nanotechnology, of quantum computing, of self-designing machines), Philosophy (the analysis of Binding Problem, of Symbol Grounding Problem, of concepts such as Coeherence and Consciousness). All these developments evidenced how, contrarily to what expected, the features of emergent phenomena, in a so wide range of different domains of inquiry, were not typical of the domain under study. In other words, emergence within the biological realm isn’t intrinisically different from emergence, for instance, within the cognitive realm or within the physical realm. This allowed for a unified, and by its very nature transdisciplinary, treatment of emergence without a specific reference to a particular domain.
Here Systemics enter into play. Namely, when dealing with systems, we must be aware of the existence of different levels of inquiry, which can be listed as follows:
a) the phenomenological level, in which we observe and describe the particular behaviors of a particular system, whose nature must be defined in advance in a detailed way, as a function of particular goals of the observer;
b) the modeling level, in which we build a specific model of a given system, or of a previously defined class of systems, and we use mathematics to derive from the model itself precise forecastings about system behavior;
c) the meta-modeling level, in which we reason about general properties of a wide class of models, aiming TO discover general features characterizing each model of the class, independently from the domain to which it refers.
These different levels are connected both by bottom-up and top-down relationships. The bottom-up influences consist in the fact that, without experimental observations of particular systems, we cannot build models, and, in turn, without the previous existence of particular models, we cannot reason about model classes. The top-down influences are evidenced by the fact that, without having a model, not only experimental observations are devoid of any meaning, but we cannot even decide what observations should be made. Moreover, the decisions about what model to choose, between the virtually unlimited number of possibilities, and how to interpret its forecasting depend only on the results of general analyses done on large classes of models.
I claim that the main role of Systemics is to give contributions at the meta-modeling level. And emergence is just a property pertaining to this level. Without this contribution every scientific progress will be blocked: the science and the engineering of the future will be almost exclusively the science and the engineering of emergent phenomena – in a word, of systems, and not of aggregates. Such a circumstance calls for a stronger engagement of Systems community in a deep reflection about a number of questions relevant to emergence. I will list in the following some of them:
1) what are the possible definitions of emergence?
2) is emergence a feature only of systems described by quantum theory or even of systems described in classical terms?
3) is it possible to build many-level models, having more than the two usual levels of description (macroscopic and microscopic)?
4) is it possible to build a many-level model in which, given two immediately neighboring levels, their relationship be exactly the same as the one holding between any other different pair of immediately neighboring levels?
5) is it possible to observe two different levels at the same time?
6) how to deal with medium-size systems which are not infinitely large nor infinitely small?
7) up to what point is it possible to control the occurrence of emergence?
8) how would it be possible to build a general theory of the interaction between the observer and external world, able to contain, as a particular case, quantum theory, but without using the traditional formalism of quantum theory?
I think that Systemics should contribute to answering these questions both in a conceptual and in a technical (that is mathematical) way. Namely every general proposal should be implemented in a particular model (which, of course, could be also a ‘toy model’). It’s time for Systemics to continue a tradition lasting to researchers such as Mesarovic and Rosen, and to re-enter into the mainstream of actual scientific inquiry, by assuming a leading role.
TOWARDS A SYSTEMICS OF EMERGENCE
Gianfranco Minati
Associazione Italiana per le Ricerca sui Sistemi (AIRS)
Via P. Rossi, 42 20161 Milano
Tel.: +39-(0)2-66202417 gianfranco.minati@airs.it
Ludwig von Bertalanffy (1901-1972) was one of the founding fathers and vice-president of the Society for General Systems Research introducing the General Systems Theory. The impact of the introduction of this explicitly new approach (it may be considered implicitly contained in many previous philosophical contributions) has been very important in many cultural and disciplinary fields.
Thereafter the concept of system has become of great importance in many disciplinary researches, but it has often been kept separated from crucial concepts of the General Systems Theory.
In engineering, for instance, we had the introduction of Systems Theory intended as theory of automata and control systems.
The concept of system is broadly used, but most often without the cultural and philosophical implications outlined by Von Bertalanffy.
The concept of system had to be unavoidably adopted and deeply studied in a number of disciplines, such as physics, because of the nature of the disciplinary problems (like transition between phases of the matter like gas, liquid and solid).
It has already been introduced the concept of emergence (see point What AIRS is).
The concept of emergence is very important in any cultural field because it reduces the classic objectivistic approach without assuming a merely relativistic one, but supporting and inducing a constructivist one based on Cognitive Science.
Properties, reality, behaviors are more effectively intended as emergent.
Because of that the General Systems Theory is becoming more and more a Theory of Emergence.
On one side more and more scientists working in the so-called hard sciences (such as physics, chemistry, biology, ecology, mathematics, information sciences, engineering, cognitive sciences, linguistics, artificial intelligence, artificial life and so on) are deeply using the concept of system.
Very important new disciplinary specifications have been introduced, such as ergodic behavior while studying collective phenomena and self-organization, which can be applied in other scientific contexts, such as economics. The process of generalization arises from advanced disciplinary researches.
On the other side who (the systems movement made world wide by systems societies) is supposed to have the mission to theoretically and rigorously activate the process of generalizing (supporting adoption of inter- and trans- disciplinary approaches) seems to be less and less interested in (particularly scientific) disciplinary results.
In a symmetrical way scientists react exhibiting less and less interest in the systems community.
As a result the most advanced researches on General Systems Theory are carried out by research institutes using “disciplinarily” the concept of system.
My opinion is that in recent years the focus of many systems societies has been too unbalanced towards social sciences only, neglecting contributions from hard sciences, such as physics, chemistry, biology, ecology, mathematics, information sciences, engineering, cognitive sciences, linguistics, artificial intelligence, and artificial life.
Systems thinking received from the 'hard sciences' substantial contributions in terms of disciplinary ideas, approaches, methodologies and theories.
The crucial misunderstanding, in my opinion, is to assume that because of the great complexity and richness of the social, human and humanist issues, General Systems Theory may preferably flourish on this side, while considering 'hard sciences' just as professional and disciplinary bounded.
I think that we should be context sensitive for the theoretical framework, be evolutionary, and apply systems thinking to ourselves, in the same way that Cognitive Science, science applies to itself.
I would like to introduce examples of disciplinary systemic results that have to be generalized (that’s “trans-disciplinarized”).
As it is well known Emergence refers to the core theoretical problems of the processes from which systems are established, as introduced in Von Bertalanffy’s General Systems Theory by considering the crucial role of the observer, together its cognitive system and cognitive models.
Emergence is not intended as a process taking place in the domain of any discipline, but as “trans-disciplinary modeling” meaningful for any discipline.
We are now facing the process by which General Systems Theory is more and more becoming a Theory of Emergence, seeking suitable models and formalizations of its fundamental bases.
Correspondingly, we need to envisage and prepare for the establishment of a Second Systemics -a Systemics of Emergence- relating to new crucial issues such as, for instance:
- Collective Phenomena;
- Phase Transitions, such as in physics (e.g. transition from solid to liquid) and in learning processes;
- Dynamical Usage of Models (DYSAM);
- Multiple systems, emerging from identical components but simultaneously exhibiting different interactions among them;
- Uncertainty Principles;
- Modeling emergence;
- Systemic meaning of new theorizations such as Quantum Field Theory (QFT) and related applications (e.g. biology, brain, consciousness, dealing with long-range correlations).
The urgency to review and improve the systemic culture arises from the need to deal with some disciplinary and locally inter-disciplinary problems and results such as those listed above, having a level of architectural abstraction requiring a re-formulation with a systemic view.
Such a re-formulation has to be suitable for trans-disciplinary usage, rather than being just metaphorical generalized, or for popularization purposes. This is a fresh challenge for systems thinking.