# Is the size of the territory of a country purely random? [closed]

I read a blog post that attempts to show that the size of the territory of a country is purely random and therefore bears no major historical significance.

The argumentation (simplified a bit) is as follows:

• If we plot sizes of countries on a logarithmic plot the distribution is linear:

(Each bar is a country, the Y axis is the size of countries in km2; blue bars are "large" countries, orange bars are "medium" countries and grey bars are "small" countries. Countries are labelled "large", "medium" or "small" because, according to the blogger, the exponential distribution has different arguments depending on whether the country is below or above 10000 km2, so in fact there are two distributions, as shown by the straight lines in the plot)

• Therefore, if plot on a linear plot, the distribution will be exponential
• Exponential distribution occurs if the events are purely random
• If it was not random whether a country gained or lost a territory we would expect a Pareto distribution instead

This seems to make sense, but... at the same time it is very counteintuitive. Does it follow that a country competently governed is equally likely to lose a territory than a failed country? The blogger himself says this is not the case, but he claims that a country that is governed competently is likely to be governed far worse after a few generations and will start losing territory. One very profound conclusion after another...

Plus, the blogger himself claims that his findings are contrary to established knowledge.

Is it purely random if a country gains or loses a territory?

• "Exponential distribution occurs if the events are purely random" whilst true, not all white birds are swans. There are many claims here. Commented Oct 9, 2021 at 18:17
• Until such time as they define what they mean by “purely random”, this sounds like complete nonsense (to me, a mathematician with expertise in probability theory). It’s common to describe such claims as “not even wrong”. But this is based on your description, which I cannot verify since I can’t read the original post. Commented Oct 9, 2021 at 18:22
• I second @DanRomik's point: "purely random" is not a thing. From your other context, what the blog is probably trying say is that land area is the product of many small independent pieces of noise? However, I'd then expect a lognormal distribution rather than exponential? Since, there are an almost unlimited number of explanations that would result in very different distributions, this whole exercise is rather pseudo-scientific. You really can't reach any general conclusions with any certainty based on this limited evidence. At best, you can say an extremely specific hypothesis is wrong. Commented Oct 10, 2021 at 0:08
• (The hypothesis that it's not "purely random" is woefully underspecific not least because "purely random" is not a defined term.) Also, Enrico Fermi's quote applies here: "with four parameters I can fit an elephant, and with five I can make him wiggle his trunk." Commented Oct 10, 2021 at 0:11
• @gaazkam There is no way of winning an argument with those who have fully inoculated themselves against logic and reason. Commented Oct 10, 2021 at 14:12

Contrary to popular belief, no "natural boundaries" exist. All borders between states, with the exception of really great countries, which occupy a large part or even the whole (Australia) of continents, are purely random.