Politics and statistics

There’s been discussion for years now about how useful political terms like ‘left’ and ‘right-wing’ are. ‘Socialist’, ‘labour’ or ‘left-wing’ political parties nowadays seem to be more ‘conservative’ than their right-wing rivals, who keep changing society into a privatised paradise and should therefore be classed as ‘progressive’. Historically, left-wing parties emphasised human equality against more class-sensitive conservatives. But today it’s the right wing that has brought a new ring to the concept of equality: everyone is individually responsible for his own fate and therefore equal. It’s the left-wingers that protest against this by pointing out that not everyone is equally able or capable, either by nature or by nuture, to take on the full load of his or her personal responsibility. Those who used to be the champions of equality, now emphasise human inequality. So should we really keep the terms ‘right’ and ‘left’?

I still find them useful, provided they’re defined anew. Luckily statistics offer a very useful tool: type 1 and type 2 mistakes. In statistics you test a null-hypothesis, which is either accepted or rejected. A type 1 mistake occurs when you reject an hypothesis that is true. A type 2 mistake is accepting an hypothesis which is false. That can be translated into something more societal. An example: in our legal system an accused person is considered not guilty until proven guilty. That’s the ‘null-hypothesis’. If a judge makes a type 1 mistake (rejects the correct hypothesis ‘not guilty’), he convicts an innocent suspect. When he makes a type 2 mistake (accepts the false hypothesis ‘not guilty’), he sets a criminal free.

Every judge -and statistician- knows that avoiding both mistakes is impossible. At best you’ll avoid one, at worst you’ll make both. So society has to choose which way to set up their legal -and loads of other- systems, and which type of mistake it is more willing to accept. Ever since the enlightenment, western culture has preferred avoiding the risk of sending an innocent person to jail over avoiding the risk of setting a criminal free. Up until the seventies the same was true for lots of other questions in society. We preferred giving some of our welfare money to profiteers who really could work over refusing welfare to people who should have been entitled to it. We allowed refugees in who we knew weren’t always kosher refugees instead of sending home the really desperate.

After that time, in all three examples mentioned, a shift can be observed in the west. There are more protests against suspected criminals set free because of ‘technicalities’, even when we know it’s exactly those ‘technicalities’ that make the difference between the rule of law and the rule of random order. More and more people needing welfare are denied government help, even to the point of homelessness, because they’re considered ‘fit for work’. Refugees are pushed into an existence as ‘illegal aliens’, even when we know their return is actively prevented by their homelands. We’re moving from avoiding type 1 mistakes to avoiding type 2 mistakes.

Historically, the preference for avoiding type 1 mistakes is associated with left-wing politics. The present shift towards avoidance of type 2 mistakes can be ascribed to political movements more on the right. So my redefinition of ‘left’ and ‘right’ would be: ‘left’ means a preference for avoiding type 1 mistakes, ‘right’ means a preference for avoiding type 2 mistakes.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s