Behind Door #23: Not So Random Random Numbers

If I wait long enough, some blogger will almost inevitably write about a question I’ve long puzzled over. The fine folks at Language Log, seem almost clarivoyant in this regard.

I’ve often wondered why it is that when we pick an ostensibly random number to use in a situation where we’re estimating an exaggerated amount, we often choose the same number or variations on the same number. Yesterday Arnold Zwicky provided a pretty compelling argument for why that number is often seventeen—for American anglophones, at least.

As interesting as I find his discussion, I frequently find myself reaching for numbers starting with eight, though: eighteen, eighty, eight hundred thousand—depending on what order of magnitude I want to exaggerate. When I lived in Sweden, I noticed that Swedes use a made-up number—femtioelva—fifty-eleven.