Read on Twitter
Anyway Tweeps I am ready for a short thread. Earlier I tweeted ‘We could not find any evidence of institutional racism’ is racism. It is kind of racism that is epistemic and I would say typically English. A few reflections. I give this example in many of the trainings I deliver.
A couple of years ago, I did some research to find out the proportion of race related complaints which had been made against the police force, in this a particular year (don’t ask...)The response to a freedom of information request was helpfully available in the public domain 
In this particular year (I will try to dig the data) hundreds of racism complaints were received (I think it was well over 1000 - don’t quote me but again, the point would stand even if there was only 100 complaints received and investigated).
In this particular year...a grand total of exactly ZERO complaint was upheld. Not 0% which may mean a few complaints made it, actually zero. I thought I was misreading. I checked again and again. I was reading correctly.
In this particular year...(I suggest this year is likely to be representative of other years) for every single person of colour (or possibly white bystander) who reported racism was told either 1) it did not happen or 2) most likely - ‘we could not find any evidence of racism’.
Now it’s interesting how these days I return again and again to my (rudimentary) understanding of statistics and probability. Anyone with a basic understanding of stats will understand that such an outcome is very, very, very, very unlikely to ever happen by chance alone.
Now to obtain such an outcome (or a similar one), year after year and to have it replicated within other institutions makes the chance of this outcome reflecting reality...
Virtually null. Speaking statistically, we may say the probability of making a type II error is virtually 1 (or we may say 100%). In stats type II errors occur when we accept the null hypothesis (here there is no racism) but that hypothesis is actually false.
Let me try to simplify for those who may struggle to understand what I mean. Generally, if things are not skewed...we can generally assume outcomes of experiment to be equally likely (unless things are skewed).
That is IF there is no skew...and so if you are investigating a situation (or testing a hypothesis) with only two possible outcomes (here there is racism vs there is no racism) we would expect by luck alone...that outcomes (of investigations)would be more or less equal.
That is to say in short...we would expect 50% of complaints to be upheld, again...if there was no skew. Now why do we get such extraordinarily outrageous of outcomes of 0%?
And have our intelligence insulted again & again as we are forced to accept this as reality...cough (institutional racism enters the chat).
Anyone should be alarmed by the fact ‘they can’t find any evidence of racism’ not 10, not 100, not 200, not 500 but over 1000 times year after year. Anyone with an ounce of reason & rationality (attributes many love to claim...) would stop & think...what the hell is going on!
And clearly see there is a skew in the system. Or that clearly those two outcomes are not equally weighted or unconditional. This my friend is logic. It is pure logic. Not idealogy*. Not ‘liberal wokeness’. Just simple maths.
So to return to the main point... why I say ‘we could find any evidence of racism’ is in fact racism...is simply because such an outcome is simply the product of systems designed not to ‘see’ evidence of racism.
And systems which can pass this outrageous state of affairs not only as non-eventful but as reflective of reality.
In other words systems which can twist & turn to ensure their preferred & distorted version of reality is presented as ‘factual’ and scientific hence claims re: ‘evidence’ (I did a thread before about how such a move is a power move).
That is why such (epistemic) manipulations function to reinforce the illusion of truth & ‘objectivity’ but if you scratch the surface you find what? Utter irrationality, deception & lack of basic logical thinking - fin 
