Skip to main content
No. 1367:
Struggling with Abstraction

Today, we think about thinking abstractly. The University of Houston's College of Engineering presents this series about the machines that make our civilization run, and the people whose ingenuity created them.

a.jpgbstraction is such an off-putting word! An object-oriented society turns it into a bad word. It's good to be down-to-earth; it's effete to be abstract. Yet we live by abstractions all the time. I can say a scarf is green without showing anyone the scarf. You know the color of leaves, or the backs of dollar bills, or moldy cheese. You've abstracted vegetation and money to the point where you understand green apart from any object.

We engineers would be hamstrung without abstraction. We absolutely have to reduce real machinery to frictionless pulleys, point masses, and perfectly insulated bodies. If we didn't, we'd drown in intractable detail. We'd never get around to making anything.

Here's an example: Heat flows through a slab of steel heated by boiling water on one side and cooled by ice cubes on the other: The boiling water agitates atoms on one side -- makes them jiggle faster. Those atoms agitate the atoms next to them, and so on. On the other side, the ice slows the moving atoms and absorbs their energy. As you move from hot to cold, each atom passes energy on to the next one. Finally all that motion passes into the ice. And the ice gradually melts away.

It's completely impractical to calculate heat flow by describing such a complicated process, so we do a strange thing. We go back to an 18th-century abstraction. Before we had an atomic theory, we imagined that an invisible substance called caloric flowed from hot objects to cold ones. There is no such material, but no matter. You can describe heat flow just fine by imagining caloric. Our mathematical theory of heat flow still comes to rest on an 18th-century abstraction.

Hardly anyone remembers that invisible and mystical essence, caloric. Yet we still use the idea. We abstract reality in a way that's unrealistic. When would-be realists catch on to that, they shrink back, even though we'd flounder without it.

So realists with good sense do set up abstract parallel universes. When an invented world truly runs parallel to the messy world around us, then a problem solved in the imagination is also a problem solved in the real world. The public, uncomfortable with the ways of science, fights that kind of down-home good sense.

A person might chance thousand-to-one odds of having a car accident on vacation, yet be terrified by million-to-one odds of getting cancer from Alar. But then statistics is one more abstraction of human experience that we use to make sense of reality -- one more locus of discomfort if we need things cast in concrete. Shaping worthwhile abstractions means extracting the essence of reality from the clutter of sense data. It is, quite literally, a matter of seeing more -- by knowing how to see less.

I'm John Lienhard, at the University of Houston, where we're interested in the way inventive minds work.

(Theme music)

Love is blind
And being blind,
Sees less,
And seeing less,
Sees more.

(Source unknown)