Sunday, 14 October 2012

Post-Tragedy Rationalization

I'm gearing up for another spout of posting blog posts and in the coming weeks I plan to write a few posts on the internet, the limits of knowledge and things such as that. These posts will likely be from a different vantage point then what has been posted lately and is probably more on course with what the intent of this blog was at the beginning. The first of these posts will be a lead into what I call the limits of knowledge and the effects of complex systems on the stability of human society.

When we hear of a teenage suicide (as many of us has in recent news) we often launch into discussions of the tragedy of the situation. After the suicide information bubbles to the surface and we eventually create what we perceive to be a fairly complete picture of the causes of the individual's suicide, how we could prevent these causes and what to do in the future. We walk away from the tragedy thinking that we are now better equipped to deal with the reality of teenage suicides and that we can see the bigger picture clearer.

The reality of the situation is that we can't. The suicide of an individual is an unpredictable event that although warning signs tend to manifest before the suicide these warning signs are insufficient to produce the realization that the individual will actually commit suicide. Even if a plethora of warning signs is produced prior to the individual's suicide the warning signs will be dispersed across a medium of people who alone may not realize the significance of the individual events but the sum of these events is significant to trigger suicide. Even then it is likely that there exists unknown unknowns that are lost with the person's death that will never illuminate the situation and will prevent us from recognizing that these unknown unknowns may have been the factor that pushed the individual over the edge.

In the end we are left with an incomplete picture but tend to rationalize the event after the fact and come to believe that we understand the situation. We believe that we understand what causes suicide and swear to move forward in attempts to prevent it. In reality we are likely just as equipped as we were before the suicide in preventing suicides but walk away with a rationalization that we can prevent such things in the future (or at least reduce the likelihood of such an outcome). We overestimate how much we know about the situation since the alternative is a far more terrifying realization. Our prediction capabilities are far from robust and this lack of robustness is a severe limitation of human knowledge.

The reality is that we have far less control over the outcome of unpredictable situations but we are very capable of making the outcome look predictable in the end. We rationalize the outcome of an event by correlating what seems to be correlating factors with an outcome. We look at these correlations, claim these correlations are what we need to address in the future and then move on with our lives. In the future the event is repeated with the sum of the correlations spread across a medium of individuals related to the tragic event and we are not even capable then of preventing a similar outcome.

Ignoring the problems of dispersed evidence for unlikely events we also encounter another problem with predicting such events in the future. A large part of the evidence for predicting such events is lost in the annals of time with the transpiration of the event and we will never be able to access that information again. In the case of a teenage suicide we also lose the primary witness of these events as the primary witness is also the victim of cruel circumstance.

The captain goes down with their ship and with it the most accurate narrative of why the ship sunk.

Even if we could communicate with the captain through say... a magical psych-scope that allows us to communicate with the dead we would encounter yet another problem. The person who committed suicide likely has holes in the reasoning about why they did what they did. They may never realize historic events that were significant in influencing the events that led up to their suicide and that these holes are crucial to our understanding of the event.

Events such as suicides are part of a class of events I would identify as black swan events. Unpredictable events that appear predictable when viewed in the rear view mirror but still remain as unpredictable as they were before we saw them. With further increasing layers of complexity in a system unpredictable events increase in frequency and despite our best attempts to control these unpredictable events we still fail to deal with them.

Dealing with black swan events is like dealing with an oncoming hurricane. We can't stop an oncoming hurricane no matter how we try but we can recognize that such events happen and learn to adapt our understanding of the world to accommodate for the unpredictable. In the end it'll be a long time before we become apt at preventing teenage suicide despite policy attempts to reduce such things. Instead we should focus on building robust teenagers who are better equipped at dealing with the challenges of junior high and high school.

Until one realizes that the limits of one's knowledge, one is forced to make the same mistake in trying to prevent. You can't stop a hurricane... but you can mitigate its impact. As globalization increases and communication technologies spread through society like wildfire we must learn that we will find ourselves staring into the maw of hurricane lane... and that no matter how hard we try there will be hurricanes.