remarks on the passing of time

8/12/2019 Remarks on the Passing of Time

1/17

Remarks on the Passing of Time

Author(s): Tim MaudlinSource: Proceedings of the Aristotelian Society, New Series, Vol. 102 (2002), pp. 259-274Published by: Wileyon behalf of The Aristotelian SocietyStable URL: http://www.jstor.org/stable/4545373.

Accessed: 25/06/2014 12:21

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at.http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

The Aristotelian Societyand Wileyare collaborating with JSTOR to digitize, preserve and extend access to

Proceedings of the Aristotelian Society.

http://www.jstor.org

This content downloaded from 169.237.113.67 on Wed, 25 Jun 2014 12:21:10 PMAll use subject to JSTOR Terms and Conditions
http://www.jstor.org/action/showPublisher?publisherCode=blackhttp://www.jstor.org/action/showPublisher?publisherCode=aristotelianhttp://www.jstor.org/stable/4545373?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/4545373?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=aristotelianhttp://www.jstor.org/action/showPublisher?publisherCode=black


2/17

XIV* REMARKSON THE PASSINGOF TIME

by Tim MaudlinI

This essayis the firstact of a two-actplay.My ultimateaimis to defend a simple proposition:time passes. To be moreprecise, I want to defend the claim that the passage of time isan intrinsic asymmetryin the structure of space-time itself, anasymmetry hat has no spatial counterpartand is metaphysicallyindependentof the materialcontents of space-time.It is indepen-dent, for example, of the entropy gradientof the universe. Thisview is part of common-sense,but has been widely attacked byphilosophers. The passage of time, we are told, is a myth, an

illusion, even an incoherentnotion. Becausethe notion that timepassesis common sense,it perhapsrequires ittlepositive defence;if there are no weighty objections to the view, it ought to beaccepted. So the first, and more important, act of the play isdefusingthe argumentswhich have been used to cast doubt onthe passage of time. I have positive argumentsto give, but nothaving space for them here, I will confine myself to an examin-ation of the common philosophical argumentsthat have beenused to cast doubt on the passage of time.The claim that time passes has no bearing on the question ofthe 'reality'of the past or of the future.I believethat the past isreal: there are facts about what happened in the past that areindependentof the presentstateof the world and independentofmy knowledgeor beliefs about the past. I similarlybelieve thatthere is (i.e. will be) a single uniquefuture.I know what it wouldbe to believe that the past is unreal(i.e. nothing ever happened,everything was just created ex nihilo) and to believe that thefuture is unreal(i.e. all will end, I will not exist tomorrow,I haveno future). I do not believe these things, and would act verydifferently f I did. Insofar as belief in the realityof the past and*Meeting of the Aristotelian Society, held in Senate House, University of London,on Monday, 10th June, 2002 at 4.15 pm.

This content downloaded from 169.237.113.67 on Wed, 25 Jun 2014 12:21:10 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp


3/17

260 TIM MAUDLIN

the future constitutes a belief in a 'block universe',I believein ablock universe. But I also believe that time passes, and see nocontradiction or tension between these views.Not infrequently, he notion of the passageof time is discussedunder the rubric 'time's flow', or 'the moving now'. These loc-utions tend to be favoured by authors intent on dismissingthenotion, and certainly subserve that purpose. For time to 'move'or 'flow', they say, there must be some second-order time bymeans of which this movementor flow is defined,and off we areup an infinitehierarchyof times.Quiteso. Exceptin a metaphor-ical sense, time does not move or flow. Rivers flow and loco-motives move. Of course, riversonly flow and locomotives onlymove because time passes. The flow of the Mississippi or themotion of a train consists in more than just the collections ofinstantaneousstateswhichhave differentrelativepositionsof thewaters of the Mississippi to the banks, or different relativepos-itions of the train to tracks it runson. The Mississippiflows fromnorth to the south, and the locomotive goes from, say, New Yorkto Chicago.The directionof the flow or motion is dependentonthe directionof the passageof time.Changeand flow andmotionall presuppose the passage of time, so the reality of change isbound up with the reality of time's passage, but we will avoidsaying that time itself changes or flows.There are three sorts of objections to the passage of time,which we may group as logical, scientific, and epistemological.Logical objections contend that there is something incoherentabout the idea of the passageof timeper se. Scientificobjectionsclaim that the notion of the passageof time is incompatiblewithcurrent scientifictheory, and so would demand a radical revisionof the account of temporal structureprovided by physics itself.Epistemological objectionscontend that even if there were sucha thing as the passageof time, we could not know that therewas,or in which direction time passes.For expository purposes, we are fortunate to have a recent textwhich touches on all of these sorts of objections. The passageoccursin Huw Price's book Time'sArrowand Archimedes'Point,under the heading 'The Stock Philosophical ObjectionsAboutTime'.' The heading is apt. These are, indeed, the very objections1. Price (1996), pp. 12-16.



4/17

REMARKSON THEPASSINGOF TIME 261which arise spontaneously in any discussion of these matters,and, with one exception, they are not original to Price himself.He has, however, done us a service in collectingand presentingthem in such a clearway, and I will use his expositionas a guide.

IILogical Arguments.The stock logical objection is presentedbyPrice as follows:

If it made sense to say that time flows then it would make senseto ask how fast it flows, which doesn't seem to be a sensibleques-tion. Some people replythat time flows at one second per second,but even if we could live with the lack of other possibilities,thisanswer misses the more basic aspect of the objection. A rate ofseconds per second is not a rate at all in physical terms. It is adimensionlessquantity, ratherthan a rate of any sort. (We mightjust as well say that the ratio of the circumference f a circleto itsdiameterflows at ir seconds per second )2

I think the 'more basic' aspectof this problemis indeed an orig-inal contributionof Price. It is, in any case, new to me. Let's dealwith the originalobjectionfirst.Let'sbegin by considering he logic of rates of change.If some-thing, e.g. a river,flows, then we can indeed ask how fast it flows(how many miles per hour, relative to the banks). To ask howfast a river flows is to ask how far it will have gone when acertain period of time has passed. If the Mississippi flows at 5m.p. h. to the south (and maintainsa constant rate), then afteran hour each drop of waterwill be 5 miles farthersouth. It willbe 5 milesfartherfromCanadaand 5 miles closerto the equator.On this basis, if we ask how fast time flows, i.e. how fast timepasses, we must mean to ask how the temporal state of thingswill have changed after a certain period of time has passed. Inone hour's time, for example, how will my temporal positionhave changed?Clearly,I will be one hourfurther nto the future,one hour closerto my death and one hour fartherfrom my birth.So time does indeed pass at the rate of one hour per hour, orone second per second, or 3,600 seconds per hour ...2. Price (1996), p. 13.



5/17

262 TIM MAUDLIN

What exactly is supposed to be objectionable about thisanswer?Price says we must 'live with the lack of other possibil-ities', which indeed we must: it is necessary, and, I suppose, apriori that if time passes at all, it passes at one secondpersecond.But that hardlymakes the answer eitherunintelligibleor mean-ingless. Consider the notion of a fair rate of exchange betweencurrencies.If one selects a standardset of items to be purchased,and has the costs of the items in various currencies,then onemay define a fair rate of exchange between the currenciesbyequalityof purchasingpower:a fairexchangeof euros for dollarsis however many euros will purchase exactly what the givenamount of dollars will purchase, and similarlyfor yen and yuanand so on. What, then, is a fair rate of exchange of dollars fordollars?Obviously,and necessarily,and a priori, one dollar perdollar. If you think that this answeris meaningless, magineyourreaction to an offer of exchange at any other rate. We do notneed to 'live with' the lack of otherpossibilities:no objectionableconcessionis required.What of Price's 'more basic' objection?This, I fear, is just aconfusion. A rate of one second per second is no more a dimen-sionless number than an exchange rate of one dollar per dollaris free of a specified currency.Price seems to suggest that theunits in a rate can 'cancelout', like reducinga fraction to sim-plest terms.Any ratedemands that one specifythe quantitiesputin the ratio:without the same quantities, one no longer has thesame rate.Suppose, for example, we are bartering loor tiles for liquoricesticks.The agreedrate of exchangemight be given in squarefeet(of tile) for feet (of liquorice).3We are exchanging squarefeet ofone thing for feet of another. It is simplya mistake to think wecan 'cancelout' one of the feet in each quantity and say that theexchangeis reallyin units of feet ratherthan squarefeet perfoot.Of course, the rate will transform like a linear measure if wedecide to change units:expressingour barterin terms of squareinches of tile for inches of liquorice, we multiply squarefeet by144and feet by 12.The realnumberused to expressthe rate (i.e.the real numberwhich stands in the same ratio to unity as the3. The exchange rate might also be given without any standard units at all, as thismuch tile (holding up a tile) to that muchliquorice (holding up a strand).



6/17

REMARKSON THE PASSINGOF TIME 263amount of tile stands to the amount of liquorice, in the givenunits) will be multipliedby 12, just as if we were simply trans-forming a linearmeasurefrom feet to inches. But still the unitsof the barter are square feet per foot, not feet.Similarly,Xrs defined as a ratio of a length (of the circumfer-ence of a Euclideancircle)to a length(of the diameter).The ratiois length to length:lengthdoes not 'cancelout'. By contrast, therate of passageof time at one secondper second is still a rate:it,unlike r, is a measure of how much somethingchanges per unittime.Price also mentions a logical objection to the notion of thedirectionof the passageof time:

If time flowed,then as with any flow-it would only make senseto assign that flow a directionwith respect to a choice as to whatis to count as a positive directionof time. In saying that the sunmoves from east to west or that the hands of a clock move clock-wise, we take for grantedthat the positive time axis lies towardwhat we call the future. But in the absence of some objectivegroundingfor this convention, there isn't an objectivefact as towhich way the sun or the hands of the clock are 'really'moving.Of course, proponents of the view that there is an objectiveflowof time might see it as an advantage of their view that it doesprovide such an objectivebasis for the usual choice of temporalcoordinate.The problem is that until we have such an objectivebasiswe don't have an objective sensein whichtime is flowing oneway rather than another.4

This objectiondemandssome untangling.First, it will help greatlyhere to insist that properlyspeakingtime passes, ratherthan flows, and that properlyspeakingany-thing that does flow only flows because time passes. The pointabout directionalityof flow is then exactly correct: flows onlyhave a directionbecause the asymmetry nherentin the passageof time provides temporal direction: from past to future. Thenatural thing is now to turn Price's Modus Tollens into a ModusPonens:since thereobviouslyis a fact about how the Mississippiflows (north to south) or how the hands of standardclocks turn(clockwise)there is equally a real distinctionbetween the future4. Price (1996), p. 13.



7/17

264 TIMMAUDLINdirection n timeand the past direction.The remarkabout choos-ing a convention for the 'positivedirection of time' is a red her-ring: t is, of course,merelya convention that our clocks typicallycount up (i.e. indicatelargernumbers as timepasses)ratherthancount down. Nothing in the objective nature of the passage oftimeprovidesan 'objectivebasis' for that choice. A societywhichhappened to build clocks that count down rather than up is notmaking any sort of mistake:attachingnumbers to moments oftime clearly requires purely arbitrary conventions. One whobelieves in the objectivepassage of time does not think there isan objective fact about which sort of clock is counting 'right'andwhich 'wrong', merelythat there is an objectivefact about whichis counting up and which down. Up-counting clocks show highernumbers in the future direction, down-counting clocks lowernumbers.To deny that there is an objective distinction betweensuch clocks is to deny that there is any objective distinctionbetween the future direction and the past, and that is preciselyto beg the question.This exhausts our examination of the Logicalobjections to thepassageof time.

IIIScientific Objections. Scientificobjectionsto the passageof timestem from two sources. One is the spatio-temporal structure pos-tulated by the Special and General Theories of Relativity, andthe other is the so-called Time Reversal Invarianceof the funda-mental laws of physics.Let me take these in turn.Kurt Godel providesa clear example of someone who thinksthat the space-time structure of Relativity is incompatible withthe passageof time. After a descriptionof the familiar'relativityof simultaneity' n the SpecialTheory, G6del writes:

Following up the consequencesof this strange state of affairs,oneis led to conclusionsabout the nature of time which are very farreaching ndeed.In short,it seems that one obtains an unequivocalproof for the view of those philosopherswho, like Parmenides,Kant, and the modernidealists, deny the objectivityof change andconsiderchangeas an illusion or an appearancedue to our specialmode of perception. The argument runs as follows: Changebecomespossible only through the lapse of time. The existence of



8/17

REMARKSON THE PASSINGOF TIME 265an objective apseof time, however,means(or at least is equivalentto the fact) that reality consists of an infinity of layers of 'now'which come into existence successively. But, if simultaneity issomething relative in the sense just explained, reality cannot besplit up into such layers in an objectivelydeterminedway. Eachobserverhas his own set of 'nows', and none of these various sys-tems of layerscan claim the prerogativeof representing he objec-tive lapse of time.5

Godel then goes on to describe his famous solution to theEinstein field equations which not only cannot be 'split up intolayers' (i.e. foliated into spacelike hypersurfaces)in a singleobjective way, it cannot be foliated into spacelike hypersurfacesat all.So we are left with two questions. First, does the passage oftime imply a foliation of space-time, i.e. does it imply that thefour-dimensionalspace-time structure is split into a stack ofthree-dimensional slices in an observer-independent way?Second, if it does, does this set the notion of passageof time indirectoppositionto the account of space-timeofferedby our bestscientifictheories?To the first question, I can find no justification for Godel'sblank assertion that the 'objective lapse of time' is 'equivalent'to the fact that reality is a stack of 'now's. The passage of timeprovides,in the first instance, a fundamentalobjectivedistinctionbetween two temporal directions n time: the direction from anyevent towards its future and the directionfrom any event towardsits past. If we want to distinguish,for example, an asteroid goingfrom Earth to Mars from an asteroid going from Mars to Earth,what do we need?We may focus completely on the world-line ofthe asteroid in question. Everyone agrees that one end of theworld-linehas Earthin the near vicinity of the asteroid, and theotherend has Marsin the nearvicinity:these facts do not requirea foliation of the space-time.Does addinga foliation help to anydegreeat all in determiningwhetherwe have an Earth-to-Marsor a Mars-to-Earthtrip?No. For even if we were to add thefoliation, the crucial question of which events come first andwhich later would be unsettled. So the 'lapse of time' cannot beequivalentto the existence of a foliation.5. Godel (1949), pp. 557-8.



9/17

266 TIMMAUDLINPerhaps the 'lapse of time' is a foliation plus a specificationofwhat the past-to-future direction is? This will certainly allow us

to distinguish two trips from each other- but it is the direction,not the foliation that is doing all the work. Give me the past-to-futuredirectionand I'll tell you where the asteroid is going with-out referenceto any foliation at all.In a fully relativistic space-time, the obvious mathematicalgadget one needs to distinguish one direction in time fromanother is not a foliation but an orientation.All relativisticmod-els already employ orientablespace-times:space-timesin whichthe light-conesare divided into two classes, such that any con-tinuous timelike vector field contains only vectors which lie inone of the classes. In orderto account for direction of flows, orothermotions, all we need to do is to identifyone of these classesas the future light-cones and the other as the past light-cones.Once I know which set is which, I can easily distinguisha Mars-to-Earth asteroid from an Earth-to-Marsone.What then of an orientation?Is that somethingwhich physicsalone can do without, which is being added to our account ofspace-time for merely 'philosophical' reasons? Is it, further,somethingthat would sit uneasilywith the spatio-temporal truc-ture posited by physics alone?The treatment of this question is one of the most peculiarinthe philosophicalliterature.The usualapproachsets the problemas follows: the fundamentalphysical laws have a feature called

'Time Reversal Invariance'. If the laws are time-reversalinvariant,then it is supposed to follow that physics itself recog-nizes no directionalityof time: it does not distinguish,at the levelof fundamental aw, the future-direction rom the past-direction,or future light-conesfrom past light-cones. Therefore, it is said,any such distinctionmust be groundednot in fundamental aw,or in the intrinsicnature of the space-time tself,but in contingentfacts about how matter is distributed through space-time. Thedirection of time, we are told, is nothing but (e.g.) the directionin whichentropyincreases.The philosophical puzzle is then howto relate various other sorts of temporal asymmetry(the asym-metry of knowledge, or of causation, or of control) to theentropic asymmetry. (Paul Horwich'sAsymmetries n Timepro-vides an exampleof this form.)66. Horwich 1988).



10/17

REMARKS N THEPASSINGOFTIME 267This problematic is peculiar because it fails at every step.Tobegin with, the laws of physicsas we have them (even apartfrom

worrying about a coherent understanding of quantum mech-anics)are not Time Reversal Invariant.The discoverythatphysi-cal processesare not, in any sense, indifferent o the directionoftime is importantand well known:it is the discoveryof the viol-ation of so-called CP invariance, as observed in the decay of theneutral K meson. These decays are not invariant if one changesa right-handed or a left-handed spatial orientation(parity)andchanges positive for negative charge (charge conjugation).According to the CPT theorem, any plausible quantum theorywill be invariant under parity-plus-charge-conjugation-plus-time-reversal, so the violation of CP implies a violation of T. In short,the fundamental laws of physics, as we have them, do require atemporal orientation on the space-time manifold. So the argu-ment given above collapses at the first step.

How do philosophers respond to this difficulty? Horwich, hav-ing noted the problem, writes:

However, this argumentis far from airtight. First, the predictionhas not been directlyconfirmed.And, even if it weretrue, it couldturn out to be a merelydefJmctosymmetry,whichdoes not involvetime-asymmetricalaws of nature. Moreover neither the exper-imental nor the theoretical assumptions nvolvedin the predictionarebeyondquestion.For the frequencydifferencebetween the twoforms of neutral K mesondecayis not substantialand will perhapsbe explainedaway. Anyway, the assumptionthat these processesare spatialmirrorimages may turn out to be false ... Finally, theso-called 'CPT theorem', though plausible, may be false. Sincethere are so many individuallydubious assumptionsin the argu-ment, we may regardtheirconjunction as quite implausible.7There is a certain air of desperation about this passage. Thereis no dispute in the physics community about the reality or impli-

cations of this effect: Nobel prizes have been awarded both forthe theoreticalwork and for the experimentalverificationof theeffect. Insofar as philosophers of physics are looking to actualscientific results to base ontological conclusions on, this is a clearcase of the science testifying in favour of a temporal orientation.7. Horwich (1988), p. 56.



11/17

268 TIM MAUDLIN

The only plausible reason for Horwichsuddenly to turn scepticalis that the failureof T invariancespoils his argument.There is a somewhat better responseavailable.That would beto admit that the laws of physics are not Time ReversalInvariant, and that there is, indeed, a physical orientation tospace-time,but to insist that the physical processes sensitive tothis orientation(like neutral kaon decay) are too infrequentandexotic to explain the widespread and manifest distinctionbetween the past-directionand the future-directionhatwe beganwith. We will examine the plausibility of this response in duecourse. But we should at least acknowledge that the admissionof an orientation to space-time s not, per se, wildlyat odds withpresent physical theory since present physical theory alreadyadmitsone.But let's set aside the observed violations of CP invariance.Even apartfrom these, it is not at all clear that the acceptedlawsof physicsareTime ReversalInvariant n a way that suggeststhat

there is no intrinsicdirectionof the passage of time. This pointhas been well argued by David Albert in his recent book Timeand Chance.8Here's the problem. The way that time-reversalhas beenunderstood since the advent of electro-magnetictheory is not:for every physicallyallowedsequenceof instantaneousstates,thesame set of states in the reverse time order is also physicallyallowed. This can plausibly be argued to hold for Newtonianmechanics,9but beyond that one needs to do more than simplyreverse the time order of the states: one has to perform a 'time-reversal'operationon the instantaneousstates themselves.So thetheoremis now: for every physicallyallowablesequenceof states,the inverse sequence of time reversed states is also physicallyallowable. More precisely: f states To, T1, T2, ..., TN are physi-cally allowable as time runs rom Toto TN nsomedirection, henthe sequence T*, ..., T*, T*, To is also allowed, where the *representsthe time-reversaloperation as applied to the states,and where time runs from T* to To in the same direction as itruns froin Toto TN. As stated, this resultdoes not even suggest8. Albert (2000), chapter 1.9. One has to be clear about what it means for a state to be instantaneous: instan-taneous states, for example, do not include velocities or particles, or, more generally,the rate-of-change-with-time of any quantity. See Albert's discussion.



12/17

REMARKSON THEPASSINGOF TIME 269that time fails to have a direction at all. Indeed, the necessityof invoking the time-reversaloperation on instantaneousstatessuggestsjust the opposite: it suggests that even for an instan-taneous state, there is a fact about how it is orientedwithrespectto the directionof time.Given certain facts about the time-reversaloperator (in par-ticular,giventhat particlepositionsor fieldvaluesdo not changeunderthe operation),Time ReversalInvariance,as statedabove,is still a very importantfeature of physicallaws. For, as Albertinsists, this sort of Time Reversal Invarianceimplies that in acertainsenseanythingthat happensin the universecan, as a mat-ter of physicalprinciple,happen'in reverse': f ice cubes can meltinto puddles, then puddles can spontaneouslyfreeze back intoice cubes, for example. But notice that this sense of 'in reverse',far from delegitimizingthe passage of time, presupposesit: themelting of an ice cube 'in reverse'requiresthat the puddle stageprecedethe ice cube stage. So to get from Time ReversalInvari-ance of the laws (in this sense)to the objectiveabsenceof a direc-tion of time would require some extensive philosophicalargumentation n any case.

IVEpistemicObjections.Let's returnfor moment to the violation ofCP invariancedisplayedin neutral kaon decay. We noted abovethat this phenomenonseems to imply that the laws of naturearenot Time Reversal Invariantin any sense, and hence that thelawsthemselvesrequirean intrinsicasymmetry n timedirections,and hence that space-timeitself, in order to support such laws,must come equipped with an orientation. But one might stillobject that this orientation (let's call it Kaon Orientation)hasnothing to do with the supposed large-scaleand manifestasym-metry involved in the passage of time. For it is only in ratherspecialand reconditecircumstances hat Kaon Orientationmani-fests itself, while the passage of time is evident and ubiquitous.More directly, even if there is an intrinsicKaon Orientationtospace-time, in most normal circumstancewe would not be in aposition to determinewhat it is.We can call this sort of objection an epistemic objection:itdoes not directlydeny the logical or physicalacceptabilityof the



13/17

270 TIM MAUDLINexistenceof a fundamentaltemporalasymmetry,but insists thateven if such an asymmetryis postulated,we would not be ableto tell what it is. In the case of the Kaon Orientation,the objec-tion is merely practical:kaon decay experiments or other physi-cal phenomena whose outcomes depend on the KaonOrientation) are relatively rare and inaccessible.But the epis-temic objection can be raised to an even more thoroughgoingand radical manner, in form which, if valid, would make thepassageof time epistemicallyopaque even if physicallyreal.Price raises the objection in one way, D. C. Williams inanother. Let's begin with Price's formulation:

In practice,themost influentialargument n favourof the objectivepresent and the objectiveflow of time rests on an appeal to psy-chology-to our own experienceof time. It seemsto us that timeflows, the argument runs, and surely the most reasonableexpla-nation of this is that there is some genuine movement of timewhich we experience,or in whichwe partake.Arguments of this kind need to be treated with caution, how-ever. After all, how would things seem if time didn't flow? If wesupposefor the moment that thereis an objectiveflow of time, weseem to be able to imagine a worldwhich would be just like ours,except that it would be a four-dimensionalblock universe ratherthana three-dimensional ynamic one. It is easy to see how to mapevents-at-times n the dynamic universeonto events-at-temporal-locations in the block universe.Among other things, our individ-

ual mental states get mapped over, moment by moment.But thensurelyourcopiesin theblock universewould have the same experi-ences that we do-in which case they are not distinctive of adynamicaluniverse after all. Things would seem this way, even ifwe ourselveswere elements of a block universe.10The short diagnosis of the foregoing is that it is an argumentby 'surely': Price simply asserts that the crucial conclusion fol-lows from his premises even though it is by no means evidentthat it does. The point can perhaps be made more clearlyif weswitchto the form that Williamsuses.That form begins by grantingthat there is a directionof time,that the past-to-futuredirection differs from the future-to-pastdirection. But now the observationis made that we also accept

10. Price (1996), pp. 14-5.



14/17

REMARKSON THE PASSINGOF TIME 271

the Time ReversalInvariance let'sset asidethe kaon interactionsas irrelevant here) as stated above: for physically possiblesequence of states To, T1, ..., TN running from past to future,there is physically possible sequence TN, ..., T1, To running frompast to future. For example, giventhe actualsequenceof physicalstates of your body over the last ten minutes, the time-reversedsequenceof time-reversed tates is also physicallypossible.Some-where on some other planet (as far as the laws of physics go)some such sequencecould exist, unproblematicallyime reversedrelative to the sequence of states which make you up. Let's callthis sequenceof statesyour time-reversedDoppelganger.But, theobjection goes, thereis an obvious one-to-onemappingfrom theDoppelganger's states to yours. So the Doppelganger wouldsurelyhave qualitatively denticalexperiences o yours, only withthe whole process oppositely oriented in time. Here is Williams'sdescription:

It is conceivable too then that a human life be twisted, not 90?but 180?,from the normal temporalgrain of the world. F. ScottFitzgeraldtells the story of BenjaminButton who was born in thelast stages of senility and got younger all his life till he dies adwindlingembryo.Fitzgerald maginesthe reversal o be so imper-fect that Benjamin'sstream of consciousnessran, not backwardwith his body's gross development, but in the common clockwisemanner. We might better conceive a reversalof every cell twitchand electronwhirl,and hencesuppose that he experiencedhis ownlife stages in the same order as we do ours, but that he observedeverythingaround him moving backward from the grave to thecradle.

If we accept that the relevant physics is Time ReversalInvariant,then we accept that your time-reversedDoppelgangeris physically possible. Let's suppose, then, that such a Doppel-ganger exists somewhere in the universe.What should we con-clude about its mental life?The objector, of course, wants to conclude that the mentalstate of the Doppelganger is, from a subjectiveviewpoint, justlike ours. So just as we judge the 'direction of the passage totime' to go from our infant stage to our grey-haired, o too withthe Doppelganger. But that direction, for the Doppelganger, is" Williams (1951), p. 113.



15/17

272 TIM MAUDLIN

oppositelyoriented to ours. So the Doppelgangerwilljudge thatthe temporaldirection into the futurepoints opposite to the waywejudge it. And if we insist that there is a directionof time, andwe know what it is, then we must say that the Doppelgangerisdeceived, and has mistaken the direction of time. But now webecome worried:the Doppelganger seems to have exactly thesame evidenceabout the direction of time as we do. So how dowe know that (as it were) we are not the Doppelgangers,that weare not mistaken about the direction of time. If there is a direc-tion of time, it would seem to become epistemically naccessible.And at this point, it seems best to drop the idea of such a direc-tion altogether.But is this correct?In order to facilitatethe discussion,I will referto correspond-ing bits of the Doppelgangerwith a simple modification of thetermsfor parts of the originalperson. For example, I will speakof the Doppelganger's neuron*s: these are just the bits of theDoppelganger that correspond,under the obvious mapping, tothe original's neurons. We can unproblematicallysay that theDoppelganger's neuron*s fire*, meaning that they undergo thetime-reverseof the process of a normal neuronfiring.It may bethat neuron*scan be properlycalled neurons and the firing*maybe properlycalledfiring,but we do not want to presupposethatat the outset, so one must remainalertto the asterisks.So the first question is: given the physical descriptionof theDoppelganger that we have, what can we conclude about itsmentalstate? The answer,I think, is that we would have no rea-son whatsoever to believe that the Doppelganger has a mentalstate at all. After all, the physicalprocesses going on the Doppel-ganger'sbrain* are quite unlike the processesgoing on in a nor-mal brain. Nerve impulses* do not travelalong dendritesto thecell body, which then fires a pulse out along the axon. Rather,pulses travelup the axon* to the cell body*, which (in a ratherunpredictableway) sends pulses out along the dendrite*s.Thevisual system*of the Doppelganger is also quite unusual:ratherthan absorbing light from the environment, the retina*s emitlight out into the environment. (The emitted light is correlatedwith the environment n a way that would seemmiraculous f wedid not know how the physical state of the Doppelgangerwasfixed:by time-reversinga normalperson.) There is no reason tobelabourthe point: in every detail, the physicalprocessesgoing



16/17

REMARKS ON THE PASSING OF TIME 273

on in the Doppelgangerare completelyunlike any physical pro-cesseswe have ever encounteredor studied in a laboratory, quiteunlike any biological processes we have ever met. We have noreason whatsoever o suppose that any mental state at all wouldthe associated with the physical processesin the Doppelganger.Given that the Doppelgangeranti-metabolises,etc., it is doubtfulthat it could even properlybe called a living organism (ratherthan a living* organism*),much less conscious living organism.Now the response is likely to chime in: the Doppelganger'sphysical state is not unfamiliar: t is just like ours save for thedirectionof time. That is so: but the difference s no minor one.It turns emission into absorption, normal metabolic processesinto weird and unexampledanti-thermodynamic nes.No, no, the responseinsists: since the gradientof the entropy(or whatever)has been reversed, the directionof time itself hasbeenreversed,and, oriented to the thus-defineddirection of time,the physical processesare just like normal ones. This response,of course has a name:petitio principii.The aim of the argumentit to show that there is no intrinsicdirection to time, but only,say, an entropy gradient.But it achieves its aim only if we areconvinced that the Doppelganger has a mental state 'just likeours', and the only way to make that claim even vaguely plaus-ible is to assert that the Doppelganger's physical state is not,in any significantsense, time-reversed relative to any physicallysignificantdirection of time) at all. And that is preciselyto begthe question.Havingworkedthroughthe responseto Williams,the responseto Price is even more stark.He imaginesa Doppelgangerwhichis not just reversed n time, but a Doppelgangerin a world withno objective low of time at all, i.e. (accordingto his opponent)to a world in which there is no time at all, perhaps a purelyspatial four-dimensionalworld. So it not just that the nervepulse*s of this Doppelganger go the wrong way (compared tonormal nerve pulses), these nerve pulse*s don't go anywhere atall. Nothinghappensin this world.True, there is a mapping frombits of this world to bits of our own, but (unless one alreadyhasbegged the central question) the state of this world is so unlikethe physical state of anything in our universe, that to supposethat therearemental states at all is completelyunfounded. (Evenpure functionalists,who suppose that mental states can super-vene on all manner of physical substrate, use temporal notions



17/17

274 TIM MAUDLIN

in defining the relevant functional characterizations.Even purefunctionalistswould discern no mental states here.)All the participants n this debate accept the supervenienceofthe mental on the physical. If you believein the passingof time,then that is among the relevant physical characteristicsof thesubvenience base. To simply suppose that it is not, that thecharacter of mental states cannot possibly depend on how timeis passing for the brain states, and to disguise this criticalassumptionwith no more than an offhand 'surely', s to miss the

natureof the dispute altogether.So none of the argumentsfor the epistemic inaccessibilityofthe direction of the passageof timegoes throughwithout alreadybegging the questionat hand. And havingexhausted the logical,the scientific and the epistemic arguments, there are no stockarguments left. The usual philosophical arguments,which haveinduced Price and Williams and many others to rejectan objec-tive passage of time, have no force whatsoever.Departmentof PhilosophyRutgers University26 Nichol AvenueNew Brunswick,NJ 08901USA

REFERENCESAlbert, D., 2000, Timeand Chance Cambridge,MA:HarvardUniversityPress).Godel, K., 1949,'A RemarkAbout the RelationshipBetweenRelativityTheoryand the IdealisticPhilosophy', n P. Schilpp(ed.),AlbertEinstein.Philosopher-Scientist (La Salle, IL: Open Court), 557-562.Horwich, P., 1987,Asymmetries n Time(Cambridge,Mass.: MIT Press).Price, H., 1996,Time'sArrowandArchimedes' oint (Oxford:Oxford UniversityPress).Williams, D.C., 1951, 'The Myth of Passage', Journalof Philosophy,48. Pagenumbersgiven are for the reprintedversion in R. Gale (ed.), The Philosophyof Time(GardenCity, NY: Anchor),98-116.

remarks on the passing of time

Documents