What is time?

Time is an oddity. In mystic thought, it is often cyclical. In classical Newtonian physics it was thought of as a steady march onward. In Einstein’s universe, it is wrapped up with the fabric of spacetime and can dilate in unexpected ways.

One thing we know about time is that it waits for no one. Time marches on (although not in a steady way, thanks to Einstein) and we experience it passively. Without any effort on our part, future becomes present and recedes into the past, even while we sleep. We can’t feel the passing of time like we feel the wind against our face, although change is happening imperceptibly, and after a lifetime, we realize that its passing has ravaged our fragile bodies.

Where did the time go, we ask? As if it is a thing that moves.

Our human experience of time is very subjective and unreliable. When we are having fun, time vanishes, which seems rather unfair, as if the world is conspiring to cut short our enjoyment and prolong our agony. We much prefer enjoyable experiences to take place in the present or the future, and consign our bad experiences to the past. Most of what motivates us in life is geared to trying to make this happen.

Time isn’t something we can measure easily. We can’t put physical markers on it. To grasp time, we must remember past events, and attach labels to those events. But those events vanish in an instant and no longer exist. We must trust to our memory, or written records, to recall them.

Time is something that we must measure indirectly and relatively. There is no start time or standard time that we can refer to. The only way to measure time is by intervals between regular repeating events. The rising of the sun on the horizon is one of those events. We can call it a day. We can subdivide it according to our whim into hours and minutes, and combine days into weeks, months and years. But the length of a day isn’t a constant. The sun rises and sets at different times throughout the year.  The number of days in a year isn’t a whole number. Even the rate of rotation of the Earth is slowing inexorably as the Moon pulls it towards an eventual halt, making the day longer by a millisecond or so every century. Time based on such crude measures would be hopelessly inaccurate, except for simple purposes.

To measure time more accurately, we need a regular repeating event that never changes. Atomic clocks, based on energy level transitions of caesium atoms, measure time more accurately than any astronomical observations ever did, with an accuracy of one second every billion years. This is the best we can do with our current technology.

Yet still we are stuck with a definition of time as “what a clock measures.” It sounds like the opinion of a three-year-old, yet it’s not a bad definition, apart from being circular (pun intended.) After all, what is a clock, other than a device for measuring time?

Can science not do any better than this?

According to Einstein, time is what separates events in space. And Einstein wasn’t being vague when he used the word event. Events are not just arbitrary human-invented entities like “my birthday” or “the battle of Hastings in 1066”. They are also the fundamental interactions between elementary particles that make up the matter in this universe. If a photon is absorbed by an electron, or if two electrons scatter off each other, that is an event, and it is observable.


What about the direction of time? Some people have suggested that it is the irreversibility of physical events that give time its direction. For example, a hot object will cool, never spontaneously heat up, unless we apply some kind of heat source. It is argued that this arrow of time, defined by the scientific measure called “entropy” (a precise definition of the amount of disorder in a physical system) is what makes time go forward, rather than back. Entropy always increases.

But even reversible events have a before and after, so I don’t think that can be true. If the universe ever stopped expanding and started to contract again, entropy would continue to increase. We wouldn’t all reappear, running our lives in reverse order, from death to birth. The direction of time is real, not a perception.

The microscopic events that map out the passage of time are real and objective, and so time must be too. We can measure it with clocks, and define how it appears to different observers using Einstein’s theory. And that, however unsatisfactory, is probably as close as we can get to grasping what it is. At the present time, anyway. In the future (whatever that is), who knows?

33 responses to “What is time?

  1. Time really is a fascinating subject, isn’t it?

  2. Simple: it’s what stops everything happening at once. 😉

  3. Time is observed as energy dissipates in a medium where the observation is relative to distance from the medium. If the observer is in the medium time happens relative to the observer…. Or fairly closely to.

  4. Sometimes. Energy need not be dissipated for time to flow. Time flows regardless of whether anything happens.

  5. Time’s relation to entropy feels somewhat circular to me. A while back, I tried to understand exactly what entropy is. (Everyone called it “disorder”, but that just seemed like a value-laden term; “disordered” according to who?) Eventually, after lots of discussion, I came to the conclusion that entropy is the state that natural systems will move toward in time, the lowest stable energy state.

    But if so, then entropy is emergent from time. But some people seem to think that time is emergent from entropy 😕

    • Yes, I’m sure that it’s a fallacy.

    • “Everyone called it “disorder”, but that just seemed like a value-laden term”

      You’re saying it has multiple values, but the definition(s) of entropy are quite precise. Disorder (under Boltzmann’s definition) is the log of the number of indistinguishable macro states of the system (the formula, famously, is on his gravestone).

      The way I explain entropy is to consider a collection of books sorted by author, title, date, and however many other criteria necessary to insure a perfect sort. When sorted, the collection has almost no entropy (in terms of sort order). Any other state of the system is immediately recognizable. There is only one indistinguishable state.

      Now dump all the books in a pile on the floor and stir them up. The order of books in one random pile of books is fairly indistinguishable from another. If someone rearranged several books, you’d be hard-pressed to spot the difference. There are lots and lots of piles of books all basically indistinguishable from each other. A pile of books has high entropy.

      (FWIW: Sean Carroll’s latest blog post is about entropy.)

      • I just meant that I found the “disorder” label by itself unhelpful. Your description coupled with Boltzmann’s definition make sense. Maybe another way to put it is the degree of heterogeneity without repeating patterns?

        • Ah, gotcha, yes, the word “disorder” is vague. Repeating patterns are definitely low in entropy. The digits of π, for example, should have very high entropy as a digit string, but as the value of π they have extremely low entropy since any change would make the value no longer π.

          One thing: The ordered set of integers has very low entropy, but every digit is distinct, so in one sense the heterogeneity is high, and there are no obvious repeating patterns. At an analytical level we find the string is trivially described (“start with 0, keep adding 1”) and ordered so its entropy is very low.

          I really think the key to the general description of entropy is the “indistinguishable (macro) states” part. Grasp that and you’ve got it. (The “macro” really just means “at the level you’re describing the system.”)

        • Right, but the “indistinguishable” part, similar to “disorder”, makes me want to ask, indistinguishable for who?

        • LOL! 😀

          A serious answer: Any theoretical observer capable of making accurate observations. Implicit is the idea that the order is quantifiable (in my books example, a perfect sort) and that any state not matching that order is quantifiable in its difference or distance from the ordered state.

          Mathematically, there are N possible ways to take one book and insert it out of order. The number depends on M, the number of books, and it’s fairly small compared to moving multiple books, let alone all of them.

          For a set of M books, there are M! (M factoral) possible combinations, the log of which is the entropy for an M-sized fully random (indistinguishable) collection.

  6. Good post. (I’m taking partial credit in getting you thinking about time. 😀 )

    “As if it is a thing that moves.”

    That’s a good question. Is time like a medium that we’re moving along, or is it something that flows past us? It’s actually a little odd to think of ourselves as moving in time. What started that movement? Is there “time friction” or is time a “frictionless” medium?

    If we say that things happen “because of time” (e.g. rust), then we seem to be saying time has power of some sort. It doesn’t seem quite as odd to think that something with power flows. But maybe the whole concept of movement is a non-starter. Time doesn’t move, it ticks!

    “Time isn’t something we can measure easily.”

    I think maybe you mean “define easily” more than “measure easily.” You describe a variety of ways of measuring time. Ultimately, as you said, it involves finding an interval we can use to mark off units of time. Days are really sloppy, but sometimes sufficient. We once used candles and dripping buckets for smaller units. The clock escapement was better, and now we use vibrating atoms and bouncing light beams.

    We do the same thing measuring space. We went from “cubits” and (literal) “feet” to wavelengths of light, but all we’re doing is finding an interval that we use to mark off units of space. Those are identical situations. So we’re also “stuck with a definition” of space as “what a ruler measures” (likewise, voltage is what a voltmeter measures).

    So if it’s true that time isn’t something we can measure easily, then nothing is. All measurements involve finding an appropriate measure to use. Given that current technology is measuring time at the pico-second (if not femto- or atto-second), we’ve gotten pretty precise!

    I think the way out of the conundrum is that a measurement is not a definition and that we don’t have a good definition of time (which is a point of your post). We do have a definition of voltage, for example. Time isn’t something we can define easily. Or so far, at all!

    “But even reversible events have a before and after, so I don’t think that can be true.”

    I don’t quite follow your logic there. If an event is reversed, ‘before’ becomes ‘after’ and vice-versa. I agree that, if the universe underwent gravitational collapse towards a “big crunch” time would not run backwards. That’s not what’s meant by reversible events, though.

    You showed a Feynmann diagram with an arrow of time pointing upwards. A characteristic of Feynmann diagrams is that you can rotate and reverse them. (Under rotation the particle interactions are different but still valid.) If you reverse time on a Feynmann diagram it still describes a valid interaction. It’s obvious the one shown in your post describes the same interaction if time flows backwards.

    You can even view that diagram as describing the interaction of the anti-particles of the original flowing backwards in time. It can be interpreted as showing positrons moving downwards (and still exchanging a photon since photons are their own anti-particle).

    That’s the reversibility conundrum. That particle interactions work regardless of which way time flows. And since they do, everything does. Since it appears, as you say, that the “direction of time is real,” the question is: Why should that be (since physics doesn’t require it)?

    One best-guess answer is, “Um,… maybe entropy?” XD

    • Yes, credit is due to you, Wyrd 🙂

      As you say, we move effortlessly through time, and indeed we move effortlessly through space too. Only friction slows us down. That infuriating Einstein fellow showed us that if we start moving through space, then our movement through time slows down. Space, time, and motion are tightly bound together like an over-wound watch spring.

      Is space a medium? No. A medium is something that fills space. Implicitly it fills time too. So time isn’t a medium, but a dimension, which is really saying that we can measure it but don’t know what it is. 🙂

      Why does time move forward? Don’t know. But I don’t buy entropy. The Feynman diagram was supposed to be illustrating why. At the microscopic level there is no entropy, yet time still moves in one direction only.

      • “Space, time, and motion are tightly bound together like an over-wound watch spring.”

        Well that’s a lurid description! 😀

        (Point of clarity: When you move through space, your movement through time speeds up and you emerge in the future once you stop.)

        “So time isn’t a medium, but a dimension,…”

        And we’ve already talked about how it’s not really a (spatial) dimension, either (it is in the sense of being a degree of freedom and a number we assign to an event).

        I’m not convinced by your analogy with space. What if time is a medium that fills space?

        “At the microscopic level there is no entropy, yet time still moves in one direction only.”

        But what is also the point of the Feynman diagrams is that, while time does happen to move in one direction, there’s no requirement it do so. All Feynman diagrams work in reverse.

        The belief is that: If time is emergent from entropy, then perhaps it drags particle physics along in one direction thus accounting for that preferred direction.

        • What if time is a medium that fills space?
          That’s quite a Newtonian perspective though.

          The Feynman diagrams show time at its most fundamental level, stripped of macroscopic effects like entropy. As you say, its symmetric, with no preference for direction in time. The thing about time is that it just keeps chugging along. Once set in motion, it’s never going to stop or change direction (as far as we know.) So, what if there’s nothing special whatsoever about the direction of time, and it just happens to be going in this direction because of initial conditions? Rather like the rotation of the Earth?

          Then the question of the direction of time ceases to become interesting. In fact, if time was running in reverse, we probably wouldn’t be able to tell the difference (although maybe there would be other effects that would be reversed, such as the dominance of matter vs antimatter. Just thinking out loud.)

  7. Wait, did you just say that your birthday is an event comparable in magnitude to the Battle of Hastings? 🙂 Your conclusion that time is “real and objective” seems persuasive, but other interpretations may be so too (I’m not sure yet). Could this be a “begging the question” fallacy? Linear time is not reversible (we can’t live our lives backwards), but that presupposes that it was linear in the first place. Could it be an illusion that our lives are right now being lived out along a linear string of time? Might that just be our subjective way of making the unbearable simultaneity of all events intelligible? Yours from the event horizon, the vanishing point of time, Gary (aka William to your Harold)

    • LOL, Gary. All human events should be treated with equal importance, unless you’re a historian or social commentator or something. As for the rest, I refer you to Einstein and his annoyingly non-linear view of time.

  8. and if Mr. Steve some time ago appeared in a television culinary program… (even in Poland that was shown – several months ago). but i’m not entirely sure if it was you… i turned on too late tv… but tall, handsome, astrophysicist from Oxford … everything points to you?!

  9. so i disappear… hope that with this “trial text” something (meaningful) can be done.

  10. good article, embracing many aspects of this problem. although with some issues i wouldn’t agree. but this isn’t important because in general science understands time in not too skillful way. but something else attracted me: “although I tend to find questions I can’t answer to be infuriating rather than fascinating. That’s my problem”. is it just my “morbid” curiosity, or maybe we have really a serious problem?!

  11. so i got the distinct impression there will be another wonderful text on these topics!?

  12. Hey very nice one… I really liked your blog… Please do make some time to visit my blog, I do write on science and tech… And I did writer on NATURE OF TIME😜😜

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.