Multi-level engineered systems are technologies that involve many different emergent levels to their overall structure. Urban infrastructure systems are such an example requiring many different levels to their design in order for them to function effectively; office development requiring telecommunications infrastructure, which rests upon an electrical power grid, which is dependent upon a national energy architecture, which in turn depends upon a transport infrastructure etc. Emergence gives rise to what are called integrative levels and multi-tier systems that have distinct, irreducible levels to them. This process of emergence likewise can lead to macro-scale fractal patterns of organization as can be seen in urban networks.
Emergence is a process whereby larger entities or patterns of organization arise through synergistic interactions among smaller or simpler entities that themselves do not exhibit such properties. And thus, emergent properties are those that belong to whole systems. They only arise when all the parts are put together in a particular manner. The functionality of a system of technologies that has any degree of complexity is in many ways an emergent property. It is only when we put all the parts to our car together that we get the functionality of a vehicle of transportation. And many properties to this car such as safety, security, and reliability are emergent properties, meaning we have to wait until the car is fully put together before we can start to do testing on it. There is no point in taking the car for a test drive when it is half built because the functionality that we wish to test has not emerged yet. These emergent properties cannot be studied by physically decomposing a system and looking at the parts in isolation, what is called reductionism. They can, however, be studied by looking at each of the parts in the context of the system as a whole. A microprocessor can only be properly understood by looking at its function within the whole computer system it is a part of.
Emergent properties are a product of the interaction between the components within a system and typically cannot be deduced by reference to the properties of the parts. Thus, emergence typically produces novel phenomena that we could not have predicted until we ran the system and all the parts have interacted. An example of this might be our concerns around technologies like genetic engineering or algorithmic trading. We can fully analyze and understand how an individual trading algorithm behaves, or what effect altering the genes in a plant has on that plant in isolation. But because ecosystems and financial markets are complex systems where the behavior of the whole system is an emergent product of the interaction between its parts, we do not know what emergent behavior will arise from having many different algorithms interacting or differently genetically modified plants coevolving within a whole ecosystem. The net result is an emergent phenomenon and we cannot deduce it from analyzing the parts in isolation.
Emergence leads to one of the key concepts within complexity theory, that of uncertainty. The fact that the future emerges is a key source of the fundamental uncertainty within complex systems. If we take something like the Internet, we don’t know what future technologies will be built on the network or more importantly, how those technologies will combine to form new possibilities. In this world of complexity, the future is not just unknown. It may well be in fact unknowable, and this fundamental uncertainty changes our whole approach to the future.
Emergence gives rise to new levels of organization, what is called integrative levels. The theory of integrative levels describes how new levels of organization emerge out of lower levels of complexity. To understand the relevance of this to technology, we might think about how we needed to have the agricultural revolution before we could have the industrial revolution, and in turn needed to have both of these to have the information revolution. A mobile phone without any farms to feed us would be of little value. The terms platform technology or multi-tier architecture are used to capture how technological systems are built on top of and enabled by others. A functioning urban center that provides a high quality of life to its citizens is an emergent property of multiple layers of technological infrastructure. Each layer needs to be properly integrated to enable the technologies it supports.
Although emergent properties can arise without self-organization, as in our former example of the car, emergence is also a product of self-organization. With self-organization individual components interact, synchronize to form patterns and out of this emerges a new level of organization. This process of emergence doesn’t just stop at one level. Elements can interact and self-organize with new levels of organization emerging, but then this new system starts to also interact with other systems in its environment, with the net result being that another level emerging. The parts in our car give rise to the global functionality of the car, but then this car is put into operation within a transportation system and interacts with other cars as we get the emergence of traffic. Whereas the car was produced through a formal design process, traffic emerges through the self- organization of all the individual cars interacting. Thus, through this process of emergence, we get the development of multi-tier systems.
This multi-tier, hierarchical structure can be seen in the formation of urban centers. From the perspective of technology analysis, urban networks are really the fabric of our engineered environment on the macro scale. They are dense concentrations of integrated infrastructure, technology and services that have emerged over a long process of evolution, and they have a hierarchical structure to them. From villages to towns to cities to metropolitan areas, this hierarchy is described by central place theory, a geographical theory that seeks to explain the number, size, and distribution of urban centers.
It describes how certain differentiated services emerge at a certain scale threshold. A village can provide some set of basic services, with a collection of villages, we can get the emergence of a town that will provide certain differentiated services in its functioning as a regional hub. And again with a dense enough concentration of towns, we will get the emergence of a city and so on, all the way up until we get globally differentiated metropolitan areas where one can access certain advanced services that are not available anywhere else, as would be the case with global cities like New York, London or Tokyo. This is an example of an emergent multi-tier system through distributed self-organization.
As we go up this hierarchy through different tiers, the economic infrastructure fundamentally changes, from serving the function of primary production to manufacturing to services, information and knowledge activities. This is the three-sector hypothesis to the development of our economic infrastructure and it is the macro-scale structure to our global economy. Each level to this hierarchy allows for greater specialization and differentiation. A small village serving a few hundred people can only really provide the basic services that the mass of people need, but a global city like Singapore can provide highly specialized financial products because it plays a differentiated role within South East Asia and the global economy as a trade and financial hub.
As we go up this hierarchy, there will be thresholds and tipping points beyond which we get the emergence of a new phenomenon. One way to think about tipping points is that many emergent phenomena are discrete, meaning either you have them or you don’t, either a city has an airport or it doesn’t. You can’t have half an airport. But many factors are also continuous, like the population of a city. You don’t go from one million people to two million. There are many small steps along the away. When we combine these two metric systems because one is changing continuously and the other in a discrete fashion, we get tipping points. To illustrate this, imagine the government in a country makes a policy that once a city reaches a threshold of a million people then they will fund the building of an airport. The result is that the population may be growing at a steady state for many decades or even centuries without any airport. And then just a few more people are added to get a million and we suddenly get a flip within the discrete variable from a city without an airport to one with an airport, and this flip came about through a very small change to the continuous variable. This is a somewhat stylized example but it should help to illustrate the dynamics behind thresholds and phase transitions
This hierarchical structure that emerges within complex systems often creates patterns that repeat themselves at various level of scale, what are called fractals such as can be seen in sea shells, the emergent patterns of a snowflake and the macro-scale structure to our engineered environment. The emergent pattern created by the central place theory is a fractal that has this scale-invariant property where we find the same network pattern on the micro level of an individual village as on the macro level of a national urban network.
Characteristic of these fractals are power law distributions that describe a power relationship between the frequency of the occurrence to a phenomenon and the scale of that occurrence. Urban networks have been shown to follow this power law relationship between the size of a city and how many there are of them.3 This is quite remarkable, that through a somewhat chaotic, self-organizing, evolutionary process we get this macro-scale pattern of organization that has a quantifiable regularity to it. This fractal structure is a very economical way to create a macro-scale pattern. Through iteration of some simple rule, we get the same structure on the macro and micro level.
But this scale invariances does not mean that the system is the same on its different levels. It is simply a global structural pattern that emerges within nonlinear systems. Through emergence and phase transitions properties describing one level of a complex system do not necessarily explain another level, despite how intrinsically connected the two may be. At each level of complexity, new laws, properties, and phenomena arise with their own internal dynamics that are specific to that level of organization and cannot be reduced to simple aggregates of lower level phenomena. Different functional levels to our economic infrastructure run on very different principals. The primary sector, industrial sector, and services sector are all governed by their own internal dynamics and set of rules, thus applying an industrial logic to services simply doesn’t work. One may emerge out of the other, but they are based on fundamentally different rules.
Bottom-up & Top-down
With this process of emergence and the creation of a multi-tier system, we get a complex dynamic forming between the bottom-up process of organization and the global pattern that has emerged, as it feeds back to enable and constrain the components on the local level. Being part of a city and that macro-scale pattern of organization both enable us to do more, as we are enabled by a vast technology infrastructure, giving us access to a wide array of services. But it also constrains us, as there are global standards and rules that have to be followed in order to coordinate the system on the macro scale.
As an example, we might think about the phenomenon of urban gardening in Detroit, USA, where due to a mass exodus of people, there is a significant amount of unused land within the city. Locals have moved in to start small garden farms on these open spaces. This is a bottom-up self-organizing process where people are simply reacting to local phenomena, but it is in strong tension with the macro-scale pattern of organization, as an industrial city like Detroit plays a differentiated role within a region as a manufacturing, commercial, residential and economic hub. The macro-scale pattern of organization that has emerged is not designed to accommodate this local self-organizing phenomenon. Because of the irreducible nature of emergence, this tension between bottom-up and top- down organization is a fundamental phenomenon with complex multi-tier systems and trying to resolve it is a key design engineering challenge.
1. (2017). Cambridge.org. Retrieved 6 July 2017, from https://www.cambridge.org/core/services/aop-cambridge-core/content/view/B8FC1B4924E931732FFD5773F4ED74E1/S0890060414000481a.pdf/design-of-complex-engineered-systems.pdf
2. (2017). Old-classes.design4complexity.com. Retrieved 6 July 2017, from http://old-classes.design4complexity.com/7701-S14/reading/critical-thinking/Types-and-Forms-of-Emergence.pdf
3. (2017). Arxiv.org. Retrieved 6 July 2017, from https://arxiv.org/pdf/cond-mat/0411241.pdf