CC BY
116
UDC 1(091) Haig KHATCHADOURIAN
STARTING-POINTS IN PHILOSOPHY
The question of the so-called starting-point of philosophy has exercised many minds since the beginning of philosophical thinking. Even those philosophers who have not written about it have of course had to take something or other as a starting-point in their own philosophy. Still, not many philosophers have been clear as to what sort of starting-point they were thinking of, advocating or employing as a point of departure. For one thing, they often did not distinguish factual and normative starting-points: where a philosopher simply happened to start, and where he or she should start, respectively (cf. Hahn, 1958).
From a methodological standpoint, the latter is obviously more the important of the two, and has the more important consequences for philosophy, in relation to its nature and aims. It is in this more important sense that I shall consider the question of the so-called startingpoints of philosophy.
The question we want to raise and attempt to answer therefore is: "What should be the starting-point of philosophy or what is the 'proper' starting-point of philosophy?" But the question raises a number of other questions. First of all, what should we understand by a "starting-point," leaving aside for the moment the distinction between a normative and a factual starting-point? Shall we say that Descartes' starting-point was his (alleged) skepticism regarding everything he had believed to
* This article was written by late Haig Khatchadouri-an in October 31, 2015.
be true, at the time he ostensibly wanted to launch upon philosophical inquiry? Or shall we say that his starting-point was "really" the Cogito ergo sum? Or shall we go even further back than his supposed "universal doubt" and say that his starting-point was "more really" or "more truly" those beliefs that had been taught or had been unconsciously acquired, and with which he gradually or suddenly became dissatisfied? Clearly each of these can, in some sense, be considered Descartes' "starting-point." We might say, for example, that the first was his negative or critical starting-point, while the second was his positive starting-point; that the first was his starting-point in the sense that it was the point at which he began to clear the ground for his constructive views, and that the second was the "starting-point" of his positive pronouncements about reality. We can state the matter differently by saying that the first was the starting-point of Descartes' philosophizing, the latter, of his philosophy. What then about the third presumed "starting-point"? That might be called his pre-critical "starting-point." These distinctions would leave some people dissatisfied; and I think for a number of good reasons. Some may very well begin by objecting to the last-mentioned "starting-point." They would say that the beliefs Descartes temporarily repudiated when he launched his so-called methodological skepticism cannot be considered a starting-point. And they would go on to explain in what sense their objection is justified; in what way that was not the, or even, a starting-point
in Descartes' thought. I am not here concerned with whether the objection would be valid. I am concerned with one important point which this and our illustration as a whole brings out; namely that it does not make sense to speak of this or that as an actual (or as a normative) starting-point unless one specifies in what sense or respect, in relation to what, one's starting-point is a starting-point. But the dissatisfaction with the above distinctions may arise from a different source. It might be pointed out that when we speak of something as a starting-point, we should distinguish a starting-point in the sense of (a) certain "data" we start with or from, and one in the sense of (b) a certain method of investigation we utilize at the start, or in the general pursuit of our philosophical inquiries. In addition, that under (a) we should distinguish three sorts of starting-points: (i) certain questions or problems we start with; (ii) certain facts in the sense of actually existing objects or states of affairs, linguistic habits, and the like; and (iii) certain propositions believed to be true. If we make these distinctions we shall find that, in the case of our historical example, the first-startingpoint falls under (b) if the "universal doubt" is considered as a philosophical method of doubt, and not as a purely psychological process of doubting. The second "starting-point," by contrast, then falls under (a) (iii). Or, if we think of the starting-point as the fact or alleged fact that I (Descartes in the actual example) think and therefore the fact or alleged fact that I (Descartes) exist, rather than as the proposition "I think therefore I exist," our starting-point will fall under (a) (ii).
The preceding does not exhaust all possible senses of 'starting-point.' Even in the case of a normative starting-point, a temporal distinction is necessary. To say that in doing phi-
losophy we should start with or from this or that thing - a given method, certain empirical facts, a given proposition or set of propositions, a given problem - means that we should start with or from it chronologically.
In the case of a normative starting-point in our sense (b) above, the temporal and logical starting-points of philosophical inquiry coincide. A starting-point in the sense of a method of inquiry enables us to discover evidence for the truth of philosophical propositions we arrive at through its employment. It would also help us to arrive at evidence for or against philosophical propositions entertained by other philosophers; or propositions we ourselves entertained before we employed the method. In regard to a starting-point of type (a) above, the same would be true in the case of (a) (ii) and (a) (iii), but not of (a) (i). This, I think, is clear in the case of (a) (ii). As to (a)(iii) the propositions we start with or from would furnish us with evidence for or against certain other propositions provided they themselves are considered to be true - as Descartes believes the proposition "I think therefore I am" to be true. But that supposition is already involved in its being considered as a startingpoint; though, I need not add, others may deny their truth and reject them as a proper startingpoint.
The thesis I wish to affirm - a familiar thesis in contemporary philosophy - is, following the Later Wittgenstein, that the proper-starting point of philosophy is language, and the analysis of language. What I precisely mean in terms of the foregoing distinctions may be stated under the following heads: 1. That language is the proper starting-point of philosophical inquiry in our sense (a) of 'starting-point'; that is, that language is the "datum" with and from which we
have to begin the activity of doing philosophy. By that I mean that actual usage (ordinary usage in the case of ordinary language, and technical usage in the case of technical languages) constitutes the original raw material of philosophical inquiry.
2. That the analysis of language, both ordinary and technical, constitutes the first proper method of philosophical inquiry.
3. That, as a consequence of (1), language as datum is the logical, proper starting-point of philosophical inquiry; and as a result of (1) and (2) the nature of language insofar as it is known to us, furnishes the logically first (kind of) criterion of truth or falsity of the results that may be arrived at through its analysis.
Each of the preceding three assertions raises some important as well as some less important issues. And at least some of the former issues must be dealt with if a plausible case is to be made for our assertions. Thus, it will be said, first of all, that those assertions require support; and second, that the elaboration of this support is logically prior to the analysis of language as the "starting-point" of philosophical inquiry. The real starting-point is the justification of these so-called starting-points themselves!
That the views propounded here require support I fully agree; and I may point out that the analysis of the uses of the term 'philosophy', as it has been employed in the history of philosophy, would provide part of this justification. Part of the justification must also come from a study of the actual practice of philosophers, past and present, as philosophers. And that is not a matter of linguistic analysis. Indeed, some account of the nature of philosophy
is necessary if our discussion of starting-points is to make sense. For instance, it is obvious that the nature of an inquiry's starting-point in sense (b) depends on the nature of the inquiry and its aims. It is true that one and the same method may be capable of serving different ends; but one general sort of method suits one sort of inquiry, and another general sort of method another. Thus, the conceptual analysis of ordinary language, however exhaustive or profound, cannot furnish, say, scientific knowledge of the physical world. Similarly, the nature of a starting-point of philosophy in the sense of initial data or raw material depends on the nature and aims of philosophy.
But if the starting-points of philosophy have to be determined in light of the nature and aim of philosophy, is not the specification of the latter logically prior to the former; and in that sense at least, if not also temporally, the proper starting-point of doing philosophy? That it is a curious and rather disturbing fact that after more than two millennia of philosophy the very nature and aims of philosophy -let alone the proper method or methods of philosophical inquiry - are still a matter of debate and controversy. This frequently makes it desirable for philosophers to temporally start with a conception of philosophy and its functions arrived at through an empirical analysis of what other philosophers say about philosophy, and, more importantly, practice as philosophers. True, many philosophers do not actually begin with such an inquiry; they plunge into the business of raising and attempting to answer specific issues; and as long as they have a clear idea of what they are trying to do as philosophers, that is perfectly in order. What is imperative is that the philosopher needs to have a clear idea of what he or she is trying to
do as a philosopher, and of the ways in which that is like or unlike what other philosophers have professor or practiced. Otherwise, he or she is in danger of using improper tools.
A major, if not the most important reason why a considerable number of contemporary philosophers reject the traditional method(s) of doing philosophy and, so, the conclusions they reach is their changed or changing conception of the nature and functions of philosophy. A main example is the jettisoning of metaphysics, the bette noire of much present philosophy, due in good measure in my view, to the great advances in physical sciences and cosmology in the 20th and 21st centuries. Indeed, since the rise of modern science in Galileo, Newton and beyond, the relation between philosophy and science have been continually changing, with much that was traditionally considered to be part of philosophy being taken over by the sciences. A recent example is the appropriation of speech-act theory, which originated in John Searle's combining J. L. Austin's Wittgenstein-inspired How To Do Things With Words with the later Wittgenstein conception of a practice, by the empirical science of linguistics. Another recent example are the questions and controversies, in contemporary physics and cosmology, of the so-called "strong anthropic principle," first formulated by the cosmologist B. Carter in "Large Number Coincidences and the Anthropic Principle in Cosmology," which recasts the traditional teleological argument for God's example in terms of the current scientific theorizing concerning the origin and nature of the universe (see Khatchadourian, 1955). In the continually changing relationship between philosophy and science, various originally philosophical issues have now become empirical, thus more exact
scientific questions, and, at the same time, philosophers of science and even philosophers or religion are drawing on contemporary scientific knowledge and theorizing to answer certain erstwhile philosophical questions. The ongoing interplay between philosophy and science, science and philosophy, have so far had, and should continue to have in the coming years, a continually-evolving healthy relationship between philosophical analysis and scientific inquiry, with philosophy critically reacting to and appropriating the theoretically and empirically well-grounded results of the various sciences, to their mutual benefit.
Where so much depends on a philosopher's conception of philosophy, it is necessary that he or she start with a clear understanding or conception of it. I do not mean that the use of a given method, or of certain questions, facts, or propositions as starting-points, necessarily presupposes or "implies one particular view of the nature and objectives of philosophy. A given starting-point, in any of the senses of this term distinguished, may be compatible with a number of more or less different conceptions of philosophy and its objectives. This is of considerable importance, as will become clearer later on. Yet, in practice, it is better to be as clear as one possibly can - and imperative to have some idea - about what one wants to do and where he or she wishes to go, before he or she actually starts on his or her way.
There is an important sense in which one can have what philosophy can, properly speaking, be and achieve - as against what one may conceive it actually to be - only by practicing what, in the history of philosophy as a whole, has been more often than not considered as philosophy. In that sense, it is necessary to start with asserted questions that are or were tradi-
tionally considered to be philosophical in nature, and to attempt to see what one can do with them; and whether they can be answered at all.1 The careful study of the attempts of other philosophers to answer these questions is also enlightening. A considerable number of contemporary philosophers are doing these things in attempting to define or redefine philosophy and its function. The present writer's views here concerning the forms and methods of inquiry proper to philosophy are mainly the result of his reflections on the nature of the utterances of philosophers and their philosophical practices.2
Now the analysis of the past and present technical philosophical uses of 'philosophy' (the ordinary, non-technical uses are of secondary importance here) is in line with the present author's view that language constitutes the proper starting-point of philosophical inquiry in sense (a) of this 'proper starting-point of philosophy.' However, since the analysis of the extant body of writings generally considered as philosophical in nature can likewise be claimed as a proper substantive starting-point of philo-
One example: Whether, given our limited scientific knowledge of the nature of consciousness and its relation to the brain and the body as a whole, any credible conceptual-empirical evidence exists for individual survival after death. There is of course a fairly extensive body of questions or problems that past philosophers have considered as philosophical in nature. That is less true at present. It is notorious that the 20th century logical positivists rejected as pseudo-problems a large number of traditional questions as nonsensical. Other 20th century philosophers, e.g., the Oxford School, have re-interpreted and dealt with them in a novel way. There is certainly much more agreement in the history of philosophy on questions and problems then on the answers to them. Again, there has been, on the whole, more agreement or less disagreement on questions than on the methods of dealing with them. Hence despite the important problems traditionally considered as problems have a strong claim to being taken as the starting-point of philosophical inquiry in
sense (a) (i).
sophical inquiry in the same sense (a), we must qualify our earlier three theses. For in our proposed analysis of these writings we are not solely or even primarily concerned with the language in which they are couched but chiefly with the nature of the questions they deal with, the methods actually utilized to deal with these questions, and the results arrived at. Therefore, we should now add that language and its analysis are indeed our proper-starting-points, but only with respect to the investigation of specific philosophical questions, and of specific terms and concepts that may be or are of philosophical significance. That is, that they are the proper starting-points of philosophy as against meta-philosophy. That will, in an obvious sense, also take care of our third way of discovering the nature of philosophy; i.e., through different attempts to resolve assorted philosophical problems. This view of the proper starting-point of philosophical investigation of specific problems requires support in terms of an account of the nature and objectives of philosophy.
Two points need now to be made in connection with our statement about the analysis of the corpus of philosophical writings, and the attempt to resolve assorted philosophical problems so as to discover the exact nature of philosophy. First, I am well aware that under the impetus and inspiration of the movement called "Linguistic Analysis," the works of an increasing number of philosophers in the past have been subjected to semantic analysis, and this tendency has been to interpret - or rather, to re-interpret - their utterances in purely semantic terms, as purely semantic pronouncements rather than as true or false statements
2
about the world (see White, 1955). From the standpoint of the history of philosophy, I consider that as illegitimate, if and when it is asserted or implied that past philosophers actually intended their works to be simply and solely collections of semantic utterances. Such a view would distort the intentions of these philosophers and what they thought they were saying or doing. It would be a dramatic example of reading meaning into their works; or turning back the clock and attributing to them a conception of philosophy they did not hold, or did not hold in the thoroughgoing form their interpreters sometimes attribute to them. It is, I think, undeniable that many if not all philosophers in the past considered as part of philosophy the analysis of ordinary concepts such as knowledge, courage, virtue, mind, matter, free, determined. Also, that they did attempt the analysis of these and numerous other concepts. But they certainly did not think that this constituted the whole of philosophy, or that their analyses were purely and solely analyses of (ordinary) language; that they were not at the same time making discoveries about the "facts" in the world, adding to our knowledge of the world - the "nature of the world," the "nature of reality," as they usually put it. The situation would be quite different if we consider the contemporary philosophers' analysis of the view of a Berkeley, a Plato or an Aristotle simply as a re-interpretation of them in the sense of being a conception of what they (puta-tively) are really about; what the interpreter thinks they must be if they are to be considered philosophical in nature. Whether any particular re-interpretation of past philosophy or of phi-
Also compare (and contrast) that to the discussion and evaluation of David Hume's theory of personal identity, and other essays.
losophy as a whole, it is certainly theoretically possible that philosophy is different, perhaps quite different, from what it was, or even is now, thought to be. Thus, there is a real need for the discovery of what philosophy is and can be through the actual attempts of philosophers to do philosophy. Putting it more strongly and justly, what really matters in the last analysis is the degree or extent to which philosophy can attain those objectives it has traditionally professed; not what philosophers, past and present, optimistically or pessimistically thought or think it capable or incapable of achieving. That, by the nature of the case, must logically and temporally follow on the required investigation and is not, in either sense, prior to it. Here again I am advocating the use of the general method of semantic analysis as a startingpoint. To arrive at a satisfactory view of philosophy itself and not only for the investigation of particular philosophical issues in their own right, we need, among other things, to start by investigating these specific issues with the help of the method of conceptual, semantic analysis. A fortiori, since I am claiming logical and temporal priority for conceptual analysis (startingpoint in sense (b)) as applies to specific issues (our starting-point in sense (a) (i)), only with respect to philosophy and not also to meta-philosophy, the foregoing does not conflict with our position. I might add that the attempt to answer a given philosophical question, whether as a means or as an end, requires the analysis of the key concepts involved, hence the analysis of the particular meaning(s) of the key expressions in the question. Where these expressions and concepts occur in ordinary discourse (and in the case of philosophy, many of them do), the proper starting-point in the sense of the primary datum, is in my view ordinary language itself. In
3
addition, the starting-point in sense (b) is the method I am referring to as semantic or conceptual analysis.4
The reader may cry out that she cannot see the value of all this talk about starting-points. She may protest that this talk about different sorts of starting-points is more confusing than enlightening. She may feel that philosophical inquiry is not only being placed in a straitjacket by maintaining that ordinary language - or anything else for that matter - is the proper startingpoint precisely to the type of abstract sweeping pronouncements to which philosophers have been traditionally prone, and from which contemporary philosophers are trying to free philosophy. And why should there be a single starting-point for all philosophical inquiry, in any of the senses of 'starting-point' distinguished? Would we not, in maintaining that, be committing something like the "fallacy of essentialism" if not that fallacy itself?
It is admitted that rigidity is a real danger and must be carefully avoided. The attempt to clarify issues by means of logical distinctions may defeat its purpose by going too far, by turning into a logical game of hair-splitting. At the same time, the making of clear distinctions - whenever the particular subject-matter permits it - is essential if philosophy is not to be -and in some ways to remain - a set of vague and confused practices and a set of vague and confused statements. Additionally, the drawing of proper distinctions here is of considerable utility. For what a philosopher takes as his
In Philosophical Analysis: A Critical Study in Method (Khatchadourian, 1967) (later published by Wiley) is attempted to show that even when a given philosophical issue or statement is couched in technical language, and even where it cannot be rendered in ordinary nontechnical language - for instance, "Are a priori synthetic judgments possible?" - the safest if not the only road to them passes through the thoroughfare of ordinary language.
"primary datum or data," and the method with which she starts, will have an important bearing on the kind of results she will reach. The history of philosophy shows that it makes quite a difference whether we start with, say, ordinary language and the analysis of ordinary concepts, facts drawn from the natural sciences, or the "common sense" beliefs of the "man in the street" about cabbages and kings.5 It also shows that a considerable number of the errors and confusions of past philosophy are due to the employment of the wrong sort of phenomena as primary data, or the wrong method or methods: a method or methods inappropriate to philosophy, or one(s) that, though appropriate for it, can only be properly used after some other method has done its share. And that, I am maintaining, partly stems from an inadequate or misguided conception of philosophy and what it
One question concerning the "correspondence," if any, between the "commonsense view of the world" - if such a thing exists - and ordinary language. To the extent that they may be "correspondence" between the two, determines the precise nature of and the reasons for the "correspondence." For example, when we say that physical objects continue to exist when not perceived, is that "true" by virtue of the way we ordinarily use the expression 'physical object,' as that anything that would disappear when nobody perceives it cannot be (called) a physical object? If the answer is "yes," is the reason that ordinary language itself, the way we talk, a reflection and product of our commonsense view of the world: in this case what we call objects? If that is true, the fundamental question becomes: "Why is the com-monsense view of the world, for example, what it is, and what evidence we have or can have in its support?" Further, why is it that we are so terribly reluctant to give up or even to modify it even a little? Or do we modify and even abandon commonsense beliefs? The answer is clearly yes with respect to, for example, the earth and the universe as a whole; e.g., the long discarded belief that the earth is flat, that it is stationary, that the sun and the planets revolve round the earth, or that the universe came into existence (following Old Testament genealogy) some four thousand years ago. All these early "com-monsense beliefs" were abandoned under the influence of science. So after all, what is a "commonsense belief"?
5
4
can or cannot do. Modern science is fortunate in these respects, and provides an interesting difference between it and philosophy. But science was not always immune to the sort of danger or error I am speaking about, as the history of astrology and alchemy for instance shows.
It might add that what many contemporary philosophers consider as a revolution in philosophy has resulted precisely from the way in which those associated with or inspired by it have differed from many past philosophers in their use of philosophical methods, and in what they take as the data of philosophy. They have also differed from it in other respects as well; but the points I mentioned are certainly part of the difference.
With respect to the question of whether different types of philosophical questions or different branches of philosophy require different starting-points, my position is as follows: what I said about ordinary language and its analysis is meant to apply to any and all philosophical issues in value theory in general; e.g., in ethics and aesthetics, in semantics and the philosophy of language, epistemology and ontology, and to the philosophy of religion.6 The disciplines to which it does not apply are (1) formal logic and the philosophy of logic (logical theory), (2) philosophy of science, social, political and economic theory, philosophy of mathematics, philosophy of education: in short, in branches of philosophy in which the object of philosophical inquiry does not itself fall within philosophy as historically understood and delimited, or as restricted, extended or re-mapped in contemporary philosophy.
6 For the crucial role played by contemporary physics and cosmology in various recent discussions concerning, for example, the origin and nature of the universe see The New Design Argument and God (Khatchadourian, 2016).
(1) It is clear I think that the construction and manipulation of purely formal, un-interpreted logical calculi can be undertaken without appeal to ordinary ("natural") language; and certainly without starting with ordinary language and its conceptual analysis. But the application of formal logic to everyday pieces of reasoning does require an understanding of the nature of relevant ordinary concepts. Similarly, the elaboration of a logical theory that, among other things, attempts to determine the exact nature of logic and its relation to (a) science, (b) to philosophical inquiry and to (c) ordinary discourse, requires no appeal to ordinary language and to analyses of it, whether as a starting-point or no, as far as (a) is concerned. On the other hand, it is necessary to appeal to them in relation to (c). For without some knowledge of the nature and articulation of ordinary discourse in all its bewildering complexity, it becomes impossible to discern the relation of formal calculi to it. Finally, if it is true that philosophical inquiry respecting specific issues and concepts should properly start with the analysis of ordinary language, some analytical knowledge of it is necessary for a correct and adequate understanding of the relation of logic to philosophical inquiry. But an understanding of the method of semantic analysis is not arrived at by any semantic analysis of that method; though the nature of ordinary language is discoverable through the conceptual, semantic analysis of ordinary language. It does not follow from this - and it is also not true - that ordinary language and its analysis constituted the, or even, a - the proper starting-point of any attempt to trace the relation of formal logic to philosophical inquiry, using 'philosophical inquiry' as against 'meta-philosophical inquiry.' For the tracing of the relation of logic to philosophical inquiry lies in
meta-logic and meta-philosophy. The analysis of ordinary language is therefore necessary but not sufficient for it. We also need an analysis of formal logic itself: and that is not semantic analysis for the simple reason that formal logic is not ordinary language in any ordinary, un-extended sense of 'ordinary language.' The situation is different with respect to so-called "informal logic," the logic of ordinary language. There the logical structure of ordinary lan-guage7 itself is the object of inquiry; and that is discoverable only by an analysis of ordinary language. So, an understanding of "informal logic" needs to start off if not also end up with the analysis of ordinary language.
The situation is similar in the case of the branches of philosophy enumerated under (2) earlier. Since the subject-matter of inquiry is each case something other than either philosophy or ordinary language, though ordinary language is part of that subject-matter in some instances, the latter cannot be necessarily taken as the proper or even as an actual starting-point in sense (b). Nevertheless, for a complete understanding of some of these disciplines, semantic analysis is possible and useful, nay indispensable. But it would be arbitrary to maintain that such analysis must be taken as the starting-point. There is a language of religion, a language of politics, a language of economics, and, similarly, of law. Insofar as that is so semantic analysis is perfectly applicable to them. But the language of politics or econom-
7 The logical structure of ordinary language must be distinguished from the grammatical structure, e.g., of sentences the particular language, elaborated by grammarians. During the time I myself taught English grammar, I found that sentence-analysis in English grammar textbooks is not always in line with the logical analysis. In other words, one cannot teach English grammar using logical analysis of sentences. Whether that is also true of other languages I do not know.
ics or law is by no means the whole of political science, of the science of economics, and so on. In addition, there exist the political, economic, or legal practices of individuals and groups, that form part of the subject-matter of economic, political or legal philosophy. In the case of religion, too, we have the various practices in Wittgenstein's sense that form part of the subject-matter of the philosophy of religion.
Finally, the analysis of language, both ordinary and technical, forms without doubt an important part of any thorough attempt to understand the activities we call learning and teaching: of linguistic communication in general. But the kind of analysis relevant to them is mainly if not wholly scientific analysis: psychological, sociological, philological, and the like. In addition, it seems to me that the analysis of language, whether scientific or philosophical, does not form part of the Philosophy of Education. Semantic analysis of applicable, however, to the language Education as a discipline itself employs. But since that language is becoming increasingly technical, its analysis falls outside the scope of ordinary language analysis.
Semantic, conceptual analysis is applicable to the language of mathematics. But that language is almost wholly technical, and so lies almost completely outside the scope of ordinary language.
REFERENCES
Hahn, L. E. (1958, March). What is the Starting Point of Metaphysics? Philosophy and Phenomenological Research, XVIII(3), 293-311.
Khatchadourian, H. (1955, December). Proteins
and Probability: A Criticism of M. Pierre Lecomte Du nouy's Argument for Teleology Based on Some Probability Estimates. Philosophy and Phe-nomenological Research, XVI(2), 223228.
Khatchadourian, H. (1967). A Critical Study in Method. The Hague: Martinus Nijhoff.
Khatchadourian, H. (2016). The New Design Argument and God. WISDOM, 6(1), 46-66.
https://doi.org/10.24234/wisdom.v1i6.63 White, A. R. (1955, December). A Linguistic Approach to Berkeley's Philosophy.
Philosophy and Phenomenological Research, XVI(2), 172-187.
