One easy answer would be that in practical purposes, it is a very special sort of parameter. This point comes up quite clearly when we do the derivation for the work-energy theorem:

$$ W = \int\mathbf{ F} \cdot \text d\mathbf s = \int m \left(\frac{\text dv}{\text dx}\cdot \frac{\text dx}{\text dt} \right)\, \text dx = \int m v\, \text dv =\frac12 mv^2 +C$$

When we go from result of first equality to the one of second equality, we are implicitly choosing a very specific parameterization which is of the actual time experienced ticking as the object moves through the path, but, on a mathematical level, no matter how fast we would have run through the force field (assuming force is indep of time), it would be that the work integral gives same answer.

So, this made me wonder, what properties does time have beyond just being a parameter when doing integrals in Newtonian mechanics?


2 Answers 2


Here is one way to address your question.

"Time is defined so that motion looks simple." - Misner, Thorne, and Wheeler in Gravitation, p.23.

Continue through to p. 26 where they say
"Good clocks make spacetime trajectories of free particles look straight".

Misner, Thorne, and Wheeler (Gravitation, p.26)

Look at a bad clock for a good view of how time is defined. Let $t$ be time on a "good" clock (time coordinate of a local inertial frame); it makes the tracks of free particles through the local region of spacetime look straight. Let $T(t)$ be the reading of the "bad" clock; it makes the world lines of free particles through the local region of spacetime look curved (Figure 1.9). The old value of the acceleration, translated into the new ("bad") time, becomes
$$0=\frac{d^2x}{dt^2}=\frac{d}{dt}\left(\frac{dT}{dt}\frac{dx}{dT}\right)=\frac{d^2T}{dt^2}\frac{dx}{dT}+\left(\frac{dT}{dt}\right)^2\frac{d^2x}{dT^2}$$ To explain the apparent accelerations of the particles, the user of the new time introduces a force that one knows to be fictitious: $$ F_x=m\frac{d^2x}{dT^2}=-m\frac{\left(\frac{dx}{dT}\right)\left(\frac{d^2T}{dt^2}\right)}{\left(\frac{dT}{dt}\right)^2} $$ It is clear from this example of a "bad" time that Newton thought of a "good" time when he set up the principle that "Time flows uniformly" ($d^2T/dt^2 = 0$). Time is defined to make motion look simple!


For a similar argument,
refer to p.415 in Trautman's "Comparison of Newtonian and Relativistic Theories of Space-Time"
in Perspectives in Geometry and Relativity, Essays in honor of V. Hlavaty, ed. by B. Hoffmann, 1966


(item 95 from http://trautman.fuw.edu.pl/publications/scientific-articles.html )

  • 3
    $\begingroup$ I'm very impressed this was discussed in MTW. that book has everything. I think this is the answer, however, I'll wait and see what others will say before accepting $\endgroup$ Jun 5 at 22:27
  • $\begingroup$ I have a question, how do you define what is a straight path? $\endgroup$ Jun 7 at 11:05
  • $\begingroup$ @Aplateofmomos "Straight trajectory" in this context is referring to the particle's worldline through a graph of spacetime being a straight line on the graph. Specifically, it means that the particle's measured velocity with respect to our clock is constant if that particle has no forces exerted on it. In other words, measurements made using the clock must satisfy Newton's first and second laws. $\endgroup$
    – Jivan Pal
    Jun 7 at 11:21
  • $\begingroup$ I mean, to define what exactly is straight on the space time - manifold, one would need to choose a connection. So, the question in a mathematical sense is , what sort of connection is being choosen? $\endgroup$ Jun 7 at 11:30
  • $\begingroup$ @Aplateofmomos Truly periodic processes, needed for time measurement, don't exist. Only the ideal clock used in GR has this property. So the clock with least diverging period causes the straightest line. $\endgroup$ Jun 7 at 12:35

In my opinion the idea of time is related to a fixed relation between some physical events. The numbers of full moons, or day/night cycles between the repetition of a sky configuration, which we call year for example.

The idea is regularity. The oscillation of the same pendulum can be compared with the period of a day with a good precision. But our pulse rate fails the test. Even the frequency of atomic clocks are based on the same principle.

It is this parameter that works in the Newtonian equations.

  • 1
    $\begingroup$ This begs the question: "regularity" is defined in terms of time. Are you saying that time is that that can be measured by the oscillations of a pendulum?, Why not say that it is that which is measured out by a particle travelling a constant speed passing equidistant points? Its is a circular definition $\endgroup$
    – Psionman
    Jun 6 at 15:16
  • $\begingroup$ Regularity means: after approximately 365 day/night cycles, the starry sky has the same appearance. And it happens again and again. It is a constant numerical relation between certain types of events. $\endgroup$ Jun 6 at 15:24
  • 3
    $\begingroup$ How do you know it's constant except by using other "regular" processes? $\endgroup$
    – Psionman
    Jun 6 at 15:33
  • $\begingroup$ The point is: there are some events that keep a fixed (within a reasonable margin of error) numerical relation between then. The majority of events of course fails for this job. $\endgroup$ Jun 6 at 18:21
  • $\begingroup$ @Psionman: You don't. That's pretty much the point, though. If t is a good clock, then t'= c*t also is. But t'=t*t isn't. The reason for this is that any periodic process (per Fourier) is described by a sum of sines and cosines. And for a time-periodic process, those terms have the form sin(2*pi*t/T). Using a clock that differs by a constant factor wouldn't make a difference, because the period T would also scale by the exact same constant. $\endgroup$
    – MSalters
    Jun 7 at 14:52

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.