Mysteries include questions of the form “Who did that?” and “Why did that happen?” This latter case can be lumped in with troubleshooting, but we are free to separate cases where our unmet expectations are so surprising that we think of the case as mysterious rather than troublesome. (Note the distinctions we make here guide our investigations; the distinctions do not determine the course of our investigations. We determine the course, which is to say, we still say we “troubleshoot” some mysterious behaviours.)

Mysteries involve breaches of protocol or transgressions of law. We don’t expect to find patterns, and so we don’t expect to find systematic causes. Instead, in the case of mystery, we aim to find a perpetrator or ghost in the machine. (The “ghost in the machine” is a metaphor introduced into philosophical talk in 1949 by Gilbert Ryle in his “The Concept of Mind”. The metaphor expresses an impulse to investigate unseen, immaterial forces that might control the behaviours we see in the world. When we have trouble explaining behaviours in the world, we might be tempted to attribute those behaviours to such unseen forces. But here we take those cases to be explainable by seen, public forces, and within our system, we aim to describe those forces and give an account of how they work, thereby solving a mystery.)

In some cases, mysteries are deeply unmet expectations, and in this sense, the distinction is one of degree, not of kind. The value of the distinction between troubleshooting and mystery, then, is to indicate the heavy burden explanations bear in these cases.

For example, my friend worked at a restaurant next to an old cemetery. The restaurant closed late, and sometimes we would help her to close the place before we all went out. We would walk through turning off lights as we went, checking that doors and cupboards were locked. Reliably, when we got outside and looked back at the restaurant, one particular light would have turned back on. She said it was a ghost, an unseen force, a mystery. Obviously none of us expected a light to turn back on after leaving the building, but we all saw it happen. We could say this is a problem that stands in need of competent troubleshooting. But here our expectations are so deeply unmet we’re not even sure where to begin investigating. That’s a mystery.

Example 2.2
The Ghost Light

LQ: Why does the light come on after we leave the building?
EE1: After turning off the lights, one always comes back on when we leave the building.
RA1: The building is haunted.
RA2: Vibration from closing the front door causes the light to come back on.
ER1: There is a cemetery nearby.

Here we don’t have much to go on. We include R1 because, for as far-fetched as it might seem, at least there’s something to get the investigation under way. We offer A2 as a way to connect the front door to the light, but we’re equally as justified in saying that when we step on a certain floorboard the light switches back on, or that the front door is, itself, a switch, wired as such as a hoax. The point is: it’s unclear where to even start, given how little evidence we have, and how deeply unexpected the evidence is.

Commonly, mystery solving, which we call “detecting”, involves a greater amount of evidence and resources than troubleshooting anomalous behaviour, partly because it isn’t always clear what the problem is or where to begin an investigation. Given this, we will treat troubleshooting and detecting separately.

(We never figured out why the light would go back on, and A1 is still a favourite explanation among this group of friends.)

Last modified: Wednesday, 7 February 2018, 9:49 PM