I'm sorry if this has been asked before and if the title is unclear; I'm not sure how to search for this question. Is it proper to write a function that is a function it's variables derivative as:


instead of


where $x=x(t)$? This (possibly) creates a contradiction in that


Thanks in advance!

  • 1
    $\begingroup$ I'm not exactly sure what is your question; people write $\dot{x} = f(x)$ informally to mean the differential equation $\dot{x}(t) = f(x(t))$ for all $t$ on some domain. $\endgroup$ – Qi Zhu Oct 25 '19 at 17:22
  • $\begingroup$ That's a good point. Here's an example. If I'm trying to define the force $f$ due to friction, I would say something like $f(\dot{x})=-k\dot{x}$. But I could also write $f(x)=-k\frac{dx}{dt}=-k\dot{x}$ couldn't I? $\endgroup$ – eball Oct 25 '19 at 18:29
  • 1
    $\begingroup$ Well, the two equations you've written are different. The first takes $\dot{x}$ as the parameter, the second $x$. I think you probably mean the second equation. (Yes, you can write the second equation, here $f$ is a function of functions $x$ and you apply the differential operator on the right which is in itself a function.) $\endgroup$ – Qi Zhu Oct 25 '19 at 22:12
  • $\begingroup$ @QiZhu, I see what you're saying. Would you agree that both $f(\dot{x})=-k\dot{x}$ and $f(x)=-k\frac{dx}{dt}$ are acceptable ways to write it, but maybe the second way is canonical? $\endgroup$ – eball Oct 28 '19 at 11:48
  • $\begingroup$ No, I‘m saying that these two mean different things. $\endgroup$ – Qi Zhu Oct 28 '19 at 12:51

I cannot imagine a function that is a function of its own derivative. Sounds like a circular reference.

Perhaps you mean this. Consider a function $x = x(t)$ and you can have a function of both $x(t)$ and $x'(t)$ at the same time? For example, consider an iteration of Newton's method, which computes $$ N(t, x(t), x'(t)) = t - \frac{x(t)}{x'(t)}. $$ This is well-defined as a map $N : \mathbb{R}^3 \to \mathbb{R}$.

If you want to map functions to other functions, it's a different story. For example, consider the space $\mathcal{S}$ of all one-variable functions, differentiable infinitely many times on $[0,1]$, e.g. $\cos(x)$ or $e^x$ or any polynomial. Then you can have a map $D : \mathcal{S} \to \mathcal{S}$ defined by $$ D[f(x)] = f'(x), $$ thus mapping $e^x \to e^x$ and $\sin x \to \cos x$, for example.

| cite | improve this answer | |
  • $\begingroup$ I think I meant "function that is a function of its variables derivative". Overall, I think what I'm asking is if it makes sense to write $f(x)=\dot{x}$. From your answer, it sounds like no, but I'm not quit sure why. $\endgroup$ – eball Oct 25 '19 at 17:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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