Chapter 2: Argument—Fine Classification, Coarse Reasoning¶

This chapter passes no judgment on doctrine; it looks only at how the book gets to its doctrine. The backbone claim in one sentence: the capacity for classification and the capacity for argument are not of the same order, and the gap between them has a fixed cause. One and the same set of distinctions is a precision instrument in the classificatory work and never takes the field at all in the argumentative work. The conclusion is not that the book cannot think. The passage on memory in fascicle twelve, omniscience and free will in fascicle eleven, the dilemma of the atoms in fascicle one hundred thirty-two are the book in its best condition. The weakness is in the argument, not in the intelligence. The sixteen sections that follow are a sampling of sixteen ways of proceeding; every section names the fascicle its sample comes from, so that each can be checked.

Section one: the measure, first. The standard is taken from the book itself. Chapter 1 established the point: this book takes itself to be conducting a debate, and it declares as much of itself in more than a hundred places in the formula about stopping other schools and displaying its own thesis, so that measuring it by the standards of debate is taking it at its own word. Begin by quoting outright the bluntest of these self-declarations, in fascicle sixteen: these six causes are not stated in the sūtras; the sūtras state only that there are four conditions; the six causes are stated here because one wishes to distinguish the conditions by way of cause.¹ This sentence has to be read before the seventy-five charter sentences and not after. The charter says that the scripture did not say everything and that the treatise will supply the rest; this one says that what the treatise has here is not in the scriptures at all. With the measure set that high, two positive specimens are wanted as calibration, lest the fifteen sections following be read as a complaint. Fascicle eleven, on omniscience and free will, has both sides of the exchange reasoning properly. Fascicle seventeen breaks an analogy, and breaks it exactly at the relation the analogy has to carry, which is an accurate stroke.¹ The main body of the positive specimens is in section fourteen; here they serve only to fix the scale.

Section two: the two levels of reductio. A reductio has two levels. One rests on common sense, and anyone at all must grant it; the other rests on the doctrine of one's own school, and is effective only for those who already believe that doctrine. In the arguments of fascicle ten the two levels are used interchangeably, and if the opponent stands outside the fold the second level cannot move a single step.¹ Mixing them is no crime in itself; the failure to distinguish them is the problem. Nowhere in the book does it stop to ask whose absurdity a given reductio is reducing to.

Section three: the inertia of the table. Once the grid has been drawn it has to be filled. Fascicle thirteen fits one instance apiece to six formulas of the pattern "because such and such is predominant," and all six occur once in the book and never again. In the same fascicle a single "why" is answered by eight "whats" in a row. The female faculty and the male faculty each take a cell on the table, and the existence of the cell precedes any reason for it.¹ The leading instance is in fascicle thirty: the strength of Nārāyaṇa is set out as a ladder of nine grades multiplying upward, every grade with a name to it and not one grade with a source.¹ The three lists of kinds of existence that stand side by side in fascicle nine without being adjudicated were treated in full in Chapter 1 of Part 15, and belong to the same family. Tables breed of themselves.

Section four: false dilemmas and question-begging. Fascicle nine, arguing whether an intrinsic nature can know its own intrinsic nature, sets out from a premise implicit in the school's own doctrine in order to prove a presupposition of that same doctrine, and comes round in a circle to its own starting point.¹ The hardest figure is this one. Every place in the book where a ruling is issued after the announcement that all the positions on offer are at fault (twenty-seven cases in all, read through one by one) divides as follows: seventeen pronounce the faults and then take one of the branches back again; five discard both branches and set up a third item; three revise the finding to say that both branches have their reason; one announces that the opponent's copy of the sūtra is miscopied; one concedes in part. The judgement is for the most part a performance for the reader.

Section five: example in place of analysis; similes broken but never unpacked. The simile of the gushing spring in fascicle ten is never analysed from beginning to end. The four graphs rendered "citing a present-day worldly simile" do three mutually contrary jobs across three fascicles, and the one simile put to two different challenges in fascicle thirty is the fourth. Fascicle seventy-three, explaining how the sphere of dharmas came by its name, cites eight precedents in a row in place of one analysis.¹ In this book an example is not an auxiliary to argument; very often it is the argument.

Section six: the slippery slope. "What is to prevent…" occurs eleven times in the whole book, seven of them crowded into a single passage of fascicle ten.¹ Every step of the slope pairs a number with a stock of dharmas, and the pairing runs out at six; the words "a hundred, a thousand, past all counting" stand bare at the end of the sentence with not one step of argument behind them. The place in fascicle seventeen is the counter-example, and has already been assigned to section one as calibration.

Section seven: arguments from a missing option, and the terminus of no-liberation. A number of possibilities are listed, they are eliminated one by one, and what is left over is the answer; but why the list should be complete is never explained. Reasoning of this kind recurs throughout the book, and its heaviest form is a closing sentence: if that were so, there would be no liberation. As though the unacceptability of a consequence were by itself enough to establish a premise.

Section eight: the double standard, in four movements. The rule is laid down. Fascicle nine states expressly that one may not use a present-day worldly simile to raise difficulties against the dharmas of the noble ones, the dharmas of the noble ones being one thing and worldly dharmas another; and in the same place it dismisses a simile of the Mahāsāṃghikas outright—it need not be reconciled, since it is not comprised in the three baskets.¹ The apparatus is built. Fascicle fourteen subdivides this prohibition into four pairs of categories, and the instrument grows finer the more it is fitted out. The rule is broken. Fascicle twelve uses a present-day worldly simile to explain its own school's position, and fascicle thirteen appeals to the ridicule of ordinary people in order to fix the number of places the material faculties occupy. The fourth movement is the most telling of all. "Said according to the conventional and not according to the ultimate" occurs in exactly two places in the whole book, fascicle nineteen and fascicle seventy-three, and in both the two truths are brought in to get the book out of a difficulty, not brought in to analyse a problem.¹ The instrument exists; it comes on stage only when an escape is needed.

Section nine: three answers stacked one on another. The chief instance is in fascicle thirteen: three answers set side by side, mutually exclusive without the book noticing it, and the best of the three buried in the middle. Two subsidiary instances: the difficulty about sequence in fascicle ten stacks three answers of which two cancel one another, and fascicle twelve gives three answers to a single question about Devadatta and Yajñadatta.¹ The convention of "further" queues answers up without ranking them, and the longer the queue grows the less likely it is that anyone will ask whether they can be made to fit together.

Section ten: "it should not be pressed," a sentence that never counts. "The author of the treatise wished it so" occurs sixty-seven times, "it should not be pressed" seventy-two, "it should not be objected to" thirteen.¹ The literal sense of all three is that the matter stops here. Their actual use is that having said so, the book goes on to answer whatever needs answering; at the place in fascicle thirteen, "it should not be pressed" is said and the reasons then supplied are more numerous than the reasons supplied where the sentence is not said at all. This makes an exact pair with the next section: there the book really does stop, and here the mouth stops while the hand goes on.

Section eleven: the stopping points of argument, in three forms. Wherever a question touches the perfection of the Buddha, the method changes from reasoning to something else. The three forms make a complete set. Scripture is taken as the standard—"because scripture is the authority," fascicle twelve. There is flat denial—"this is not reasonable," fascicle eleven. And there is the appeal to how things simply are—"such is the nature of things," fascicle thirteen, with which go "it does not conflict with the nature of dharmas" and "everything gives rise to doubt, and therefore it should not be pressed."¹ What the three stopping points have in common is that not one of them explains anything; each merely declares that the question terminates here.

Section twelve: "for the sake of following the text." One place in fascicle twelve: a metaphysical difficulty receives two answers set side by side, one of them the principle of the two truths, the other "for the sake of following the text," that is, in order to keep the wording running smoothly.¹ A reason of style and a reason of philosophy are placed in the same cell and treated as being of one weight. This is the smallest section in the chapter, and its posture says more than its size.

Section thirteen: the stretch of name and entity. When a name is a new dharma and when it is merely another name for an old one—the criterion is never declared. One passage in fascicle twelve lists twenty names and allows only fifteen substances; yet at possession and non-possession a single absence is enough to be set up as a real dharma.¹ The rate of exchange between name and substance floats as need requires, and the floating is itself never explained.

Section fourteen: proportion, and what the book does well. The main body of the positive specimens belongs here. The passage on memory in fascicle twelve classifies finely and observes accurately (forgetting is not the same as failing to call to mind), and it keeps a sure hold on how far name and substance may each be stretched. The dilemma of the atoms in fascicle one hundred thirty-two is a genuine dilemma, with three counter-questions besides; the three difficulties about the sphere of the tangible in fascicle seventy-three are another writing of the same topic, and reading the two fascicles together shows at a glance where the book's sense of proportion lies.¹ Assigned to this section along with the positive specimens is the mechanism of instruments built and never used. The book has made a number of excellent instruments of discrimination, and those already named are the four modes of answering, the two registers of speech, the four kinds of remoteness, the five kinds of existence in fascicle nine, and knowledge from the teaching—this last a term used once in the whole book and dropped as soon as it has been used.¹ It belongs to this section and not to section eight, because the two are not the same thing. A double standard is a rule laid down and then broken by the one who laid it down; instruments built and never used is a good instrument made and thereafter never brought onto the field. It is the latter that supplies the working mechanism for the sentence in this chapter's title: one and the same set of distinctions is a precision instrument in the classificatory work and never takes the field at all in the argumentative work.

Section fifteen: evidence placed out of reach. Two of these. A scripture that has been lost: fascicle sixteen says that the Ekottarāgama originally increased from one item up to a hundred and that only one to ten now remain, and that the venerable Kātyāyanīputra, by the power of the knowledge born of resolve, beheld the place in the sūtras where the six causes are taught, and composed the treatise on that basis; this is the only occasion in the whole book where the knowledge born of resolve is used to certify, after the fact, the source of a scripture no longer extant.¹ Past lives: fascicle eighteen brings forward three previous-life stories in a row as evidence.¹ What the two have in common is that the form the evidence takes makes refutation impossible in principle, and that for the same reason the evidence does not constitute support either.

Section sixteen: substance first, argument after. The section that closes the chapter. Fascicle thirty supplies the axiom (because the intrinsic nature of all dharmas is fixed), the one time in the whole book that this premise is spoken out loud; and the same fascicle pushes superior and inferior through flat as fixed natures, rewriting a relation into a constituent.¹ Fascicle one hundred fifty-eight supplies the picture of that axiom run all the way to its end: three questions and one answer on the infinite regress in the severing of the defilements, eleven points of difference between possession and non-possession, and at the last the setting up of the possession of possession—in a single moment there are three dharmas only, that dharma, its possession, and the possession of the possession, and since they possess one another in turn there is no infinite regress.¹ The frame has four movements: an axiom, an inference, a cost, and the cost renamed as doctrine. The three difficulties about the sphere of the tangible in fascicle seventy-three, and Śāriputra's symmetrical derivation by six times two, belong here as well: first the grid is laid out evenly, and then the problems that come up inside the grid are dealt with. This section is the first beam of the verdict in Chapter 9.

Closing. Taken together, the sixteen sections say one thing: the book's apparatus of reasoning is far weaker than its apparatus of classification, and the weakness has a definite shape. It does not lack the power to discriminate; what it lacks is a channel by which that power can be brought up to the front line of an argument. What blocks the channel is a question Chapter 3 answers by half: wherever the argument cannot go the whole way, something less troublesome is available.

It is handed on to Chapter 3. With the actual state of the argument laid out, the question falls of itself upon the rulings: where argument cannot settle a matter, what does.

References¶