Thursday, January 18th, 2018

Understanding connections … in plain English


The systems view looks at the world in terms of relationships and integration” -Fritjof Capra

Okay, now we probe to the very core of systems thinking: connectivity.  You’ll recall that in our first discussion, we defined a system as “an integrated set of elements that perform a desired function.” That’s not the only definition of a system, but it shares with all others some critical pieces. First, there are parts, a “set of elements” in our construction. Second, the parts relate to one another in some fashion; they are integrated. Third, there is a rational logic to the integration of the parts; a common purpose or function. If something is comprised of all three of these – parts that relate to one another for a common purpose – we are in the presence of a system.

  • Parts
  • Relation (or connectedness)
  • Purpose

If indeed these are the three critical elements that comprise a system, let’s further note that it is the second element – relation – that gives a system coherence and therefore meaning. A faucet, a hose, and a sprinker head are three parts manufactured separately. It is only when the three are connected that “they” become an “it.” A sprinkler system. Parts. Relation. Purpose.

What is true for simple systems is also true for complex systems. The human body, as we’ve noted, is a classic system of systems. But we can’t really conceive of a human body in which one major part – a principal system – is present but not another. Imagine a body with all the major systems – cardiopulmonary, nervous, endocrinal, digestive, circulatory – minus one, the skeletal. Remove one system from the network that is the body and our very notion of the human being quite literally collapses in on itself. Parts. Relation. Purpose.


In our earlier discussion of systems, we noted that another word for a system of systems is a network. Interestingly, although the word ‘network’ has a modern technical connotation, it’s provenance is actually the mid-1500′s. It’s a compound word made up of ‘net’ (“a bag or other contrivance of strong thread or cord worked into an open, meshed fabric”) and ‘work’ (“an exertion or effort directed to produce or accomplish something”). Let’s unpack the definition quickly: A network is a contrivance of strong thread worked into an open, meshed fabric  -  a web – and directed to accomplish something. Do you see it? Parts. Relation. Purpose. Based on both the textbook definition and our own discovery about parts, relation, and purpose, we can say that a network is a web of parts connected for a common purpose.


Okay, let’s move beyond definitions and begin working directly with connections. As we’ve noted, connection gives a system coherence and purpose. But how? Clearly, by acting as a conduit for the flow of a stock between parts. That stock may be water (in the case of the sprinkler system), gasoline and air (in the case of an automobile engine), or data, as in the case of a digital network. In a digital network, the connection points – the parts – are called ‘nodes.’ It’s a useful word because while “part” can have multiple meanings, there really are no nodes apart from networks. So let’s use it.

In a network of human relationships such as a family, each member is a node. Stocks – love, memory, information, money, time – flow from one node to another in order to achieve a common purpose. (And not always “good” stocks, either. We all know, for example, that families can also channel resentments, jealousies and other less than desirable substances.) But not all connections have the same strength or efficacy. In the family network, a son may be estranged from his father. There is still a connection, but it may be weaker and less fruitful than the connection, say, between the son and his mother, which may remain strong. In this instance, we can say that the father and son are loosely coupled while the mother and son are tightly coupled.


So, nodes on a network may be tightly coupled or loosely coupled (or any degree of coupling that falls in between!). A tight coupling means that the connection is strong and stocks move efficiently between nodes. A loose coupling means that the connection is weak and stocks move sluggishly between nodes. This is simply the description of a relative condition, not a value judgment. In some cases, a loose coupling may a good thing. In others, a tight coupling may be a bad thing. It really depends on the quality of the substance being moved through the connection. If I have a connection to someone from whom I receive gossip and disinformation, it is probably to my advantage to make or keep that connection a loosely coupled one. By contrast, if I am only loosely coupled to someone who can provide me with a steady stream of solid, useful information, I may be missing an opportunity by not tightening that coupling.


In specialized digital networks, some nodes are designed as unique broadcast devices that convert mulitple inputs into a single stream of data. These devices, known as hubs, serve an important routing, amplifying and filtering function. Thinking creatively, we can see that all networks, whether by design or evolution, likewise contain nodes that serve the same function as hubs. In a family network, it may be the one person everyone confides in and trusts unreservedly. In a small business, it may be that key player whose advice everyone, including the owner, seeks and of whom it is said “the place couldn’t run without him/her.”  Hubs connect discrete or isolated portions of the network to the broader complex itself. They are sources of diverse, filtered, and authoritative streams of data. In human networks, we might call that data “information,” or even “knowledge,” or even “wisdom.”

Diagnosing Failure

Whenever any system is experiencing poor performance, there are two possible causes: component failure or connectivity failure. In an automobile engine, for instance, a difficulty may be caused by a piston rod (component failure) or a hose (connectivity failure); a pump (component failure) or just a relay switch (connectivity failure). In human networks, the components are much hardier, more durable than even those found in an automobile engine. So it makes sense to begin any diagnosis of poor performance in a human network by examining connections rather than components. Are key connections present and operating? Are they bi-directional? Is there a critical mass of tightly coupled connections within the network?  Are there working hubs, and are they receiving diverse streams of information?

We’ll return to our discussion of connections in later posts. In fact, it’s fair to say that in one form or another we will return to connections in every subsequent post, and whenever we discuss systems thinking in any form. Because systems thinking is all about connections: understanding how the parts of a system relate to one another in order to acheive the purpose for which the system was designed or evolved.

Speak Your Mind

Tell us what you're thinking...
and oh, if you want a pic to show with your comment, go get a gravatar!