It seems to me there are two different interpretations of a symbol $f(y)$. I will explain what I mean:

Suppose I have a function $f(x) = x$. (I took the identity map to have a simple example). Also suppose I have a dependence between $x$ and $y$ which is another variable. Say $x = 2y$.

I seems to me I can interpret $f$ and particularly the symbol $f(y)$ in two different ways:

  1. $f$ is strictly a map and it taken whatever variable we give it and maps it accordingly. In this case $f(y) = y$ as $f$ is the identity map so just maps $y$ to itself.

  2. $f$ is a variable dependent on $x$, since I've defined $f(x) = x$, then when I denote $f(y)$ I can interpret it as the variable $f$ but now expressed in terms of $y$ instead of in terms of $x$, so in this case $f(y) = x = 2y$. In this case I viewed $f$ as already being defined in terms of x and the symbol $f(y)$ merely gives me this predefined variable $f$ in terms of $y$.

My question is what is the usual interpretation, and if there is convenient notation and/or common terminology which distinguish the two.


It is better to consider the composition of functions specifying range and domain. Let us work in the easy setting:

$$f:\mathbb R\rightarrow \mathbb R,~~z\mapsto f(z):=z, ~~(*)$$


$$y:\mathbb R\rightarrow \mathbb R,~~x\mapsto y(x):=\frac{1}{2}x.$$

When you write $f(y)=y$ in $1.$ you are just considering the function $f$ on any element $y$ (call it $y$, $z$, $x$ etc...it is the same: you are using $(*)$).

When you write $f(y)=x=2y$ in $2.$ notation is confusing, instead. I believe you should introduce the composition the composition $g:=f\circ y:\mathbb R\rightarrow\mathbb R$, where $$g(x)=f(y(x))=y(x)=\frac{1}{2}x.$$


I'll restrict this answer to the case where $f$ is a real function defined on the whole real line, ($f\colon \mathbb R\to \mathbb R$).

Since the domain of $f$ is $\Bbb R$, its input are real numbers. Therefore, the symbol $f(y)$ is meaningless if $y$ is not a real number.

In your example you set $f=\text{id}_\mathbb R$. Let $y$ be a real number. If it is the case that there exists $x\in \mathbb R$ such that $y=2x$, then $f(y)=f(2x)=2x$, by virtue of the definition of equality and of $f$.

If $y$ is a function (for simplication, let $\text{dom}(y)=\mathbb R$ and let it take only real values), then, as said before, $f(y)$ doesn't make sense. Something that makes sense is $f\circ y$ and, given $x\in \mathbb R$, $f(y(x))$ also makes sense because $y(x)\in \mathbb R$.

Now answering your question, the usual interpretation is that $y$ is a function and that you should look at $f(y)$ as $f\circ y$ (accompanied by a deep sigh), which is the convenient notation. Common terminology is saying that '$y$ is a function of $x$', which doesn't excuse the wrongness of $f(y)$.

  • $\begingroup$ In some disciplines, mainly PDE theory and Calculus of Variations, it is customary to write $f(u)$ instead of $f \circ u$. However, I agree that this is an abuse of notation. $\endgroup$ – Siminore Dec 8 '13 at 12:47
  • $\begingroup$ @Siminore Unfortunately it is common. I thought I made that clear in the last paragraph. In any case your comment makes it clear, if it wasn't already. Thanks. $\endgroup$ – Git Gud Dec 8 '13 at 12:50

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.