Naked Time ( Is there such a thing ?)

Question

I have lately been concerned with time and whether or not it is an illusion. For argument's sake let's assume it is. Then time should stand on its own.

Can anyone describe, (or point me to a paper that describes) time without referring to something else.

For example, defining time without the mention of a measuring device.

Try stating--precisely--what would be different between two worlds, one of which had "illusionary" time and the the other with "real" time. Take your ahem time. I'll wait. — dmckee --- ex-moderator kitten, Apr 21 '15 at 16:06
That is not what I am asking. I don't know why folks insist on flagging a question without reading the dang thing.
I want to know if anyone can define time with out referring to some other process ( ie with out referring to the phase or cycle count of the vibrations of a super chilled cesium atom.) — user33995, Apr 21 '15 at 16:11
Time is the name we assign to the fact that two distinct events can occur in the same place without being coincident. — dmckee --- ex-moderator kitten, Apr 21 '15 at 16:15
This question has been treated pretty massively by the ancient Greeks and pretty much any natural philosopher since. I think it would be helpful for you to at least read on the Aristotelian notion of time, which does not explicitly rely on measurement. Hume also had useful things to say about this. I'm sure Kant did too, but it's been so long since i"ve read his impenetrable nonsense, that I don't remember. — Zo the Relativist, Apr 21 '15 at 16:45
Yes but what is TIME? the +1 in 3+1. As a dimensional reality. Can you describe it as a stand alone entity? — user33995, Apr 21 '15 at 16:45
No. it's part of the underlying spacetime geometry. It cannot be meaningfully separated from space in relativistic theories. — Zo the Relativist, Apr 21 '15 at 16:58
And you can come up with observer-dependent notions of time, but these will ultimately reduce down to one of the classical notions that one of those philosophers I cited used. — Zo the Relativist, Apr 21 '15 at 17:00
Related: http://physics.stackexchange.com/q/15371/2451 , http://physics.stackexchange.com/q/17056/2451 and links therein. — Qmechanic, Apr 21 '15 at 20:24
@Zach466920: can you describe length without a measuring device? Yes, easily. I caught a fish that big. I can hold my hands up a yard apart. I can show you a length. I can waggle my hands and show you motion too. But you can't show me time. — John Duffield, Apr 22 '15 at 16:08
@JohnDuffield Isn't your hand (or other comparative device) the measuring device??? — Zach466920, Apr 22 '15 at 16:55
Not really. My two hands are position markers, that's all. You use your experience to gauge (=measure) that they're a yard apart. The important thing is that they're separated by a distance, a space, and you can see it. You can see motion too, which is what clocks really clock up. But not time. In the science fiction movies, when some guy has some gizmo that can "stop time", what it really stops is motion. People say you need time to have motion, but in truth it's the other way around. — John Duffield, Apr 22 '15 at 17:07
@JohnDuffield Of course you can see distance, but you can't see time, its not a visual phenomena! Also, if you replace hands with rulers all of a sudden it becomes obvious that you're using a measuring device. In addition, can you describe weight, coulombs, or tesla, without a measuring device? I'm pointing out that all units can't be defined without a measuring device. — Zach466920, Apr 22 '15 at 17:17
@Zach466920: time is a "motion phenomena". And whilst a clock is a measuring device, what it actually does is tot up some kind of regular cyclical motion rather than the literal flow of time. IMHO it's important to appreciate what a clock really does - it relates to what Einstein was talking about with his operational definition, and is crucial to understanding relativity. — John Duffield, Apr 23 '15 at 09:21

score 1 · Answer 1 · answered Apr 21 '15 at 18:48

Can anyone describe, (or point me to a paper that describes) time without referring to something else.

No. Nobody can. However I can point you to Presentism, and to A World Without Time: The Forgotten Legacy of Godel and Einstein. I can also point out that relativity accurately models our world using spacetime and worldlines and geometry, but that the map is not the territory, and that we live in a world of space and motion, not a block universe. I can also point you towards a clock, and tell you that the inner workings of that clock are quite aptly called a movement. I can then point out the patent blatant fact that a clock clocks up some kind of regular cyclical motion and shows you a cumulative display that we call the time. Whether it's a mechanical clock with a spring-driven rocker, a grandfather clock with a pendulum, a quartz wristwatch with a vibrating crystal, or any other kind of clock, that clock "clocks up" some kind of regular cyclical motion and shows you a cumulative display called the time. The big hand moves, and the little hand moves. That's no illusion. But it doesn't literally measure the flow of time like some magic cosmic gas-meter gizmo. Moreover I can hold my hands up a foot apart. Can you see that gap, that space between them? Yes. Now watch as I waggle my hands. Can you see that motion? Yes. Because space and motion are empirical. But can you show me time? No. You cannot. Nor can anybody else.

You've probably read Minkowski's introduction to Space and Time. He said "space for itself, and time for itself shall completely reduce to a mere shadow, and only some sort of union of the two shall preserve independence". I would venture to volunteer that had Minkowski lived longer, if he hadn't died before his time, he would have ended up writing a new paper. Called space and motion.

Comments are not for extended discussion; this conversation has been moved to chat. — Manishearth, Apr 24 '15 at 02:42

score 0 · Answer 2 · answered Apr 21 '15 at 20:20

0

Can anyone describe, (or point me to a paper that describes) time without referring to something else.

The principal relevant description is surely the one which Einstein put at the foundation of the (special) theory of relativity:

"[... that instead] of ``time´´ we substitute ``the position of the little hand of my watch´´."
[Punctuation marks as in the German original: Ann. Phys. 17, 891 (1905)]

Note that the mere appearances and the distinguishability of "positions of the little hand" of some particular participant's watch, or generally: of "indications" of some particular participant, does not imply any particular durations, or duration ratios, between pairs of such indications. Consequently, Einstein's quoted description doesn't mention or involve any "measurment device" (such as any "ideal clock" whose construction is presented in MTW, §16.4).

In turn, this is of course an acknowledgement that "time" can be described only in referring to any particular identifiable participant (MTW Box 13.1 chose the description "principal identifiable point"); i.e. not exactly "naked".

p.s.

Unfortunately, IMHO, Einstein missed the opportunity to likewise dispense as plainly and rigorously with the notion of "naked place(s)", e.g. by substituting "you", or "me", or any particular participant ("just as you and me", at least for the purpose of thought-experimental description).

answered Apr 21 '15 at 20:20

user12262

4,258
17
40

-1. A watch is a measurement device, one that measures time. – David Hammen Apr 22 '15 at 05:58
@David Hammen: "A watch is a measurement device" -- No. A watch is a (typically easily portable) generator of distinctive, conspicuous indications (to watch and remember). The measurement of durations of a given watch (resp. its "wearer"), between pairs of its indications, is (only, by definition) accomplished by "ideal clocks", such as sketched in MTW §16.4. (p.s. Sorry for not having replied earlier; I nearly missed your comment ...) – user12262 Apr 23 '15 at 04:44
A real clock is an approximation to an ideal clock, just like a real ruler for measuring distance is an approximation to an ideal ruler, a real scale for measuring mass is an approximation to an ideal scale, etc. All continuous physical quantities are assumed to have a "true" value which real physical instruments can only tell us to some degree of approximation. – Hypnosifl Apr 23 '15 at 13:55
@Hypnosifl: "A real clock is an approximation to an ideal clock," -- Sure: any real/observed set (or already even: sequence) "$\mathcal S$" of watch indications together with pretty much any coordinate assignment "$t : \mathcal S \rightarrow \mathbb R$" constitutes some "approximation to an ideal clock". Concretely, for any three (distinct) indications $S_b, S_p, S_w~\in\mathcal S$ there sure is some real number value $u$ such that $$(t[~S_w~]-t[~S_p~]) = u~t[~S_p~] - t[~S_b~])~\frac{\text{Duration}{\mathcal S}[~S_w, S_p~]}{\text{Duration}{\mathcal S}[~S_p, S_b~]}.$$ [continued] – user12262 Apr 23 '15 at 17:22
It's of course the task of (experimental) physicists to quantify or to delimit, trial by trial, how bad an approximation. (W/o that you don't even have any serious estimates; right?). Here: to evaluate the (otherwise unknown) number $u$, case by case. And this requires knowledge and appreciation of how to determine duration ratios properly in the first place. "All continuous physical quantities are assumed to have a "true" value" -- All physical quantities have likewise an operational definition; and thus a definite range of values. But as you surely know: not all such operators commute. – user12262 Apr 23 '15 at 17:24
OK, but are you arguing that there aren't good ways to determine which real clocks are better approximations of ideal clocks? If so I think experimental physicists would disagree. And the word "operational" in "operational definition" has nothing to do with the definition of "operators" in QM, so I don't really understand your last point. If it helps, what I said about every quantity having a "true" value would in QM refer to the "true" state of the wavefunction (which can be determined to arbitrary accuracy), I'm not making any assumption of "true" simultaneous values of non-commuting vbls. – Hypnosifl Apr 23 '15 at 17:33
Hypnosifl: "OK, but are you arguing that there aren't good ways to determine which real clocks are better approximations of ideal clocks?" -- Hardly: I just wrote down the relevant formula for you in my above comment; and the answer on which we're commenting lists a (the!?) reference of how to determine duration ratios: MTW. "the word "operational" in "operational def." has nothing to do with the definition of "operators" in QM" -- Only for a sufficiently narrow conception of "having to do". The point being: certain applicable operational/thought-experimental definitions don't commute. – user12262 Apr 23 '15 at 18:29
I don't understand your formula--I don't know what you mean by "indications", or what $u$ and $t$ are supposed to represent (does either represent the "true" time an ideal clock would measure?), or what $Duration_S [ S_w , S_p]$ represents (what do the brackets mean here?). In any case, is your formula supposed to tell us which of several measurements by non-ideal clocks is likely to be closer to the unknown "true" value of time that would be measured by an ideal one? And would you agree non-ideal clocks measure time (approximately) just as non-ideal rulers measure distance (approximately)? – Hypnosifl Apr 23 '15 at 22:36
(cont) maybe by "indication" you mean what people usually call a clock reading? (e.g., "3:56 pm") And then would $S_w$ and $S_p$ be two different readings at two different events (whether on the same clock or distinct clocks), so $Duration_S [S_w , S_p]$ just means a calculated time interval between the events based on those readings, i.e. just $S_w - S_p$? Or would $Duration_S [S_w , S_p]$ represent the true duration between the events, distinct from the readings? And would $S_b$ represent a different clock's reading at one of the same two events, or some clock's reading at a third event? – Hypnosifl Apr 23 '15 at 23:16
@Hypnosifl: "maybe by "indication" you mean what people usually call a clock reading? (e.g., "3:56 pm")" -- Well, by an "indication" (denoted e.g. "$S_p$") I mean (rather) that what is being read; the observed output of any (analogue or digital) indicator such as a "clock face/dial"; I mean in all generality what Einstein in a famous specific case called "the position of the little hand". I do not mean any numeric (or boolean, or alpha-numeric) value which might be (subsequently) assigned or determined of an indication; such as the/any coordinate time $t[~S_p~]$. – user12262 Apr 24 '15 at 02:07
@Hypnosifl: "And then would $S_w$ and $S_p$ be two different readings" ... indications ... "at two different events" ... yes ... "(whether on the same clock or distinct clocks)" -- No-no! Above you and I were referring to "A (one specific) clock"; so in my understanding this refers to the (ordered) set (e.g. "$\mathcal S$") of indications of one specific identifiable participant, along with one specific coordinate assignment "$t : \mathcal S \rightarrow \mathbb R$". N.B.: [contd.] – user12262 Apr 24 '15 at 02:07
We can simply calculate an arithmetic difference between two $t$ values. But determining/evaluating (as a real number value) a particular duration ratio (of the specific participant under consideration; for some of her/his/its specific indications) is a more demanding task (cmp. again MTW §16.4). There's no "given plain arithmetic" way to evaluate "the difference between" any two distinct observed "positions of the little hand" (or for ratios of those). We must first seek, define, and then apply some unambiguous operational method for measuring. That's daily work/fun for (exp.) physicists. – user12262 Apr 24 '15 at 02:09
@Hypnosifl: "And would you agree non-ideal clocks" ... in technical jargon one (also) speaks of "bad" clocks, in distinction to "good" clocks (cmp. MTW Fig. 1.9) ... "measure time (approximately)" -- If (big IF) the "degree of approximation", here: the value "$u - 1$", is measured along with that. In other words: strictly, no. That's why we distinguish "measuring" from ... "gauging". Btw., saying "to measure time" (instead of "to measure duration") is a bit of a category error; like saying "to measure cars" while meaning "measuring their gas-milage". – user12262 Apr 24 '15 at 02:32
Concrete examples would help, since your way of speaking is idiosyncratic--if $S_w$ could be something like "the clock reads 45.2 seconds", the common English word for this would be the clock's "reading", I've never heard any one refer to it as an "indication" (and if that's not what you mean by 'indication', please clarify). And what does "one specific identifiable participant" mean? Do you just mean one specific physical clock, or does "participant" mean something different? When you say $t$ is a coordinate assignment, do you just mean any arbitrary coordinate system, or an inertial frame? – Hypnosifl Apr 24 '15 at 04:12
Finally I don't know what you mean by "duration ratio", is this supposed to be the ratio that different co-moving clocks show between the same pair of events, or something else? And if you're suggesting that in relativity the "duration" a clock measures between two events along its worldline (as opposed to the 'true' duration of proper time along its worldline between those points) is found by any procedure more complicated than subtracting the time it showed at the first event from the time it showed at the second, then I think you're confused, and MTW §16.4 says nothing like this. – Hypnosifl Apr 24 '15 at 04:19
@Hypnosifl: "Concrete examples would help" -- Well, consider indications of a "biological clock (of a person)"; or of a "forensic clock (of a corpse)" ... Or pretty much anything distinctive having to do with "hands, on a dial". "what does "one specific identifiable participant" mean?" -- Hmm ... What did Einstein mean in writing (about) "recognizable material points"? What did you mean in writing: "you" !? ... That's "(Exp.) Physics/Ontology 101". – user12262 Apr 24 '15 at 05:56
@Hypnosifl: "[...] opposed to the 'true' duration [...] along its worldline between [...]" -- Why "opposed"? No, I (merely) object to this quantity being referred to as "(proper) time". Einstein declared at the outset (Ann. Phys. 17, 891-892) that "time" is synonymous to "position of the hand of my watch", so "proper time" is a misnomer for some (related) quantity to be measured. The correct common name is: duration. "MTW §16.4 says [...]" -- MTW Box 16.4 presents the sketch "how to" even twice. (However: How to determine whether a given participant had been "free/geodesic"? ...) – user12262 Apr 24 '15 at 06:16
You didn't answer my question about if your idiosyncratic use of "indications" means the same thing as the standard terminology of "clock readings". And when I asked for concrete examples I didn't mean verbal examples of other types of clocks like a "biological clock", I meant something like "for example, let $S_w$ be the event of a clock reading 45.2 seconds, and $S_p$ be the event of the same clock reading 53.7 seconds, and let $S_p$ and $S_w$ be assigned coordinate times $t=98 s$ and $t=107 s$ respectively"...i.e. explaining how these symbols of yours would be applied to a specific example. – Hypnosifl Apr 24 '15 at 12:24
As for "recognizable material points", he just meant some physical object that we can assign a localized position coord. to at different times, like a particular marking on a ruler, or a particular particle. Then if we have two events A and B that occur next to this object at different moments, there is a unique "proper time" that is measured between A and B *along the object's own worldline" (proper time is specific to particular worldlines), which is just the time interval between A and B that would be measured by an ideal clock which remained next to the object (shared the same worldline). – Hypnosifl Apr 24 '15 at 12:28
...and that proper time measured by an ideal clock is what I meant by "true" duration. I don't know why you think "time" is synonymous with a particular reading of a watch rather than an interval between two events, and in general you make mutual understanding more difficult if you insist on substituting your own terminology for agreed-upon terminology (as I said I'm still not sure if your 'indications' means the same thing as 'readings', or if by 'durations' you mean what physicists normally call the interval of 'proper time' between events on a worldline, or coordinate time, or what). – Hypnosifl Apr 24 '15 at 12:32
@Hypnosifl: "You didn't answer my question about if your idiosyncratic use of "indications" means the same thing as the standard terminology of "clock readings"." -- I addressed your question above; but since you don't find that satisfying you should elaborate your idiosyncratic use of "the standard terminology of "clock readings"", please. Is there something specific that needs to be done before knowing/uttering, for instance ""3:56 pm"" ? Btw.: Let's hope that "the chat" will be supporting MathJax rendering soon, then allowing us to actually consider moving our correspondence there – user12262 Apr 24 '15 at 14:10
@Hypnosifl: "[...] for example [...] let $S_p$ and $S_w$ be assigned coordinate times $t=98s$ and $t=107s$ respectively" -- Hmm ... What's putting me off there, at the moment, is the letter "$s$" attached to the suggested $t$-coordinate assignments. (I'm more used to encoutering/challenging $t$-coordinate assignments that are strictly real numbers, as shown above; and generally coordinate tuples being elements of $\mathbb R^4$.) So what you suggest is apparently not meant as "just any (arbitrary) $t$ coordinates", but perhaps (even) as good coordinates of the given set $\mathcal S$. – user12262 Apr 24 '15 at 14:26
Let us continue this discussion in chat. – Hypnosifl Apr 24 '15 at 18:02

Naked Time ( Is there such a thing ?)

2 Answers2