since 05 February 2011 :
View(s): 1972 (8 ULiège)
Download(s): 194 (0 ULiège)
Claudio Cormick

Against the “view from nowhen”: A Merleau-Pontyan contribution to Dummett’s approach to McTaggart’s paradox

(Volume 10 (2014) — Numéro 2)
Open Access


This paper will attempt to explore the fecundity of Merleau-Ponty’s analysis of time, by means of showing how it can be linked with a problem whose origin is external to the phenomenological tradition: Michael Dummett’s approach to McTaggart’s paradox. With this purpose, I will make explicit the striking parallelism between the Merleau-Pontyan “situational” conception of time (that is, his tenet that time can only exist for a subjective perspective situated in time itself) and Dummett’s view that time can not appear in a “complete description” of reality, in other words, that the flow of time vanishes if we try to describe reality without any point of view (a description that would amount to what has been called “a view from nowhen”). I will try to show (via an incidental polemic with recent inter­pretations, such as Hoy’s, which turn Merleau-Ponty into an “idealist”) how the French phenomenologist’s analysis contributes to support Dummett’s tenet concerning the tension between time and a non-situated description of reality, as well as to prove that his notion of time’s synthesis as a transition-synthesis makes it possible to include, in the same account, a “pluralist” element along with a combinability of the different temporal perspectives.

1. Recalling McTaggart’s paradox

1As it is known, McTaggart reconstructs the problem of temporality on the basis of a distinction between two series, “A” and “B,” into which time can be analyzed. Whereas the relations of the second series (relations of before and after) characterize two events in an invariable manner (given that if an event has ever been previous to another one, it will always be, no matter if they are present, past or future), the determinations of the A series must, precisely for there to be a “flow” of time, be subject to constant change.1 The first part of the argument of the British philosopher, let us recall, attempts to show that these transformations of A-determinations (the future becoming present, and the present past) constitutes the only “change” we can strictly think about, and, insofar as time requires change, the A series is, then, a condition of possibility of time.2

2But we have to add to this variability of A-determinations the fact that they are mutually excluding.3 As a consequence, any moment M will have to acquire the determinations of being present, being past and being future, even though their conjunction is inconsistent. And this leads us to a third point, which is actually the core of McTaggart’s paradox: in which way can A-determinations be attributed to a moment; are they absolute or relative? For instance, given their variability, it would seem incomplete simply to say, about a moment M, that it is future; insofar as it belongs to an A-series, it also has to be present or past. The obvious solution to this problem would apparently be to declare that this moment M, which is future, will be present and then past, and consequently the three determinations are “distributed” in a way that no inconsistency appears. Nevertheless, this amounts to establish­ing that M is future from the point of view of some particular moment, but from other points of view it will be present or past. If we obtain here any unequivocal A-determinations (if we can say, in our example, that a certain moment is future, and not attribute to it another temporal character), it will be according to a certain relation with some other moment in time, but precisely this other moment from which we attribute a determination, the moment that functions for us as a reference point, makes us face again the initial problem: this other moment has to be, according to where we consider it from, both present and past or future. And so on, indefinitely.4 If we attribute an A-determination as relative to another moment in time, the same inconsistency presents itself with respect with this other moment, which is homogeneous to the first.

2. Dummett’s way—out?

3In a classical article included in his 1978 book Truth and other enigmas,5 Michael Dummett has argued that the problem of the consistency of A-determinations can be traced back to a question which he conceives as more basic: the one concerning whether we can abandon the requisite of complete­ness in a description of temporal reality. If this requisite was abandoned, this line of argument follows, the inconsistency would vanish.

4The argument appears in the context of what the title of the article calls a “defense” of McTaggart’s problematization insofar as it shows its specifically ontological pertinence and depth, which Dummett believes must be acknowledged even if we do not agree with its counterintuitive result. In particular,6 his defense attempts to dispel the appearance that McTaggart’s tenets do not constitute a problem beyond that of the correct use of deictic terms. This defense is not (or attempts to be), by itself, a solution to the paradox. And, remarkably, between the two great halves in which McTaggart’s argument is divided, Dummett does not focus on the second one (which attempts to prove, let us recall, why an A series is contradictory), but on the first, namely, why specifically time (in apparent contrast with other realities as space, as Dummett remarks) is only describable from a point of view situated in time itself, in virtue of requiring “situational” deter­minations as those of the A series.7 He considers that the key to the problem, and what consequently is metaphysically interesting in McTaggart’s work, is that the latter “is saying that […] a description of events as taking place in time is impossible unless temporally token-reflexive expressions enter into it, that is, unless the description is given by someone who is himself in that time.”8 That is, it is not possible to describe a temporal sequence from a “God’s Eye View”; instead, we need a “someone” in relation with whose perspective we can say that certain events happened, others are happening and others will happen. Or, in other words, someone who experiences events in succession (which is how temporal phenomena are presented to our consciousness) needs, in order to describe this experience, to distinguish these events by means of deictic expressions as “now.”9 According to Dummett, we only get to reconstruct time when we include in the description the “movement” of our consciousness, that is, the circumstance that it experiences successively different moments as present ones. Dummett illus­trates this with the following analogy:

We are […] inclined to assume that what we observe at any one time is a three-dimensional segment of a static four-dimensional physical reality; but as we travel through the four-dimensional structure we observe different three-dimensional segments at different times. But […] the fourth dimension can no more be identified with time than the road down which someone travels can be identified with the time that passes as he travels down it. If our hypo­thetical observer observes only the four-dimensional configuration without observing our movement—the movement of our consciousness—through it, like someone observing the road but blind to the traveler, he does not see all that happens. But if he also observes our passage through it, what he is observing is no longer static, and he will again need token-reflexive expres­sions to report what he observes.10

5In other words: it would be possible to conceive of the totality of the real as composed by four dimensions, and in this way, in the same manner as in respect to space we assume that there exists an already given three-dimensional reality through which we simply travel, we could also think of the succession of our experiences as a “movement” through a static four-dimensional reality, a reality which (for someone who was able to observe it under that four-dimensional presentation) does not change, but of which we perceive different three-dimensional “segments” each time. Nonetheless, according to Dummett, the fourth dimension of this static reality would not be time, if time is precisely what is distinctive of our experience, insofar as this experience is not static. We could say that this fourth dimension “re­presents” time, in a certain “model,” but an observer who held to this static model would not have the experience of time itself.11 To recover that experience, that is, to recover change, we need to abandon a non-situated description of reality and refer to reality using deictic expressions, distinguish what happened, happens or will happen, which presupposes to situate oneself in a particular temporal perspective, that of the present. (As it is obvious, the character of present must be attributed to ever varying moments, but we will always be describing from a present, not from the non-place of a subject who was able to observe all reality as simultaneous).

6The problem to be explained is, clearly, the conclusion that, from this first result, is drawn by the author of the paradox; that is, how it is that once that “part one of McTaggart’s argument establishes that what is in time cannot be fully described without token-reflexive expressions,” then part two can “enable us to pass from this to the assertion that time is unreal,” instead of assuming, on the contrary, that the first one of them has already demons­trated “the reality of time in a very strong sense,” since time “cannot be […] reduced to anything else.”12

7Now, this is when Dummett puts in conjunction with a first very im­portant tenet (the one we have seen according to which McTaggart’s analysis shows the impossibility of describing time without a situated observer), an­other, complementary, tenet, which points out that, according to McTaggart, reality must be something describable without those resources. In Dummett’s words,

McTaggart is taking for granted that reality must be something of which there exists in principle a complete description. […] The description of what is really there […] must be independent of any particular point of view. Now if time were real, then since what is temporal cannot be completely described without the use of token-reflexive expressions, there would be no such thing as the complete description of reality. There would be one, as it were, maxim­al description of reality in which the statement “The event M is happening” figured, others which contained the statement “The event M has happened,” and yet others which contained “The event M is going to happen.”13

8The distinction between these two sorts of descriptions—a single “complete” description and a variety of “maximal” ones—could perhaps be better under­stood if we call the first “absolute,” to do justice to Dummett’s words in calling it “independent of any particular point of view.” In other words, the problem with time resides in that a description of reality as temporal will never be able to directly dispense with points of view to show us reality as given at once, but can only multiply them. The event M will be describable in a variety of ways, as present or as past or as future, according to “from where” we look at it, but this is all we can obtain; not a description in which M lacks all deictic determinations. Thinking about reality in this way is incompatible with the (plausible) tenet that our finite, situated perspectives are perspectives on something real which should be describable having disposed of all of them. On the basis of considering this general ontological assumption, Dummett believes that he has finally been able to reconstruct the foundations of McTaggart’s argument:

I personally feel very strongly inclined to believe that there must be a com­plete description of reality […]; that of anything which is real, there must be a complete—that is, observer-independent—description. Hence, since part one of McTaggart argument [this is, that time requires an A series] is certainly correct, his conclusion appears to follow that time is unreal.14

9In this way, McTaggart’s argument on time’s contradictoriness would be better understood, according to Dummett, once we give it the following form: if time cannot be described without deictic expressions, and what is real must be described without them, then time cannot be real.

10Once he has reached this point, the last step in Dummett’s line of thought (in order not to have to conclude, with McTaggart, that time does not exist) does not consist of thinking an alternative reconstruction of McTaggart’s argument, which might show some flaw in it. Dummett simply states, instead, that the problem of proving the unreality of time is that it does not account for, at least, the appearance, the illusion, that phenomena are temporal; does not account for our apprehension of them, even if the world itself is static.15 As a consequence, in any case, if we cannot conclude that time is unreal, “then McTaggart’s argument shows that we must abandon our prejudice that there must be a complete description of reality.”16

11Dummett’s reading, then, attributes the following tenets to McTaggart:

121. If there are temporal phenomena, then there exist phenomena which require a situated description, supported on the point of view of a subjectivity (“someone”) within time itself.

132. But, on the other hand, everything which is real must be susceptible of a complete description, that is, without any point of view. There cannot exist phenomena that require a situated description.

143. It is in virtue of this tension between the nature of time and the requisites that define reality (in other words, in virtue of the fact that we cannot accept the consequent of the conditional expressed in (1)) that time cannot be real.17

15Dummett’s point, then, is ontologically challenging. According to William Lane Craig, his “defence” of the ontological depth of McTaggart’s argument would have shown that “the reality of tense”—that is, of the determinations of present, past and future—

implies that there can be no complete and consistent characterization of a possible world in terms of true propositions and states of affairs. If pro­positions are held to be tensed, then any characterization of a possible world in terms of a maximal conjunction of consistent propositions is bound to be radically incomplete, indeed, to characterize a possible world only at t.18

16Nevertheless, on closer inspection, Dummett’s article involves at least two important points that we must now face.

17First, we know that (as stated in (1)) if a reality is going to be genuine­ly temporal, then it requires to be described from a certain point of view; it is not susceptible of a “complete description.” A non-situated vision of reality would not be able to reconstruct time; without a perspective which is, itself, temporally situated, the moments described would neither be present, past or future. In spite of that, Dummett is imprecise as to whether, once this incom­patibility between time and a “complete description of real­ity” has been made explicit, this provides us with the solution for the problem of the un­reality of A-determinations, or not. This is to say, we do not know if we must abandon the idea of a “complete description” or not (that is, abandon premise (2)).

18In other words, we find on the one hand the problem of the necessity of a situated description if that which we are going to describe is genuinely temporal, and, on the other hand, the problem of the possibility of that description. Dummett poses a conditional statement to face the problem of how time should be described, but then fails to clarify whether he himself considers that the consequent of the conditional is acceptable. As we have seen, he hesitates between declaring that “he feels inclined” to reject this consequent and suggesting that, however, we might abandon the “prejudice” of a complete describability of the real; that is, that we might accept the consequent of the conditional. This ambiguity, in fact, has generated in the literature a curious situation in which, at the same time, there exist authors who speak about a “McTaggart-Dummett argument,” attributing to the latter an agreement with the premise which was implicit in McTaggart’s argument, and other interpreters who, on the contrary, consider that Dummett is bringing forth a possible solution by means of quitting that premise.19

19But the second point to consider in Dummett’s approach is that, aside from not being straightforward in what concerns abandoning or not the onto­logical principle of a complete describability, it has been questioned whether this abandonment is not simply an ad hoc solution for McTaggart’s paradox. According to Paul Horwich, McTaggart’s argument “depends on exposing a contradiction between facts ‘E is past’, ‘E is present’, and ‘E is future’, all of which must […] belong to the totality of absolute facts in the world.” Dummett’s point—Horwich goes on—is that

this argument requires the assumption (which […] is questionable) that there is such a totality of facts. If there is no such thing—if the facts change from one temporal perspective to another—then the only troublesome contra­dictions are contradictions from a particular temporal perspective. But a ‘moving now’ does not require that E be past, present, and future from a single temporal perspective. So if there is no time-neutral body of absolute facts, there is no contradiction. Thus, by denying the assumption of this total­ity, McTaggart’s objection can be sidestepped.

But only at substantial cost. For the crucial move—denying the assumption that there is a totality of facts—seems quite bizarre, unless it is independently motivated.20

20As we will argue, the tenet that the structure of temporality reasonably chal­lenges the pretentions of a “complete description” of reality does turn out to be tenable, and, as a consequence, a relevant consideration for the treat­ment of McTaggart’s paradox. If, based on Dummett, all it is clear is that there is no chance of both affirming the existence of time and attempting to describe it from a non-situated point of view, to go beyond this dilemma our analysis will move to Merleau-Pontyan phenomenology. In this way, we will be able to reconstruct a phenomenological argument as to why it is necessary that temporal reality be described from a present point of view, in absence of which the description simply ceases to have A-determinations—or be temporal in any sense.

3. From Dummett back to Merleau-Ponty: some “situational” theses on time

21As a consequence, we will reconstruct Merleau-Ponty as a source of argu­ments in favor of the idea that time in general, and A-determinations in particular, depend on a situated subjectivity. Let us begin, then, by establish­ing the key continuity between Dummett’s analysis and Merleau-Ponty’s theses on time. In this perspective, a passage from the beginning of the chapter “Temporality” in Phenomenology of perception turns to be par­ticularly revealing:

The ‘events’ are shapes cut out [sont découpés] by a finite observer from the spatio-temporal totality of the objective world. But on the other hand, if I consider the world itself, there is simply one indivisible and changeless being in it. Change presupposes a certain position which I take up and from which I see things in procession before me: there are no events without someone to whom they happen and whose finite perspective is the basis of their individuality. Time presupposes a view of time.21

22Under the light of these propositions, a series of Merleau-Pontyan theses can be identified: the idea of a “fragmentation” appears in parallel with a reference to the “individuality” of events as “founded” in a finite perspective, whereas, in the absence of such perspective, there is nothing more than “a single indivisible being.” The meaning of the passage is, then, evident: it is only in virtue of the subjectivity having a finite perspective that, instead of perceiving “a single indivisible being,” it fragments that totality turning it into a succession of individual events. In this context, it is also clear (on the basis of the reference to the “place where I am situated”) that we only perceive time from a certain point of view in the totality of the temporal flux itself; we are “in” a moment, which is associated to a “point of view,” in an analogous manner to the way we are in a spatial place, and from that moment it is only a finite part of reality what we can perceive. “Time exists for me,” Merleau-Ponty insists, “only because I am situated in it, that is, because I become aware of myself as already committed to it, because the whole of being is not given to me incarnate [ne m’est pas donné en personne].”22 In the same way, we can recall, the recovery of the problem of time in the “Work­ing notes” in The Visible and the Invisible leads Merleau-Ponty to state that “every analysis of time that views it from above is insufficient. Time must constitute itself—be always seen from the point of view of someone who is of it”23; “it is graspable only for him who is there, at a present.”24

23Surprisingly, this essential “situational” aspect of Merleau-Ponty’s approach to time has been overtly obscured by some interpreters who, having attempted to analyze the French phenomenologist in relation to other philosophical reconstructions of the problem of temporality, tended, how­ever, to mistake him for an idealist, which would simply include time in an account according to which, like any event, time only exists as the correlate of a subject. This is, in fact, the most outstanding problem of the reading of Merleau-Ponty in David Couzens Hoy’s The Time of Our Lives, according to which “Merleau-Ponty maintains that for there to be events, there must be someone to whom the events happen. Similarly, for there to be time, there must be an observer. […] If this were the entire story, he would thus be a temporal idealist.”25 Even though idealism “is not,” on the word of Hoy, “the entire story”26—because the French phenomenologist disputes, following Heidegger, the status of subjectivity in opposition to the world27—, it is clear that, in this interpretation, time is subject-dependent in a way similar to that in which events are, because, in an idealistic fashion, Merleau-Ponty would assert that “there must be someone” to witness them—an assertion that only concerns the existence of such phenomena, not their finite, multiple struc­ture. Actually, the key lines we have quoted prove that the Merleau-Pontyan approach to time is not reducible to a realism-idealism dispute; that the point is not subjectivity but a situated subjectivity with a finite perspective which “fragments” the totality of reality.28

24But, once this mistake has been corrected, we can perceive the clear implications of Merleau-Ponty’s philosophy for the understanding of the sense of the A-determinations of present, past and future. Indeed, even though he does not name them according to McTaggart’s vocabulary, it is clear that the existence of such determinations is only conceivable, according to Merleau-Ponty, from a perspective in time itself. This position appears quite explicitly when the phenomenologist, criticizing the approach of transcendental philosophy (for which the moments in time would all be objects constituted by consciousness, only in virtue of which a synthesis of different moments in a single temporality would become possible), objects:

[The constituting consciousness] travels freely from a past and a future which are not far removed from it, since it constitutes them as past and future, and since they are its immanent objects, to a present which is not near to it, since it is present only in virtue of the relations which consciousness establishes between past, present and future. But then has not a consciousness thus freed lost all notion of what future, past and even present can possibly be? […] Time as the immanent object of a consciousness is time brought down to one uniform level, in other words it is no longer time at all. There can be time only if it is not completely deployed […]. Constituted time [...] is not time itself […]. It is spatial, since its moments co-exist spread out before thought; it is a present, because consciousness is contemporary with all times. It is a setting distinct from me and unchanging, in which nothing either elapses or happens.29

25Merleau-Ponty’s position in the face of a “non-situational” reconstruction (or, in Dummett’s words, a “complete description”) of time is clear: if we approached time, not from a “point of view” which was itself temporal, but as the object of a “freed” consciousness, which “travels freely” between different moments, which “is contemporary” of all of them, which is not “far away” from past and future or, correlatively, “near” the present, which finds time “deployed” before it, then this consciousness can give no sense at all to the alleged distinction between present, past and future, given that the moments that would have to exemplify the differences between these three determinations appear as “brought down to a uniform level” in their simul­taneous appearance to consciousness.

26The thesis of the “cutting off” effected by subjectivity on reality, as the support for the existence of time, also means (as we have begun to see) a certain “negativist” thesis about the non-present. That “not all being is given to me in the flesh” means that future and past are, precisely, defined by their character of excluded from the present perspective; by remaining outside the “cutting off” that each finite perspective exerts on reality. It is in this line of thought that, deepening the analysis, Merleau-Ponty points out:

If we separate the objective world from the finite perspectives which open upon it, and posit it in itself, we find everywhere in it only so many instances of ‘now’. These instances of ‘now’, moreover, not being present to anybody, have no temporal character and could not occur in sequence.30

27Once again: without a finite perspective which allows to establish a difference between the “fragment” of being that is given “in the flesh” and what is not, we can no longer distinguish present from past and future; we have only but instances of “now.” But, naturally, without a subjectivity be­fore which they can “succeed” each other, the different instances of “now” “would have no temporal character,” would not be able (as the dif­ferent moments before a transcendental constituting consciousness) to be dis­tinguished following the characters of present, past and future.

28In this way, we find, schematically, that Merleau-Ponty brings forth two “situational” theses with respect to time: in the first place, we find a strong thesis, according to which the very presentation of the real as a multiplicity of events—and, therefore, the circumstance that even facts with only B-determinations are possible—is unconceivable without a “fragmenta­tion” of the real produced by a situated standpoint.31 But, in the second place, we find a weak situational thesis (which, even though it is a part of the same kind of argument, becomes central in passages such as Merleau-Ponty’s reference to transcendental philosophy): the tenet that, specifically, the notions of present, past and future can only be attributed to moments in time in virtue of the relations that these moments hold with a situated subjectivity.

4. Merleau-Ponty’s more explicit espousal of pluralism. A comparison between space and time

29We have been led, as we have anticipated, to the problem of whether it is acceptable to pluralize our descriptions of reality, to abandon the idea of a single “complete description” and move, on the contrary, to a multiplicity of “maximal descriptions,” all of which is required if we are going to remain describing time. But we see in this case that, unlike Dummett, Merleau-Ponty explicitly accepts these requisites. According to the phenomenologist, who affirms in this respect a parallelism between time and space, an object “is seen from all times [est vu de tous temps] as it is seen from all directions and by the same means, namely the structure imposed by a horizon [la structure d’horizon].”32From all of them, not from none: “the house itself is not the house seen from nowhere, but the house seen from everywhere.”33

30Let us appeal, at this point, to an analogy. As it is evident, my own body, center of my perspectives, can be seen as instituting a qualitative difference, constitutive of our lived space, between “here,” from which I currently perceive, and the places which are not “here” (this is, the different “there”).34 But these other places do not acquire in absolute terms the role of non-here; on the contrary, those are places from which I could perceive; places with which the “here” can be, as stated by Étienne Bimbenet in an analysis of Merleau-Ponty’s approach to space, “laterally exchanged.”35 As a consequence, the asymmetrical relationship in virtue of which, within each perspective on reality, there is unavoidably some spatial point differentiated from the others (as a “here” is differentiated from a series of “there”) does not eliminate the symmetrical relation between spatial points, in virtue of which we can change our “here,” change the particular spatial point that has that role (let us say, the point a) and from which another one (let us say, the point b) is a “there.” But, having conceded this, it is not the case either, on the contrary, that the symmetrical relation ends up absorbing, suppressing, the asymmetrical one. Because, at most, we can vary which particular point (if a or b) is “marked off” with the role of “here,” but we cannot, given the nature of space, suppress all “here.” Keeping the first of these two relations at expenses of the second one would mean assuming an experience of the world in which the different places in space were leveled, not potentially but actually. The possibility of adopting different points of view is not one that we can find already actualized; we can explore the spatial reality, but not all sights are already displayed before us.

31Now, it is precisely this point which is analyzed by Merleau-Pontyan “pluralism” with respect to space, by the thesis that “the world” does not emerge from suppressing every particular perspective and observing reality with no point of view, but from multiplying the points of view, the situated descriptions. Bimbenet introduces the point in a synthetic manner by pointing out: “We can escape from our situs without, however, escaping every situs; this is what has to be understood.”36

32Now, a similar (not exactly the same) sort of pluralism can be found in Merleau-Ponty’s treatment of time, and this will help us bring his position closer to Dummett’s appeal to a plurality of “maximal descriptions” instead of a single “complete description” of reality. Nevertheless, we must not make our parallelism between spatial and temporal “pluralisms” simpler than it should be, by means of misreading the passage concerning the house seen “from all places” and “from all times.” An exact parallelism between time and space would imply that if, at a certain time (say, 2014), we were asked to provide a description of the totality of temporal reality, we would have to include, on an equal footing, the world as it is seen from that moment, and also the world as it is seen from a previous time (say, 2013). Only in this manner could our temporal pluralism claim a strict sameness with the spatial pluralism according to which the house seen from the front is on an equal footing with the house seen from behind. And, on the other hand, this reading of Merleau-Ponty would seem to find a textual basis when, going on with the example of the house, the phenomenologist insists that “if it should collapse tomorrow, it will remain for ever true that it existed today.”37 That is, the present is not more real than the past, it “replaces” it but does not refute it as a true tenet refutes a false one, and the past remains a part of the totum of spatiotemporal reality. However, it is not clear that the French philosopher is thinking, in this beginning of Part I of PhP, on a pluralistic “equating” of the different temporal perspectives, the present and the past ones. Rather than an irreducible equivalence of these perspectives, analogous to the spatial equi­valence we have considered, Merleau-Ponty seems to be thinking, instead, on how every present “collects up” all the past without a loss, integrates the past inside itself, and, in this way, achieves a primacy over the past that has no parallel in the case of different spatial points of view. The correct description of the totality of time could be, then, that which we give from the present point of view (2014, continuing with our example), and a previous per­spective (the one from 2013) would simply be subsumed under it. Let us reread the key passage from Part I:

It is true that I see [the house] from a certain point in my ‘duration’, but it is the same house that I saw yesterday when it was a day younger […]. It is true, moreover, that age and change affect it, but even if it should collapse to­morrow, it will remain for ever true that it existed today: each moment of time calls all the others to witness […]; each present permanently underpins a point of time which calls for recognition from all the others […]. The present still holds on to the immediate past without positing it as an object, and since the immediate past similarly holds its immediate predecessor, past time is wholly collected up and grasped [repris et saisi] in the present.38

33Therefore, acknowledging the past existence of the house that collapsed, not reducing reality to the present perspective, does not imply considering past and present on an equal footing, as we do have to consider two different places, because there is in the case of time a phenomenon of “collecting up,” of “recovery” of the past (by means of the horizon of retentions) with no spatial parallel.

34In spite of these lines, there remains a clear sense in which we can speak of Merleau-Pontyan situational conception of time as a pluralistic approach to temporal reality, just like in the case of Dummett. The reason is actually rather obvious: even if at each time (in our example, 2014) there is, by virtue of temporal synthesis, one privileged description of reality, the present one, which outbids the pretentions of descriptions that could be made from the point of view of past moments, it is a part of the same dynamics, nevertheless, the fact that the present description is not final, definitive, the synthesis of temporal reality. It is bound, instead, to be displaced by a future description (the one made from the point of view of 2015, in which 2014 will appear as past), and in this manner pluralism will remain being unavoidable.39

35Summing up, we have reached the following partial result: Merleau-Ponty’s approach to time, same as Dummett’s, involves a treatment of A-determinations that requires a plurality of descriptions of temporal reality, and in this way achieves to preserve such determinations without falling into some sort of inconsistency as that denounced by McTaggart. There can exist different descriptions, each one of them with their corresponding “zero point”—this is, their corresponding “now,” as in the case of space is “here.” Therefore, McTaggart’s claim that there does not exist, in absolute terms, such a thing as the present, given that each moment M can be attributed each of the three A-determinations according to the perspective from which it is considered, can be simply replied by stating that M will be a present moment in a description, and past or future in others, and no particular description of reality will need, as a consequence, to commit itself to a contradiction. But, in this way, the burden of the problem moves from that of situated descriptions to that of plural descriptions: is it that simple to renounce the idea of the unification of the different descriptions of reality?

5. Once again beyond Dummett: the problem of pluralism as temporal non-combinability

36However, concerning the problem of combinability, our incidental reference to the problem of temporal synthesis in Merleau-Ponty has also begun to take us beyond Dummett’s version of temporal pluralism. The point here is that phenomenological temporal synthesis—unlike the Kantian transcendental counterpart criticized by Merleau-Ponty—is a “synthesis of transition,” and as a consequence a unity through multiple perspectives.

37Indeed, although we have identified in Merleau-Ponty’s approach to time a pluralistic “phase” (which allowed us to relate him with Dummett), the phenomenologist’s analyses do not lack, however, a worry for the re­covery of a unity of time. The problem, then, turns to that of the conditions of possibility for the existence (to put it in Merleau-Ponty’s terms) of “time,” in singular, not only “times.” “Time must be understood as a system that embraces everything–Although it is graspable only for him who is there, at a present.”40 In other words, the rejection of a “view from nowhen” is not, for Merleau-Ponty, incompatible with the requisite of understanding time as a “system.”

38By equally contrasting his position with the transcendental solution and the “fragmented” temporality of empiricism, Merleau-Ponty slips from speaking about a subject which is in time to a subject that is time, this is, to point out that, whereas the subject cannot perform the unifying synthesis of moments from “up” or “outside” time, it cannot, either, be “inside” time in the sense of an intra-temporality. The succession of different presents is, for Merleau-Ponty, a chaining which is identified with the display of subjectivity itself: we cannot unify a temporal flux but by, precisely, living it, by gradually “effecting” the transition from a present to another (the presents through which we pass), rather than “contemplating” it as a finished process. Unifying the different presents cannot mean overlapping them one another into simultaneity, but precisely experiencing them as parts of the same flux, a flux within which each perspective leads to the following one:

the unbroken chain of the fields of presence […] has the essential charac­teristic of being formed only gradually and one step at a time; each present, in virtue of its very essence as a present, rules out the juxtaposition of other pre­sents and, even in the context of a time long past, I can take in a certain period of my past life only by unfolding it anew according to its own tempo. […] Once again, time’s ‘synthesis’ is a transition-synthesis, the action of a life which unfolds, and there is no way of bringing it about other than by living that life […].41

39It is in virtue of this that (from another point of view) it could not be objected that the transition-synthesis is, in fact, a synthesis. The unity of time is not jeopardized by the circumstance that it can only be experienced “from within,” from a present, and this is precisely because of a point that the reconstruction of temporal “pluralism” by Dummett completely overlooks. Namely, the multiple temporal perspectives need not be, ipso facto, discrete perspectives; each present is surrounded by its horizons of past and future which allow the “unbroken chain” of moments, and which imply, on the other hand, that separating a perspective from other ones requires an act of intellectual abstraction. Returning to the problem of A-determinations: the perspective in which a moment C is present is, certainly, a different one from that in which it has acquired the character of past, but this form of plurality does not break the unity of time. Merleau-Ponty illustrates this point in the context of a reference to Bergson, who, he says,

was right to stick to the continuity of time as an essential phenomenon. […] Instant C and instant D, however near they are together, are not indistinguishable, for if they were there would be no time; what happens is that they run into each other and C becomes D because C has never been anything but the anticipation of D as present, and of its own lapse into the past. This amounts to saying that each present reasserts the presence of the whole past which it supplants, and anticipates that of all that is to come, and that by definition the present is not shut up within itself, but transcends itself towards a future and a past. What there is, is not a present, then another present which takes its place in being, and not even a present with its vistas of past and future followed by another present in which those vistas are disrupted, so that one and the same spectator is needed to effect the synthesis of successive perspectives: there is one single time which is self-confirmatory, which can bring nothing into existence unless it has already laid that thing’s foundations as present and eventual past, and which establishes itself at a stroke.42

40Given the very nature of time, the diagram of time from the point of view of C-as-present (accompanied by D-as-future) is not contradicted, but confirm­ed, by another diagram of D-as-present (accompanied by C-as-past); C and D, as temporal moments, never ceased to be bound to this dynamics. To be a present, as Merleau-Ponty remarks, means to be an imminent past. As a consequence, one point should remain clear: whereas Dummett’s argument simply introduces the notion of a variety of “maximal descriptions” without reaching some sort of unification of them (which would be needed to explain, among other things, how it is that we have a single temporal experience), Merleau-Pontyan phenomenology provides us with a complementary ap­proach by means of the kind of unity through the multiplicity which is involved in the notion of synthesis of transition (in which perspectives are not suppressed, but run into one another).

6. Conclusions

41Let us recapitulate the results we have obtained.

42i. The classical article by Dummett which identifies the pretention of a “complete description” of reality as a key assumption provides us with a valuable approach to McTaggart’s paradox, but does not get to present a solution for the paradox. Dummett poses a dilemma between either affirming the reality of time or committing oneself to the assumption that reality must be the object of a complete description, but he does not clearly adjudicate the dilemma in favour of abandoning this assumption.

43ii. On the contrary, an approach that counters the weaknesses we have found in Dummett’s reading can be found in Merleau-Ponty’s phenomeno­logy of time. The key point of our parallelism is that the alleged paradox dissolves when we understand reality as the result of a sum of perspectives of reality, all of them from a present. The inconsistency in which every moment would allegedly be past, present and future cannot be found in any descrip­tion of temporal reality, but can only exist betweendifferent descriptions.

44iii. If a “perspectival” approach to temporality rejects the idea of a “complete description” of reality, it does not require, in the other extreme, a commitment with non-combinable temporal perspectives. The “middle way” that is associated to the concept of synthesis is, precisely, that of a unity through the multiple, a unity in which the diverse perspectival presentations fall into each other following their own horizons. In other words, the fact that temporal reality cannot be known without a point of view does not imply that these are discrete points.


1 Cf.McTaggart, J.M.E., The Nature of Existence, vol. II, Cambridge, Cambridge University Press, 1927, p. 10; cf. also the original formulation in McTaggart, “The Unreality of Time,” Mind, Vol. 17, No. 68, Oct., 1908, p. 458.

2  Cf. “The Unreality…,” pp. 459-461; The Nature…, pp. 11-13. The second part of the argument, as we shall see, attempts, in turn, to demonstrate why the A series is intrinsically inconsistent, and, therefore, so is time itself (Cf.“The Unreality…,” pp. 467-470; The Nature…, pp. 20-22).

3  “Past, present and future are incompatible determinations. Every event must be one or the other, but no event can be more than one” (“The Unreality…,” p. 468; The Nature…, p. 20).

4  Cf. The Nature…, p. 21; “The Unreality…,” p. 468.

5 Dummett, Michael, Truth and other enigmas, Harvard University Press, Cambridge, Massachussets, 1978.

6  And as a consequence of which Dummett received replies such as those by Lowe (Lowe, E. J., “The Indexical Fallacy in McTaggart’s Proof of the Unreality of Time,” Mind, No. 381 (1987)) or Macbeath (Macbeath, M., “Dummett’s Second-Order Indexicals,” Mind, No. 385 (Jan., 1988)), who consider that the paradox does lie on a mere linguistic misunderstanding.

7  In fact, Dummett’s point about what McTaggart would be demonstrating does not need to refer to time in contrast with the phenomenon of space; it could also be thought that not being describable without a situated standpoint is an interesting result even if it is referred to time as well as space (which seems to be the case, and in favour of which Thomson (Thomson, J. J., “McTaggart on Time,”Noûs 12 (2002), pp. 229-252) has argued cogently). As a consequence, we can avail of these tenets by Dummett even if, below, we will analyse this character of time in analogy with its spatial equivalent.

8 Dummett, Truth…, p. 354.

9  Cf. ibid. Of course, this also applies to the corresponding expressions for future and past, which are also bound to the “perspective” of the observer, such as “just,” or “soon.”

10 Truth…, p. 355. Emphasis mine.

11  Cf. Dummett, Truth…, p. 354.

12 Dummett, Truth…, p. 356.

13 Dummett, Truth…, p. 356. Emphasis mine.

14 Ibid.

15  Cf. ibid.

16  Dummett, Truth…, p. 357.

17  Dummett’s attribution of these tenets to McTaggart is quite explicit, though it admits of degrees: about (1) it is said that the author of the paradox “is saying” it; about (2), that he “is taking [it] for granted.” Dummett does not get to attribute to McTaggart, with the same explicitness, the idea that it is this tension between (1) and (2) that leads him to state the unreality of time (Dummett might simply be analyz­ing what the argument in fact reveals, no matter if McTaggart was aware of it or not, and this is why he considers the possibility that the paradox proves the falsity of the general ontological assumption stated in (2), instead of the unreality of time), but the fact that he refers to what McTaggart “takes for granted” emphatically sug­gests that he is actually attempting to reconstruct not only the philosophical problem itself but also the position of the author of the paradox—a paradox which, by the way, Dummett does not give us any elements to understand in any other way. Con­sequent­ly, it seems legitimate to consider that Dummett’s reading on McTaggart’s position attributes the tenet (3) to the latter.

18 Craig, W. L., The Tensed Theory of Time: A Critical Examination, Dordrecht, Kluwer Academic Publishers, 2000, p. 206.

19  For two recent examples, consider, on the one hand, Falvey, K., “The view from nowhen: The McTaggart-Dummett argument for the unreality of time,” Philosophia 38, 2010, and, on the other hand, Le Poidevin, R., “Time, Tense, and Topology,” in Meister, J. C. and Schernus, W., Time. From Concept to Narrative Construct, De Gruyter, 2011.

20 Horwich, P., Asymmetries in Time: Problems in the Philosophy of Science, MIT Press, 1987, p. 27.

21  Merleau-Ponty, M., Phénoménologie de la perception, Paris, Gallimard, 1945, p. 470; Phenomenology of perception, tr. Colin Smith, London-New York, Routledge, 2002, p. 477.

22  Merleau-Ponty, M., Phénoménologie..., p. 484; Phenomenology of Perception, p. 492.

23 Merleau-Ponty, M., Le Visible et l’invisible. Suivi de Notes de travail, Paris, Gallimard, 1964, p. 235; The Visible and the Invisible. Followed by Working Notes, tr. Alphonso Lingis, Evanston, Northwestern University Press, 1968, p. 184.

24  Merleau-Ponty, M., Le Visible..., p. 245; The Visible..., pp. 190-191.

25  Hoy, David Couzens, The Time of Our Lives. A Critical History of Temporality, Cambridge-London, MIT Press, 2009, p. 67; emphasis mine.

26  Ibid., p. 68.

27  Cf. ibid.

28 Actually, in the contemporary Anglo-Saxon philosophical production, not only Hoy has been guilty of turning Merleau-Ponty into an idealist. Exactly the same mistake had been made by Stephen Priest, according to whom the idea of events being “découpés” by an observer simply refers to a “selection” of events, “governed by our pragmatic interests” (Priest, S., Merleau-Ponty, London, Routledge, 1998, p. 129), which does not exclude the possibility of events “being selected from a non-ideal totality of events with which the perceiver […] could […] be realistically acquaint­ed” (ibid., p. 127). Once again, we must insist: according to Merleau-Ponty, events do not appear as such in absence of a situated, finite perspective, and therefore can not be a multiplicity out of which to “select.” (For a further discussion of Priest’s reading, cf. our article “Cómo recibir a Merleau-Ponty en la ‘theory of time’ Una polémica,” Tábano, no. 9 (2013), 69-84).

29 Merleau-Ponty, M., Phénoménologie..., pp. 474-475; Phenomenology..., pp. 481-482.

30 Merleau-Ponty, M., Phénoménologie..., p. 471; Phenomenology..., p. 479. Emphasis mine.

31 In any case, this strong tenet is probably false: even though the image of “one indivisible and changeless being” is presented by the phenomenologist as a counter­factual, it is not clear in which way this “indivisible being” could condense in a simultaneity mutually incompatible states of affairs, which then a situated gaze would simply put into succession. The idea of such a “being” would have to be, literally, incoherent.

32 Phénoménologie..., p. 83; Phenomenology..., p. 80. I have slightly modified the English translation (“seen from all times” instead of “seen at all times”) to keep the analogy between spatial and temporal “points of view” from the original French text.

33 Phénoménologie..., p. 83; Phenomenology..., p. 79.

34  Concerning how the “corps propre” imposes us a perspective on the world, cf., in particular, Phénoménologie…, p. 107; Phenomenology…, p. 104. As to how space would cease to exist if our spatial situatedness could be suppressed, cf. Phénoméno­logie, 382-383; Phenomenology…, pp. 386-387.

35  Bimbenet, Étienne, “Un motif d’étonnement majeur: le perspectivisme,” Alter. Revue de phénoménologie, No. 16, 2008.

36 Ibid., p. 103.

37  Phénoménologie…, p. 83; Phenomenology…, p. 79.

38 Phénoménologie…, p. 83; Phenomenology…, pp. 79-80.

39  This necessary incompleteness of our synthesis of the world is made explicit in the third chapter of Part II of PhP: “If the synthesis could be genuine and my experience formed a closed system, if the thing and the world could be defined once and for all, if the spatio-temporal horizons could, even theoretically, be made explicit and the world conceived from no point of view, then nothing would exist; I should hover above the world, so that all times and places, far from becoming simultaneously real, would become unreal, because I should live in none of them and would be involved nowhere. If I am at all times and everywhere, then I am at no time and nowhere.” Phénoménologie…, pp. 382-383; Phenomenology…, p. 387.

40  Merleau-Ponty, M., Le visible et l’invisible. Suivi de Notes de travail, Paris, Gallimard, 1964; The visible and the invisible. Followed by working notes, tr. Alphonso Lingis, Evanston, Northwestern University Press, 1968, pp. 190-191.

41  Merleau-Ponty, M., Phénoménologie de la perception, Paris, Gallimard, 1945, pp. 483-484; Phenomenology of perception, tr. Colin Smith, London-New York, Routledge, 2002, p. 491.

42  Merleau-Ponty, M., Phénoménologie..., p. 481; Phenomenology..., pp. 488-489.

To cite this article

Claudio Cormick, «Against the “view from nowhen”: A Merleau-Pontyan contribution to Dummett’s approach to McTaggart’s paradox», Bulletin d'Analyse Phénoménologique [En ligne], Volume 10 (2014), Numéro 2, URL :

About: Claudio Cormick

Universidad de Buenos Aires, CONICET

Call for papers

Permanent call-for-papers

More info