John Anderson and Twentieth-Century Philosophy

The University of Glasgow, when John Anderson entered it in 1911, was still an outpost of Absolute Idealism, as represented especially by Sir Henry Jones, friend and pupil of Edward Caird. Anderson's original inclinations were towards physics and mathematics rather than philosophy. (He graduated with first-class honours in the Mathematics and Physics School and in the Philosophy School in 1917.) He was unlikely, then, to be wholly satisfied by Jones's measured rhetoric; and his family tradition of radical politics—his father was a Socialist headmaster—set him in opposition to the solidarist tendencies of Absolute Idealism. Nor was the logic of Jones's colleague, Latta, Idealist in its general assumptions, even if Latta himself was a little more amiably disposed than most Idealists towards traditional formal logic, sufficiently rigorous or sufficiently consistent to satisfy Anderson's demands. Yet, unlike most rebels against Absolute Idealism, Anderson was not prepared to reject its teachings as wholly worthless; he took seriously the Idealist criticisms of Mill; he was not going to swing back in simple reaction from Absolute Idealism to “impressions and ideas”, however ingeniously they might be deployed in their new guise as “sense-data”.

In the theory of ideas Anderson detected not the true contradictory of Absolute Idealism, but rather a variant development of the same metaphysical impulse: the search for something ultimate, in contrast with which the complex objects of everyday experience are arbitrary constructions, mind-made. The vital issue for Anderson is whether facts are constructions; the question which divided Absolute Idealist and traditional empiricist—whether facts are hacked out of the Absolute or built up out of sensations—was for him of slight importance. Against Absolute Idealist and traditional empiricist alike, Anderson set out to show that there is no reality (whether “higher” or “lower”) other than the complex, and complexly interacting, objects of everyday experience. For Anderson as for “the vulgar”, when I assert that the book is on the table I am neither imposing abstractions upon Reality nor constructing objects and their interconnections out of simple experiences; I am taking something to be the case, really to be the case. And if there is any doubt whether it is “really the case” that the book is on the table, this is because the book might in fact be on the floor or in the book-case or because what is on the table might not be a book but a cigar-box, not because the book is neither the Absolute nor an elementary perception.

  ― x ―
Anderson is not being merely arbitrary when he describes himself as an “empiricist”, for all that he has broken so sharply with traditional empiricism. Like, say, Hume and Mill, he sees in experience the only guide to what is the case; like them, he rejects the transcendental and those special modes of cognition which are supposed to lead us to the awareness of the transcendental. But their account of experience, he considers, is metaphysical, “rationalistic”. Impressions, ideas, sense-data, function in traditional empiricism as ultimate foundations for, justifications of, our everyday beliefs; whereas for Anderson a belief can be justified only by another belief, a statement that something is the case only by a statement that something else is the case. Justification, explanation, proof, is never “ultimate” in the sense of resting on entities or principles which are transparent, whether they be Cartesian “simple natures”, Lockeian simple ideas, Platonic forms, or the axioms of classical rationalism. To believe, to experience, to know, to assert, so Anderson argues, is to take something to be the case and in so doing to run the risk of being mistaken—a risk that no degree of care can wholly rule out. There is no experience prior to, as there is no form of knowledge higher than, taking something to be the case.

In his criticism of traditional empiricism, Anderson shows the influence of William James, whom he closely studied. James rejected the view (which, once more, Absolute Idealists and traditional empiricists had shared) that relations are “the work of the mind”, imposed upon an original experience which is itself wholly disconnected. As Hume had expressed the matter: “All our distinct perceptions are distinct existences and the mind never perceives any real connexion among distinct existences”. James condemned this account of experience as “vicious intellectualism”; it rests on the supposition, he argued, that if A and B are connected, it is impossible for them also to be distinct, and if distinct, impossible for them to be connected. In fact, James insisted, things are, and are experienced as being, both connected and distinct; and there is no good reason for supposing either that their connectedness is unreal and their distinctness real, or that their distinctness is unreal and their connectedness real. This is essential to Anderson's position. For it is impossible to identify, as he does, “experiencing” and “taking something to be the case”, if experience is always of the unrelated. To take something to be the case is to point both to connections and distinctions. “The book is on the table” presupposes the possibility of distinguishing the book and the table from what surrounds them, yet it at the same time links them; unless there are both connections and distinctions in experience no proposition can be a simple statement of what we experience.

Anderson's position on this point may at times remind the reader of Wittgenstein's Tractatus Logico-Philosophicus, for which “the world is the totality of facts, not of things”. But the parallel extends only thus far: for Anderson as for Wittgenstein the starting point is facts, not elementary entities which are later put together into facts. Anderson, in contrast with the Tractatus (a book of which he was not aware when he was working

  ― xi ―
out his own position), rejects both the conception of “the world”—understood as a totality of facts—and the conception of an atomic fact. Furthermore, there is still talk in the Tractatus of elementary objects, even if they are supposed only to exist in facts. For Anderson, on the contrary, every fact (which includes every “object”) is a complex situation: there are no simples, no atomic facts, no totalities, no objects which cannot be, as it were, expanded into facts. On Anderson's view the statement that “X exists” can always be expressed in the form “some Y are Z”, whatever X may be. That is why he describes himself as a “pluralist”.

This again can be misleading; traditionally, pluralism has been the theory that there is a plurality of ultimate entities. Anderson's pluralism, on the contrary, is thorough-going: he wholly rejects the view that complexes are built out of simple entities. “Even if the world is infinitely complex”, Wittgenstein wrote, “so that every fact consists of an infinite number of atomic facts and every atomic fact is composed of an infinite number of objects, even then there must be simples and atomic facts.” When Anderson, on the contrary, asserts infinite complexity he is denying that there are simples; and far from its being the case that there must be simples, there cannot, on his view, be simples. The least one can encounter, in his own phrase, is “a thing of a certain sort”, something of a certain kind (indeed, of an endless variety of kinds) happening somewhere. “There are only facts”, as he writes in “Empiricism”, “i.e. occurrences in space and time”; within such a fact we can always distinguish what is happening from where it is happening.

Why cannot there be simples? Because, Anderson would say, the supposition that there are simples is unspeakable—as Plato pointed out in the Sophist (252c), the theory of simples is self-refuting. Or, approaching the matter in another way, because the whole point of supposing that there are simples is that the simples are the building blocks out of which complexes are constructed whereas, in fact, complexes cannot be built up out of purely simple ingredients. A table is not brownness, and darkness, and hardness, and smoothness, in the sense of being a logical conjunction of these properties. In general, as soon as we try to say anything whatever about alleged ultimates—and something has to be said about them—they turn out to be either entirely empty of content, and hence useless as “foundations”, or else to be no more ultimate than what they are supposed to support. So far as it rejects atomism Anderson would be in sympathy with Wittgenstein's Philosophical Investigations, but Anderson's spare and dialectical manner of procedure is at the opposite extreme from Wittgenstein's diary-like ruminations. Indeed towards “ordinary language” philosophy of every sort Anderson's hostility has been unmitigated. He has refused in any way to take it seriously; its rise to prominence undoubtedly did a great deal to make him feel disheartened about the prospects for philosophy.

In his criticisms of traditional empiricism, Anderson looked for support to certain of the “new realists”, both in Great Britain and the United States. However, many of the self-styled, or by-others-styled, realists were

  ― xii ―
actually phenomenalists. They attempted, as Russell did, to develop more systematically the traditional approach of British empiricism, with Mill and Mach as the immediate starting-point; with such “realists” Anderson certainly did not wish to associate himself. In his eyes, indeed, epistemological questions are always secondary to logical (or, some would say, ontological) questions. A realist theory of knowledge, he thinks, is a particular application of a realist theory of relations, flowing out of the rejection of “constitutive” relations and taking the form of an attack on what Anderson calls “relativism”. Opposing the phenomenalist doctrine that the esse of what we perceive consists in its being perceived, he argues that what we perceive must have characteristics of its own. Even the assertion: “I am perceiving a red sense-datum”, makes a specific claim: the claim to be perceiving a certain sort of object; did “the whole essence” of that object consist in its being perceived, then I should no longer be claiming to perceive it rather than something else, say, a blue sense-datum. Indeed, I should not be claiming to perceive anything at all. I should be entitled to say only: “I am perceiving”, for to claim to perceive x would at once distinguish x from the fact that it is perceived, thus denying that the whole essence of x consists in its being perceived. The phenomenalist, that is, gives no account of what is perceived, for it is not such an account to assert that I perceive “perceptions”.

To the question: “What do I perceive?” then, the answer: “A perception” will not suffice; and the same is true of such answers as “awareness”, or “consciousness” to the question: “What perceives?” Just as what is perceived cannot be constituted by its being perceived, so what perceives cannot be constituted by its perceiving; both perceived and perceiver, according to Anderson, must be complex states of affairs—variously describable occurrences. When the perceiver perceives the perceived this will simply be another complex state of affairs. “Being a book” is being an occurrence of a certain sort; so is “being a book on the table”, and so is “someone's perceiving a book on the table”.

Of course this, as a purely logical point, still leaves completely open the question what sort of complex thing is perceived and what sort of complex thing perceives. It is still possible in principle that I never perceive anything but complex states of my own mind; but that will not at all be a plausible view, Anderson thinks, once logical confusions are cleared out of the way. Similarly, logic cannot by itself determine what are the distinctive features of mind. It can however, rule out the supposition that mind is wholly constituted by its apprehension of objects, or that it is a mere “consciousness”.

Equally, Anderson thought, logic can rule out the supposition that mind has a mode of existence that is somehow “higher” than that of other things; minds, too, must be complex occurrences in space and time. At this point the decisive influence was Alexander's Space, Time and Deity, which Anderson heard as Gifford lectures in the University of Glasgow (1916-18). Alexander profoundly stirred Anderson's philosophical

  ― xiii ―
imagination; those who heard his lectures on Alexander felt that they were being led into the very heart of Anderson's philosophy.

Of course, as “The Non-Existence of Consciousness” makes clear, Anderson came to be anything but a whole-hearted disciple of Alexander. The crucial thing Anderson derived from Alexander was the doctrine of a single spatio-temporal medium within which everything had a place and a time, whatever its specific characteristics. However, Alexander's Space-Time is both a medium and a stuff; for Anderson, it is a medium only. Alexander, too, had taken over from Lloyd Morgan a theory of emergence (involving a distinction between degrees or levels of complexity) which Anderson wholly rejected. But for Anderson, as for Alexander, there are no “special”, no “privileged”, entities. All distinctions are distinctions in characteristics, not in status. There is no special realm of “mental entities”; in remembering, in imagining, in expecting, in desiring, we are concerned all the time with independent states of affairs. Any other view, Anderson thought, built impassable boundaries between the mental and the non-mental—barriers which yet, it had to be admitted, constantly were passed through in our everyday dealings with things. His objection to Alexander was that in his theory, too, not all the barriers are down; if, as Alexander had argued, mental processes are “enjoyed” and non-mental processes “contemplated”, then it is impossible in principle, Anderson thought, to give any coherent account of our observation of the connection between the two. For that connection itself, by the nature of the case, can neither be contemplated nor enjoyed.

Alexander taught Anderson, however, that knowledge is a way of striving with things rather than a simple reflection of them. Quite as vigorously as Collingwood, Anderson rejected what Collingwood called the “transparency” theory of knowledge and ascribed to his “realist” contemporaries. Knowledge—using this word in its broadest sense to include every case where a proposition is taken to be true—is never, according to Anderson, the bare reception of a given object by an act of awareness. Rather, it is an attempt to come to terms with ourselves (in self-knowledge) or the things around us. The transparency theory, he thought, could give no account of belief, whether true or false, and no account, either, of the selectivity of knowledge, or of error. In any adequate theory of knowledge the knowing mind must be regarded as a complex entity with its own demands, which are partly satisfied by, partly encounter obstacles in, the complex behaviour of other things, including other people and other tendencies within the same mind.

Unlike Alexander, however, Anderson does not adopt a merely conationalist theory of mind; for Anderson the mind is a complex of feelings or passions, these being the things that strive. Such an approach to mind is by no means unique. It is more than suggested by a number of eighteenth-century moralists; it is explicit in Nietzsche when he speaks of mind as “the social structure of instincts and passions”; Anderson was led to it by his reading of McDougall and Freud. But most contemporary philosophers will no doubt regard it as one of the oddest features of

  ― xiv ―
Anderson's philosophy; the general philosophical tendency has been to regard passion as lying quite outside the mind with which the epistemologist concerns himself—as something which perhaps ought to be discussed, as Aristotle discusses it, in connection with ethics but ought to be strictly excluded, as merely psychological, from epistemology. (Anderson's Cook-Wilsonian contemporaries were particularly insistent upon this apartheid policy.) If, like Hume, philosophers have argued for the complexity of mind, they have taken it to be a complexity of “perceptions”, not of passions. But certainly that will not be a satisfactory view once it is recognised that what Hume refers to as “perceptions” are simply what we perceive, and that what we perceive is not, ordinarily, part of our mind at all. The question then has to be raised afresh: “Of what is the mind a complex?”

Anderson would not be content with the answer suggested by Ryle in The Concept of Mind, that, to put it roughly, a person's mind is wholly describable in terms of his actions and dispositions to act. When Anderson speaks of a “motive”, he thinks of it as actually being a moving force; the familiar expression “he is moved by curiosity” is to be taken quite literally: curiosity, understood as a complex mental structure distinguishable from, but continuous with, other mental structures, can be the actual agent in our attempts to understand the things around us. Curiosity is not to be thought of as operating upon “the mind” and inducing it to act in a particular way; for “the mind” is nothing but a complex of such complex structures. Thus for Anderson, it is curiosity itself which moves us, and psychology will describe the interplay of such mental structures just as sociology describes social structures. It is to such mental complexes, too, that ethical predicates apply: our motives (not in the sense of our objectives but of our moving forces) are good or evil.

Anderson describes these motives as “tensions”; they are, of course, physical as well as mental structures. Such ways of talking Ryle would no doubt condemn as “para-mechanical”. This accusation would not disturb Anderson, but would rather help to confirm him in his belief that he is thinking along sound lines. Anderson's approach is essentially that of a generaliser; he expects to find that methods of approach, principles of explanation, modes of description, which have proved fruitful in one field of inquiry will also prove fruitful in others. Of course, he would freely admit that they might not, in a given case, turn out to be appropriate. What he is looking for, all the same, are modes of description which will be as widely applicable as possible—in opposition to the differentiating approach of so much contemporary philosophy, which seeks to make distinctions rather than to establish connections. Thus if, as McDougall maintains, the influence of mind on mind turns out to be parallel to electrical induction—or, one should rather say, to exemplify the same sort of general relationship as is also to be found in electrical theory—that is exactly what Anderson would expect to be the case.

Certainly, Anderson would wish to reject any account of mind which was wholly dispositional. To assert of any person that, say, he “has a

  ― xv ―
happy disposition”, is, on this view, equivalent to asserting that he displays certain characteristics under unspecified, but specifiable, circumstances. Disposition-statements are reducible to categorical statements about the actual behaviour of complex entities. This is connected with Anderson's general logical doctrine that hypothetical assertions are simply a rhetorical variant of categorical assertions. “If…, then…” is used, he would admit, in a variety of ways, but always as a way of asserting that some actual occurrence takes place, or some actual relation holds. Thus, for example, “if he were to come, I should be astonished” asserts an actual connection between the sort of circumstances under which he would come and my being astonished; it might be expressed as: “The arrival of people of his sort at ceremonial occasions always astonishes me”.

In general, Anderson's logic, like his psychology, will strike strangely on the ears of his contemporaries. With his mathematical bent, Anderson might have been expected to pick up and develop the Russellian mathematical logic; alternatively, he might have been expected to associate with his pragmatic theory of belief (“We believe what satisfies us”) the characteristic pragmatic criticism of formal logic. He did neither of these things. Of course, he was aware of, and was affected by, Russell's work, taking over, for example, Russell's conception of a “propositional function”. But it was the traditional formal logic which he chose to expound and to develop, defending it against its critics, whether they were Russellians, pragmatists or idealists.

To put the matter thus, however, is certainly to underestimate Anderson's contributions to logic. For, if his logic is traditional in its allegiance to the “four forms” and its emphasis upon syllogism, no topic in the traditional logic comes out of his hands quite as it entered them. His logic is philosophical, thought through consistently, as the traditional logic of the text-books is not. For example, he brings into the open the existential presuppositions of traditional logic; he rejects both the class interpretation and the substance-attribute interpretation of the proposition; he denies that the validity of a syllogism depends either upon a dictum or on rules about distribution; he sees that to assert that a term is distributed is identical with, not a necessary preliminary to, the recognition that certain forms of syllogism are valid and others fallacious; he develops and generalises the traditional theory of opposition; he describes and systematises a number of forms of non-syllogistic inferences; he considers at length, if he finally rejects, the claims of relational, hypothetical and disjunctive reasoning to be regarded as wholly distinct from syllogism. If his logic is traditional, the tradition is worn with a difference.

What he would with special vehemence oppose is the doctrine, now almost universal, that logic is a calculus. He would not be prepared to grant, for example, that his own logic is simply an alternative version of the traditional predicate calculus. Logic, as Anderson sees it, describes the general structure of facts, including the relationships between facts. What are sometimes regarded as the supreme examples of logical truths, e.g. the principle of identity, are not, on his view, truths at all; they say

  ― xvi ―
nothing. But he does not agree with Wittgenstein that all the propositions of logic are tautologies. This is not true, he would say, of a proposition which asserts the validity of a type of syllogism or even the equivalence of “No X are Y” and “No Y are X”. Such propositions tell us something; we have to learn that “No X are Y” and “No Y are X”, unlike “Some X are not Y” and “Some Y are not X”, are equivalent.

Logic, as Anderson conceives it, incorporates what Mr Strawson has recently described as “descriptive metaphysics”. It is neither about forms of language, nor about reasoning processes, nor about special-status entities, e.g. universals; it is about the most general features of facts. There are, indeed, no “universals”, in the sense that there are no entities which are simply properties. But equally there are no “particulars”, in the sense in which logical atomism presumes the existence of particulars. Every proposition, for Anderson, is about things of a certain description and offers a further description of them. Both subject and predicate, indeed, are things of a certain description; but the primary function of the subject, Anderson argues, is to “locate” i.e. to be the centre of reference which a predicate describes. The function of the predicate is, on the other hand, to describe. There are not two classes of entities: pure locations (“particulars” or “substances”) and pure descriptions (“universals”). Any entity is both specific and general. It occurs somewhere and somewhen but it also behaves in a regular way and stands in certain relations to other entities; it is of a variety of descriptions. To talk of an “entity”, a “fact”, an “occurrence”, or to describe a proposition as “true”, is in each case to say that something happens in a certain place or in certain places.

The logician, it will follow, is already discussing universals, individuals, identity, space-time, causality, when he describes the function of subject and predicate in propositions and the logical relations between propositions. Thus, so Anderson argues in “The Problem of Causality”, to discuss causality is not to describe a special sort of metaphysical entity—a “necessary connection”—but is rather to draw attention to the form of the universal propositions asserted by “A is the cause of B”. The logician is always concerned with “what is”. He does not deduce an ontology from logic or a logic from ontology; rather, in discussing logic, he is already discussing ontology.

Anderson also rejects the view, now almost universally held, that a mathematical “truth” is “true” only within a calculus, its “truth” consisting in its derivability from a given set of postulates. Mathematical truths, too, are concerned with the general types of relationship holding between facts; and what we take to be a mathematical truth may turn out to be false. Quite generally, on his view, there are no “analytic truths”; if a proposition says anything at all it can be false. And he is not to be persuaded that mathematical propositions “all say the same thing i.e. nothing at all”, as Wittgenstein had suggested.

It does not follow, however, that mathematical propositions are “empirical generalisations”. Anderson totally rejects Mill's view that universal

  ― xvii ―
propositions are generalisations, and the “inductive logic” that goes with it. According to the traditional empiricist doctrine, an immediate experience is always particular, in a sense of “particularity” which prevents it from also being universal; a universal proposition is derived by generalisation from such facts. But this way of regarding universal propositions (as derived by some special sort of inference from experience) depends upon a conception of immediate experience which Anderson completely rejects on the ground that, so conceived, it would allow no such inference. His criticisms of induction are in some respects very like although quite independent of Karl Popper's; but Popper has not wholly broken with the traditional doctrine of immediate experience. When somebody says: “I know by immediate experience that the book is on the table”, this is just a way of claiming, according to Anderson, that I found out that the book is on the table by being there with it and looking for myself, as distinct from hearing somebody say that the book was on the table, or deducing that it would be on the table from our knowledge of the habits of the person reading it. It is not a way of saying that I first had “an immediate experience” and then somehow (but how?) inferred from this experience that there is a book on the table. And I can know by experience, Anderson would add, that fire burns, that when two lines meet they make four angles, or that a Boeing 707 has four engines. If, indeed, I could not know this sort of thing by immediate experience, then equally I could not know that a particular thing is a fire, or that there are two lines on the piece of paper, or that what confronts me is a Boeing 707.

In a way, Russell realised this; he saw that Mill's “particular propositions” contained general descriptions and that if immediate experience must be of “pure particulars”, then we cannot properly be said to know by immediate experience that, say, “Socrates is dead”. But whereas Russell went off in search of what could be known by a completely particularised immediate experience (in formal terms, he looked for a “logically proper” name) and ended in bankruptcy, Anderson denied there can be immediate experience as Russell defined it. There are, on his view, no merely “given” particulars and no “logically proper” names—names, that is, which refer to a thing without being potentially descriptive.

Thus, on Anderson's account of universal propositions, the truth of mathematical propositions can be a “matter of experience” without the conclusion following that their truth depends on the validity of some sort of inductive inference; they, like all other universal propositions, can be derived from other universal propositions, can be tested in experience, can be simply taken to be the case. Of course they might, when taken to be the case, turn out to be false; but so might any finding of ours. I may wrongly take it to be the case that all fires burn, but so, too, I may wrongly take it to be the case that this is a fire, that it burns, that my hand is hurt—or whatever else I can take to be the case.

What, on Anderson's view, does a universal proposition assert? It

  ― xviii ―
certainly does not say that there is “a general connection” between being A and being B if by there being such a general connection is meant anything more than the simple fact that all A are B. Nor does it assert that some sort of class relation holds between A and B. For one thing “All A are B” does not unambiguously point to any such relationship between classes—it could be the case either that B includes A or that A and B are coextensive; for another thing, to assert a class relationship is just a way of saying that certain propositions are true. Classes are not entities. Nor, as Bradley and Russell thought, does “All A are B” assert only a hypothetical connection between descriptions. For although “All A's are B” does not assert that “there are A's”, it certainly presupposes that there are, and if somebody convinces me that there are no A's I shall not continue to assert that “All A are B”.

In a way, Anderson could only reply to those who offer such interpretations of the universal proposition that “All A are B” says what it says; it is just because he thinks that alternative modes of formulation fail to bring out the full force of that proposition, and no more than its full force, that Anderson prefers this—the traditional—way of asserting a universal proposition to any of the alternatives that have been suggested. But “All A are B” is not, of course, about some totality describable as “All A”; it asserts of each and every A, that it is a B. This comes out, Anderson considers, in the fact that we recognise as valid the argument “All A are B, this is an A, therefore this is a B”.

Indeed, the validity of such a syllogism is, in a sense, Anderson's starting-point in logic. In such a syllogism, we see exhibited, he thinks, such basic logical facts as that what is a subject in one proposition can be a predicate in another—in opposition to the view that subjects and predicates are names for different classes of entity; that a singular and a universal proposition are of the same general structure, since the logical form of a syllogism will not be disrupted by the replacement of one by the other; that such expressions as “Jones is a man” and “Jones is human” have the same logical form (here once again Anderson is in opposition to most of his contemporaries), since either can appear indifferently as minor premise in an argument in which the major premise refers either to “men” or to “human beings”.

“Being a man”, “being human”, “being a human being”, “being a member of the class ‘man’”, “having the attribute of being human”, “being truly describable as human”, if Anderson is right, are different ways of offering precisely the same description of something. One mode of speech is, no doubt, more natural than another in a given context; a noun is most often the subject of a sentence, an adjective its predicate. But the logical subject is what we are talking about and the logical predicate, i.e. the description we are offering of it, will be the same whatever of these forms of words we select. (The choice is determined by rhetorical, not logical, considerations.)

Thus, once more, Anderson seeks to generalise, in an age of differentiation. The distinction between singular and universal propositions; between

  ― xix ―
categorical, hypothetical and disjunctive; between the actual, the possible and the necessary; between class-membership and predicative propositions—all of these turn out, if he is right, not to be differences in logical form but only in modes of expression. The logical form of any statement will reveal it to be asserting some matter of fact (whether truly or falsely), to be offering a description, that is, of a certain kind of thing.

What about ethical and aesthetic statements? Here again Anderson begins with an attack on “relativism”. Traditional ethics abounds in such conceptions as that whose nature it is to be an end (the “intrinsically desirable”). All such conceptions Anderson at once rejects, on the ground that nothing can have its nature in being an end or in being binding. We do, certainly, have desires, we do, certainly, make demands upon ourselves and upon other people, but we desire something, or we demand of ourselves that we act in a certain way, because the objective or the act has such-and-such characteristics, not because it is its nature to be sought after or to be incumbent upon us.

For Anderson the most important of ethical writings since the Greeks is Moore's Principia Ethica. Of course, Anderson wholly rejects the view that good is a non-natural, simple, quality, that “good” is different from “the good” or that there is a “naturalistic fallacy”—except in so far as Moore was attacking relativism under that name. He admits, too, that Moore had by no means freed himself from the presumption that good is an end, however inconsistent this might be with his more general doctrine that to call something “good” is just to characterise it. All the same, Anderson saw in Moore's work a fundamental contribution to what he calls “a positive ethics”.

Ethics, of course, must for Anderson be concerned with facts; there are no “values” above facts. To call a mental activity “good” (it will be remembered that, on his view, it is to such activities, to certain “spirits” like the spirit of inquiry, that ethical predicates apply) is to describe it in a way which it is one task of ethics further to elucidate. Not, however, its only task. Ethics is not simply “the analysis of ethical predicates”; it is an inquiry into ethical facts, which may lead us to the conclusion that many conventional ethical judgments are false. Anderson himself tries to show that whatever is good is a form of enterprise, that it is productive, capable of developing in a special way, by means of what he calls “communication”. In saying, for example, that human affection is good, Anderson is not asserting that it always ought to be pursued, or that it is obligatory to bring it about; good activities, he thought, are not even the sort of things that we can be obliged or (sensibly) exhorted to pursue. Far from its being the case that good activities are by nature ends, they are, unlike economic goods, the sort of thing that is not ordinarily achieved by being pursued; one does not come to love, or to be courageous, or to develop a spirit of inquiry, by taking these activities as one's objectives, but rather by “catching” them in the course of one's membership of social groups.

When it comes to the other concepts with which ethical theorists

  ― xx ―
have commonly concerned themselves—right, obligation, duty—Anderson's position is quite different. Ethical theorists like Ross, in developing Moore's ideas, had argued that “right”, too, ought to be treated as a special sort of quality. Anderson is convinced that it cannot be; that the notion that something ought to be done is, as it were, “written into” the description of an act as obligatory, or as right, or as our duty; and that it certainly cannot be a quality of an act that it ought to be done, whether prima facie, as Ross thought, or after consideration. Indeed, he works out what might be described as a “sociological ethics”, so far as these expressions are concerned. A form of social organisation, he argues, develops regular habits of action, and these come to be thought of as being obligatory upon those who adhere to such an organisation even when they would prefer to act otherwise. If, then, we describe an act as right, or as our duty, this will be easily understood by those who are involved in the same mode of life. We will think of the act in question as being in some absolute sense our duty, as possessing an intrinsic ought-to-be-doneness; but the theorist can understand why we take one act rather than another to be our duty, not by simple examination of the act in question but by psychological and sociological inquiry.

Indeed, whereas many ethical theorists have seen in duty the central ethical concept and have thought that to act from a sense of duty is to be impelled by the best of all motives, Anderson argues that if we act only because we “feel obliged” to do so, this is a clear indication of the absence of goodness. Where there is goodness, there is free participation, spontaneity (although not in a metaphysical sense—Anderson rejects the metaphysical idea of “free will”), and certainly not a feeling of constraint. A person who lives under the shadow of conscience, in a desperate effort to avoid feelings of guilt, is the very type of the neurotic.

Anderson's most fully worked-out defence of the logic of his procedure is in “The Meaning of Good”. If somebody asserts, and somebody else denies, that an expression qualifies rather than relates, the issue cannot be easily settled. Anderson originally directed his argument mainly against those who tried to hold both that good is a quality and that it is its essence to be pursued. Rather different problems arise when the prevailing view is that “good” ought to be analysed much as Anderson analysed “right”—as a way of indicating our adherence to a way of life. Anderson is attempting what is now sometimes called a “rational reconstruction” of ethics; he sees in traditional ethical theory a partly confused, partly enlightening, attempt to draw attention to the characteristics he has himself emphasised. The question for him is not, it should be observed, “whether there is such a thing as the property of goodness”, but whether it is possible to discover distinguishing characteristics in certain of the things which have ordinarily been called “good”, characteristics sufficient and necessary to mark them off from the morally indifferent and the evil.

In his lectures on ethics, Anderson devoted a good deal of attention to establishing this sort of link between his theories and those of his

  ― xxi ―
predecessors. But also he concerned himself with describing the activities he took to be good. Love, courage, the spirit of inquiry, aesthetic creation and appreciation—these too stand in need of “rational reconstruction”. Thus Anderson is not prepared to restrict himself to considering “the logic of ethical statements”. For him the ethical theorist is discussing forms of human activity—not, primarily, what people say about these activities; it is quite natural for him to appeal for confirmation of his views to writers as diverse as Sorel, Freud, Marx, Joyce, Marshall, Arnold. In this respect, his approach to ethics links him with Continental or American speculation rather than with the general stream of British moral thought.

In aesthetics the story is much the same. There, again, he is looking for the distinguishing characteristics of beautiful objects; he has been particularly concerned to battle against relativism; he has been influenced not only by professional aestheticians but at least as much by novelists and by critics. Aesthetics, for him, is the direct consideration of works of art (and of nature) with the object of discovering their general characterising properties; and that procedure naturally demands a close attention to the works themselves. So, again unlike most philosophical aestheticians, he has written a number of special literary studies—most notably, perhaps, on Joyce's Ulysses—as well as attempting, in Some Questions in Aesthetics, to present his general ideas about the character of aesthetic inquiry.

Similarly, too, he does not admit a distinction between political theory and political philosophy, whether the latter is understood as “the application of ethics to politics” or as an investigation of, say, “the vocabulary of politics”. The philosopher who takes up the study of politics, he would freely admit, will naturally be particularly concerned with the logical points raised by political theory. At the same time he will have to look concretely at what happens in human society.

Naturally, Anderson rejects both the traditional atomism of British political theory, for which a society is “really” a collection of individuals and the State a deliberate arrangement, and that Absolutism for which there is only “Society”, which the State expresses. Human beings, as he sees them, are neither self-originative centres of activity, nor instruments in the development of Absolute Spirit. The subject matter of political theory, like the subject matter of any other theory, will be complex structures which both act and are acted on, which enter into relations with one another and with other things and contain relations within themselves. Once more, this still leaves open the question (which logic by itself cannot settle) what these structures are. Anderson takes them to be such complexes as traditions, social movements of one sort or another, considered as giving rise to and as being themselves shaped by a variety of social institutions, of which the State is only one—although one of particular concern to political theory. Here again, this is not the ordinary starting-point of British philosophers; his views grew out of his reading of Marx and of the political pluralists, especially in the great

  ― xxii ―
days of Orage's journal The New Age, rather than out of the empiricist tradition, to the individualism of which he is profoundly opposed.

Anderson places great emphasis upon social conflicts, as opposed to the traditional emphasis on social unity. Like Heraclitus he believes that “the hidden harmony is better than the open”. Pre-Socratic philosophy, in general, greatly attracts him—he thinks of Burnet's Early Greek Philosophy as one of the most philosophically enlightening of all books—but he was particularly fascinated by, and made particularly fascinating, the philosophy of Heraclitus. On Anderson's interpretation of him, Heraclitus taught that a thing's permanence, its stability, as well as its growth and development, had their source in the counter-poising of opposite tendencies, not in the subordination of every force within it to a single objective. This, according to Anderson, is as true of the human mind and of human society as it is of the candle-flame.

Thus we ought not to ask of a social institution: “What end or purpose does it serve?” but rather: “Of what conflicts is it the scene?” That is the way in which we shall come to an understanding of its mode of operation. Anderson is particularly interested in educational institutions and in the struggle within them between what he takes to be the classical tradition (the tradition of critical inquiry) and the utilitarian attempt to use educational institutions for merely vocational training or for “social adjustment”. Similarly, in the political institutions of a society—in the State, for example—he sees neither “an expression of the general will” nor the instrument of a particular class, but rather an arena within which conflicts are fought out and compromises reached. Particular social forces could no doubt be dominant within the State at a particular time, but the mere existence of legislative and judicial institutions clearly indicates that their domination is not absolute. To glorify the State, to attempt to subordinate all social activity to its powers, is in his eyes wholly to misunderstand what gives a society vitality—the free play of a variety of traditions, of diverse modes of life, in their conflicts and co-operations with one another.

Similarly, Anderson is opposed to a religious view of things, whether understood as laying down a principle of conduct to which all human activities are to be subordinate, or as determining some goal towards which “the whole Universe” moves. In religion, as in State-worship, he detects a fundamental servility; unlike many critics of Christianity, he is as opposed to what he takes to be its moral standpoint (not in all, but in very many, respects) as to its theological doctrines. So far as it offers consolation, exhorts man to meekness and humility, emphasises individual salvation, Christianity cuts across Anderson's conviction that struggle is unceasing, that every good activity is a form of enterprise, that what is of first importance is the continuance of tradition. Not by subordination to a person, human or divine, he would say, but only by critical participation in forms of productive activity can men be “saved”.

If Anderson does not much like being described as an “atheist”, this is primarily because of the negative suggestion of that description—as

  ― xxiii ―
if it were a matter of primary concern to him to deny that God exists. For what Anderson argues against is any conception of a total system, an ultimate end or a final principle of explanation, whether or not it is regarded as being, or as being related to, an omnipotent person. Thus he is as critical of Haeckel as of Aquinas, of the conception of a man-centred, as of a God-centred, scheme of things. Indeed, in so far as some forms of religion emphasise that our life is not of our making, that it goes on, in large part, in independence of our plans, he much prefers them to any form of sentimental humanism. He is as sympathetic to the view that “providence moves in mysterious ways” as he is to Hegel's “the cunning of history”, but he is not prepared to admit that providence or history is a supranatural entity with plans of its own.

Similarly, Anderson's determinism and his materialism are each of them of a rather special character. He is not a “reductive materialist” i.e. he does not believe that there is some single entity “matter”, of which minds, for example, are a confused appearance. For him minds have distinctive characteristics; they are not only physico-chemical structures, they are also passionate. His determinism, similarly, is not Laplacean; there can be no question for Anderson of “giving a complete description”, whether of the present or of the future. It amounts only to this: that whenever a change takes place, it does so under sufficient and necessary conditions. There are genuine novelties, new characteristics appear, but always under determinate conditions. Indeed, we must always, in Heraclitus's phrase, “expect the unexpected”; we cannot make the future safe for ourselves, any more than we can make the world safe for democracy. Struggle, uncertainty, risk, disappointment, are not accidental features of human life, remediable by the exercise of sufficient good will; they arise inevitably out of the general structure of things, including our own structure.

No total scheme, no simple units, no first principles, no ultimate objectives, no modes of being, no necessary truths—these, not the rejection of God, are the fundamental negations of Anderson's philosophy. To put it positively, there is, on his view, a single way of being: the complex activity of a spatio-temporal occurrence, within which discriminations can be made and which is itself discriminable within a wider system. To explain, to prove, is to draw attention to relationships which occur between such occurrences; to assert a proposition is to take something of a certain kind to occur; any proposition can be false; science proceeds by the critical examination of hypotheses; any objective has a variety of characteristics and it can always be pursued as part of a procedure for getting something else.

These are definite and general doctrines. In taking it to be his task as a philosopher to enunciate, and to argue for, propositions of such a kind, Anderson is in total opposition to the view that philosophy is simply analysis, or that its object is purely therapeutic. So far he is an untypical figure in recent British philosophy. Yet he has no sympathy either, for the rhetorical methods of so much contemporary Continental philosophy.

  ― xxiv ―
Like his Continental contemporaries, he wants to discuss large issues, but critically, analytically. A common presumption of our time is that this cannot be done; in England, that large issues either fall within the province of a particular science, or else are matters of decision, not of argument; on the Continent, that large issues certainly belong to philosophy, but that careful analysis, close criticising, has no place there. Yet there are signs that the tide is turning, in England at least. Many positions which Anderson in the early 'thirties was almost alone in defending, are now at least respectable. In the United States, there has been, indeed, a continuing interest in the critical discussion of large issues as a result of the influence of such philosophers as C. I. Lewis and Morris Cohen. Even within the better sort of “ordinary language” philosophy, I should say, the same interest is clearly exhibited—notwithstanding, sometimes, the explicit pronouncements of its proponents—as it most certainly is in the philosophy of Karl Popper. Their presumptions and their arguments Anderson would certainly wish to challenge, but they belong, at least, to the same controversial tradition. To the presentation and continuance of that tradition, Anderson has hoped to contribute; to its judgment, he submits his case.