Statues and Statistics

             Among the dramatic events that happened in 2020 was the groundswell of support for removing statues of people deemed prejudiced against, primarily, Black Americans.  This groundswell arose from the demonstrations against the shooting of Black Americans by police officers - the "Black Lives Matter" demonstrations.  It was a tangible, and probably lasting, outcome of these demonstrations.

 

             I'm not sure if the toppling of statues really helped reduce the ongoing problem of prejudice against Black Americans or against minorities in general, but I do understand the desire to have some immediate tangible outcome of the movement.  Social change is very hard to accomplish, generally occurs slowly, and is often hard to measure.  Thus, when there is significant social angst and a great desire to "change things now," it is tempting to look for tangible (i.e. measurable), immediate impacts.  The removal of statues of people deemed prejudiced or demeaning to Black people met that criterion.  A crowd of people can band together and by sheer force of will and determination, topple a statue.  There is something invigorating about being part of a group that acts in unison with ropes and brute strength and accomplishes a task that, by all rights, should have required a crane and a bulldozer.  And, further, when you are all done, there is an empty base of the statue left, providing tangible evidence of what you accomplished.  You can point to this spot in the future and tell people that you were part of the group that removed the statue that used to be here, a statue of someone who stood for bigotry.

 

             This blog entry is not about the Black Lives Matter movement per se, but I want to focus on the problem of establishing a benchmark of human behavior that is inherent in the removal of statues.  Of course, the actions of a random group of angry people - the "mob mentality" - is not likely to result in a well-reasoned social act and, frankly, it's easy to pick on such actions after the fact.  In the heat of the moment, it seemed like a good idea.  I'm sure it felt exciting at the time.  But, in the aftermath, society as a whole is now left to grapple with the "benchmark problem."  The problem is, we can't address the benchmark problem without understanding human nature and without a little understanding of...statistics (!).

 

             The "Benchmark Problem" is this:  where do you draw the line?  What is the benchmark behavior that qualifies a human being to be memorialized in a statue?  As more and more statues were toppled, we were left with questions about whether statues of people like George Washington or Abraham Lincoln should be toppled.  Recently I heard that there was a movement to topple certain statues of Jesus (the white, non-middle-eastern-looking ones).  Once you start down that path, where do you stop?  That's the Benchmark Problem and it includes questions such as:

·        How comprehensive should you be in evaluating the life of a person to determine whether they should have a statue?

·        What key issues of an individual's life should be evaluated?

·        What are the criteria by which the individual's life should be evaluated?

·        What defines a 'good' person or a 'bad' person?

·        Should the prevailing social views of the time be taken into account when evaluating people who lived in the past?

·        What level of evidence is sufficient to establish an individual's good-ness or bad-ness?

             These are tough questions to answer and we certainly can't expect a mob to stop in the midst of their angst and carefully evaluate and think through these questions.  But the "social consciousness" that was aroused by these events continues on and allows the opportunity for more reasoned discourse.  Personally, I think this is a good thing and, at least for the remainder of this blog entry, I'd like to focus specifically on the Benchmark Problem.  Especially the issue of deciding whether someone is "good"...or at least good enough to warrant being enshrined in a statue.  And, though it may seem odd, I'd like to use some simple statistical principles to illustrate the problem.

 

           I'm going to start by suggesting that, in general, we human beings tend to view our society as being separable into three groups, as illustrated in Figure 1.  These three groups I called: 1) the really good people, 2) most people, and 3) the really bad people.  This diagram is just meant to illustrate the sense I have from listening to people talk and trying to understand the basic underpinning of their views.  To be honest, I don't know that most people would articulate the concept in something so concrete as a diagram like Figure 1.  I think it is an unstated and overlooked foundational belief.  In other words, I think most people don't necessarily know that this is really how they view society, but if you could peel back their beliefs and statements sufficiently, I think you would find this concept is a pretty foundational and strongly held belief. 

 

             The problem with the concept shown in Figure 1 is that it is certainly not true.  In every other human characteristic than can be measured in some way (I'm acknowledging that "goodness" is very hard to quantify and measure), human traits generally follow a Gaussian distribution, or the "bell-shaped curve", as shown in Figure 2.  I've shown examples of people's height, IQ score, and autism spectrum score.  The details are not important here, but what I'm focusing on is the general pattern of the curves.  They do not follow the distribution that would be implied with the scenario in Figure 1 (that graph would result in three distinct and fully separate peaks with no overlap between the groups).  Instead, there is one continuum with a big peak in the middle and outliers at both extremes.  There is every reason to believe that, regardless of the yardstick you use to measure "goodness", you're going to get the same kind of distribution.  There will be lots of people who fall in the middle and then there will be a continuous distribution of people with higher and lower scores, ultimately bounded by some maximum and minimum scores.

  

             I added the autism graphs because I thought they were particularly relevant to the point I want to make.  When I was younger, autism was a thing you either had or you didn't.  I assume it was based on the diagnosis of a health professional somewhere.  I'm sure they didn't always use scores and I'm sure there was plenty of subjectivity in that diagnosis.  But as the field progressed and as more measurement techniques and more data became available, the field of autism study realized that autism was not at all like Figure 1 but instead it was much more like every other human characteristic - more like Figure 2.  A diagnosis of something like Asperger's became more common as a "milder" form of autism.  And, eventually, everyone realized what was probably obvious from the beginning:  there's a whole "spectrum" of autism.  So, now we talk about "being on the spectrum" which, to some extent, still holds on to the Figure 1 idea that there are some people who are "on the spectrum" and some who are not.  This illustrates how hard it is for us to let go of the Figure 1 concept, but the fact is that everyone is "on the spectrum" of autism, just like everyone is "on the spectrum" of human height or IQ level.  Or...  "goodness" rating.

 

             The reality is that human goodness must follow a spectrum from awful to great.  It is like Figure 2, not Figure 1.  Which, of course, begs one of the biggest questions of all time:  where do you draw the line along this continuum and declare someone good?  We have to face the fact that there is no good answer to that question.  Any answer we give is going to be arbitrary.  There will be some people who are just barely below the "good" line and they will be virtually indistinguishable from those who are just above the "good" line and it won't be fair to separate them into "good" and "bad" categories.  It can never be fair.  Wherever the line is drawn, it is an entirely arbitrary "benchmark".  So, if the mob decides that how you treated minorities is the benchmark for deciding if you should be honored with a statue, while they ignore other character qualities, such as whether or not you were a womanizer, well, then, that's the arbitrary benchmark and down come the statues.  Abraham...you're out.  Martin and John...you get to stay.  Is that going to be fair?  No.  There is no way setting a benchmark anywhere along a continuum is going to be fair.

 

             Actually, there would be one fair way to decide on the benchmark for goodness:  a benchmark of "perfection" would work.  There is a demonstrably fundamental difference between perfection and everything else.  That criterion can work as a yes/no category.  We're you perfect?  You get a statue.  Not perfect?  No statue.  But that criterion fails when you consider the teaching of a majority of major religions ("all have sinned") and the commonly accepted view of just about everyone else ("nobody's perfect").  In fact, I would suggest that the real distribution of "goodness" for the human race is about what is shown in Figure 3.  You might think of it similarly to the distance human beings can swim compared to the distance needed to swim across the Pacific Ocean.  Perfection isn't a high bar - it's an impossible bar.  There should be no disputing that fundamental fact of human nature.  So, while perfection might be the best benchmark, it's kind of useless for determining who gets a statue or who doesn't.

             I'm going to finish by extending the Benchmark Problem to its logical conclusion.  A benchmark for goodness (or any other human quality) is a difficulty that pervades a lot of the decisions we have to make in society, not just selecting statues.  It affects how we view ourselves, how we view others, and ultimately how we view our place in the universe.  Even the meaning of life espoused by many religions (both spiritual and secular) seem to lean heavily toward the Figure 1 view of humanity.  In my opinion, this is a fundamental problem with most religious and secular concepts.  If these concepts have, at their core, some semblance of the Figure 1 view, then they are based on a false assumption.  Thus, if there is some stated or unstated performance level (i.e. benchmark) in order to obtain or achieve some positive goal (heaven, life success, etc.), then such a concept can never be fair.  Any such system will have to have some arbitrary dividing line and some people will barely achieve the goal and some barely miss it.  That is a much more significant problem than deciding who gets a statue and who doesn't. 

 

             It is the Benchmark Problem that was key in driving me to explore Christianity.  Christian teaching takes a unique (as far as I have found) approach to the Benchmark Problem because it uses the perfection benchmark as the foundation.  Christian teaching says that no one qualifies for heaven.  "All have sinned."  That fits with my observation of human nature (Figure 3).  No one is even close to perfect.  As I have discussed elsewhere, that's why I consider Christianity to be reasonable.  This does not, of course, prove that Christianityis true in its entirety.  For that you have to examine other aspects.  But any religious or secular view that is based on Figure 1 is, for me, a non-starter.

 

             I doubt we will ever really agree on a benchmark for who gets a statue and who doesn't.  But I do think that an honest, open discussion of the Benchmark Problem and our underlying concept of human nature is a good thing for all of us.