When Random Isn't Random

This post is partly inspired by the distant memory of an ex-something of mine who would insist; "There's no such thing as random." Despite the fact there was greater scope for intelligent discourse with my big toe than with the individual in question, the statement raises an interesting point to consider. As the concept of random deserves a dedicated post, this is a precursor to my next post in the Karma series.

Define 'random.'
References are good. In order to avoid having a sea of links or pages of text, I've edited out additional definitions from the sources below where the meaning is adequately captured by another source.

I've included some definitions that I personally believe are wrong. Common usage or not, in terms of capturing the core concept of random - dead fucking wrong. I've done this, partly to poke fun at particular professions, but also to highlight the extent to which this term has been applied to so many different meanings, some being mutually exclusive, that it has essentially become a meaningless term without a specific context.

Apparently, random is...

(1) "lacking any definite plan or order or purpose; governed by or depending on chance."
Princeton website

(2) "a lack of order, purpose, cause, or predictability." (more)
Wikipedia

(3) "is often used in statistics to signify well-defined statistical properties, such as a lack of bias or correlation."
Wikipedia

(4) "an event for which there is no way to know, before it occurs, what the outcome will be. Instead, only the probabilities of each possible outcome can be stated."
Cliffs Notes

(5) "the equal chance of any occurrence."
Marketing and Geography education resources

(6) "an undefined, unknown or unimportant person; a person of no consequence;"
Wiktionary

(7) "(describes) a happening or event due to chance and not determined by other factors."
A Dictionary Of Epidemiology

(8) "sometimes used as a colloquialism for nonsense, e.g. for outburst that are non sequitur."
Wikipedia dispute page for 'randomness'

And the list goes on. Others include, "haphazardly", "at great speed", "an undesirable person."

Etc, etc, etc.

Wow, you're right. The term is all but meaningless.
Yes and no. Whilst it may be near impossible to neatly box it up into a single definition, it does express a concept and idea that has to do with unpredictability and, by extension, unfathomability. So let's see if we can't tighten up the definition some by referencing it to that with a little insight and integrity...

Those wacky statisticians
As per definition (3), a random distribution of events is defined as a series of events that follows a uniform and predictable pattern, where the greatest number of events occurs in the middle of the range of options (ie, the average, aka mean), and each of the extremes (variables that are the largest number of standard devs away from the average) represents the fewest number of occurrences. Also known as a Gaussian distribution or 'normal' distribution. (Interesting, no?)

This is relevant because, in the absence of any other information, statisticians assume in their modelling, with some success that a distribution is 'random.' Why does this work? Because, when you're looking at a series of data, no matter what the data is (length of cats claws, height of humans, etc), the greater your sample is, the more likely it is that the pattern of events will resemble this kind distribution. Whilst this has some very important ramifications for random as a whole concept, I'm going to largely leave the technical aspect behind as we clear out some of the guff associated with this term.

The Elimination Game Time eliminating definitions for the purposes of sanity and maintaining the scope. As you will see from below, the following definitions are, in a word, bollocks;

Random as a synonym for the non sequitur, the nonsensical, the absurd: As can be observed, there are a variety of terms that are a) more descriptive of the event and b) intrinsically less confusing for the listener that ought be used in this context instead of random. *definition lands in the trash with a swoosh.*

Random as a way of describing strangers, friends of friends, casual observers, people that you may or may not have met yet: As for previous.

Random as a way to describe something "with no cause": This concept is a myth. I say with an enormous amount of confidence that every single event that we are ever likely to experience in this lifetime has a cause. If it doesn't have a cause, it has several causes. Whether there's any intelligence, behind the cause(s), whether we can accurately discern the nature of these causes - irrelevant. The cause is there. In the same way that it doesn't work logically to put the cart before the horse, you can't have an effect without a cause. Busted.

Random as a way to describe an equal chance of any occurrence: What a load of crap. And yet, I will recognise that it is common usage to describe the event of any number coming up on a dice as 'random' despite the fact that the chances of any specific number coming up are equal.

Whilst I am guilty of the offence of using the term in one or more of the following incorrect contexts, I do not condone it. The definitions as per above are symptomatic of lazy thinking, and are best avoided.

When *is* random not random?
In to assist dissecting delve into the more woolly interpretations of the term, I will quote one more source;

This article presumes a priori that randomness means irreducible randomness, and 'essentially' pretends that reducible randomness doesn't exist. This is in spite of the fact that reducible randomness is the only form that can be definitively demonstrated, in special cases.
Wikipedia dispute page for 'randomness'

From this and the disparity between some of the remaining def'ns, it would seem that you can divide random into two categories;

Irreducibly random - ie How the fuck did that happen? I had no basis for any reasonable assumption that that was going to occur.
Reducibly random - ie OK, there was a likelihood that this would occur. Maybe I felt like I knew enough to put a number on it. But I didn't know it was going to happen, indeed, it wasn't even likely, given the number and/or probability associated with the alternatives. And yes, the existing uncertainty associated with the (supposedly) random object.

This is all very well and good, as far as it goes. Only one problem - the absence of parameters. Due to the level of reasonable uncertainty which we determine and predict things as human beans, anything can be classified as reducibly random. My car starting in the morning could be random. The sun rising in the morning could be construed as reducibly random. (It could get hit by another universe and be wiped out of existence before tomorrow.)

My question would then be - at what point do we say, 'Although we cannot perfectly predict the outcome relating to this scenario, we are sufficiently familiar with the causes relating to the situation, and have a reasonable ability to foresee what's going to happen (ie when I roll a six-sided dice, it will come up with an integer between 1 and 6) such that it has sufficient predictability that it can no longer be considered random'?

I take the point that we can't prove whether something is irreducibly random, only when it's not. However I believe there's a danger in assigning too much to the term 'reducibly random.' Surely when we have certain level of knowledge of related causes and effects, whatever the outcome turns out to be, we can say 'Yes. I hereby declare this thing un-random." Stamped, sealed and delivered.

Either that, or we need to start coming up with some more terms to describe the various degrees of random. Clarity of language people. I demand clarity of language.