Hacker News new | ask | show | jobs
by dsg42 3696 days ago
You're right, that isn't really the best statement of the paradox. Try this:

Imagine you have a heap of sand. Now, if you remove a single grain of sand from a heap, it's still a heap, right? But if you remove every grain of sand, then eventually, you're left without any sand, which is definitely not a heap. So somewhere along the way, when removing individual grains of sand, the heap stopped being a heap. But where, exactly, did that happen?

5 comments

It happened at the exact moment when none of the grains were supported by any of the others. A flat, single grain layer of sand on the ground is not a heap no matter how many grains are in it.
Soros in Greek does not have the structural connotation of heap in English (which I infer from your reply, given English is not my mother tongue).
It stopped being a heap at the moment at which there were no longer both:

(1) more than one vertical layer (the requirement for it to be a pile), and

(2) each grain not directly in contact with the ground being above a point on the ground within the triangle defined the points on the ground directly underneath three lower grains (distinguishing a heap from other piles).

(Of course, you may have a different definition of a heap, but if once you make concrete what you mean by a "heap", the problem is resolved. If you view heap as a fuzzy concept, then there is a continuous-valued fuzzy membership function, and there is no crisp point at which the collection stops, or starts, being a heap.)

Just use a different example, e.g. baldness, or the pumpkin one from http://philpapers.org/archive/FRAQTV.pdf
For very blocky, stackable grains of sand, it could've happened at 2. It undoubtedly happened somewhere below 9. That's as exact as this needs to be, unless you are a pedant. In which case, fuggedaboudit!
To describe this concept I've always used a boat as the subject.

Imagine you have a boat. Now take a piece off of it. Is it still a boat? How far can you go until it no longer holds the distinction of being a boat?

I imagine 'when it sinks' would work for most people.
So if I understand correctly, then twigs and leaves could be boats? :) I guess if a boat was originally constructed of a bunch of twigs (some of which still had leaves on them), but caught on fire. One lucky leaf was knocked off and managed to survive the fire. Everything else sunk. The single remaining artifact of the boat was that one leaf. And it's still floating... Everyone looked at me perplexed when I told them all that we were looking at a boat! :)
And if it is constructed entirely from wood that floats?
Then it may still function as a boat. As in all of these examples, X-iness is context-dependent, but for some boats, unlike the other examples here, there actually is something approaching an objective decision procedure.

Yes, I know that submarines are colloquially known as (and are in fact) boats.