Trends in analysis and why they tend to be wrong

I am a big Isaac Asimov fan (and a big science fiction fan in general, although I don’t get to read as much as I’d like what with all this terrorism to look at).  In his classic Foundation series we are introduced to a character right at the outset named Hari Seldon, a mathematics professor who develops a new field of study called psychohistory.  Prof. Seldon claims to have created a series of scientific algorithms that allow one to predict the future along probabilistic lines.  The series is a set of events that unfold just as he foretold.

We are obsessed with the future and a lot of people profess to have found a way to know it.  Some time ago, I was asked as part of a team to outline national security threats 10-15 years down the road.  I was not impressed with this task and if I recall correctly I wrote that I thought Al Qaeda-like terrorism would still be with us in 2020.  I look pretty prescient, don’t I?  Not really.  Anyone could have made a similarly brilliant prediction based on what we knew then.

What worries me is the tendency to proclaim new trends based on few data points and even fewer ‘time’ points.  Here is an example.  In the wake of a number of electoral successes of far right or populist parties, especially in Europe, many were calling for a wave of intolerant, and in some cases blatantly racist governments.  Marine Le Pen and the National Front in Europe.  Brexit.  Geert Wilders in the Netherlands.  Right wing parties in Denmark, Sweden, Austria, Poland, Hungary and Germany.  Donald Trump.  Buckle up folks because it is going to get rough.

Except that maybe it won’t.  In a spate of recent elections – the UK, Germany, Austria, France, and now Italy – the parties described as populist or right of centre have not done as well as predicted.  While part of this bears on a worrisome trend in inaccurate polling data, the other has to do with how and why we forecast things.  Some people are constantly saying that they know what is going to happen next whether or not they have any data to back up their theories.

And the same goes for terrorism analysis.  We are  inundated with authoritative statements like:

  • Al Qaeda is on the wane.
  • Islamic State is on the wane.
  • Boko Haram is on the wane (this is a favourite of Nigerian authorities, usually in the weeks leading up to Christmas).
  • ____ is on the wane (take your pick: Jemaah Islamiyah in Indonesia, Al Shabaab in Somalia, Al Qaeda in the Islamic Maghreb – the list is truly a long one).
  • it is getting harder to attack hard targets.
  • terrorists now have to resort to simple plots.
  • and, my choice for hubris, the fifth wave of terrorism is on the horizon.

None of these have proven to be very accurate.  This last one refers to a grand theory of political violence developed by American political scientist David Rapoport known as the wave theory.  You can read the link to get a better idea but the crux is that since all previous waves had a life span, so will the fourth ‘religious’ wave and we need to start looking out for its replacement.  In other words, we must predict the future so we can better prepare for it.

The bad news here, aside from my point that the prediction of human behaviour is not always very good, is that the current wave has a lot of energy to it.  I would not spend too many resources worrying about what comes next when we are still neck-deep in what we have now.  The truth is that we will see more attacks attributed to individuals and groups that share the same essential ideology we see in AQ ,IS and others.  And this will happen for the foreseeable future, which we really should learn is not very foreseeable.


By Phil Gurski

Phil Gurski is the President and CEO of Borealis Threat and Risk Consulting Ltd. Phil is a 32-year veteran of CSE and CSIS and the author of six books on terrorism.

Leave a Reply