The human brain is hard-wired to look for agency and intentional design where there is none. It's an evolutionary remnant of our hyper-active agency detection device (HADD). It's helps explain why conspiracy theories are so popular, why we see zombies on our fish-sticks, and why we tend to assume that if something looks designed, that it just must have had a designer.
Of course by now we should know that our HADD is just playing tricks on us, finding purpose where there is none, giving way to what Ullmann-Margalit calls our "artificer bias". We should recognize that just because something looks like it was purposefully constructed doesn't necessarily mean that it was. But it's not easy to ignore hundreds of thousands of years of evolutionary instinct, and people still routinely make the same mistakes.
Ullmann-Margalit defines the natural realm as consisting of everything that is the result of neither human action, nor design. But that hasn't stopped people from letting their HADD-induced artificer bias get the best of them, looking at nature for evidence of action and design from their imaginary friends in the sky. (Richard Dawkins coined the term "designoid" to describe natural phenomena that look "intelligently" designed, but in reality are not.)
Things get a little more complicated when we introduce the artificial, or "social" realm though. Now we're dealing with phenomena that are the result of both human action and human design. What happens when the two realms overlap? Adam Smith's Invisible Hand, the decentralized coordination of the market process that results from the pursuit of our individual self-interest, is the result of our collective actions, but not our collective design or intention. Thus in a way Invisible Hand Explanations represent the intersection between natural and social phenomena.
Economists devote the bulk of their efforts to the study of these issues. This might help explain why even though they typically describe themselves as "social" scientists, economists (probably more-so than other "social" scientists) tend to fancy themselves practitioners of the "harder" sciences.
The article doesn't really deal with this, but I wonder what types of phenomena we could consider the result of human design, but not human action? Can such a thing exist? If so what would we term it?