One of the most remarkable facts about the well-known ten or fourteen "unexplained points" (avyaak.rta vastuuni), which occur in numerous places in the Sutta Pi.taka, is the quantity of learned discussion that they have engendered, not only among classical Buddhist writers but also among modern scholars. This testifies to the general appeal of mysteries, puzzles, riddles, paradoxes and nonsense. Zeno's paradoxes and the doctrine of the Trinity have exercised a similar effect upon the minds of thinkers. If a literary critic were to happen to read the volumes of scholarly prose devoted to the unexplained points, he might well see the history of the problem as a variant of the romantic quest, dogged by the failure preordained for all quests after the unattainable. A dialectical theologian might seize upon this case as an. instance of the paradoxical character of man's relation to being, a viewpoint that shows through Hermann Oldenberg's rather poetic conclusion to his discussion of the unexplained points:
There is a path out of the world of the created into unfathomable infinity. Does it lead to the highest being? Does it lead into nothingness? The Buddhist creed maintains itself on the knife-edge between the two. The longing of the heart that aspires to the eternal does not have nothingness, and yet thought has no something which it might hold onto. The thought of the endless, the eternal, could not disappear into a distance more remote from belief than here, where, a gentle breath on the point of merging into nothingness, it threatens to evade the gaze. 
A more prosaic reason for the continuing investigation of the unexplained points is that no explanation to date has succeeded in fitting them tidily into the conventional schema of philosophical classification. Does early Buddhism have a concept of the absolute? Is it agnostic? Is it nihilistic? Is it pragmatic? Is it metaphysical? Inquiries of this sort spring not from attraction to an elusive answer shimmering, like Brahman in the Kena Upani.sad, on the boundaries of the visible, but from the mundane motives of scholarly housekeeping.
The aim of this article is not to explain the unexplained points, but rather to work out a critique of the methods and techniques by which their significance has been and may be investigated. This is not going to be a historical critique of "Avyaak.rta vastuuni-Forschungen," but a theoretical approach in which particular scholars' work is adduced to illustrate principles. Nevertheless, a study of methods presupposes a concern with the nature of the subject matter to which the methods are applied. Consequently, my questions are: 1. What did the unexplained points mean to early Buddhists? 2. What methods facilitate the best answer to this question? 3. What postulates validly support answers to this question?
1. Buddha: His Life, His Doctrine, His Order (London: Williams and Nogent, 1882), pp. 283-284 (German edition, p. 328).
Before elaborating on an explanation of these questions, let us consider what questions have been asked by Buddhologists hitherto. The principle that the questions tend to prejudice the conclusions is no less important for being obvious. Oldenberg asked, "Is Nirvana nothingness, annihilation?"  Thus the dominant theme of his discussion is the relation between the Highest Goal and being and nonbeing. Henry C. Warren asked, "What was the Buddha's precise attitude towards these questions "  This question involves the problem whether or not the canonical accounts represent the ideas of Gautama, a point which Warren passes over lightly. Warren's discussion, in keeping with his question, deals chiefly with Gautama's motives rather than his philosophy. Louis de La Vallée Poussin asked: "Was Gautama agnostic about these questions?"  This question was imposed on him by the fact that he was writing an encyclopedia article on Buddhist agnosticism, and the article conveys the impression that he was really more interested in several other aspects of the problem. E. J. Thomas says: "The only real question is what conclusion did the Buddhists draw and what for them was the logical inference."  His inquiry is directed scrupulously and exclusively to the place of the unexplained points in early Buddhist thought. This is necessary and commendable, yet Thomas's account, for all its reasonableness and objectivity, fails to achieve a clear representation of the early Buddhist thought-world. It may be suggested that his question is not the only real question, and that though necessary, it is not a sufficient starting point for inquiry. Other questions must be asked, too. Shooson Miyamoto asked, "Is it true that the Buddha himself did not come to any definite conclusions about the truth of such questions?"  This question, unlike some of the preceding ones, calls for a yes-or-no answer. Miyamoto's subsequent discussion is, however, not a mere answer to this initial question, but a full-dress study of the whole problem of the unexplained points. Troy Organ asked: "Why was the Buddha silent on these metaphysical issues?"  This question incorporates a factual error; the canonical accounts say that Gautama remained silent in answer to these questions on certain occasions, but in the majority of instances he did speak, and answered at some length. He simply refused to answer "yes" or "no." The question can be rephrased, "Why did Gautama, as represented in the Nikaayas and the AAgamas, refuse to affirm or deny the unexplained points?" This question, like Warren's, concerns Gautama's motives and his inner attitude. Such things are
2. Ibid., p. 271 (German edition p. 312).
3. Buddhism in Translations (Cambridge: Harvard University Press, 1896), p. 111.
4. "Agnosticism (Buddhist)," ed. James Hastings Encyclopaedia of Religion and Ethics (ERE) 13vols. (New York: Charles Scribner's Sons), 1:220b-225a (see also "Scepticism [Buddhist]." ERE Vol. 11, 231b-232b).
5. The History of Buddhist Thought (London: Routledge & Kegan Paul, 1933), p. 128.
6. Chuudo-shisoo oyobi sono Hattatsu (Kyoto, 1944), p. 194.
7. "The Silence of the Buddha." Philosophy East and West 4 (no. 2, 1954): 127.
difficult to investigate because they are so subjective. The question, though, has the merit of not prejudging any philosophical issue. Franklin Edgerton asks: "Did the Buddha have a system of metaphysics?"  This clearly presupposes some definition of "system," and a classificatory slot labeled "metaphysics." It is a question of the philosophical housekeeping class, as the subsequent argumentation confirms. It should be noted that Edgerton took the question from Glasenapp, whose views he was criticizing. Nonetheless, the question is invidious, since it structures investigation as a process of selecting between alternatives proper to the problematic of Western, and particularly modern, philosophy.
The first methodological problem to be broached is whether typological categories are the best means of dealing with early Buddhist conceptual articulations of doctrine. One outstanding detect of the typological approach is that its objective is to classify rather than to explain. It aims at attaching descriptive labels to a system on the basis of theses, which it maintains on a limited number of diagnostic topics. It ignores structural differences, the relative importance of different aspects of the system, the problematic that the system attempts to deal with, and the routes by which conclusions are derived or proven. In classifying a. system, it eliminates almost everything that is of interest in it. Another defect is that typology assumes the commensurability of philosophical systems. This is as suspect as the now generally repudiated assumption that there is such a thing as universal grammar. Just as each term -- noun, sentence, vowel, for instance -- has to be defined anew for every language -- so each philosophical term -- being, cognition, virtue, for example -- has to be defined in terms of the system under description.
One result of this assumption that common terms and categories underlie all philosophies is to obscure the lines of historical development. The standard labels are non historical -- they do not provide designations for systems in which no distinction is made between subjective and objective phenomena, for systems where causality is not distinguished from inherence, for those which classify cognition and motor activities and physiological processes together, or for the many other salient features of sundry early Indian philosophies. A further defect in typologies is that they are often redundant if clearly defined and misleading when not. For instance, Schweitzer's distinction between world-affirming and world-denying, which says nothing factual except that philosophies that find the highest good in certain kinds of mundane experience find the highest good in that kind of experience, but which emotively has the added meaning, "and right-thinking people like Schweitzer, prefer world-affirmation to its opposite." The factual sense of such characterizations tells
8. "Did the Buddha Have a System of Metaphysics?", Journal of the American Oriental Society 79 (no. 2, 1959): 81ff.
little about the system; the emotive meaning is alien to scholarly purpose.
Typologies generally rest on a series of binary selections. Is the system optimistic or pessimistic? Is it nihilistic or not? Is it realistic or idealistic? Is it monistic or pluralistic? Is it rationalist or intuitionist? Is it positive or negative? These dichotomies have value connotations which cannot help prejudicing the inquiry. Equally bad, they contain presuppositions about the structure and problems of philosophy which cause too much trouble to warrant their use in the description of early Indian thought.
A preferable method for handling early Buddhism is to investigate the texts inductively, allowing their own problems and concepts and relational structures to appear, and shaping the description in the systems themselves. The resulting description may perhaps afford an answer to questions drawn from the history of Western philosophy, but the former will certainly not be dependent on the latter. In this process, the first operation is to work out a text-critical approach.
The unexplained points occur in each of the five Nikaayas of the Paali Sutta Pi.taka, with the greatest concentration of occurrences in the Diigha, Majjhima, and Sa^myutta Nikaayas. Most of the suttas also occur in the Chinese AAgamas, though they are grouped somewhat differently, and there are some variations in content between the Chinese suttas and their Paali counterparts. The three Diigha suttantas in which the unexplained points occur -- the Brahmajaala (DN.1), the Mahaali (DN.6), and the Po.t.thapaada (DN.9) -- are all late in composition but contain early materials, according to current ideas on stratification in the canonical literature.  This means that the lists of views and points are probably early, while the expanded explanations of them are perhaps later. Majjhima 63 is classed as early and Majjhima 72 as composite, according to G. C. Pande.  The supposed later part of No. 72 is a set piece on the skandhas, and this judgment rests on the view that the skandha is a late development. Similarly, in the Sa^myutta, Pande picks out the skandha doctrine as a hallmark of late suttas in the Avyaakata Sa^myutta.  This may be right, but there are reasons for skepticism about Mrs. Rhys-Davids's theories about the development of skandhavaada, so that if possible it will be expedient to dispense with this criterion of age. Even by Mrs. Rhys-Davids's own theories, the skandha passages might well be later editors' insertions into older texts. Thus the stratification theory as it now stands confirms that the unexplained points are early, but leaves in doubt the age of the material in the suttas in which they occur.
Another critical procedure for separating early from later is to study the variations and interpret their patterns. Two principles that apply here are:
1. When the framework narratives reveal a variety of interlocutors and places, at least some of them infrequent names (for example, Uttiya and Koka-
9. Studies in the Origins of Buddhism, Ancient History Research Series, no. 1 (Allahabad: University of Allahabad, 1957), pp. 81-90.
10. Ibid., pp. 179, 127, 156.
11. Ibid., p. 223.
nuda), then the framework belongs to an early period. Conversely, if all the locales and speakers are favorites, then there is no guarantee that the nidaanas are primitive (e.g., the Squirrels' Feeding Ground, Vacchagotta, and AAnanda).
2. In the doctrinal core of the suttas, the part that is common to several suttas is more certain to be early than the parts that differ from sutta to sutta, though this in itself does not prove the variant parts to be later.
A further principle is that a doctrine which occurs in numerous variations is likely to be early, and was probably important at an early stage. It is not legitimate to hold that a standardized exposition is less ancient than one that exhibits variants; it may simply have been better preserved or have received rigorous formulation earlier.
Applying these principles to the suttas concerning the unexplained points, one discovers that only one group of suttas -- those concerning Vacchagotta -- shows a common denominator in their nidaanas. Majjhima 72, the Aggivacchagotta and M.73, the Mahaavacchagotta are placed in the Chinese AAgamas in the Sa^myutta, together with the other Vacchagotta-suttas, which altogether constitute a cycle. This cycle will be examined shortly, but first it should be pointed out that the primitiveness of the introductory settings of the suttas outside the Vacchagotta series is clearly established because of their diversity and because of the rarity of some of the persons and places. The common core of these suttas is a general concern with the unexplained points. However, the number of points and the way in which they are stated varies considerably. Thus the common denominator is the problem underlying the unexplained points rather than the specific propositions themselves. Another common factor in many of the suttas is the tetralemma (catu.sko.ti) construction -- for example, "the tathaagata exists ... does not ... both exists and does not ... neither exists nor does not..."
Coming now to the Vacchagotta cycle, we find that in suttas 957 to 964 in Gu.nabhadra's Chinese translation of the Sa^myukta-aagama, the questioner is Vacchagotta, and the place is the Ve.luvana, in the Kalandaka-nivaapa at Raajagaha. But the respondents vary -- Gotama in 957; Moggallaana in 958; Gotama again in 959, 960, 961, 962, 963 and 964; Kaccaayana in the second part of 959; and AAnanda as second respondent in 961. This last sutta belongs to a set which extends beyond the Vacchagotta series. When Gotama remains silent in 961, AAnanda, who has been fanning him, asks Gotama for an explanation. In 965, when Gotama declines to say anything further, AAnanda, who has been fanning Gotama, offers a parable to Uttiya. In 967, Kokanuda meets AAnanda at a bathing place and asks him to explain the unexplained points. The common theme here -- AAnanda in the role of personal attendant and privileged confidant of Gotama -- looks rather like a literary stereotype but might well be simply a historical fact preserved as a stock literary motif.
Returning to the Vacchagotta cycle, we can observe a story line that continues throughout the series as arranged in the Chinese Sa^myukta. In 957, Vaccha's concern is with the sameness or difference of soul and body, and the problem underlying his concern is how Gotama can reveal (vyaakaroti) that his disciples after death are born in such and such places. At the end of the sutta, Vacchagotta is converted. In 958, Vaccha asks Moggallaana why, unlike other teachers, Gotama does not assert (vyaakaroti) one of the four propositions about the existence of the tathaagata after death. He is told that Gotama understands the skandhas yathaabhuutam, while the other teachers do not. The tathaagata, being unattached to such propositions, is profound and immeasurable. Vaccha reportedly was satisfied. In 959, Vaccha asks Gotama the same questions that he had asked Moggallaana in 958 and marvels that he gets exactly the same answer. This looks as if an editor was confronted with two suttas differing only in that Moggallaana was the respondent in one and Gotama in the other, and that the editor put his own astonishment into Vaccha's mouth. In the second half of 959 and in 960, the same questions and the same routine recur. Vaccha first asks Kaccaayana, then asks Gotama, then marvels that the two answers agree exactly.
So far, Vaccha's questions have all concerned survival after death, either for the ordinary soul or for the tathaagata. In 961, Vaccha asks Gotama whether there is an aatman. He asks three times, and each time Gotama remains silent. The question is asked simply in the affirmative, not in doublet or tetradic form. Vaccha goes away unsatisfied, whereupon AAnanda receives an explanation why Gotama remained silent. Here, it is to be noted, Gotama explains his own silence, and does so on the grounds that to affirm an aatman would only confirm Vaccha in his false views, while to deny the aatman would increase his perplexity. One alternative would have confirmed Vaccha in eternalism and the other in annihilism. But Gotama abides in the Middle Path and teaches dependent coarising. An exposition of the twelve nidaanas follows.
Oldenberg in translating the first part of this dialogue, disregards the concluding exposition of pa.ticcasamuppaada, then says, "Through the shirking of the question ... is heard the answer, to which the premises of the Buddhist teaching tended: The ego does not exist. Or, what is equivalent: Nirvaa.na is annihilation."  Thomas objects: "It is certain, however, that this is a conclusion which the Buddhists never drew. In this very sutta annihilationism is rejected. It is not really to the point to say that the Buddhist premises tended to this conclusion."  There will be occasion to consider this sutta again shortly in connection with philosophical approaches to the unexplained points.
Sutta 962 in the Gu.nabhadra Sa^myukta-aagama corresponds to the Paali
12. Oldenberg, op. cit., pp. 272-273 (cf. German edition, p. 314-315, which differs slightly).
13. Thomas, op. cit., p. 128.
Aggi-Vacchagotta, Majjhima 72. For the first time in the Vacchagotta series, the unexplained points are fully listed. Fourteen points are given in 962 as against the ten of the Paali tradition. It should be noted that the only proposition not given in tetralemma form is that concerning the relation between soul and body. This is stated with two alternatives. In fact, it is noteworthy that this proposition never, throughout the canon, occurs with four alternatives. This fact, too, will be noted later among the philosophical considerations. Sutta 963 begins much the same as 962 -- the same place and people, the same fourteen points. But Vaccha asks, "Through ignorance of what do people hold such views?" The reply is that it is through ignorance of the skandhas. The whole sutta looks like a mere mechanical expansion of one fragment of the materials constituting the Aggi-Vacchagotta. We may refuse to consider the skandha doctrine a mark of lateness, but there is nothing whatever to guarantee the earliness of this sutta, and here are strong grounds for suspecting it to be quite late.
Returning for a moment to 962, we should note that the locale of this sutta differs from that of its Pali counterpart, the Ve.luvana at Raajagaha versus the Jetavana at Saavatthi. The two Chinese versions (T. 100, 444c ff) agree with each other. The Paali, though, is almost certainly the more authentic, since it is more likely that the editors of the Sa^myukta would have brought the locale of this sutta into line with that of the others in the series than that the editors of the Pali Majjhima would have changer the locale so that it differs from that of the Mahaavacchagotta which follows immediately after, Majjhima 73.
Sa^myukta 964, the Mahaavacchagotta, does not deal with the unexplained points at all, but it must be considered in a study of the points because it is the climax and conclusion of the Vacchagotta cycle. The introduction to the two Chinese versions seems to be a late patch-up, since it has Gotama thinking, "I will now teach him according to the Abhidharma and the Vinaya." It lacks the reference to previous conversations, a long time before, with which Vaccha opens conversation in the Pali version. It is most likely that his allusion belonged to the earliest recensions, since the Paali has less reason to invent a "continuity" piece like this; but if this utterance is early, it must have fallen out of the Sa^myukta version before the Sa^myukta was arranged in its present order. Reference to previous conversations would have added a great deal to the literary unity of the cycle, and a compiler would not have excised it.
The main feature of this sutta, in contrast with the others in the cycle, is that Vaccha asks questions for which he receives explicit answers. He is told unequivocally that certain actions are good and their opposites are evil. He is told that large numbers of Gotama's followers have achieved each of the ariyaphalaani, sotaapanna, sakadaagaamin, opapaatika (e.g. anaagaamin), and arahan. And thus by implication it is affirmed that there is karman, that rebirth takes place for all except the arahan, and that these truths, their metaphysical char-
acter notwithstanding, can be expressed plainly in mundane language. Vaccha practices the teaching and becomes an arahan. Thus, success rewards his quest when he stops asking the wrong questions and asks the right ones.
This sutta consists largely of rigmarole permutations of stock items within a framework of categories. It gives a decidedly inflated impression, and in its present form is more likely the words of a sutta-reciter than those of Gotama and Vaccha. But the basic lines of the story seem credible enough, and the declaration that Vaccha became an arahan is probably based on a sound tradition. The question then arises, to what extent is the Vacchagotta cycle history and to what extent literary invention? The most highly structured, intricate, and complete of the suttas dealing with the unexplained points is the Aggi-Vacchagotta (MN. 72). Possibly this is the most faithful and complete record of an actual dialogue, but more probably it is a synthetic composition made out of fragments that occur in the Sa^myutta and A^nguttara suttas and Vaccha. There are counterparts in simpler suttas for all the constituents of Majjhima 72, but the wording is generally different. This indicates that the materials are early, but tends to throw doubt on their form. We must assume a good deal of literary invention in the composition and wording of Majjhima 72. This assumption is confirmed by the considerable variations in wording between the Chinese and the Paali versions. Even allowing for rather free translation, the Chinese must repose on an Indic original worded rather differently from the Paali text.
Some, but not all, of the suttas concerning the unexplained points contain parables. The Aggivacchagotta has the famous fire parable, another version of which is to be found in Sa^myutta 44.9 (Sa^myukta no. 957). From the text-critical standpoint, these two fire similes should be compared with the various other fire illustrations, and the theme interpreted in terms of its variants and their textual locations. The Cula-Maalu^nkyaputta-sutta, Majjhima 63, is roughly the same as the Aggi-Vacchagotta, except for the simile of the arrow, and this account has made it a perennial favorite of those who contend that Gotama condemned theoretical inquiry and espoused a purely pragmatic viewpoint. The original parts of this sutta are the introduction -- Maalu^nkya does not figure elsewhere in the suttas concerning the unexplained points -- and the parable. It is fairly certain, then, that Maalu^nkya and this parable belong together in the early tradition. But it is by no means certain that the parable originally applied to the ten unexplained points If, as the comparative study of the Vacchagotta cycle suggests, there is a direct correlation between degree of literary invention and the number of unexplained points listed, then the composition (not necessarily the materials) of the Maalu^nkya includes a good deal of artifice and embellishment. On the other hand, the parable applies well to its purpose of elucidating Gotama's refusal to affirm or deny the ten propositions, therefore, the connection between the unexplained points and the parable could be authen-
tic and early. However, we cannot assume that in its early form all ten points were listed. Consequently we cannot be sure whether the parable applied to the cosmological points -- whether the world is eternal, whether the world is finite -- to the points on human destiny -- whether soul and body are different, whether the tathaagata exists after death, or to both classes of points. The net result is that one cannot lean too hard on the specific content of this sutta, though, as Pande says, "The nature of the problem, which Buddha must have had to face often enough, the simple and practical character of the answer, the centrality of the parable, all these support the impression of earliness which the sutta as a whole carries." 
The Udaana treatment of the unexplained points (Udaana iv. 6, pp. 66-69) not previously mentioned, gives the parable of the blind men and the elephant. The story is told with an abundance of literary embellishment, indicating that the form is not primitive, but of course the story itself is sufficiently timeless in character that it could easily have been told by Gotama. The views listed in this are the standard ten unexplained points. The following section (vi. 5), gives a variant list: the self and the world are eternal, not, both, neither.
the self and world are self-made, etc.
sukha and du.hkha, self and world are eternal, etc.
sukha and du.hkha, self and world are self-made, etc.
Neither of these lists is simple, and both are probably marks of an advanced stage in systematizing. Pande says of this sutta, "It expresses Buddha's opposition to Ekaa.m`savaada so well, correlating it with his Avyaakatavaada, that it is difficult to resist considering it as old as Buddha himself. It stands on the same level as the parable of the poisoned arrow, to which it is a valuable supplement."  This argument, though, presupposes that something well said was probably said by the Buddha; it is an argument from doctrine rather than a strictly textual and formal argument. Consequently, conclusions based on it cannot be used to support philosophical reasoning without the danger of circularity.
That several different parables are used to illustrate the unexplained points indicates that the points themselves were important in early Buddhism but this fact also shows that the interpretation of these points was not fixed in detail. In the suttas, even in that stratification which the experts consider early, there is evidence of composition, systematizing, expansion, and interpretation. Furthermore, this is the sort of artifice that a redactor rather than an original preacher would employ. Consequently, we must resign ourselves to not considering the suttas the very words of Gotama, though the problems and themes are likely to be his. What we can now proceed to investigate from
14. Pande, op. cit., p. 127.
15. Pande, op. cit., p. 75.
the philosophical point of view is fairly certainly the early Buddhist tradition, and from this we can perhaps project to what Gotama's own position may have been. Thus the outcome of these text-critical considerations is both cautionary and reassuring. We must beware of taking the word of the text to be the word of the Buddha, but we need not despair of getting behind the textual formulations to some appreciation of primitive Buddhism.
The philosophical approach begins with an examination of the texts as they stand, with a view to determining what they mean. This approach is not dependent on the form-criticism approach outlined in the preceding: pages, a fact which will prove convenient when it comes to combining the results of the two approaches.
At the outset, we should try to determine what each of the propositions means in the standard list of ten unexplained points. The fullest explanation in the canon is in the Brahmajaala-sutta, D.N. 1. Here, the first point, "soul and world are eternal" is interpreted to mean, "soul is eternal, and world, giving birth to nothing new, is steadfast as a mountain peak, as a pillar firmly fixed, and though these living beings transmigrate and pass away, fall from one state of existence and spring up in another, yet they exist for ever and ever" (SSB, 2:28). This view is not an idle speculation, but an inference from the `srama.na or braahma.na's recollection of past lives. Point three of the fourteen point list, "Soul and world are both eternal and noneternal," is interpreted as meaning, "Some things are eternal, and other things are noneternal." Brahma, the Creator, is eternal, while created beings are noneternal (SSB, 30-32). The body and senses are impermanent, but the soul is eternal, (SSB, 34). The fifth point in the list of fourteen, namely, "world is finite," is taken to mean, "This world is finite, circumscribed" (antavaa aya^m loko pariva.tumo). Those who maintain this believe that in samaadhi they have touched the limits of the world. Thus rejection of this view by Gotama is as much a repudiation of the testimony of the yogin (yogipratyak.sa) as of metaphysics. The proposition "the world is infinite" is interpreted as simply the denial of the preceding point, because the experience of another yogin and samaadhi shows no limits to the world. The proposition, "This world is both finite and infinite" means that the world is limited up-and-down but unlimited sideways, in other words, that it is limited in one sense and unlimited in another. The proposition "This world is neither finite nor infinite" is explained only as a denial of the preceding three propositions, based on the reasoning and argumentation of the denier. But the formal structure is the same as that of the eel-wriggler's case (SSB, 38) "I don't take it thus. I don't take it the other way. But I advance no different opinion. And I don't deny your position. And I don't say it is neither the one, nor the other." This position is branded as equivocation and is attributed to fear of being found out in one's errors. Not knowing the answer and knowing that they do not know, they do not make assertions (Paali verb:
vyaakaroti) on the subject. Thus the fourth lemma seems to have meant equivocation to early Buddhists. The rejection of this lemma, together with the explicit statements attributed to Gotama and his disciples to the effect that he knew what was to be known, should dispel the view that Gotama refused to assert the unexplained points because he was agnostic about them.
The Brahmajaala account reveals why Gotama is said to have disapproved of the points on which he refused to declare an answer. "The Tathaagata knows that these view-points (di.t.thi.t.thaanaa) ... will have such and such a result, such and such an effect on the future condition of those, who trust in them" (SBB, 40). Thus the question is not whether these views are true or false, but whether they lead to good or evil rebirth, or to freedom from rebirth. The underlying principle, one often overlooked by modern investigators, is that you become what you know, that what happens in thought affects what happens in existence. These di.t.thi are not mere speculations for those who hold them, but vidyaa, gnosis. He who knows the secret attains the goal, because knowing gives him control over the laws of destiny. It is not a question of metaphysics versus pragmatic wisdom, but rather one of which metaphysics is the most efficacious in attaining an existential objective.
At this place another method ought to be discussed, namely, that of historical bracketing. The principle is that an idea is not likely to be less sophisticated than its prototype in an earlier period within the same intellectual community, nor more sophisticated than the views of later commentators on the idea itself. Thus the primitive Buddhist view on the relation between knowledge and existential events is fairly sure to be at least as sophisticated as that of the early and middle Upani.sads, but not as advanced as that of Naagaarjuna, Buddhaghosa, or Candrakiirti.
Upani.sadic texts on vidyaa are numerous indeed. One example will have to suffice. Kena iv.9: "Whoever knows this (secret upani.sad), having overcome sin, in the end becomes established in the supreme heaven-world." The vidyaa (upani.sad) in question is that brahman is the moving force behind the cognitive and vital processes. But knowing this arcanum is not to be a matter of mere factual recognition. "Tad-vanam ity upaasitavyam," "That brahman is the prize is to be meditated upon, ritually pondered over" (Kena iv.6). Thought is still something of an adjunct to ritual -- the idea and the act are not divorced. Przyluski drew attention to the deep significance of the terms derived from upa-aas: "Upaas, upaasana, upaasaka, bears witness to a fund of common ideas, of an ancient teaching in which deliverance is the fruit of the act.)"  In primitive Buddhism, upaasaka seems to have meant a devotee or adherent either in the household life or in the homeless life. This is why
16. Jean Przyluski and Etienne Lamotte. "Bouddhisme et Upani.sad," Bulletin de l'Ecole Francaise d'Extreme Orient 32 (1932): 147.
Vacchagotta, at the end of the Aggi-vacchagotta, can be said to have become an upaasaka. Upaasakam-ma.m bhava.m Gotamo dhaaretu..., Horner's translation, "May the revered Gotama accept me as a lay-follower," would mean that Vaccha ceased to be a wanderer. His upasampadaa (full ordination) is recounted in the Mahaavacchagotta (Majjhima 73), and so would have been later than his taking refuge. But his becoming an upaasaka involved a commitment to a teaching which he accepted. His goal, according to the Vaccha cycle, was arhatship, and he eventually attained it by a process of combined moral and mental cultivation, which is essentially the Upani.sadic upaasana, except that it concentrated on a different gnosis -- the fourfold truth and the twelvefold dependent coarising, rather than the identity of Brahman. As Przyluski says: "Buddhist gnosis and Vedaanta have for a foundation the same substratum of magic beliefs."
Another philosophical approach that clarifies the problem of the unexplained points is to ascertain from texts in the Buddhist canon and from other texts that bracket the canon temporally what basic concepts such as being (bhaava, sat) and nonbeing (abhaava, asat) meant to early Indians. The Hymn of Creation (.Rgveda x.129), which we may safely consider pre-Buddhist, says "There was not the non-existent nor the existent then... Was there water, unfathomable, profound? (gahana^m gambhiiram)." The two things to note here are: first, that the existent (sat) and the nonexistent (asat) do not jointly constitute the totality of the cosmos. Both are evidently determinate, limited. Second, the primal something, antecedent to sat and asat, is conceived as an ocean, and described in a manner reminiscent of the Aggi-Vacchagotta-sutta: "Freed from denotion by form (ruupasa^nkhaavimutto) is the Tathaagata, Vaccha, he is deep (gambhiira), immeasurable (appameyya), unfathomable (duppariyogaaho) as is the great ocean" (MN.i.487, trans. 166).
The stock canonical passage on being and nonbeing is the Kaccaayana-ovaada, Sa^myutta 12.2.15: "This world, Kaccaayana, usually bases (its view) on two things: on existence and on non-existence. Now he, who with right insight sees the uprising of the world as it really is, does not hold with the non-existence of the world. But he, who with right insight sees the passing away of the world as it really is, does not hold with the existence of the world... Everything exists: this is one extreme. Nothing exists: this is the other extreme. Not approaching either extreme, the Tathaagata teaches you a doctrine by the middle (way): Conditioned by ignorance activities come to pass, etc., etc."
This sutta is referred to by Naagaarjuna, MMK 15.7: "In the Instruction of Katyaayana the Blessed One denied the proposition that (things) exist, that (things) do not exist, and that (they) both (exist and do not) and thus explains being and non-being."
The Kaccaayana introduces the problem of change versus identity which does not appear in the creation hymn. It also defines existence and nonexistence as absolute -- what passes away does not exist in the real sense. But there is no evidence of a dual theory of truth -- an apparent versus a real truth, laukika versus paramaartha. The early teaching seems to have been monosystemic and free from the epistemological dualism of later `Suunyavaada.
Thomas observes that for the early Buddhists bhaava is something perceptible to the senses.  This should be taken together with Schayer's point that in ancient Indian discussions existence is always spatial.  Thus the question "Does the tathaagata exist (hoti or atthi) after death?" means "Does the deceased tathaagata have a spatial location, and is he perceptible to the senses?" Early Upani.sadic asseverations place the realm of the immortal, the liberated variously in the brahmaloka, svargaloka, or the trans-solar region. It is quite literally and spatially the highest cosmic plane. In cosmological suttas such as the Devaddha (DN), however, the paradise of the god Brahma is merely a devaloka, and devaloka is not the abode of immortality. The question in the Devaddha is "Where do the great elements -- earth, water, fire, etc. -- not occur?" The answer -- in the vi~n~naa.na, the spirit of the liberated man -- in effect answers the question about the destination of the tathaagata after death. It is the nirodhadhaatu (DN 111.215), otherwise called dhammadhaatu (dhamma.t.thiti), which transcends the triple world (tiloka). Przyluski maintains that in original Buddhism the three dhaatus were (1) ruupadhaatu, (2) aruupadhaatu, (3) nirodhadhaatu, but that later the kaamadhaatu was inserted as the lowest plane, when the numerical basis shifted from three's to four's ("Canonical Buddhism adds a supplementary dhaatu," p. 159).  I am skeptical about so rigid an application of the numerical theory, and furthermore, it is hard to explain away the fact that the formula continued to be called the three dhaatus. It seems preferable to assume that in primitive Buddhism the nirodhadhaatu was nonspatial, that kaamadhaatu is earth, ruupadhaatu is atmosphere (antarik.sa, antalikkha), and aruupadhaatu is sky. Since the atmosphere is the domain of the visible, this explains the otherwise puzzling name, "ruupadhaatu", "ruupa" here is not 'solid object' but 'visible object'.
I also believe that Przyluski has overlooked the metaphysical usage which words for "place" take on in canonical Buddhism, and the latish Upani.sads such as the Ka.tha and `Svetaa`svatara. 'Pada,' 'aalambana,' 'sthaana,' 'sthiti' and 'dhaatu' acquire, with the growth of Upani.sadic notions of a nonspatial Absolute, a nonspatial meaning. Thus there is no need to fit 'nirodhadhaatu' into a
17. Thomas, op. cit.
18. S. Schayer, "Das mahaayaanistiche Absolutum nach der Lehre der Maadhyamikas," Orientalische Literaturzeitung (1935): 401-415.
19. Przyluski, op. cit.
spatial cosmology; in fact, the whole puzzle as to whether it exists or does not exist can be solved simply and plausibly by assuming that it is nonspatial, "aprati.s.thita."
La Vallée Poussin and Przyluski both maintained that the early Buddhist nirvaa.na was eternal bliss in a paradise, and certainly brahmaloka is sometimes used with the apparent meaning of 'nirvaa.na dhaatu.'  Here we may suppose, though, that brahmaloka is "loka of brahman" rather than "loka of Brahmaa." The equation brahman and dhamma/dharma is established by the frequent epithet of the arahan, brahmabhuuta, dhammabhuuta, "having become/realized brahman, having become/realized dhamma."
This leads on to a reconsideration, of the parables in suttas dealing with the unexplained points. These are: the fire (MN 72 and SN), the blind men and the elephant (Udaana), and the arrow (MN 63).
Well, where does a fire go when it goes out? The answer depends on what one thinks that fire is. Neither Vaccha nor Gotama thought that it was a particular kind of oxidization. For them, it was one of the four material elements (mahaabhuuta). For many of their contemporaries it was the god Agni, dwelling in fuel, latent in the fire-sticks, sometimes unmanifested, and sometimes manifested. As the `Svetaa`svatara Upani.sad (almost certainly post-Gotama) says:
As the form of fire when latent in its source is not seen and yet its seed (li^nga) is not destroyed, but may be seized again and again in its source by means of the drill, so it is in both cases. The self (s.b. brahman) has to be seized in the body by means of the syllable aum (I. 13).
By making one's body the lower friction stick and the syllable aum the upper friction stick, by practising the drill (or friction) of meditation one may see the God, hidden as it were (I. 14).
As oil in sesamum seeds, as butter in curds, as water in riverbeds, as fire in friction sticks, so is the Self seized in one's own soul if one looks for Him with truthfulness and austerity (I. 15).
Thus the answer is, "When Agni goes out, he goes home (asta^m gacchati), returns to his abode in the unmanifested state. When the tathaagata goes out, he casts off spatial limitations, and 'goes' to the aayatana dhaatu where there is neither earth, water, fire, air nor aakaa`sa."
Now, how about that favorite of the pragmatists, the Parable of the Arrows? Here one must apply yet another well-known principle of method: the rule of the minimum hypothesis. The seemingly simple pragmatist interpretation makes several complex and unwarranted assumptions: (a) that an opposition between theory and practice was formulated by Gautama; (b) that the d.r.s.ti
20. Przyluski, op. cit., p. 154.
are "metaphysical"; (c) that Gotama's teaching (four truths, twelve nidaanas) is not metaphysical.
None of these is so.
Simple answer: Gotama objected to the d.r.s.ti because they were ineffectual.
Sammaad.r.s.ti are effective.