The mathematical ‘t’

The last page left one habit unfinished. Day and year are steady enough that a scale built from them feels like a feature of the world. Mathematics takes a further step, and physics has often followed it past the point where the step is still only a convenience.

A great deal of mathematics writes a single letter, *t*, and lets it run. Position is some function of *t*. A rate of change is how one quantity differs as *t* differs. The letter is universal inside the equation because the equation has only one of it. Every process in the sum is scored against the same parameter. Nothing in the algebra requires that parameter to be a physical constituent. It is a label that keeps the changes in order and lets the differences be written down.

That is a powerful economy. One steady index can carry a pendulum, an orbit, a cooling cup and a population of atoms, and the relations among them can be stated without inventing a new index for each. The economy is the reason the letter is there. It is not, by itself, a discovery that the world contains one flowing thing with that name.

Physics inherited the device, and had reasons to trust it. The Earth’s rotation and orbit really are regular. Clocks built to echo that regularity really do let distant events be compared. Laws written with a single *t* really do predict. From those successes it is a short slide to a stronger sentence: there is a universal time, the same everywhere, flowing at one rate, the backdrop against which every physical process occurs. Newton’s absolute time, in the Scholium, is that sentence said plainly: time flows equably of itself, without relation to anything external. Later physics complicated the rate, and in relativity abandoned the single shared reading, but the habit of speaking as if *t* named a physical item often remained. The parameter had been acquired — taken over from the calculation and treated as a constituent of the world. “Acquired” is this argument’s name for that step. It is an interpretation of the step, offered for assessment.

The constancy is what makes the acquisition feel like an observation. If the measures we live by were erratic, a single universal flow would be a strange thing to posit. Because day follows day, the single letter looks like a description of what day is doing, rather than a choice of how to score it. Most change, as an earlier page noted, is not regular. The parameter hides that. Inside the equation everything is already paced by *t*, so unevenness appears only as a complicated function, never as a challenge to the idea that one time sits underneath.

A shared label is not a shared substance. Two clocks can be assigned the same coordinate *t* and still accumulate different amounts of change along their paths. The coordinate keeps the account in order. The readings record the histories. Treating the coordinate as the thing that flowed, and the disagreement as that thing slowing, confuses the label with the change. Relativity already restricts the stronger reading. The restriction is empirical. It is not, by itself, the ontological conclusion.

Nor does the letter cause what it orders. The cup cools because heat leaves it. The equation may write the cooling as a function of *t*. The difference that produces the new state is physical, not alphabetical. In the cases already traced, remove the physical difference and the change does not occur, however the parameter is allowed to run.

A universal *t* is also a poor fit, on this account, for an unchanging object considered alone. The diamond between its own rare changes has no internal process for the letter to score. The number we attach — millions of years — is the external series, written in the common parameter. The parameter makes the number easy to enter in a table. It does not soak the stone in a flow.

None of this retires the letter. A single parameter is still the right tool when a single comparison is what is wanted: appointments, navigation, the statement of a law in a chosen frame. Use *t* as a label. Do not, merely because the label is unique in the equation, conclude that a unique physical time has been found. The second sentence is the claim. The first is a concession.

The mistake, if it is one, is the acquisition, not the mathematics. A convenient universal notion inside a calculation was treated as a universal physical thing. Constancy of the Earth’s motions made that treatment look like common sense. Whether the changes required it is the residual question, not a result declared by pointing at the letter.

If the parameter is only a parameter, the conditional claim of these pages is complete enough to state. What is seen is change. Time, on the uses distinguished here, names either the gathering of that change or the scale used to score it. Space, in the sense that matters here, is the frame for static place. Space-time is the frame for motion. Period and duration are comparisons along those frames. A mathematical *t* is one way of writing the comparison down. In none of these jobs has a further flowing entity been shown, by this argument, to do work that the changes and the relations do not already do.

What would count against that is not the usefulness of the letter. It is a residue: some empirical result or some explanation that still cannot be recovered once the parameter is put back in its place as a label. That test is stated here. It is not settled here. Settling it would mean looking at particular theories, not at the word.

Optional — for specialists.

The claim on this page is narrower than a rejection of temporal representation. A single parameter *t* is indispensable for writing a large class of dynamical laws. Indispensability for representation does not, by itself, decide ontology. The further question is whether, once the relevant physical states, relations and dynamics have been specified, a universal temporal substance still does empirical or explanatory work that the parameter and those structures do not. If it does, the posit stands. If it does not, the parameter remains a label.

This is a test, not an elimination result. It is compatible with treating the metric, or other dynamical structure, as physically real. It is not compatible with inferring a universal temporal constituent from the mere presence of *t* in an equation.

Newton’s absolute time is the acquisition stated outright: time flows equably of itself, without relation to anything external. The page treats that step as a metaphysical addition to the parameter, not as a misreading of the *Principia*’s usefulness.

Spacetime functionalism asks which roles associated with spacetime are required, and what can realise them (Lam and Wüthrich, 2018). The question here is the next one. Once ordering and measurement are recovered, does the role-filler add a constituent? Dynamical approaches already account for inertial structure from the behaviour of bodies rather than from a container (Brown, 2005). Effective-geometry accounts treat spacetime structure as what is recovered at the relevant scale (Knox, 2013). None of these is the thesis above. They are why the thesis can be stated as a residual question rather than as a rival theory.

Relativity already restricts the stronger reading. Proper time is accumulated along a worldline; two histories that share a coordinate label need not agree when brought back together. That restriction is empirical. It does not, by itself, answer the ontological question. It removes the simplest excuse for a single universal flow.

It does not say that laws written in *t* are notational mistakes, or that their predictions fail. It does not say that vacuum curvature or radiative degrees of freedom are unreal. It does not say that no temporal or spatiotemporal posit could ever meet the test. A theory in which a temporal degree of freedom made a contribution that could not be recovered from the states, relations and dynamics already in use would count against the claim. The criterion, and its relation to functionalism, is set out in full in the accompanying paper.