Predictable Rate of Change

A shortened version of this post appears on the South Africa HR Strategy website of

It is clear from observation of the world around us and a study of history of human development that things are different today than they were yesterday. That much is obvious. It is also obvious that the rate of change is different. The amount of change between any two equidistant points in time is not the same; it is more. There thus seems to be, at present, an increasing rate of change. Most writers are agreeing on this.

The point now is to consider whether the observable trend is part of a cyclical pattern or an exponential one. It seems from our point of view, and from all the evidence that we have, that never before in human history has the rate of change been this high. Three possibilities here: evidence; boundary conditions; and snapshot.

Firstly, we may not have sufficient evidence to make this judgement. The tools that we have at our disposal are the writings of historians and scholars, archaeological evidence and present artefacts. Whether these constitute enough evidence upon which to base a prediction of the the future rate of change of not is the point in question. It is debatable, but probable that we do have enough.

Secondly, if we take the first point as probable, we should consider whether or not we are in a cyclical pattern, but we are at the first iteration of the cycle, and hence at the lowest boundary of the time scale (perhaps we could consider a graph with t=0 at the start of the cycle and we are presently at t=1). Then we would not know whether this was cyclical or not. In fact it is impossible to tell. We would have to be able to look back past t=0, and we presently cannot do this with our observation methods.

Thirdly, we are biased by living memory. Psychologists refer to recency bias which the phenomenon that events chronology closest to us in time have the greatest impact on us in terms of confirmation bias. Confirmation bias is the tendency for us to locate evidence to support a hypothesis and to ignore evidence that does not support this hypothesis. The same could be said of the rate of change. There are domains where human development is speeding up certainly, but there could be domains where human development is either slowing down or is static.

So now we have conversations with one another on the drive to the airport, through infrastructure improvement, when surfing the net, reading the news and listening to the curricula of our children. We often say things like, "Wouldn\'t my grandparents or great-grandparents just be blown away by the things we have and do today! But I can\'t imagine the world of my grand-children and great-grandchildren."

So what do we make of that?

Perhaps there is another way to look at this, that is not based on past history or the nature of the historical change curve, cyclical or otherwise. There is a way of predicting, not the possible futures, but the nature of the rate change towards many possible futures, at least for the a generation or two. I put forward three interacting systems: present nature of any one domain of human development; the number of domains of human development and the rate of change of these; and the nature of the interaction between the domains.

So let us look at each of these in turn. What we will see is that even if we consider the rate of change of the first two factors to be precisely static from this point forward, the total rate of change will be increasing.

Firstly, the separate domains. If we consider some examples, such as technology, biology, theoretical physics, literature and philosophy, culture, sociology and others, we can see that we have extremely complex systems at play. By complex systems we mean that the number of nodes and their connections is high enough for us to be unable to create sufficiently accurate simulations and mathematical models to be able to forecast with any degree of certainty what the effect of an event will be. The nature of these systems lends itself to outcomes that are often disproportionate to the magnitude of the event itself; sometime surprisingly high for small events and sometimes negligible for larger events. In short, the present domains are inherently unpredictable.

Secondly, the there in an increasing number of domains of human endeavour. Technology in the last century was typically limited to a few domains such as production, flight and electricity. Semi-conductors changed all that. Increased specialization has resulted in the subdivision of domains - for example the study of human behaviour used to be moral or physical. Now we have psychology, psychiatry, neurology, esoteric, personality, biological, cultural, and so on. Each of these fields has grown into a complex domain in their own right and therefore inherently unpredictable, but at present we observe them to be increasing in rate of change, or very least, changing towards greater complexity.

Thirdly, the actual contents of some these domains has led to a greater degree of interactions between the domains and their elements. In particular the speed of global travel and communication has allowed previously unrelated people and systems to interact to create unpredictable outcomes at a macro level. Medicine and technology are interacting and emerging. Space exploration and biology are interacting and emerging. Culture and religion are interacting and emerging.

The result of all of this is that the combined effect of these three notions, change in a domain, the number of domains, and the interaction between domains means that even for linear rates of change in each, the overall effect will at least be exponential to the third order, but since they all all complex systems, it is really anybody's guess.

So for any single node in the uber-system such as a human being, the present experience is unlikely to same the the future experience. One dire warning in 1947 comes from the noted scientist whose theories now pervade our sciences, "I know not with what weapons the third world war will be fought, but world war four will be fought with sticks and stones."

We can't predict what the future will be like, or how fast things will change. But what is within our grasp are these faculties: self-awareness of our limitations in the science (or art) of prediction; and self-regulation to develop new, more adaptive tools for leading ourselves and others. There are three particular practices that are imperative for leaders to adopt:

  1. Experiment
    Complex systems are best examined by being inside them. Start the clock on some probing initiatives that are ?safe-to-fail?, and don?t be too worried about them being fail-safe.
  2. Find Patterns
    Learn from the experiments and sense the patterns that they reveal about the nature of those systems.
  3. Respond
    Desirable patterns should be amplified by more confident experiments. Undesirable patterns should be dampened by withdrawing the insightful, but unhelpful, experiments.

Confident judgement on the part of leaders peering through the half-light of unpredictability eases our anxiety, and passes to our great-grandchildren a batten of hope.

UPDATE: The founder of Drupal, the open source content management system (CMS) made a similar comment about the nature of "safe-to-fail". An extract from the interview he did with Danish IT media Version2 is quoted below (emphasis, mine): "If you are to look at just the language, yes, then Java is the better language in terms of language design but PHP definitely gets the job done and in many ways it is faster to write PHP code than it is to write Java code. No language is perfect. I still use Java and still write software in Java. I think it is great for a backend system. But for web stuff one of my big philosophies, and that is also an open source philosophy, is to fail often and fail cheap. I think PHP lends itself to that very well."