Bertrand Russell's Construction of the External World

Charles A. Fritz, Jr.,

• Publisher: Routledge
• Year: 2014
• eBook ISBN:
  9781317830986
• Topic: Philosophy
• Subtopic: Philosophy History & Theory

First published in 2000. This is Volume III of six in the International Library of Philosophy looking at the area of Nineteenth and Twentieth Century Anglo-American Philosophy. Written in 1952, it focuses on Bertrand Russell's Construction of the External World, which covers a wide variety of topics, attempts to answer many of the problems traditionally associated with philosophy.

CHAPTER III

The Problem of the External World

THE problems I considered in the preceding Chapter were specialized logical or mathematical problems for whose solution Russell developed a technique which he was to find of great usefulness in attacking more extensive problems. In this Chapter, I shall discuss several general philosophical problems with which Russell is greatly concerned in his later writings. Critically examining common-sense beliefs will reveal philosophical problems, Russell believes, and lead us to determine the extent to which such beliefs are sound, and what elements are really involved in them. The primary problem which he first found in this way is the common-sense belief in permanent material objects, or ‘matter’, and it is the inquiry into the possible justification of this belief that occupies an important place in his philosophical interests. In The Problems of Philosophy (1912) we find a forceful and oversimplified expression of this approach to philosophy, and the problem of material objects which he determines through that approach. He maintains it is necessary to investigate ordinary beliefs to find the extent to which they are sound, which of them, if any, are ‘certain’, which are only instinctive beliefs, or worse, prejudices. Russell compares himself to Descartes in wishing to find which of our knowledge is ‘certain’, which ‘doubtful’. In addition, it is also valuable to examine the soundness of scientific knowledge, and the basis upon which that knowledge rests. In his later writings, especially Our Knowledge of the External World (1914), The Analysis of Matter (1927), Philosophy (1927), An Inquiry into Meaning and Truth (1940), and Human Knowledge (1948), his first rather crude expression of the problem becomes more refined and transformed into one version of the problem which concerns so many philosophers today, the foundation and basis of empirical knowledge, and the relation of perception to that knowledge.

In Part I, after first briefly presenting Russell’s views in The Problems of Philosophy, I shall summarize the criticisms of common sense and scientific knowledge and the problems to which that criticism gives rise as developed principally in Our Knowledge of the External World, several other papers written about the time of Our Knowledge of the External World,¹ The Analysis of Matter, An Inquiry into Meaning and Truth, and Human Knowledge. Russell also formulated a philosophy of logic and an analysis of propositions which he called the philosophy of ‘logical-atomism’ that led him by a somewhat different route to the same problems. I shall present these views in Part II. It is a logical doctrine which can be found in writings around 1918 and for a period thereafter, most fully expressed in the lectures, ‘Philosophy of Logical Atomism’ (1918–19). Finally, in Part III, I shall discuss the need for an ‘interpretation’ of science if we are to examine the question of the validity of scientific knowledge. Russell’s views on this point are found chiefly in The Analysis of Matter and Human Knowledge.

PART I

HARD VS. SOFT DATA

It is significant that the first sentence of The Problems of Philosophy is ‘Is there any knowledge in the world which is so certain that no reasonable man could doubt it?’ (9). It is in this spirit that Russell approaches philosophy in that book, trying to find the certain elements in our common-sense knowledge, and the extent to which it is reasonable to continue to believe the remaining elements. The common-sense belief that Russell is most interested in is that in the existence of permanent material objects, or matter. Russell employs familiar epistemological arguments in an attempt to show that we do not directly perceive such objects, but must obtain our knowledge of them in some way from our own sensations, or sense-data. The existence of sense-data cannot be doubted, but that of material objects can.

Before we embark upon doubtful matters, let us try to find some more or less fixed point from which to start. Although we are doubting the physical existence of the table, we are not doubting the existence of the sense-data which made us think there was a table; we are not doubting that, while we look, a certain colour and shape appear to us, and while we press, a certain sensation of hardness is experienced by us…. In fact, whatever else may be doubtful, some at least of our immediate experiences seem absolutely certain.¹

Starting from this ‘solid basis’ do we have ‘any reason for regarding [our sense-data] as signs of the existence of something else, which we can call the physical object’?²

Some of our ‘immediate experiences’ Russell has called ‘absolutely certain’. However, the sense-data themselves are not certain in the sense that propositions may be certain, or absolutely true, but rather just are. ‘A particular patch of colour which I see, for example, simply exists: it is not the sort of thing that is true or false’ (178). Judgments about these sense-data, however, may be ‘self-evident’ in the more usual sense of the word. These ‘self-evident truths’ are of two kinds: ‘First, there is the kind which simply asserts the existence of the sense-datum, without in any way analysing it…. The other kind arises when the object of sense is complex, and we subject it to some degree of analysis. If, for instance, we see a round patch of red, we may judge “that patch of red is round”’ (178–9). Russell distinguishes between degrees of self-evidence, the highest degree of self-evidence gives an ‘absolute guarantee of truth’. Truths about sense-data possess this highest degree because we can have ‘acquaintance’ with the fact which corresponds to the particular truth, namely the particular sense-datum concerned (212). Starting from our own sensations, either sense-data or propositions about them, the problem is to determine the grounds for the belief in the existence of permanent material objects.

The solution of this problem which Russell maintained in The Problems of Philosophy is a form of the causal theory of perception. Material objects are the causes of our sense-data, are the reason why there is agreement among observers as to what they see, are the reason for whatever stability and orderliness exists among our sensations. In fact, it is the ease with which the belief in material objects explains this orderliness and stability that is one of the chief reasons he finds in favour of the causal theory. But Russell is looking for more certain grounds than usefulness and plausibility. However, the best Russell can do toward establishing this belief on other grounds is to find ‘this belief ready in ourselves as soon as we begin to reflect: it is what may be called an instinctive belief’ (37). It does not seem to be a self-evident belief like one about sense-data, but we are justified in accepting this belief, Russell feels, since it is a belief we hold strongly, and ‘there can never be any reason for rejecting one instinctive belief except that it clashes with others; thus, if they are found to harmonize, the whole system becomes worthy of acceptance’ (39).

Although we are justified in believing that material objects exist, we can infer little as to their nature. We can assume that if there is a difference between sense-data, there must be some corresponding difference between material objects. If two sense-data are related to each other as right to left, then we can suppose that the corresponding objects are related in a similar relation. But the intrinsic nature of these objects seems to be something we cannot know, at least by the senses. A red sense-datum has the quality red to our senses, but it is unlikely that the object itself is red in the same sense, since physics tells us that the red sense-datum is the result of a complicated process involving among other things light and a sense-organ (54, 59). The discussion on these points is brief in The Problems of Philosophy, hardly more extensive than my presentation of them. It would be unfair to criticize this book too harshly, since it was not intended to be a detailed work. Russell wrote most of his later works about points only briefly treated in The Problems of Philosophy, in many cases modifying the views which he expressed there. In the remainder of this Part I shall discuss his more comprehensive treatment, in works following The Problems of Philosophy, of the distinction between basic and inferred knowledge, and his belief in the desirability of grounding our knowledge on basic knowledge. As in The Problems of Philosophy, it is largely the problem of the existence of material objects to which these distinctions are applied. His solution to the problem I will leave to the following Chapter.

In Our Knowledge of the External World, the common-sense knowledge that forms the material for philosophic inquiry Russell divides into several kinds, knowledge that we have obtained from our own experience of objects and events—such as towns, people, and their actions—knowledge that we have by testimony of objects and events that have not been experienced by us, and finally scientific knowledge which systematizes our more particular knowledge by means of inclusive generalizations. This basic material for examination forms the ‘data’ for philosophic problems, as the ‘data’ for a science, like psychology, would consist in human actions and behaviour.¹ It should be noted that Russell uses ‘data’ in different senses in his writings. One use is that which I have just mentioned, the material from which any inquiry starts. Epistemological data are further distinguished into ‘hard’ and ‘soft’ data, and ‘data’ then comes to mean simply ‘hard data’. Russell makes this distinction explicit in the Monist for 1914, where he says of data in the second sense: ‘… what we have hitherto called data will be more fitly called premises.’¹ In Human Knowledge he says, ‘There is thus a distinction between beliefs that arise spontaneously and beliefs for which no further reason can be given. It is the latter class of beliefs that are of most importance for theory of knowledge, since they are the indispensable minimum of premises for our knowledge of matters of fact. Such beliefs I shall call “data”’.² Normally when Russell speaks simply of ‘data’ he means data in this latter sense. I shall follow this same practice in my discussion.

Our original data is not all of the same degree of certainty; it can be classified in respect to its relative certainty or uncertainty, and the degrees of certainty of the different facts we know are themselves part of the data of common knowledge. Not only do we begin with various facts of perception, or of history, or of science, but also with the knowledge, though not necessarily very precise, of the degree of confidence we can have of the truth of that knowledge. That data which is certain Russell calls ‘hard data’, that which after examination is found to be not certain, is ‘soft data’.³

I give the name ‘data’ or rather ‘hard data’ to all that survives the most severe critical scrutiny of which I am capable, excluding what, after the scrutiny, is only arrived at by argument and inference.⁴

The part of our knowledge which can be called ‘hard data’ is further characterized as that data which is certain, self-evident, where ‘self-evident’ means believed on its own account,⁵ or ‘known otherwise than by inference’.¹ Knowledge which is not found to be certain is ‘soft data’, even though at first sight it may seem as certain as hard data, upon examination such knowledge is found to be really ‘inferred’ or ‘derivative’ knowledge. Soft data are not certain, Russell believes, because they are based upon some other knowledge, and the inference from the prior knowledge to them might be erroneous.

The distinction between hard data and soft data corresponds in Our Knowledge of the External World to that between ‘primitive’ and ‘derivative’ knowledge.² Hard data, or primitive beliefs, are believed on their own account, while derivative, or ‘inferred’, beliefs are those ‘we only believe because of something else from which [they have] been inferred in some sense, though not necessarily in a strict logical sense’.³ Most common-sense knowledge, it turns out, is really derivative knowledge, thus the real shape of a perceived object is inferred from our particular perceptions; our knowledge of a person’s feelings is an inference from his observed actions and our experience of him and human nature; and, of course, we have knowledge which we have obtained by conscious inference from previously substantiated beliefs. For the present, w...

Table of contents

• Cover
• Title Page
• Copyright Page
• Table of Contents
• ACKNOWLEDGEMENTS
• I. THE DEVELOPMENT OF BERTRAND RUSSELL’S PHILOSOPHY
• II. THE CONSTRUCTION OF NUMBERS, DESCRIPTIONS, AND CLASSES
• III. THE PROBLEM OF THE EXTERNAL WORLD
• IV. THE CONSTRUCTION OF MATERIAL OBJECTS AND THE ENTITIES OF PHYSICS
• V. CONSTRUCTIONS AS A METHODOLOGICAL PRINCIPLE
• BIBLIOGRAPHY
• INDEX OF SUBJECTS
• INDEX OF NAMES
• INDEX OF REFERENCES