## Infinity and Paradoxes

For the discussion of math. Duh.

Moderators: gmalivuk, Moderators General, Prelates

undecim
Posts: 289
Joined: Tue Jan 19, 2010 7:09 pm UTC
Contact:

### Infinity and Paradoxes

I'm looking for interesting paradoxes involving infinity.

One example I thought of last night:

$\sum_{k=1}^{\infty}k - \sum_{k=1}^{\infty}2k-1 = 2\sum_{k=1}^{\infty}k$
$\left(1 + 2 + 3 + 4 + \cdots\right) - \left(1 + 3 + 5 + 7 + \cdots\right) = 2 + 4 + 6 + 8 + \cdots$

In other words, if you subtract all the odd terms from the series, that's the same as doubling all the terms.

As I understand it, this is roughly equivalent to the Banach-Tarski paradox

EDIT: sorry, I failed at basic arithmetic... This doesn't work with the harmonic series
Blue, blue, blue

LLCoolDave
Posts: 165
Joined: Wed Apr 11, 2007 10:17 am UTC

### Re: Infinity and Paradoxes

The difference is anything but well defined, it depends massivly on how you put your brackets. Conversly, I could argue that each number in the second series is at least as big as the corresponding one in the first sum, so I get

$(1-1)+(2-3)+(3-5)+(4-7)+...= \sum_{k=0}^\infty -k = - \infty$

The issue is that the difference of these two series is not well defined, depending on just how I construct the differences of partial sums I can get very different results in the limit, and there is most certainly no "one correct" way to do it.

This is in no shape or farm related to the Banach Tarski Paradoxon at all, it's just an ill defined term (and therefore a nonsensical conclusion).

z4lis
Posts: 767
Joined: Mon Mar 03, 2008 10:59 pm UTC

### Re: Infinity and Paradoxes

Yeah, I wouldn't say that this is roughly equivalent to the Banach-Tarski paradox... however, I would say that they have in common the idea of paradoxical decompositions of "infinity". At any rate, if you liked what you came up with, you would probably really be blown away by what Riemann had to say about bad series:

http://en.wikipedia.org/wiki/Riemann_series_theorem

It basically says that if a series converges, but that's because of cancellation (that is, the sum of the absolute values of the terms goes to infinity), then if you give me any real number, I can rearrange the terms of the original series to add up to that number. I can also rearrange them to go to infinity or negative infinity. It's a pretty strong indication about how commutativity breaks down under taking limits.

If you want more stuff on infinity and its paradoxes, there are plenty of interesting philosophical and mathematical questions floating around on it, since we've been collectively wrestling with the concept since its inception. If you haven't already, you might also want to read some set theory and learn about cardinality, where you'll see some statements about how there are different "sizes" of infinity. An "infinite" number of different sizes, of course.
What they (mathematicians) define as interesting depends on their particular field of study; mathematical anaylsts find pain and extreme confusion interesting, whereas geometers are interested in beauty.

Eastwinn
Posts: 303
Joined: Thu Jun 19, 2008 12:36 am UTC
Location: Maryland

### Re: Infinity and Paradoxes

What you've shown is manipulation of symbols without meaning. A divergent sum doesn't actually represent a value, so how can you subtract them? Just another example of how intuition seems to fail us when dealing with infinities.
http://aselliedraws.tumblr.com/ - surreal sketches and characters.

doogly
Dr. The Juggernaut of Touching Himself
Posts: 5532
Joined: Mon Oct 23, 2006 2:31 am UTC
Location: Lexington, MA
Contact:

### Re: Infinity and Paradoxes

Here's your problem, you forgot that
$\sum_n n = -\frac{1}{12}$
From there it should be easy to see where to go.
LE4dGOLEM: What's a Doug?
Noc: A larval Doogly. They grow the tail and stinger upon reaching adulthood.

Keep waggling your butt brows Brothers.
Or; Is that your eye butthairs?

ImTestingSleeping
Posts: 88
Joined: Mon Dec 06, 2010 3:46 am UTC

### Re: Infinity and Paradoxes

You can't rearrange a series which isn't absolutely convergent.

If you are looking for something strange, then I would offer the fact that for any conditionally convergent series [imath]\sum a_n[/imath], for any number x there exists a rearrangement of [imath]\sum a_n[/imath] such that [imath]\sum a_n[/imath]=x.

 Whoops, didn't notice z4lis already linked Riemann's theorem in an above post. Neat stuff.