ELSI: Cross-Domain Effects Part 2

Where This Fits

Continuing from Part 1, this unit examines more cross-domain interactions, focusing on network parameters like connectivity, redundancy, and transparency, and how they create cascading effects across ELSI domains.

---

Connectivity

Connectivity is the level at which the nodes or agents in a system are connected to one another. It’s a very basic network property. The characteristics and consequences of Connectivity are usually easy to figure out; counting your friends is easy, and if you have more friends, you have higher chances of people coming to your birthday party.

If there’s more roads between cities, travel times will be shorter overall, and possibly more efficient. Of course, each connection comes at a cost, and depending on the cost, the Connectivity property of a system may reduce or increase Efficiency, and thus Resilience and finally Sustainability. High or Low? In most cases you want a system to have a high level of Connectivity.

If connection cost (and management) is not a limiting factor, it’s usually best to have as many connections as possible. The quality of these connections, of course, also matters, which is covered in the other indicators. Diversity Diversity indicates the different types and connections the network has.

Diversity is often important to have a system withstand changes in environment, have a level of self-resolution and increase inventiveness. Diversity is leveraged on both the types of nodes, and the types of relations between each node. High or Low? Diversity is often a desired property in a system. Especially larger systems suffer when the Diversity is too low.

Diversity is often a property that wants to be balanced: not too high, not too low, but in a comfortable middle, possibly err on the side of more than less. Imagine a warehouse filled with a single product: if the demand for that one product goes down, the company that owns it suffers quickly.

If the warehouse had a lot of different products, its risks would have been spread, but its management would be more complex. Similarly with people in an organization: more diverse people increase the ability of an organization to respond to challenges, and have a broader platform of experience and perspective. Too much Diversity may lead to fragmentation and poor cohesion of a system, as well as low Efficiency.

Complexity Complexity is a network parameter combining the amount of nodes, the amount of connections and the network’s Diversity. In that sense, it is a compound indicator, but its qualities are so fundamental that we’ve included in as a base indicator for networks. Network Complexity governs a wide range of effects that are important in virtually all cases.

Complexity is also an emergent property of systems in general, one which is paired with the law of diminishing marginal returns. This important network effect, described by Joseph Tainter in his book The Collapse of Complex Societies (1988), has proven to be a major reason for human societies to collapse.

This leads us to adopt a general strategy of ‘decomplexification’. Since other network properties such as Redundancy and Connectivity tend to drive up Complexity, this parameter can explicitly serve as a check and balance element, much like the Efficiency parameter. High or Low? For a given outcome, it’s desirable for the network complexity to be low.

A high Complexity usually makes a system fragile, especially if the complexity rises above the beneficial limits of the size of the system (diminishing marginal returns). This means that there’s usually an optimum of system Complexity that lies somewhere in between its minimum and maximum properties.

Flexibility Flexibility determines the ability of the network to form new connections, and to reroute or ‘bend’ existing connections. It’s a measure of the ‘elasticity’ of the network, with an important time component: how quickly can connections be (re)established, and at what cost? How far outside of the required parameters can a system operate before it fails? Flexibility is dependent on time.

A system may be able to form new connections easily, but how long does it take to do this? High or Low? Generally speaking, Flexibility is a good thing. It’s one of the important supporters of system resiliency and allows faster changes.

A flexible system can more easily move itself into an optimal position. Awareness is necessary for a system to change when it is needed though, therefore Flexibility is often dependent on other system qualities such as Awareness and Transparency. Flexibility is also often related to Redundancy and Diversity.

For example, if there’s a low Diversity in disciplines in a company, the company can’t be very flexible on its delivery performance on a certain task, because it doesn’t have the know-how and expertise. Similarly, a city can’t house a new generation of people with different living requirements if its housing stock is limited in Diversity, and cannot be converted quickly, where Flexibility relates to the convertability of the stock.

Redundancy

Redundancy is a straightforward indicator that measures the level of repetition of nodes and relationships in the network. In societal systems, there’s always a redundancy present. In technical systems, Redundancy is often built in to increase reliability.

High or Low? Low Redundancy leads to a fragile system, which can easily fail due to the breaking of a single critical component or connection. High Redundancy has a positive influence on Resilience, but also reduces the Efficiency of a system, increases Complexity, and may affect other parameters as well.

Efficiency is usually not a goal in itself, but rather a means to an end. Therefore, if a system is sustainable it does not matter what its Efficiency is, and it’s preferred to have higher Resiliency than Efficiency. It follows that a certain degree of Redundancy is usually a good thing. Redundancy in itself is rarely a goal, but its effect on the Resilience system indicator is large.

For example, combined with a low score on the Centrality indicator, Redundancy is a major driving force behind decentralized power generation. Combined with Diversity Redundancy supports Flexibility. Centrality Centrality measures the extent with which a system is reliant on particular critical node(s) within the network.

In other words, it gauges the structure of a network and gauges what form it has, from a star-shaped form with high Centrality, or a flat hierarchy organic shape with a low Centrality. It is a highly influential aspect that can change the behavior of a system radically.

In any network, Centrality plays a role on some level. Much has been written on Centralized vs. Decentralized systems, and many mathematical models exist to evaluate networks on their Centrality, and their consequences. High or Low? Centrality is interesting, because its effects shift quite radically depending on the size of the network.

For small networks, a highly Centralized system is often efficient, fast, and reasonably reliable. That is, if the central nodes are Redundant and Flexible enough. But, when scaling up the system Centralized networks become brittle, have low resiliency, and are not efficient anymore: they lose most of their beneficial properties.

This already measurably happens on the scale of small to medium sized organizations. Small organizations, from one person up to 50 or so, stand to benefit from some from of Centralization and hierarchy in the node structure. Beyond this, it quickly becomes rewarding to adopt a less hierarchical, and less centralized organizational structure.

Awareness Awareness measures the reach of information between nodes within the system. A node may be in total oblivion (a complete disconnect), or it may be aware of only the information its own surrounding nodes have, its own sub-networks, or it may have a complete awareness of the entire system, and so on.

Note that Awareness may be high even while Validity is very low, as in the case of propaganda. Awareness often relates to Transparency whilst Validity refers to truth. High or Low? A system with a high level of Awareness is able to respond faster to events in the system. Entities in the system can respond sooner when they are more aware. The third braking light in cars is an easy example.

This light can be seen through the windshields of other cars, and alerts drivers further down the road sooner. Higher Awareness in social and cultural networks can foster greater innovative capacity within the system. In another example, low Awareness negatively impacts the Resilience of a system, since those agents that do not know about the best ways to respond to critical events may make uninformed decisions and make the situation worse.

An example is what happened at the Chernobyl nuclear disaster. Even the highest officers at the facility were not aware that the emergency stop procedure could make a situation worse, and cause a meltdown. This is what happened eventually.

---

Takeaway: Cross-domain effects are not theoretical. The Chernobyl example shows how low transparency and poor connectivity in a system can have catastrophic consequences across every ELSI domain. The next unit, "ELSI: Cross-Domain Effects Part 2 (continued)," continues with Harmony parameters and systemic implications.