This may appear a dumb question, but I was thinking about this in a deeper way.
If we are concerning about the set of real numbers, and hence we are in $\mathbb{R}^n$ (or $\mathbb{R}$ for simplicity [or even better in $\mathbb{R}^+$]), then $$\sqrt{0} = 0$$ and well defined.
Actually, this already start to have some flaw when we try to write the square root as: $$\sqrt{x} = \frac{x}{\sqrt{x}}$$ and this function is defined in $\mathbb{R}^+$ except at $x = 0$.
But besides strange ways to write the square root, it shall be "everything ok".
Now, let's pass into $\mathbb{C}$, and here we may read $$\sqrt{z} = \large e^{\frac{1}{2}\ln(z)}$$
Now, this is the fact: $\ln(0)$ is not defined at $0$. The logarithmic function is, in addition to that, a multivalued function.
But $\ln(0)$ is not even a multivalued function, and also $0$ is an essential pole for that function.
That infinity that arises in this way, is even worse, because it's a complex infinity, hence it means it can come from any direction and forming any direction in the complex plane.
This brings me to say that $\sqrt{z}$ is not defined at all in $\mathbb{C}$.
What I am now going to ask you is:
- Am I right, am I wrong?
- Did I miss something? May you add some other way to think?
Thank you in avance, I hope this isn't really a dumb question. It's, at least for me, not.
In $\Bbb R$, $\sqrt x$ is defined to be the non-negative number that, when squared, produces $x$. With this, $\sqrt 0=0$. The fact that the otherwise valid equality $\sqrt x=\frac x{\sqrt x}$ has one side undefined when $x=0$, should not bother you. In the same manner, we have $1+x=\frac{x^2-1}{x-1}$ whenever $x\ne 1$, but that does not suggest that $1+x$ has problems when $x=1$, does it?
In the complex case, $\sqrt z$ can be defined (i.e., a branch picked) nicely for all simply connected domains that do not include $0$. Even then, if $A$ and $B$ are two such domains, it may not be possible to do this in a way that agrees on $A\cap B$. Still, all this does not stand against the fact that $\sqrt 0=0$ (in fact, this is the only point where there is no doubt about the complex square root).