Reference Points

There are three arguments for mass Muslim immigration into Europe.

1. It’s the right thing to do, you heartless racist.
2. Neo-liberal growth, GDP, Keynes!
3. The European Union has over 500 million inhabitants. It can easily absorb a few million Arabs.

I want to focus on that third argument because it demonstrates an interesting rhetorical tactic.

Whether or not a socio-political phenomenon counts as a “thing” or a “problem” or an “event” largely depends on the population, in a statistical sense, used to frame the phenomenon. It depends on the reference point.

In the context of immigration, that third argument sets the population at n=all. (Yes, I know n=sample size, not population, but I’m trying to make a point here.) Maybe a better way to put it is that the third argument above situates the population of immigrants within the population of Europe, thus reducing the scope of the immigration phenomenon. A million immigrants in the context of 500 million Europeans is a blip on the radar, at .2%. Stop making a mountain out of a molehill.

The immigration skeptic, on the other hand, would situate the population of immigrants within the population of those European cities and towns that are actually housing immigrants. For example, many of these immigrants would like to make it to London specifically. London has a population of 8 million inhabitants. A million immigrants in the context of 8 million Londoners is not a blip on the radar. Nor is 100 immigrants a minor thing in the context of a tiny dorf in Germany. The impact of the immigration phenomenon is highlighted when the reference point is set in this more limited way.

We find the same expansion/reduction of impact when the discussion shifts to the social phenomenon of police killings (and killings of African Americans in particular).

It is probably safe to assume that the same immigration booster who set the population at n=all when discussing immigration would limit the population size when discussing police killings. (Likewise, the same immigration skeptic who limited the population size when discussing immigration might very well be the sort of person to set the population at n=all when discussing police killings.)

The Bureau of Justice Statistics estimates that between 40-45 million Americans come into contact with the police every year (about 18-20 million of those contacts are in the form of traffic stops). If we place the ~500 annual police killings into the context of all 40 million police-citizen interactions, then the impact of this phenomenon is seriously reduced. I’ll let you do the math. The vast, vast majority of police interactions do not end in anyone—black or white—getting killed. Stop making a mountain out of a molehill.

However, most discussions of police shootings limit the population to those individuals who have been shot by police. Even this recent NYT article, which looks at the larger context of police shootings, limits itself to comparing police killing rates of African Americans to arrest rates of African Americans.

~~~

On a long enough timeline, the survival rate for everyone drops to zero. Likewise, in a large enough context, many social phenomena cease to be “things,” “events,” or “problems.” In the context of the universe, Earth’s destruction by a meteorite would be a molehill, not a mountain.

But on the other hand, in the context of my own life, getting a cold is a major event.

Whether a social phenomenon counts as a mountain or a molehill depends on your reference point. Changing the reference point changes the scope of any given phenomenon’s impact. Want to make a phenomenon disappear? Expand your reference point. Want to call everyone’s attention to it? Reduce your reference point.