# The derivative of an empty function

Let's say we have function with empty domain $$f:\ \varnothing \rightarrow Y$$.

Is this theoretically correct to say that the derivative of this function is another function with empty domain e.g. $$g:\ \varnothing \rightarrow Y$$?

(I know it's pointless to calculate a derivative of an empty function, but I want to know whether it is formally wrong to say so and if it contradicts with the definition of the deriviative)

• The derivative isn't even defined in a general set theoretic context. – lulu Apr 14 '20 at 17:12
• @lulu sorry I added that tag by mistake. I am asking in the context of (real) analysis. – MartinYakuza Apr 14 '20 at 17:20
• Are you assuming $Y$ to be a subset of the real numbers? – Servaes Apr 14 '20 at 17:28
• @Servaes yes, but does this really matter? The set of values will be empty anyway. – MartinYakuza Apr 14 '20 at 18:08
• You're right that the derivative has empty domain. But "another" can be confusing, since all functions with domain $\varnothing$ are equal, i.e., they're the same set of ordered pairs. – Andreas Blass Apr 14 '20 at 19:11

From definition of a derivative we get that the function must be defined at the point under consideration. The defining equation is $$\ f'(x)= \lim_{h\to0}\frac{f(x+h)-f(x)}{h}$$ for sufficiently small h. So here $$\ f(x)$$ $$\left[ also \ f(x+h)\right]$$ has to be defined at some point in the domain. If the domain doesn't contain any point, then this defining equation doesn't make any sense as the functon $$\ f(x)$$ is not defined at any point. So there is no point in even considering the derivative where the main function is itself not defined.