B-SERIES TEMPORAL ORDER IN DOGEN'S THEORY OF TIME

By Dirck Vorenkamp
Philosophy East and West
Volume 45, Number 3
1995 July
P.387-408
(C) by University of Hawaii Press


P.387 J.M.E. McTaggart's controversial argument against the reality of time has prompted many counterarguments since it was first published in 1908. Richard Gale says that the responses to McTaggart are in agreement with either the so-called "A," "B," or "Either/or" theories of time.(1) Each of these theories is based on the idea that time is either "dynamic" or "static" (and only dynamic or static). If this is true, then as a group they exhaust all possible ways to conceive of time, and any theory of time must include elements of one or more of the three responses to McTaggart. Given the arguments above, and the fact that Dogen also wrote about the nature of time, one is led to the question, "Do Dogen's views of time fall anywhere within the "A-theory, " "B-theory," or "Either/or-theory" responses to McTaggart"? Because the question is too broad, as stated, for a work of this length, I propose to examine Dogen's teachings in light of the "B-theory" response to McTaggart. This essay will show that Dogen's views on time do contain elements of each of the four main tenets of the B-theory. Furthermore, in pointing out these aspects of his thought, we will find that the assumption of consistency that has characterized modern interpretations of Dogen's view of time requires reevaluation. In order to show why this is true, we will briefly examine McTaggart's original argument, and then summarize each of the four main tenets of the B-theory while asking whether or not Dogen's theory of time reflects similar ideas. The term "B-theory" is defined according to the synopsis of the B-theorist's positions provided by Richard Gale.(2) The details of that definition will follow shortly. As for the phrase "Dogen's theory of time," it is defined as those teachings found in the Shobogenzo that deal with the concepts of uji, hoi, and/or nikon. Each of these three terms will in turn be defined before they are used to support a conclusion.(3) McTaggart's Argument In order to understand the B-theory, it is necessary to summarize McTaggart's original argument. First published in 1908, McTaggart's work argues that time is unreal.(4) The argument starts by establishing two basic aspects of time. The first aspect is associated with the dynamic flux called time's passage, and with our tensed ways of both conceiving and speaking of time. According to this way of viewing time, things are said to exist in the future, then become present, and finally enter the past. It is this mode of time wherein the future and past are said to differ ontologically. McTaggart calls this aspect of time the A-series.(5) P.388 Yet, time also seems to have a static order or structure. Even if events do pass from future to present to past, they do so with an order that never seems to change. That order is the relationship between "earlier than" and "later than."(6) The "earlier than/later than" relationship is tenseless, for the statement that an event is earlier than another is not dependent upon the particular temporal frame of reference from which it issues. Here, the emphasis is on time as a set of relative relationships. Such relations, if they could be viewed from a position outside time, would present a pattern of things and times laid out like a mosaic. Even though Crunbaum holds that such a pattern does not necessarily constitute the hypostatization of time, it is nevertheless fixed as regards the temporal relationships between the individual components.(7) McTaggart called this the B-series.(8) The issue at the heart of McTaggart's work is how these two ways of conceiving of time can be related to one another. McTaggart first states that events cannot undergo changes in their B-relations. Dogen's birth is earlier than mine, and that is a tenseless fact that will not change. Accordingly, the only changes that an event can undergo are A-series changes in which the event changes from future to present to past.(9) McTaggart then argues for the unreality of the A-series based upon the notion that, assuming there are no first or last events in the series, then every event must simultaneously share the mutually incompatible attributes of future, present, and past.(10) For example, in the year 1980, the year 1981 was the present-future. Such a state is contradictory and, therefore, according to McTaggart, unreal. To claim, however, that any event has only one of these attributes at a given time and successively, that is, that an event X was future, is present, and will be past, invokes an infinite regress. According to McTaggart, such a statement means that at a past time, event X was future, and at a future time it will be past.(11) It is true that the first contradiction has been explained away, but only by creating a second-level temporal order within which the first order must exist. The statements that event X is, was, or will be some particular tense itself relies upon a tense for comprehension. Assuming that the goal is to prevent incompatible tense overlaps, then the second order itself must be related to a "higher"-level order just as the first is related to the second. At no point are we freed from this necessity, and so the regress is infinite.(12) McTaggart therefore concludes that the tenses are logically incompatible and that the A-series is unreal. Since McTaggart holds that "real time" needs both the A-series and B-series to exist, he concludes that "real time" must be nonexistent. This conclusion rests upon the notion that A-relations are more basic than B-relations because (1) all B-relations can be reduced to A-series relations with no loss of meaning, and (2) the same is not true in reverse. Once P.389 McTaggart has demonstrated that the A-series is unreal, then it follows that the B-series is unreal, and so, too, "real time" itself.(13) As previously stated, philosophers have taken three approaches to answer McTaggart. The so-called B-theory is one such answer. Richard Gale states that there are four basic tenets that constitute the B-theory.(14) They are: (1) The A-series is reducible to the B-series with no loss of meaning. (2) The passage of time, or "temporal becoming" is psychological since it necessarily involves a B-relation to a perceiving subject. (3) All events are equally real; hence the B-series is objective. (4) Change is understandable in terms of B-relations alone. While analyzing these statements according to the summary that Gale has provided, we can ask whether these tenets, or parts of of them, are found anywhere in Dogen's view of time. The First Tenet of B-theory and Dogen's View of Language Concerning the first tenet, B-theorists claim that the tenses past, present, and future do not refer to ontological times, but are actually relative to events and/or event expressions for their meaning.(15) The terms "is past, " "is present," and "is future" are actually "tenseless two-place predicates" that take as their ground some event or expression.(16) For example, the phrase "is past"--as in "1980 is past"--actually means "1980 is past at some other time." It is correct, then, for me to say in 1988 that "1980 is past," but the same sentence is false in 1978. Hence, argue B-theorists, the futurity, presentness, or pastness of an event is not an ontological fact, but only an expression of that event's relationship to another event or expression. That two-place relationship is a tenseless truth that holds for all times and frames of reference from which it might be uttered. Regardless, for example, of the year in which I make the statement, it is, and always will be true that 1980 is past at 1988. In light of this, B-theorists say that B-statements are more fundamental than A-statements and that all A-statements can be "reduced" without loss of meaning to B-statements. According to Gale, B-theorists use two primary methods of reducing A-statements to B-statements. The first is the so-called psychological reduction, and this will be considered shortly. The second is the linguistic reduction.(17) The linguistic reduction claims that statements asserting an event's tense are relative to the speaker since they necessarily indicate and rely upon the speaker's temporal relationship to the reported event.(18) Such statements are said to be "token-reflexive" because the statement "1980 is past" actually means "1980 is past relative to this utterance (token)." An important point to notice in the modified (i.e., "reduced") statement is that it makes no mention of the current temporal status of the token- P.390 reflexive. That is why B-theorists claim that the reduced statement is timelessly, or tenselessly, true and freely repeatable.(19) B-theorists who use this method claim that the sole purpose of A-series statements is to express a B-series relation between the expressed event and the utterance (token).(20) Does Dogen ever utilize a similar reduction? Certainly Ddgen never spoke of "reduced A-series relations." But, did he "reduce" events to linguistic expressions in order to recognize language itself as a "reflexive" vehicle referring to uji?(21) When we examine Dogen's statements about the nature of language and words, a type of implicit token-reflexive argument is apparent. First, on the nature of words, Dogen had this to say: If the Buddha's speech is shallow, turning the flower must also be shallow. If the Buddha's speech is only letters and sounds, that is not the words of learning the Buddha Dharma. Although it is known that speech is letters and sounds, it is not known that to the Buddha it is not just letters and sounds.(22) ... if it [enlightenment] is not speech, you cannot realize the Buddha's progress.... Therefore, when speech is manifested, that itself is the Buddha's progress.(23) The Dharma-nature spoken of by Baso is the Dharma-nature speaking the Dharma-nature.(24) and also: The capability to suggest that beings who do not leave the Dharma-nature are not the Dharma-nature may even accomplish something. It is three or four new levels of Dharma-nature. Speaking, answering, using, and acting as if it is not the Dharma-nature must be the Dharma-nature.(25) What we see in all these quotations is that Dogen does not just conceive of language as merely a verbal representation of some other feature of reality. Language is not purely symbolic of what we think and perceive and hence has some separate existence from these thoughts and perceptions. For Dogen, language has another level of meaning. At the secondary level, words are nothing except what they express. As Dogen said, even "the nature of things spoken of by Baso is the nature of things spoken of by the nature of things." Dogen emphasizes the importance of speech and words because he understands them as the Buddha-nature itself.(26) Accordingly, the dualism that characterizes our general notions of the relationship between expression events and the expressed event is absent in his view--as he states above, "there are no mere words with the Buddha." If it is appropriate to apply this expanded definition to the term "reduction," then Dogen can be said to claim that all reality is "token-reflexive." Note, though, that this claim does not entail the assertion that Dogen is giving special status to linguistic expressions. In fact, quite the contrary P.391 is true. The point is that linguistic expressions are no more nor less privileged than any other aspect of being-time.(27) Given the interrelationships between dharma-positions as elements of being-time (which will be discussed shortly), Dogen's insistence on including language as a form of expression of the Buddha-nature is in fact an affirmation of the token-reflexive nature of that Buddha-nature. Furthermore, since Dogen's concept of uji clearly intertwines being and time, we can conclude that this type of "token-reflexivity" means not just that the past and future as concepts are relative to a particular statement, but that tensed and nontensed locutions are token-reflexive, because the phrase itself must, in a sense, be the times to which it refers. In other words, the expression event's expression of past or future is exactly where these tenses (i.e., past and future) are found. The statement "1980 is past" exists at the moment it is uttered, and in that statement must lie the referenced event (i.e., "1980"), not merely as a referent, but as a factual (i.e., actually existing now) component of the event. Of course this sounds fantastical to anyone acquainted with modern Western language theory. But Dogen's thought is grounded in the concept of `suunyataa and specifically in the Hua-yen school's elaboration of it.(28) Since linguistic expressions are neither more nor less important than other phenomena, there is, then, an interdependent relationship between words and what they signify precisely because both are expressions of the Buddha-nature. If, in Dogen's view, words are in fact what they signify, then we can certainly say that not only are A-series statements reducible to B-series statements with no loss of meaning, but the act of the reduction itself adds meaning to the "token" because the token itself is then also seen as the symbolized. The Second Tenet of B-theory and Dogen's View of Temporal Becoming The second element of B-theory is that temporal becoming is subjective. According to Gale, B-theorists use one of two different methods to argue for an a priori subjective present in the A-series. The first is the token-reflexive argument, which we have just covered. The second method is the egocentric-particular analysis, which attempts to show the mind-dependence of A-determinations.(29) According to B-theorists, the tenses are relative to the perception of some subject in much the same way that they are token-reflexive. In defining or understanding what is past, present, or future, we must draw a reference to a mental event. For example, in explaining what is past, we will typically refer to a memory; for the present, a sense datum; for the future, an expectation or anticipation. In each case the tense, or that object which supposedly is tensed, is relative to a mental event. It follows, say B-theorists, that because the tenses are psychological, then so, P.392 too, is temporal becoming.(30) Is there anything similar to this in Dogen's thought? In speaking of passage, to think the place [of passage] is outside the mind, and that the dharmas that can pass, pass to the east past a hundred thousand worlds over a hundred thousand eons, that is not devoting yourself only to the study of the Buddha Way.(31) The three heads and eight arms [i.e., the state of unenlightenment] pass as my being-time. Although it seems to be "there," it is Now (nikon).(32) ... every being in the whole world, while being lined up, is an individual time. Because they are being-time, they are my being-time.(33) One must study that if there is not now the passage of my utmost exertion, then there is not the manifestation or passage of one dharma or one thing.(34) In each of the quotations above, Dogen is clearly making a statement concerning the relativity of time and passage to the individual subject. In the case of B-theorists, their contention is clear: in stating that temporal becoming is subjective, B-theorists mean that it is relative to, a product of, the individual mind and is not an objective feature of the cosmos. Although we see that similar statements can be found in the Shobogenzo, the matter is not that simple. In many places Dogen speaks of time's passage without making reference to a subject. For example: You should learn that passage occurs without anything external. For example, spring's passage is necessarily that which passes through spring.(35) Uji has the virtue of passage. It passes from today to tomorrow, it passes from today to yesterday, it passes from yesterday to today.(36) On the face of it, these passages indicate that Dogen also sees temporal becoming as an objective feature of reality unrelated to the subject. But is this the whole story? Two interrelated questions need to be answered before we can decide whether or not "subjective" temporal becoming exists in his thought. First, what does "subjective" mean? And secondly, what is the nature of the temporal interaction between subject and object? To answer these questions, it is first necessary to know what Dogen meant by dharma-positions. Dharma-positions designate the spatial and temporal characteristics of a particular, discrete aspect of uji.(37) Uji itself is a four-dimensional matrix of the three spatial dimensions and time. Each particular phenomenon is a four-dimensional point of space-time and constitutes a unique thing-time distinct from all other thing-times (i.e., dharma-positions). For example, a partial description of the dharma-position that is a rock in the garden might include the fact that it is round, about the size of a baseball, gray, located next to the fence, and that it is all these things at a particular time. The time of the rock is not defined by P.393 the movement of the hands of a clock, but by the spatial characteristics that describe it. Dogen calls these particularized, discrete parts of uji, "dharma-positions." Now, if dharma-positions are four-dimensional, and if, as Dogen indicates, they do move relative to one another, then their movement must take place within a second-level temporal order. As J.J.C. Smart has indicated, by definition, space-time entities (i.e., four-dimensional phenomena) must merely exist "en bloc" as a thing defined as having such and such characteristics at such and such a time."(38) These are, by definition, tenseless statements, and the concept of change is not a factor. But the definition of movement involves change (or at least the perception of change) in both position and time. In the case of four-dimensional phenomena such as Dogen's dharma-positions, since the definition of any particular thing's position already includes time, then movement must be defined as "something occupies spacetime-1 at ? -1, and spacetime-2 at ?-2." The question mark indicates that some other temporal order is required to account for change. Now we are in a position to answer our two earlier questions by simply asking," Is there such a second level temporal order in Dogen's views, and if so, how does that affect the interaction between dharma-positions?" You should not understand that time only flies past. You should not learn that flying past is time's only ability. If time were [only] given to flying past, it would have to have gaps.(39) Dogen states that to conceive of time as only a progression of discrete instants is erroneous. He demonstrates this by pointing out that a linear progression of radically discrete temporal instants is experientially inconsistent.(40) The problem is that if time passes us by, then when does the future become the present and then the past? In other words, when does the present moment suddenly cease to be that moment and become a past moment? Central to this issue is a "present now." Everyone intuitively knows what time is, but according to Dogen no one can adequately explain it according to a progression of radically discrete temporal instants. Attempting to do so leaves us with a time that "flies by," and which has "gaps" that cannot be experientially verified. For Dogen, the view that time only "flies by" is deficient because it fails to recognize the relationship of the past and present to the nikon. The now (konji) under consideration is everyone's [each person's] Now (nikon). Even if I make myself think of tens of thousands of pasts, futures, and presents, they are now (konji), they are the Now (nikon). Everyone's destiny necessarily exists now (konji).(41) For Dogen, the nikon is that aspect of uji that is ever present and P.394 mediating between the other tenses. In fact, as he has stated, the future and past exist within the matrix of the nikon. For example, the notion of time as a linear ontological continuum would hold that Dogen himself is in the past, while the year 2000 is entirely unconnected and in the future. But as we have seen above, the nikon is used to create a two-place relationship that makes it appropriate to say that Dogen's existence is what-Now-is-no-longer, while the year 2000 is what-Now-is-not-yet. The concepts of past and future take their meaning from the relative relationship to that Now (nikon). Consequently, the future actually exists, mediated by the nikon, as the Now-not-yet; the present exists as the Now-present; and the past exists as the Now-no-longer.(42) This can be clearly seen in the passage below, where Dogen discusses the interrelationship of the tenses. It is said, the past life has already perished, the future has not yet come, and the present does not stay. The past is not necessarily already perished, the future is not necessarily not-yet-come, the present is not necessarily not-staying. If you learn the qualities not-staying, not-yet-come, and already perished as past, future, and present, you must certainly grasp the reason the not-yet-come is past, present, and future.(43) First, with regard to the present, Dogen says that it is not necessarily "not-staying." The present is not just a fleeting instant forever passing us by and always gone as we try to "catch it." Similarly, the future and past are not just the "not-yet-come" and "already perished." Instead, all these tenses have further dimensions of meaning derived from the function of the nikon as the Now common to each. Accordingly, tensed events (i.e., dharma-positions) can be viewed as always existing in two-place relationships that take the "Now" (nikon) as the relative temporal frame of reference from which a B-series, "earlier than/later than" structure is generated. The only difference with the token-reflexive argument discussed above is that these statements each take the nikon as one of their two parts. But isn't this just another way of expressing a contradiction in terms? Doesn't this in essence only say that the past, for example, is the "present past," and, as McTaggart stated, isn't the "present past" contradictory? In Dogen's case, I think the only correct answer is yes and no. To understand why, and determine whether temporal passage is "subjective," we have to take a look at where Dogen locates the nikon. There is a clue in the quotation above: "The now under consideration is each person's nikon." In another section of "Uji," Dogen reinforces this notion: If time is not the aspect of coming and going, [then] the time of the mountain top is being-time's nikon. If time maintains the aspect of coming and going, [then] being-time's nikon is in the self.(44) P.395 Two points in this passage are particularly important here. Dogen says: (1) if time is not dynamic, then any given time is the nikon of uji, and (2) if time is dynamic, then nikon is in the self. Here the nikon is performing a twofold function that allows Dogen to speak of both two-place temporal relations and time's passage. In the first case, Dogen states that where passage is not present, then time is all nikon (i.e., uji's nikon) . Uji can then be thought of as a four-dimensional mosaic in which the temporal relationships between the individual dharma-positions are "earlier than" and "later than" (i.e., B-series relations). In this sense, because the cosmos is given en bloc as right-Now, as nikon, change is not a factor, and McTaggart's objection about tense overlaps does not apply. But, as Dogen states, there is also a sense in which time is dynamic, and in that case, the nikon is "in the self." We can recall two other passages from "Uji" that state: In speaking of passage, to think the place [of passage] is outside the mind, and that the dharmas that can pass, pass to the east, past a hundred thousand worlds over a hundred thousand eons, that is not devoting yourself only to the study of the Buddha Way.(45) and also: The three heads and eight arms pass as my being-time. Although it seems to be "there," it is the nikon.(46) When passage is the feature of uji evident to the subject (i.e., any particular dharma-position), then the nikon seats itself in, and is specific to, that particular dharma-position. In performing this role, nikon can be understood as the required second-level temporal order needed for passage, given that dharma-positions are in themselves four- dimensional space-time points. When the nikon is performing this second role, Dogen is presenting a view of temporal becoming that in one sense is similar to the B-theorists. Time's passage is "subjective" precisely because it is relative to the nikon seated in all dharma-positions. Of course, here, "subject" means only an arbitrarily chosen frame of reference, that is, any dharma-position then acting as the "subjective seat" of the nikon. This helps explain the apparent incongruence between those sections that speak of passage as relative to an individual and those, like the one about springtime's passage, where there is no mention of an individual. In both cases, however, we might say that a "subject" is present since any dharma-position can seat the nikon. This second role of the nikon, however, also creates problems in the form of an infinite regress. If the "objective" dharma-positions constitute four-dimensional phenomena passing through the "subjective seat" of the nikon, then that "subjective seat" must be a five-dimensional phe- P.396 nomenon. That five-dimensional dharma-position can then in turn only pass relative to a six-dimensional dharma-position. At no point does the regress end, and McTaggart's objection would seem to apply. In spite of this problem, however, we can conclude that while passage is not purely "psychological" in the sense of the B-theorists, the Shobogenzo does contain evidence to support the idea that it is "subjective" (i.e., relative to any given dharma-position acting as the seat of the nikon). And in the case of either a static or dynamic relationship between events (i.e., dharma-positions), Dogen does speak of the generated temporal relationship as the two-place "earlier than" and "later than." The Third Tenet of B-theory and Dogen's Dynamic Uji The third tenet of B-theory states that "all events are equally real, hence the B-series is objective." Gale states that the objective nature of the B-series is a logical entailment of the subjective nature of temporal becoming (i.e., B-theorist tenet number 2) .(47) Of course, the implication is that if temporal becoming is "only" subjective, then the "real" or objective temporal relationship must be only the B-series. As we have seen, there are places in the Shobogenzo where Dogen speaks of the passage of dharma-positions as occurring in "my" nikon, and in "my" uji. But the notion of passage also occurs in situations notable for their lack of a sentient subject (e.g., "springtime passes through springtime") . Hence, the subjective nature of beomcing for Dogen is not a statement about the "merely" psychological nature of passage, but an affirmation of the dynamic nature of all reality. For Dogen, since the nikon can and does seat itself in any and all dharma-positions, it is in the dynamic interaction of these dharma-positions that subject and object come to define on another. And so, it is precisely the "subjective" nature of becoming as relative to the nikon that constitutes a kind of objective, two-place, temporal relationship. But, the logical entailment of this conclusion is that Dogen has affirmed both an A-series (the change of events in respect to tense, i.e., temporal becoming) and a B-series. As we will see below, it is correct to wonder about the logical compatibility of these two ideas. The Fourth Tenet of B-theory and Dogen's Discrete Dharma-positions The fourth tenet of B-theory states that change is understandable in terms of B-relations alone. According to Gale, this means accounting for two types of change.(48) The first concerns changes "in" time. As Gale states, "What we mean by the change of a thing in time is a sequence of successive events all regarded as states of one thing."(49) When reality is viewed as a four-dimensional mosaic, we note that the same substance has different qualities at different times. Dogen seems to speak of this idea in the following passage. P.397 Firewood becomes ashes, and ashes cannot return to firewood. However, you should not understand that ashes are after and that firewood is before. You should know that firewood dwells in the dharma-position of firewood, and has before and after. Although firewood has before and after, before and after are disconnected. Ash is in the dharma-position of ash, and has before and after.(50) Dogen starts by saying that there is a temporal progression from firewood to ashes. Obviously the concept of dharma-positions includes different qualitative states of things that we believe are one and the same object. Because these causally related dharma-positions present us with a temporally seamless picture of change "in" time, we commonly say firewood "becomes" ashes. Of course, this should not be construed to mean that Dogen is affirming the idea of changes "in" an enduring substance. His point is simply that spatiotemporally proximate, yet discrete dharma-positions provide a continuity to experience. In this limited sense, Dogen does affirm changes "in" time. The second type of change is called changes "of" time. Changes "of" time are changes in an event's status as future, present, and past. This type of change is found in Dogen's writings, where he speaks of time as if it were constituted of a progression of discrete dharma-positions. The firewood-to-ashes section above is a good example--and some others include: Prior thought, succeeding thought, thoughts do not wait for each other. Prior dharma, succeeding dharma, dharmas do not oppose each other.(51) To tell the main point, every being in the whole world, while being lined up, is an individual time.(52) and the section seen earlier, Uji has the virtue of passage. It passes from today to tomorrow, it passes from today to yesterday, it passes from yesterday to today.(53) Sections like these suggest a recurrent temporal passage. There is a sense in which dharma-positions are unrelated; they do not "wait" for each other. When one time is here, the time "before" is no longer here, the time "after" is not yet here, and in combination they can be thought of as being "lined up" like a series. In any case, the idea is of discrete temporal moments succeeding one another in a "linear" fashion as they flow from the future, through the present, and into the past. In this sense, we can say that change is nothing but a change "of" an event's temporal status as it "proceeds" through the three times. A problem that arises here, though, is how one reconciles the idea of changes "of" time with the notion of two-place relations between disconnected dharma-positions. The difference between ash and firewood is obvious. But, just for the sake of argument, what about the difference P.398 between burning firewood at time-1 and burning firewood at time-2? The difference between time-i and time-2 can be as small a temporal span as we care to make it. To be consistent with Dogen, we have to say that each of these two times is also an independent dharma-position with before and after. But these correlated dharma-positions will not in themselves account for the experienced sense of continuity in the passage of time. When and how does time-1 stop being the time present and time-2 "become" the time present? It is true, as Dogen stated, that we do not experience "gaps" in time's passage. Of course, one might reasonably object to this by saying that if there were gaps between dharma-positions, we could never know it. In such a case (assuming an Abhidharmic-type succession of instants whose flashings are all precisely coordinated with simultaneous flashings of instants of consciousness), their existence is not merely unknown, but completely unknowable, and we, as Dogen has pointed out, experience only continuity in time's passage. Yet the original problem still remains because Dogen has, in fact, differentiated between different, discrete dharma-positions. In other words, he has said we do know that the various dharma-positions are different. So, again, we have to ask how the experience of continuity is related to discrete, two-place dharma-positions? Several Dogen interpreters have suggested that the nikon plays an important role in this regard.(54) At first thought, and in light of what we have said above, this seems a plausible answer. When discussing the second tenet of B-theory, we saw how B-theorists attempt to reduce the notion of passage to a bipartite relationship dependent upon a token. We also saw how Dogen's nikon can be interpreted as performing the function of such a "token." Depending upon its placement, the nikon both (1) mediates between the tenses as uji's nikon and (2) allows for the passage of dharma-positions through "my" Now. But, even though the nikon is common to both views, we must not confuse the two. As soon as the notion of nikon as a temporal referent for the past and the future was introduced, we stopped speaking of passage and started speaking only in terms of two-place relations. But, when Dogen spoke of the nikon of the self, that nikon performs an entirely different function. By seating itself in any particular dharma-position (i.e., the "subject"), it allows for a passage through that dharma-position's experiential Now. This is change in respect to tense (i.e., an A-series passage). The fact that these types of A-series notions not only exist in Dogen's thought but are frequently found alongside B-theory views is problematic in light of the fact that Western philosophers consider the two views logically incompatible.(55) Since Dogen clearly considers A-series passage as essential to his view of time, the A-theory phrases (and the use of the randomly seated nikon) would seem to mitigate against the B-theory views we have also found. But is this the case? P.399 First we should note that descriptively pointing out these apparently contradictory positions is not equivalent to assigning priority to either. Nor does the fact of uncovering an apparent contradiction weaken the validity of the descriptive evidence for the existence of the problem. Nevertheless, this obviously does not answer the question, and we are still left wondering about the consistency of Dogen's theory of time. The fact that the two very different types of assertions often occur in close proximity to one another (e.g., the firewood-to-ashes passage) suggests that Dogen had some sort of congruent whole in mind when he uttered them. This is certainly the position of most interpreters of Dogen. If we make the reasonable assumptions that Dogen's statements on time are consistent with one another, and probably reflect the influence of his own intellectual tradition, then several options are available in our attempt to explain the problem. The first option would involve the claim that these apparently contradictory statements simply reflect an application of Naagaarjuna's well-known tetralemma. Advocates of this view could assert that since the contradictions are a skillful application of conventional-level truths, the problem is apparent only. Unfortunately this option suffers from textual and philosophical problems that render it untenable. First, it is not obvious that Dogen had such a hermeneutic (i.e., a fourfold negative dialectic) in mind in the specific case of his discussions on time. For example, while it is certainly true that Dogen utilized such a dialectic in his discussion of koans, reading the "Uji" fascicle as a whole in light of an assumed tetralemma structure does not give ready evidence that it reflects a consistent, widespread use of such a device. In fact, the only clear-cut case of the presence of the tetralemma in the "Uji" fascicle occurs in one koan cited toward the end of the piece. But even there, Dogen's explanation of the koan does not reflect an application of the tetralemma.(56) Secondly, even if we still insist on hermeneutically applying the tetralemma, the conclusion of inconsistency still follows--but for different reasons. It would take us too far afield to discuss them in detail here, but suffice it to say that Naagaarjuna's use of the tetralemma is ultimately grounded in an inherently normative, a priori two-truths structure. That fact alone renders assertions based on it contradictory according to nonnormative criteria of validity.(57) Another possible option is to explain away the passages containing temporal discreteness by relying on other passages that express Hua-yen notions of interpenetration. Earlier it was stated that Dogen's thaught is grounded in Hua-yen thought. If that is true, then it is reasonable to view these statements in a similar light. To do so would entail asserting that the passage of discrete dharma-positions and the idea of two-place rela- P.400 tions are related in a manner similar to Hua-yen teachings on the interrelationship of li and shih. But this option, too, is ultimately untenable. The pertinent question in this case is not whether Dogen's discussions of time are based on Hua-yen notions--we have already assumed that they are (see note 28 below), instead, it concerns whether an application of Hua-yen notions of interpenetration can logically resolve the incongruencies between A-theory and B-theory phrases. The answer, in short, is no. Without straying too far from the topic at hand, the reason is that Hua-yen explanations of temporal interpenetration struggle with the experiential fact of temporal asymmetry.(58) Any assertion of a relative and interdependent relationship between A- and B-theory views in a Hua-yen context would have to be predicated upon first establishing a bidirectional, symmetrical relationship between anterior and posterior temporal events. Because of temporal asymmetry (i.e., time always "flows" toward the future), it is the effect-to-cause aspect of that bidirectional relationship that causes the problems. Fa-tsang struggles unsuccessfully with this very issue in sections of his Wu-chiao chang.(59) The Hua-yen failure on this count means that ultimately it cannot adequately reconcile the A- and B-theory views. A third possible option would involve explaining away the passages expressing notions of two-place relations by means of the passage of discrete dharma-positions (i.e., negating Hua-yen notions of mutual interpenetration and nonobstruction).(60) To do this one must assign greater significance to the sections on discrete dharma-positions rather than to those on two-place relations. But this option suffers from at least two problems. First, there is not only no evidence suggesting Dogen had such a position in mind, but in fact plenty of evidence indicating that he was not trying to subordinate either view (i.e., A- or B-theory) to the other. Secondly, this option has the added disadvantage of raising serious questions about the larger issue of the internal consistency of the whole Shobogenzo (see note 28 for further discussion). In light of the problems with each of the three alternatives above, we must be open to a fourth option: the real problem on this issue may simply lie with our assumption of consistency. Simply put, Western philosophical views concerning the incompatibility of the A- and B-theories may be correct even within the context of traditional Buddhist philosophy, and Dogen is therefore offering us an inconsistent philosophy of time. In any case, what may be most significant here is the existence of the problem itself. In this fact, Dogen shares the problem with B-theorists. Many A-theorists have accused B-theorists of positing a theory that cannot adequately account for the differences between B-series related P.401 temporal phenomena and the manner in which we experience time's passage.(61) Conclusion We have seen that there are elements of B-theory thought in Dogen's views about time: he viewed Reality as "token-reflexive," time's passage as subjective (i.e., relative to dharma-positions), and events as existing in objectively real, two-place relationships with one another.(62) Finally, he analyzed change as relating to qualitatively different states of a thing. But there are also important areas of divergence between the B-theorists and Dogen. In claiming that the tenses of expressed events are relative to the expression, Dogen does not conclude that the past and future are simply linguistic conventions. In fact, quite the contrary is true: because they are linguistic conventions, they cannot simply be symbols but must be the symbolized itself. In addition, the idea that temporal becoming is "subjective" does not mean that Dogen dismissed it as only "psychological." Instead, through the mediation of the nikon, temporal becoming is a notion that stands side by side with the idea of two-place relations between dharma-positions. In outlining some of the B-theory elements in Dogen's philosophy of time, we also discovered a fundamental incongruency that calls into question the consistency of his views. The problem concerns the relationship of dharma-positions as discrete space-time phenomena and the experience of passage as a smooth and continuous "flow." Dogen's notion of nikon allows for dynamic interaction between dharma-positions but not without entailing the apparently inconsistent notion of tense overlaps. In addition, although the nikon has a function within (1)time as a set of two-place relations, and (2) time as the passage from the future to the past, it does not reconcile the two. Since it is clear that two-place temporal relations hold between four-dimensional phenomena, temporal becoming continues to be an issue without an adequate answer. NOTES This article is dedicated to Alfonso Verdu. 1 - Richard M. Gale, ed., The Philosophy of Time: A Collection of Essays (Garden City, New York: Anchor Books, 1967), pp. 70-77. The explanation of McTaggart's position and the summary of B-theory are both based on, and draw heavily from, Gale's explanation. 2 - Ibid. P.402 3 - I do not claim that these are the only definitions of these terms. The definitions are stipulative and function only to establish some working parameters for the study. The intent in formulating categories through the use of stipulative definitions before the start of the study is to create a question that is verifiable or falsifiable according to available historical records. As Robert Baird has shown, this provides the basic framework for a nonnormative answer to a historical question. See Robert D. Baird, Category Formation and the History of Religions (The Hague, Netherlands: Mouton and Co., 1971). I am grateful to Robert N. Minor, Professor of Religion at the University of Kansas, for his suggestions on ways to use Baird's method. 4 -J.M.E. McTaggart, "The Unreality of Time," in Philosophical Studies, ed. S. V. Keeling (London: Edward Arnold and Co., 1934), pp. 110-131. 5 - Ibid., pp. 110-111, and also Gale, Philosophy of Time, p. 67. 6 - Ibid. 7 - Adolf Grunbaum, "The Status of Temporal Becoming," in Gale, Philosophy of Time, p. 322. 8 - McTaggart, "Unreality of Time," p. 111; also Gale, Philosophy of Time, p. 67. 9 - McTaggart, "Unreality of Time," pp. 113-116, and Gale, Philosophy of Time, p. 67. 10 - McTaggart, "Unreality of Time," pp. 123-126, and Gale, Philosophy of Time, p. 68. 11 - Ibid. 12 - Ibid. 13 - McTaggart, "Unreality of Time," p. 126, and Gale, Philosophy of Time, p. 67. 14 - Gale, Philosophy of Time, pp. 70-77. 15 - Ibid., pp. 70-71. 16 - Ibid. 17 - Ibid. 18 - Ibid., pp. 72-73. 19 - Ibid. 20 - Ibid. 21 - Hee-Jin Kim, "The Reason of Words and Letters: Dogen and Koan P.403 Language," in Dogen Studies, ed. William R. LaFleur (Honolulu: Kuroda Institute/University of Hawaii Press, 1985), pp. 54-82. 22 - Dogen, Shobogenzo, "Mitsugo." This is my translation from Nihon shiso taikei, vols. 12 and 13, ed. Terada Toru and Mizuno Yaoko (Tokyo: Iwanami, 1972) ; hereafter "Dogen." "Mitsugo" is from vol. 12, p. 57. 23 - Dogen, "Bukkojoji," vol. 12, pp. 300-301. 24 - Dogen, "Hossho," vol. 13, p. 84. 25 - Ibid., p. 85. 26 - Kim, for example, has pointed out that Dogen uses the term dotoku to signify the importance of words themselves. See Kim, "Reason of Words and Letters," p. 67. See also Hee-Jin Kim, Dogen Kigen--Mystical Realist (Tucson, Arizona: University of Arizona Press, 1975), esp. chap. 3. 27 - Kim has argued that Dogen is careful not to exclude linguistic expressions from expressions of the Buddha-nature. Note that there is no fundamental disagreement in our positions. My claim here is not that linguistic expressions are privileged in any way, only that Dogen views them, just as any other phenomenon, as expressions of the Buddha-nature. See Kim, Dogen Kigen, chap. 3, esp. pp. 103 ff. 28 - Note also that Candrakiirti defines sa^mv.rti three ways in chapter 24, verse 8, of his Prasannapadaa. According to Nagao, the third definition refers to "conventional terminology, manner of speaking, name." See Gadjin M. Nagao, Maadhyamika and Yogaacaara, trans. Leslie S. Kawamura (Albany: SUNY Press, 1991), pp. 14-15. Here I am, of course, assuming that Dogen's thought is grounded in Hua-yen thought. This position has been questioned, however, based on the fact that Dogen does not quote the Hua-yen ching directly, does not refer to the school itself in his discussions of time and being, criticizes the Hua-yen master Tsung-mi, and makes extensive use of other, non-Hua-yen texts. But, setting aside these objections for the moment, the claim that his thought is grounded in Hua-yen thought is much less problematic than the alternative--namely, that it isn't grounded in Hua-yen thought. But why? Assuming for the moment (1) that the reasonable (and historically verifiable/falsifiable) assumption that Dogen's thought owes an intellectual debt to past Buddhist thought, (2) that some acknowledgment of that influence is necessary in order to be able to make sense out of otherwise incongruent assertions, and (3) that the two options (i.e., Hua-yen or not Hua-yen) are mutually exclusive, P.404 then I note that asserting a grounding in Hua-yen Buddhism allows for a consistency of thought in the Shobogenzo on issues (including time--other problems mentioned aside) that cannot be found if we utilize the other claim. The alternative position (i.e., not grounded in Hua-yen thought) is not able to make such a claim, given the nature of many of the seemingly contradictory statements in the Shobogenzo. Since an assumption of consistency underlies both possible interpretations, the important question is, which of the alternatives yields a more consistent coherence to the text? The Hua-yen assumption has the benefit of being able to explain those items which would otherwise be irreconcilable (e.g., Sarvaastivaada-type notions of spatial and temporal discreteness with Hua-yen-like notions of spatial/temporal interpenetration), while the other option (including the attempt to read a Naagaarjunian tetralemma into Dogen's discussions of time) will not be as successful in this regard. This is particularly true for the "Uji" fascicle, which is difficult if not impossible to interpret consistently without assuming the Hua-yen view. This probably explains why many interpreters, myself included, read Dogen in that light. To take any other view requires the interpreter to explain how such glaring inconsistencies are reasonable. If we make the reasonable assumption that Dogen wasn't offering us a completely chaotic and incongruent view, then the alternative view becomes very difficult to maintain. The next question, then, is, does any of the evidence in the text necessitate against this interpretation. The answer is no--even that evidence pointed out above. The objections above are predicated upon the assumption that Dogen (1) would have made more reference to the Hua-yen ching and the Hua-yen school if in fact his thought was grounded in that school's doctrine, and (2) he would not have made reference to the texts/doctrines that he did--perhaps even in the manner that he did--if his thought is grounded in Hua-yen. Clearly, though, neither assumption is necessarily valid and both do not therefore constitute necessary or sufficient grounds for rejecting the Hua-yen reading. They certainly may give us pause in the historical sense--but, upon reevaluation, as above, there are good reasons to reject the alternative. There is, however, another possibility that must be addressed. A reviewer of this article has noted that the issue may not be either/or in nature, but simply a question of extent. The reviewer suggests that the consistency of the "Uji" fascicle, for example, is not dependent on whether or not it utilizes Hua-yen views, but rather where and how it utilizes them. This is certainly reasonable and requires further investigation. But, while the exact details of how P.405 and where these views apply exceed the scope of this essay, we should note that accepting this view does not change the validity of the assumption regarding Hua-yen as a basis for understanding Dogen's views. 29 - Gale, Philosophy of Time, pp. 73-74. 30 - Ibid. For the purposes of this essay, I take the term "temporal becoming" as synonymous with the notion of time's "passage," "change, " "flow," etc. 31 - Dogen, "Uji", vol. 12, p. 260. 32 - Ibid., p. 258. 33 - Ibid. 34 - Ibid., pp. 259-260. 35 - Ibid., p. 260. 36 - Ibid., p. 258. 37 - See, for example, the "Uji" and "Genjokoan" chapters. 38 - J.J.C. Smart, "Spatializing Time," in Gale, Philososphy of Time, p. 164. 39 - Dogen, "Uji," vol. 12, p. 258. Other scholars (e.g., Abe and Kim) have interpreted the term 'gap' in the sense of a separation between the event and the 'self'. There is no fundamental disagreement, though, between my reading here and their interpretatidn. Given agreement on the idea that Dogen's dharma-positions refer to particularized segments of being-time (ontological/spatial and temporal in nature), then it is not a matter of one view being correct while the other is incorrect. They are not mutually exclusive. 40 - The experiential issue being referred to here is our conscious experience of the passage of time as a smooth, seamless continuum without fragmented breaks, or gaps. For example, see the next sentence (not cited) of the passage mentioned above. 41 - Dogen, "Daigo," vol. 12, p. 123. 42 - I am deeply indebted to Alfonso Verdu, Professor of Philosophy at the University of Kansas, for his many insights and discussions on time and Zen Buddhism. He graciously let me have a copy of his unpublished article, "Zen and Time." While neither the article nor our discussions were specifically concerned with B-theory views of time, they did suggest to me the possibility of a study using this approach. The terms "now-not-yet," "now-present," and "now-no-longer" are developed by Verdu in the "Zen and Time" article. I P.406 have utilized the terminology here as proper to Dogen's views on time and the B-theory elements in his teachings. Note that the point made in this section of the essay does not lead to the conclusion that Dogen was a determinist. Within the context of Dogen's thought, stating that a B-series future is what is "now-not-yet" entails nothing further than the conclusion that what will be (no determinism implied or intended) stands in a two-place temporal relation to the present Now. Nevertheless, as we will see, this issue is an important one for another reason. B-theorists have been accused of "statizing" time in a way that is experientially inconsistent. Dogen certainly does not "statize" time, but, as shown, those passages which do speak of time's "flow" do so in an A-theory manner. While those A-theory-type passages do not challenge the descriptive conclusion that B-theory views also exist in the Shobogenzo, they do raise the issue of whether or not Dogen's views are internally consistent. 43 - Dogen, "Juki," vol. 12, p. 273. 44 - Dogen, "Uji," vol. 12, pp. 257-258. 45 - Ibid., p. 260. 46 - Ibid., p. 258. 47 - Gale, Philosophy of Time, p. 74. 48 - Ibid., pp. 74-75. 49 - Ibid., p. 75. 50 - Dogen, "Genjokoan," vol. 12, p. 36. 51 - Dogen, "Kaiinsammai," vol. 12, p. 141. 52 - Dogen, "Uji," vol. 12, p. 258. 53 - Ibid. 54 - For example, see Francis Cook, Hee-Jin Kim, Steven Heine, and Joan Stambaugh. Also, compare the following passages: (items in brackets are my additions): The processive nature of realization consists in the fact that the at-one-ness or immediate experience must be repeated over and over as each new event occurs.... However, according to him, the actual realization is timeless each time it occurs. (F. Cook, "Enlightenment in Dogen's Zen," JIABS 6 [1]: 25) Whether it be firewood or ash, birth or death, the winter or the spring--each has its own dharma-position which is absolutely discrete and discontinuous. Each has its "before" and "after, " but is cut off from those dharma-situations "preceding" and "following" [B-series relations in an P.407 A-series progression from future to past]. (Hee-jin Kim, "Existence/Time as the Way of Ascesis," Eastern Buddhist 11 [2]: 54-55) There are other examples as well, and I do not mean to imply here that the two scholars above have misinterpreted Dogen. In fact, the opposite is probably the case. The problem lies with Dogen's view of time itself. 55 - Gale, Philosophy of Time, p. 69. 56 - See Dogen, pp. 261-262. I quickly want to add, though, that this is not to claim that Dogen was not aware of, or did not employ, the tetralemma--only that it is not evident in the discussions of time used as a basis for the inquiry. One possible objection to this observation, noted by a reviewer, is that Dogen may have used a form of tetralemma filtered through the koan tradition of Chinese Ch'an Buddhism. As mentioned above, there is no disagreement on the point that there are koan in the Shobogenzo that obviously reflect the tretralemma. The point in dispute here would be the notion that, somehow, the other sections of the text also utilize it--only in a modified form that is different from the one used by Naagaarjuna. I do not want to take issue with the notion that ideas can change over time, but we have to be very careful about assertions such as these, which can amount to claiming that Dogen used Naagaarjuna's tretralemma, but it really is not Nagarjuna's tetralemma. The process of modifying such assertions can render them nonfalsifiable. 57 - The reason is that the use of such a tetralemma ultimately must involve the assertion of a proposition something like: "Dogen cannot be said to hold either extreme as his position...." But this is problematic because the assertion itself is subject to some criteria of validity. This is precisely the point made by Naagaarjuna's opponent in the Vigrahavyaavartanii (i.e., that in order to assert that he had no position, Naagaarjuna had to assert something and hence immediately contradicted himself) . Praasanghikas have attempted to defend him and themselves by invoking both the notion of nonpresuppositional negation and an a priori two-truths structure. Those familiar with modern scholarship on Naagaarjuna will know that there is an ongoing debate concerning whether or not Naagaarjuna and the Praasanghikas' answer is valid. I have argued in detail elsewhere that it is in fact inherently normative and hence cannot meet non-normative, verifiable/falsifiable criteria of validity. To defend Naagaarjuna's approach on this point would involve one in attempting a defense of the inherently normative two-truths claim. A further objection often raised by defenders of the validity of this method is that it is not appropriate to criticize Buddhists for P.408 their use of this device, because they had a decidedly religious purpose in mind. While this last point certainly necessitates a pause in order to ensure that our conclusions are historically valid, philosophically speaking it is a red herring and has no impact on this essay's conclusions. To grant Naagaarjuna (and Dogen in this specific case) a kind of intellectual "immunity" on these points because they may have had a religious purpose in mind is to abrogate our right to ask of them, "is this reasonable," and expect that in answer we will, in the secular academy at least, not rely upon unfalsifiable/normative criteria of validity. The issue, then, becomes whether the Naagaarjunian tetralemma meets this criterion. As mentioned above, I think the only answer is no. 58 - See, for example, Fa-tsang's Wu-chiao chang; T. 45, pp. 486a, 490a-b, 504b, 505c, etc. 59 - Ibid. 60 - This amounts to asserting that Dogen holds that the A-series is more fundamental than the B-series. But note that the first choice mentioned (i.e., using Hua-yen notions of interpenetration) does not correspond to the alternative of asserting that the B-series is more fundamental. 61 - For example, see C. D. Broad's "Ostensible Temporality," in Gale, Philosophy of Time, pp. 117-142. 62 - The conclusion here is not intended as a criticism, but simply a comparison. I have no interest in forcing Dogen to fit into the categories of McTaggart and Gale. Whether or not Dogen's agenda was different from that of the two men whose thought has been used as a basis for the comparison will not necessarily affect the descriptive results of asking the question "Are there any similarities? " It certainly would, however, have to figure into any discussion of why they are different or the same--but these matters exceed the scope of this essay.