5.9 Creating a Resilient Society
97
the world: the financial crisis. Again, this started locally, but ultimately had a major
global impact.
Therefore, how can we better prepare for future challenges? A whole range of
measures are at our disposal, including risk assessment, probabilistic prediction,
prevention, intervention, insurance, and hedging (which basically means to choose
a portfolio strategy rather than betting on one horse). Nevertheless, we must realize
that problems will sometimes occur and accidents will sometimes happen in a world
that is not entirely predictable. That is why we need resilient systems.
But what exactly does resilience mean? Resilience is the ability of a system to
absorb shocks and to recover from them both quickly and thoroughly. If we fall and
hurt us, we will recover quickly because our body is resilient to such shocks. So, how
to build resilient systems that are not prone to undesired cascading effects, but recover
quickly and well from disruptions? This is primarily a matter of systems design and
management. Safety margins, reserves, backups, and alternatives (a “plan B”, “plan
C”) can certainly help. Furthermore, modularization is a well-known principle to
make the complexity of a system manageable. This basically means that the organization of a system is broken down into substructures or “units”, between which there
is a lower level of connectivity or interaction compared to the connectivity or interaction within the units. This allows one to reduce the complexity within substructures
to a manageable level. Furthermore, it decreases interaction effects between units.
Well-designed systems have “engineered breaking points”, “shock absorbers”, or
“dynamic decoupling strategies”, which can counter the amplification of problems
and undesirable cascading effects. For example, think of electrical fuses in your flat
or the crumple zones of a car, which are there to protect the sensitive parts of the
system (such as our home or our life).
In principle, of course, the modular units of a system can be organized in a
hierarchical way. This can be efficient, when the units (and the interactions between
them, including information flows and chains of command) work reliably, with very
few errors. However, as much as hierarchical structures help to define accountability
and to generate power, control might already be lost if a single node or link in
the hierarchy is dysfunctional. This problem can be mitigated by redundancies and
decentralization. In particular, if the dynamics of a system is hard to predict, local
autonomy can improve proper adaptation, as it is needed to produce solutions that
fit local needs well. More autonomy, of course, requires the decision-makers to take
more responsibility. This calls for high-level education and suitable tools supporting
a greater awareness of potential problems, in particular reliable information systems.
A further important principle that can often support resilience is diversity. The
benefits of diversity are multifold. First of all, diversity makes it more likely that
some units stay functional when the system is disrupted, and that solutions for many
kinds of problems already exist somewhere in the system when needed. Second,
diversity supports collective intelligence, as we will see later. Third, the innovation
rate typically grows with diversity, too. However, diversity also poses challenges, as
we know, for example, in intercultural settings. For this reason, interoperability is
important. I will come back to this issue, when we discuss “digital assistants” and
“externalities”, i.e. external effects of decisions and (inter-)actions. Finally, using the
97
the world: the financial crisis. Again, this started locally, but ultimately had a major
global impact.
Therefore, how can we better prepare for future challenges? A whole range of
measures are at our disposal, including risk assessment, probabilistic prediction,
prevention, intervention, insurance, and hedging (which basically means to choose
a portfolio strategy rather than betting on one horse). Nevertheless, we must realize
that problems will sometimes occur and accidents will sometimes happen in a world
that is not entirely predictable. That is why we need resilient systems.
But what exactly does resilience mean? Resilience is the ability of a system to
absorb shocks and to recover from them both quickly and thoroughly. If we fall and
hurt us, we will recover quickly because our body is resilient to such shocks. So, how
to build resilient systems that are not prone to undesired cascading effects, but recover
quickly and well from disruptions? This is primarily a matter of systems design and
management. Safety margins, reserves, backups, and alternatives (a “plan B”, “plan
C”) can certainly help. Furthermore, modularization is a well-known principle to
make the complexity of a system manageable. This basically means that the organization of a system is broken down into substructures or “units”, between which there
is a lower level of connectivity or interaction compared to the connectivity or interaction within the units. This allows one to reduce the complexity within substructures
to a manageable level. Furthermore, it decreases interaction effects between units.
Well-designed systems have “engineered breaking points”, “shock absorbers”, or
“dynamic decoupling strategies”, which can counter the amplification of problems
and undesirable cascading effects. For example, think of electrical fuses in your flat
or the crumple zones of a car, which are there to protect the sensitive parts of the
system (such as our home or our life).
In principle, of course, the modular units of a system can be organized in a
hierarchical way. This can be efficient, when the units (and the interactions between
them, including information flows and chains of command) work reliably, with very
few errors. However, as much as hierarchical structures help to define accountability
and to generate power, control might already be lost if a single node or link in
the hierarchy is dysfunctional. This problem can be mitigated by redundancies and
decentralization. In particular, if the dynamics of a system is hard to predict, local
autonomy can improve proper adaptation, as it is needed to produce solutions that
fit local needs well. More autonomy, of course, requires the decision-makers to take
more responsibility. This calls for high-level education and suitable tools supporting
a greater awareness of potential problems, in particular reliable information systems.
A further important principle that can often support resilience is diversity. The
benefits of diversity are multifold. First of all, diversity makes it more likely that
some units stay functional when the system is disrupted, and that solutions for many
kinds of problems already exist somewhere in the system when needed. Second,
diversity supports collective intelligence, as we will see later. Third, the innovation
rate typically grows with diversity, too. However, diversity also poses challenges, as
we know, for example, in intercultural settings. For this reason, interoperability is
important. I will come back to this issue, when we discuss “digital assistants” and
“externalities”, i.e. external effects of decisions and (inter-)actions. Finally, using the