## 0^-1

For the discussion of math. Duh.

Moderators: gmalivuk, Moderators General, Prelates

Whirligig231
Posts: 3
Joined: Thu Oct 06, 2011 8:58 pm UTC

### 0^-1

How is 0^-1 for sure undefined? The only proof I've seen for x^-y = 1/(x^y) assumes x^0 = 1.

gmalivuk
GNU Terry Pratchett
Posts: 26824
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

### Re: 0^-1

That's not so much something to prove as it is the definition.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

antonfire
Posts: 1772
Joined: Thu Apr 05, 2007 7:31 pm UTC

### Re: 0^-1

You seem to be under the impression that 0^0 is not 1.
Jerry Bona wrote:The Axiom of Choice is obviously true; the Well Ordering Principle is obviously false; and who can tell about Zorn's Lemma?

Qaanol
The Cheshirest Catamount
Posts: 3069
Joined: Sat May 09, 2009 11:55 pm UTC

### Re: 0^-1

antonfire wrote:You seem to be under the impression that 0^0 is not 1.

The set of functions from the empty set to itself definitely has cardinality 1.
wee free kings

skullturf
Posts: 556
Joined: Thu Dec 07, 2006 8:37 pm UTC
Location: Chicago
Contact:

### Re: 0^-1

I'm of course aware of the reasons behind the convention of sometimes leaving 0^0 undefined, but notice that you need 0^0 = 1 in order for the binomial theorem to work for (x+y)^n when one of x,y is zero.

yurell
Posts: 2924
Joined: Sat Nov 13, 2010 2:19 am UTC
Location: Australia!

### Re: 0^-1

I thought 0-1 was undefined because in our axioms (or at least the set I learned) we state that every number except 0 has a multiplicative inverse. And if you assume it does have a multiplicative inverse you can prove just about anything, including contradictions.
cemper93 wrote:Dude, I just presented an elaborate multiple fraction in Comic Sans. Who are you to question me?

Afif_D
Posts: 184
Joined: Wed Oct 06, 2010 2:56 pm UTC

### Re: 0^-1

Actually(0^-1) doesnt make sense. Any negative power of zero doesnt make sense. If it did make then 0^n (n is a positive integer) would not equal zero and could be proved to be an indeterminate form.

Like

0^2=0^(3-1)=(0^3)*(0^-1)=0/0

So the step in which i make a negative power of zero itself is wrong. Thus is doesnt make any sense..

Diemo
Posts: 396
Joined: Mon Dec 03, 2007 8:43 pm UTC

### Re: 0^-1

What I thought of it is 0-1 =1/01=infinity?
In the beginning the Universe was created.
This has made a lot of people very angry and been widely regarded as a bad move.

gmalivuk
GNU Terry Pratchett
Posts: 26824
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

### Re: 0^-1

Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

eSOANEM
:D
Posts: 3652
Joined: Sun Apr 12, 2009 9:39 pm UTC
Location: Grantabrycge

### Re: 0^-1

gmalivuk wrote:But it isn't infinity. Defining it as *anything* leads to contradictions.

On the Riemann sphere it's well defined.
my pronouns are they

Magnanimous wrote:(fuck the macrons)

Yesila
Posts: 221
Joined: Sun Dec 16, 2007 11:38 am UTC

### Re: 0^-1

Diemo wrote:What I thought of it is 0-1 =1/01=infinity?

but x^{-1}= \frac{1}{x} only for non-zero x. So assuming that 0^{-1} "should" equal 1/0 is the "wrong" decision (since we long ago decided that dividing by zero is a poor choice).

In the case of 0^{-1} I think a case can be made that if it was defined (to be a real number) it "should be" 0^{-1}=0. In that way we preserve the "add the exponents" rule for multiplication of like bases after all IF 0^{-1} is defined to be a real number then

0^{-1}=0^{1-1-1}= 0^1 * 0^{-1} * 0^{-1} = 0

Similarly 0^0 "should" be zero as well.

0^0 = 0^{1-1}=0^1 0^{-1} = 0 * 0^{-1} = 0

Of course as was pointed out above we often times would rather use 0^0=1, which then necessitates that 0^{-1} is not a real number in which case the OP can define it to be whatever they want, so long as they realize that if they want to tack that new element onto R and then "do" things with it they better make some new field operations, and shouldn't expect them to follow the rest of the "rules" that they know and use with real numbers.

thedufer
Posts: 263
Joined: Mon Aug 06, 2007 2:11 am UTC
Location: Northern VA (not to be confused with VA)
Contact:

### Re: 0^-1

Yesila wrote:Of course as was pointed out above we often times would rather use 0^0=1, which then necessitates that 0^{-1} is not a real number in which case the OP can define it to be whatever they want, so long as they realize that if they want to tack that new element onto R and then "do" things with it they better make some new field operations, and shouldn't expect them to follow the rest of the "rules" that they know and use with real numbers.

Specifically, if we add some element "inf" to the field of reals, then inf needs an additive inverse, so now we add "-inf". Then, we probably need to define our operations such that inf*a is 1 if a=0, inf if a>0, -inf if a<0, and opposite for -inf*a. Also, inf+a is 0 if a=-inf, otherwise 0, and similarly for -inf+a.
Now, inf*(inf + (-inf))=inf*0=1, but we can distribute and get inf*inf + inf*(-inf)=inf + -inf=0, and this 0=1. This is absurd, so the reals must not be a field anymore.
I'm pretty sure there's not a more sensible way to define the operations with inf and -inf, so you'd basically have to redefine addition and multiplication, or give up the various field properties that you take for granted on a regular basis.

Deveno
Posts: 29
Joined: Thu Oct 06, 2011 10:14 pm UTC

### Re: 0^-1

i don't want to put 1 over 0, because then there's nothing there, and 1 will fall down and maybe hurt himself.

in a word, no. just no. don't go there, do not divide by 0, do not pass go, do not collect \$200.

you'd have to throw away so many other genuinely useful rules, that there wouldn't be any arithmetic left.

but, hey: if you'd like a picture, just look at the graph of the hyperbola 1/x. be sure to get a "complete" picture, and not one of those cheap hong kong forgeries that only show a finite part.

arbiteroftruth
Posts: 481
Joined: Wed Sep 21, 2011 3:44 am UTC

### Re: 0^-1

thedufer wrote:Specifically, if we add some element "inf" to the field of reals, then inf needs an additive inverse, so now we add "-inf".

I disagree. Similar logic would say that if 0 is a real number, then 0 needs a multiplicative inverse. Either way, we're making an unusual exception to what is a nearly universal rule. It's just as valid to make that exception in the realm of 'inf' having no additive inverse as it is to make that exception in the realm of 0 having no multiplicative inverse.

One may perhaps argue that part of our definitions includes the notion that a multiplicative inverse only exists for "non-zero numbers", but this only demonstrates that common convention has already selected one arbitrary exception over another. One can construct an entirely consistent math system where a multiplicative inverse exists for all real numbers but an additive inverse exists only for "non-inf numbers".

Robert'); DROP TABLE *;
Posts: 730
Joined: Mon Sep 08, 2008 6:46 pm UTC
Location: in ur fieldz

### Re: 0^-1

arbiteroftruth wrote: One can construct an entirely consistent math system where a multiplicative inverse exists for all real numbers...

Really? I thought that was much of the point of not defining a div 0 in the first place.
...And that is how we know the Earth to be banana-shaped.

lightvector
Posts: 224
Joined: Tue Jun 17, 2008 11:04 pm UTC

### Re: 0^-1

arbiteroftruth wrote:One can construct an entirely consistent math system where a multiplicative inverse exists for all real numbers but an additive inverse exists only for "non-inf numbers".

Actually, how well can you really do this, without giving up a lot of other important axioms?

If you add an element inf and declare that 0 and inf are multiplicative inverses then we have 1 = 0 * inf = (-1 * 0) * inf = -1 * (0 * inf) = -1 * 1 = -1, which is a contradiction. The only properties we needed to derive this contradiction were the associative property of multiplication, the fact that -1 * 0 = 0, and the fact that -1 * 1 is not equal to 1. Giving up any of these three properties sounds terrible.

Additionally, what is -1 * inf? We have -1 * inf + inf = -1 * inf + 1 * inf = (-1 + 1) * inf = 0 * inf = 1. The properties used here are the fact that 1 is the multiplicative identity, the distributive property, and (-1+1) = 0. We can then use the associative property of addition, and obtain (-1 + -1 * inf) + inf = -1 + (-1 * inf + inf) = -1 + 1 = 0, which means that (-1 + -1 * inf) is an additive inverse of inf.

This means that inf actually does have an additive inverse, but -1 * inf is not it! That really really sucks. But if we want to avoid that, we have to give up 1 being a multiplicative identity, or the distributive property, or (-1 + 1) = 0, or associativity of addition, or we have to say that -1 * inf is undefined and make multiplication no longer total. Any of these is also terrible.

This mess indicates that the choice to make zero lack a multiplicative inverse is not arbitrary. If you want something that even remotely resembles normal algebra and arithmetic with all the usual useful properties, introducing a single exception for zero is much better than the alternatives.

arbiteroftruth
Posts: 481
Joined: Wed Sep 21, 2011 3:44 am UTC

### Re: 0^-1

lightvector wrote:If you add an element inf and declare that 0 and inf are multiplicative inverses then we have 1 = 0 * inf = (-1 * 0) * inf = -1 * (0 * inf) = -1 * 1 = -1, which is a contradiction. The only properties we needed to derive this contradiction were the associative property of multiplication, the fact that -1 * 0 = 0, and the fact that -1 * 1 is not equal to 1.

You have also implicitly used the principle that inf can be meaningfully multiplied by coefficients. Like 0, with which inf is closely related, inf would have the property that multiplication by any finite number (neither infinitely large nor infinitesimally small) results in no change in value. (-1 * 0) * inf = (-1 * inf) * 0 = inf * 0 = 1. When people construct proofs of contradiction that involve hidden cancellations of 0/0, the explanation is typically that 0/0 is not allowed to occur even when hidden by rearrangement of the notation. This would be a similar situation. You have introduced a coefficient of inf and hidden it with the right arrangement of notation.

One result of this rule that inf cannot have coefficients is that inf is an exception to the distributive property, as a side-effect of the principle that it has no additive inverse. The lack of an additive inverse makes it unsurprising that another rule involving addition might change. The analogous side-effect of the traditional option of saying 0 has no multiplicative inverse is that 0 is an exception to the property that x^0 = 1, or to the property that x^a * x^b = x^(a+b). If we allow both properties to apply to 0 like it does to all the other real numbers, then 1 = 0^0 = 0^(-1) * 0^1 = 0^(-1) * 0, meaning 0 has a multiplicative inverse.

One thing the above example may bring to mind is that another possible contradiction would be: 0 = 1 * 0 = (inf * 0) * 0 = inf * (0 * 0) = inf * 0 = 1. Meaning that inf must also be an exception to the commutative property of multiplication.

The end result is that inf can be included consistently in one's math system, but the only valid operation one can ever do to it is to multiply it by 0 and get 1. So it's inclusion in the system gains you nothing, but it creates no contradiction.

Calling zero's lack of a multiplicative inverse "arbitrary" was overly strong wording. My point was that it isn't a fundamental mathematical truth, but a convention of our usage.

EDIT: Technically, one can perform operations of exponentiation without contradiction. But the inability to apply multiplicative commutativity means this is still restricted to inf^a * 0^a = 1, with no right to equate 0^a = 0 before first multiplying by inf^a. So it's still useless.

jestingrabbit
Factoids are just Datas that haven't grown up yet
Posts: 5967
Joined: Tue Nov 28, 2006 9:50 pm UTC
Location: Sydney

### Re: 0^-1

arbiteroftruth wrote:One thing the above example may bring to mind is that another possible contradiction would be: 0 = 1 * 0 = (inf * 0) * 0 = inf * (0 * 0) = inf * 0 = 1. Meaning that inf must also be an exception to the commutative property of multiplication.

You haven't used commutativity in your calculation here, you've used associativity, which lightvector already pointed out causes problems.

arbiteroftruth wrote:The end result is that inf can be included consistently in one's math system, but the only valid operation one can ever do to it is to multiply it by 0 and get 1. So it's inclusion in the system gains you nothing, but it creates no contradiction.

So you're saying that you're defining inf such that a*inf is undefined unless a is 0? That's considerably more useless than the standard treatment.

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

arbiteroftruth wrote:Calling zero's lack of a multiplicative inverse "arbitrary" was overly strong wording. My point was that it isn't a fundamental mathematical truth, but a convention of our usage.

It perhaps isn't fundamental, but it isn't far from it. If a is any element of a set where there are binary operations addition and multiplication, then

\begin{align*}0 &= (a \cdot 0) - (a \cdot 0) \qquad &\text{(existence of additive inverse)}\\ &= (a \cdot (0+0)) - (a \cdot 0) \qquad &\text{(0 is the additive identity)}\\ &= ((a \cdot 0) + (a\cdot 0)) - (a \cdot 0) \qquad &\text{(multiplication distributing over additition)}\\ &= (a \cdot 0) + ((a\cdot 0) - (a \cdot 0)) \qquad &\text{(associativity of addition)}\\ &= (a \cdot 0) + 0 \qquad &\text{(additive inverse cancellation)}\\ &= (a \cdot 0) \qquad &\text{(0 is the additive identity)} \end{align*}

so we only need 0 being the additive identity, additive inverses, associativity of addition and mulitiplication distributing over addition to get that a*0 = 0. Break any of those axioms and you're getting a fairly unstructured arithmetic.
ameretrifle wrote:Magic space feudalism is therefore a viable idea.

Deveno
Posts: 29
Joined: Thu Oct 06, 2011 10:14 pm UTC

### Re: 0^-1

of all those laws, i would miss the distributive law the most. the others? meh, lots of binary operations in the sea....

pfft! ok, since y'all are gonna be srs abt this:

i think the first thing to ask when considering a possible definition for 0^(-1) is:

"what do i hope to gain from this, and what am i prepared to give up"?

algebraically, the consequences are usually limiting, you lose more than you gain. for example:

imagine addition is of no consequence to us, and suppose we set 0^(-1) = inf, and set:

a*inf = inf for all a except 0. the question arises: how should we define 0*inf?

if we set 0*inf = 1, then:

a = a*1 = a*(0*inf) = (a*0)*inf = 0*inf = 1 ???? huh? oh darn, we just lost associativity of multiplication.

Whirligig231
Posts: 3
Joined: Thu Oct 06, 2011 8:58 pm UTC

### Re: 0^-1

The real projective line is useful to work with, but the important thing to keep in mind is that you still have to leave certain things undefined, namely 0/0, ∞/∞, ∞+∞, ∞-∞, ∞*0, and 0*∞. (∞ = infinity) These also (mostly) correspond to the "indeterminate forms" found in calculus.

thedufer
Posts: 263
Joined: Mon Aug 06, 2007 2:11 am UTC
Location: Northern VA (not to be confused with VA)
Contact:

### Re: 0^-1

arbiteroftruth wrote:I disagree. Similar logic would say that if 0 is a real number, then 0 needs a multiplicative inverse.

Nope. In a field, every element except the additive identity needs a multiplicative inverse. We can let 0 be special because it is special, being the inverse.