BERNARD BOSANQUET'S MORPHOLOGY OF KNOWLEDGE

Anthony Waterhouse, Member, on 8 January 2002

Bernard Bosanquet (1848-1923) was in Britain, second to F. H. Bradley, the most important exponent of a wholist philosophy often called Absolute Idealism. An Absolute idealist thinks that what we know is at least in part mind-dependant, and that what we may think are separate things - appearance - masks the reality, that what seem separate and contradictory aspects of experience are ultimately resolved into a coherent whole, the Absolute.

Bertrand Russell, reacting to the Absolute Idealists, at the beginning of the 20th century, set the agenda for Anglo-Saxon philosophy on positivist and analytical lines. For a few years he fell under the influence of the idealists, before reverting back to empiricism, which sees knowledge founded on sense experience, traditionally the dominant current of British philosophy. Russell, driven by the quest for certainty, tried to base mathematics on what he thought logical foundations, but he found that wholist idealism was incompatible with his objectives. He went on to adopt an outlook which was diametrically opposite, a pluralistic empiricism called Logical Atomism - knowledge was made up of discrete bits, facts. He says regarding his new and old faiths: "Hegel [the wholists] had maintained that all separateness is illusory and that the universe is more like a pot of treacle than a heap of shot. I therefore said `the universe is exactly like a heap of shot. Each separate shot, according to the creed I then held, had hard and precise boundaries and was as absolute as Hegel's Absolute."

Russell's reputation was in part established by his clear and witty attacks on the idealists. The doctrines of the idealists were often obscure. Their reputations never recovered from Russell's criticism, but within a few years Russell's ambition to put knowledge on firm foundations were shown to be flawed, as we shall see at the end of this talk. There is a certain irony for in more recent times W. V. Quine, a positivist-influenced philosopher faithful to Russell's logic, has adopted ideas that have some similarity to Bosanquet's doctrines, such as a coherence theory of knowledge.

The project for Bosanquet and Bradley was to develop a theory of knowledge based on an idealist philosophic logic as opposed to epistemology, theory of knowledge based on empiricism. This they felt more adequate and inclusive, where the human spirit was reincorporated into philosophy. An Empiricist often sees philosophy as a handmaiden to science. From the 17th century Science, knowledge based on experience and experimental methods, had become successful in expanding our knowledge of the tangible world. Before what we know as the Enlightenment, knowledge of the laws of nature and moral law were seen as connected, both being God given, the Church being his representative on earth. Science began to successfully challenge received knowledge handed down for generations by the church. There emerged a view that science was objective and rational, dealing only with facts, free from questions of value. By the end of the 19th century many believed that science would eventually explain everything, all other areas of knowledge, aesthetics, religion and ethics being secondary.

Both Bosanquet and Bradley were sons of evangelical clergymen, which perhaps helps to explain the ethic-religious sentiment that tinges their thought. Their position on religion is that of sympathetic scepticism. Philosophy informed all aspects of life for Bosanquet and, apart from his philosophic logic, he wrote a theory of the state, on ethics, religion, science, and introduced aesthetics into Britain as a philosophic subject. This all-inclusive approach, where value and knowledge brought together is inspired by a classical education. Plato and Aristotle are their - Bosanquet's and Bradley's - main teachers. They were also influenced by the German philosophy of Kant, Hegel and Lotze. At Oxford, Bosanquet and Bradley were taught by T. H. Green, one of the pioneers in Britain of the Idealist school. Both Bosanquet and Bradley learned from each other. Though dubbed Neo- Hegelians by their critics their approach is tentative, sceptical, with no intention to create grand philosophical systems. They are best defined as critics of empiricism and its consequent social theories, especially that of its principal 19th century proponent J.S. Mill.

J.S. Mill's enemy was institutionalised knowledge. He saw experience and experimentation as the only sound source of knowledge. So for him our knowledge is gained as sensations, which are projected into our mind as ideas. Facts have a corresponding idea as discrete occurrences in the mind, from this we get particulars. Our knowledge is to him built of particulars: fragments from our sense of the world. He denied that Universals, general terms as innate ideas, were fundamental to knowledge. A Universal is an idea or term that can be applied to a thing or quality. If we take the general term `donkey', for Mill our knowledge of `donkeyness' can only come from particular, single instances of `donkey', which we come to associate together as similar. Mill's logic is based on associative psychology, a forerunner of behaviorism. For him universals are just generalisations based on us grouping similar particulars together. Universals are an assumption behind Aristotle's syllogistic logic; however Mill accepted that syllogisms: e.g. all men are mammals, Jack is a man, therefore Jack is a mammal, represented the form that thought takes in making analytical inferences (based on reason). He says it is:

"a method of analysing the mental process which must invariably take place in correct reasoning".

Bosanquet thought Mill's ideas were inadequate; how can we, from atomic particulars, make up knowledge? Bosanquet thought that syllogism represented a narrow definition of inference and only described a type of linear inference. Bosanquet's wholist theory of knowledge is based on his philosophic logic, which gives logical principles priority as foundations of knowledge; as opposed to epistemology, theory of knowledge based on empirical principles. Instead of facts derived from our senses, the logical principle is his point of departure in the construction of knowledge. In his doctrine the universals take their revenge.

The title of his book Logic-The Morphology of Knowledge perhaps gives us a feel for his approach. This is a study of the forms which judgement and inference take to become our system of knowledge. It is a natural history of the abstract forms of thought and knowledge. This, he warns in his The Essentials of Logic, is a difficult subject, there is no stimulus of sense perception to satisfy curiosity or feed our interest.

In some ways his philosophic logic anticipates later developments in logic. His logic includes syllogistic logic but goes beyond it. He rejects the psychologism of Mill. Instead of raw ideas being the foundation stone of his theory of knowledge, as in Mill, Bosanquet's point of departure is Judgement; and to borrow from Bradley, judgement is:

"The act which refers an ideal content (recognised as such) to a reality beyond the act"

What this means is that ideas are not just particulars; for an idea to be anything at all it must have meaning referring to reality and therefore be to some degree a Universal. From isolated particulars Knowledge cannot be built up. Whereas modern logicians deal with propositions (in words) in the 19th century logical and philosophical theories were often cast in mentalistic terms: thus the terms `idea' or `judgement' are foundation stones which assume some form of idealism.

Bousanquet is reticent in using the term idealist; nevertheless he is an Objective Idealist - the doctrine that our knowing the world (through our minds) and the world can never be seen as separate. Knowledge is not just `given' through our senses. We are not to assume that we are shut up in our subjective minds, with an independent objective world outside, this is what the empiricists believe. They have the problem of a dualism, body and world, separate from mind. Bosanquet believes that though our minds are separate there are some unifying aspects, which bind them. There is correspondence between minds; our judgements are necessary and universal; if this was not so we could not understand each other. We also construct our knowledge of the world The objective idealist says that what we think as subjective and objective are in a sense just two extremes points of a continuum, we never start from a completely subjective point but move to the objective.

Bosanquet says: "Then the world as idea means no less than this, that the system of things and persons that surrounds all of us, and which each of us speaks of and refers to as the same for everyone, exists for each of us as something built up in our mind (the mind attached to his body) and out of the material of his own mind."

And " knowledge is the medium in which our world, as an inter-related whole, exists for us"

Bosanquet uses the term "judgement" in elastic ways. We have our continual conscious awareness of the world as all embracing judgement, which he describes as: "Judgement as a consciousness of the world". Bound up within our awareness there are differentiated judgements. The principle of identity and difference, parts within wholes, is always operative in our conceptual construction of knowledge. Referring back to the notions that judgements are ranging in type from those most subjective-like to those that move towards the objectivity, we can see that, the more particular-like a judgement is the more subjective-like it is, and the more whole-like a judgement is the nearer we move to the objective. Its `matter' or content will determine the form of judgements.

There will be simple judgements, which indicate place or time without qualifying it i.e. `this' or `that' (pointing to something). These are the most particular-like Judgements. But Judgements which point to a bare individual identity, for example a name of a person; these are what he calls "Individual universals" or "Concrete universals". In themselves they just indicate an individual identity which lives through time and space and they have some of the characteristics of a particular. But a bare identity is just a mark of identity and to flesh out this identity through time and space it needs attributes to qualify it. The Abstract universals representing all pervading qualities, for example colour, mammalness etc., or organisational principles, like natural laws, hardly point to any individual identity at all but refer to reality as a general quality that individuals may have. We might say that to give an abstract universal any meaningful life, it has to find an identity, a concrete universal. Anything which is only true of a particular time, e.g. true today but not tomorrow, is not completely true; only principles which are true in all times and places are completely true.

Perhaps an example of the continuum of judgements moving towards the objective, is the particle /wave theory of light. It was derived from two rival theories; one said light was wave-like, the other particle-like. Some evidence supported both, but they contradicted each other. Eventually both theories were brought together because that was the only way to account for all the evidence. The original rival theories were partial. The more all embracing a judgement is the stronger is its objective content and so is nearer to truth and reality We get a sense of this when Bosanquet says:

"The truth is that thought has always the nature of a system of connected members, and is an effort to form, which we may call a `world'. This is the only sort of thing which can satisfy the logical law that contradiction is a mark of unreality, or (the same law) that the truth or real is `the whole'."

Before we move on from Judgements; in Bosanquet's doctrine all judgements are to some extent synthetic and analytical, some more and some less: synthetic - meaning judgements from experience, and analytical - meaning judgements from reason alone; or by virtue of the meaning of the terms. i.e. sows are female pigs. The more particular-like a Judgement is the stronger will be the synthetic component, and the more it tends towards a Universal the more it tends to be an analytical judgement. Bosanquet is here trying to reconcile the rationalist and empiricist philosophies. I will later link this to Quine's famous essay `The two dogmas of empiricism' at the end of the talk.

I briefly flag up negation as a quality of Judgement; it is by negation that true and false judgements are sorted out.

Inference is the second important concept in Bosanquet's logic. Inference is a form of judgement where

"The problem of Inference is something of a paradox. Inference consists in asserting as fact or truth, on the ground of certain given facts or truth, something which is not included in those data. We have not got inference unless the conclusion is, (i) in the premise, and (ii) outside the premise"

and then he asks the question:

"What kind of relations of content must we have, in order to realise the paradox of inference?"

and then he goes on to say that it depends on Universals.

The important characteristic of an Inference is that we have: "novelty in the conclusion of the Inference."

Bosanquet gives the following three examples. The first is a little story from Thackeray that will illustrate one type of inference:

"An old Abbé, talking among a party of intimate friends, happened to say: `a Priest has strange experiences; why ladies, my first penitent was a murderer.' Upon this the principal nobleman of the neighbourhood enters the room: `Ah, Abbé, here you are: do you know, ladies, I was the Abbé's first penitent, and I promise you my first confession astonished him!'"

Looking at the story we can quickly see whom the murderer was. There is a universal thread going through the story, what Bosanquet calls the "Individual or Concrete Universal", which is an identity which runs through the story and allows you to make the inferential leap.

We talked about the second type of Universal; the Abstract Universal, and it is operative in our second example: Neptune was discovered in 1846. From the observations of Uranus it was found that there was a wobble in its orbit that was not consistent with Newton's Laws of Gravitation. From this the observers calculated that there must be a cause for the anomaly in the orbit of Uranus and it was inferred that the cause was probably an unknown planet exerting a gravitational pull. By calculations they pointed the telescope in the right direction and Leverrier and Adams were able to discover Neptune. Running through this Inference is the Abstract Universal principle of the Laws of Physics.

A final example: if we take a triangle as an example of a very simple system, from knowing two sides and one angle we can make inferences of the other two angles and the third side. Change the original known angle by so many degrees (leaving the known sides the same) and you can again calculate the unknown angles and side. This is based on a Universal of the nature of triangles; the inferences have a quality of `systematic necessity'. Implications are built into the system of knowledge; Bosanquet says this is vividly illustrated when different individuals make discoveries independently of each other over a short time span; such as the discovery of calculus by Newton and Leibniz or the theory of evolution by natural selection by Wallace and Darwin.

Bosanquet says that knowledge, as a whole is a system; what he is talking about is a coherent theory of knowledge and truth. Within a conceptual scheme we are driven to conclusions which are compatible within it. Judgements to be true within a system have to be logically consistent with each other. Inductive inference, as the enumeration of particulars incidences of similar type arranged to give us generalised conclusions is an inadequate description. Instead we see that universal principles run through the construction of our knowledge, in as far as it goes.

The word Absolute today has dogmatic connotations so before taking leave of Bosanquet I'd like to say to those of you who might be worried that we are running out of contradictions, making things rather dull: you need not worry, the Whole or Absolute, that which is finally, logically completely self-consistent, true or real, is so huge, that we as finite beings will never run out of surprises and mistakes. Bosanquet is what is called a `fallibilist'; in other words our knowledge has a potential to be falsified though there is convergence towards truth as knowledge becomes more universal. He hints at this in the quote:

"The postulate of knowledge then is badly stated as a Uniformity of Nature. That was due to the vulgar notion of `inductive generalisation'. It must be stated in two parts: first, "Once true always true:" and secondly, "Our truth is enough for us", that is, it covers enough of the universe for our practical and theoretical needs."

Now we can go back to the beginning of our story and see what happened to the doctrines that Bertrand Russell used to shipwreck Bosanquet's theories. As I said before Bertrand Russell tried to put knowledge onto firm foundations. There were two aspects to his project: he developed a mathematical logic that he aspired to make logically consistent and which could be used as a means of analysis of facts. Russell's utopian aspirations eventually collapsed like a house of cards. In 1931 the logician, Godel, came up with the Uncertainty Principle and showed that no logical system could be entirely consistent. The notion of facts attached to empiricism and the correspondent theory of truth became seen as untenable. This was replaced by semantic theories, instead of facts people talked of meaning and the conditions under which the meaning of statements were valid.

In the 1950s the American philosopher, W.V. Quine, a positivist-influenced follower of Russell in logic, developed a wholist theory of knowledge; his was a linguistic wholism and therefore did not include the idealist metaphysics of Bosanquet's wholism. He also wrote a famous essay: The two dogmas of empiricism, as O'Hear says on behalf of Quine:

"Quine sees the idea that each factual (or synthetic) statement can be confirmed or falsified in isolation from the system of statements in which it figures as the natural descendent of the reductionist idea that statements about the world can be analysed ultimately into statements about experience (and the obverse of the idea which he also wants to reject, that analytical statements, including those of logic and mathematics, are true or false by virtue of meaning alone, and independent of the way things are in the world). Quine proposes a generalisation of a thesis originally argued by the French philosopher Pierre Duhem in connection with scientific theories. Duhem's point is that scientific theories are never confirmed or falsified by their observational consequences taken in isolation or more accurately, that scientific theories do not have observational consequences on their own."

Do I detect some residual echo of Bosanquet's theories there?

A. R. Waterhouse

Bibliography

Bosanquet B (1895) The Essentials of logic London: MacMillan

Bosanquet B (1911) Logic Volume 1 (2nd Ed) Oxford: Clarendon press

Bosanquet B (1911) Logic Volume 2 (2nd Ed) Oxford: Clarendon Press

Bosanquet B (1912) The principle of individuality and value London: MacMillan

O'Hear A (1985) What philosophy is London: Penguin

Passmore J (1966) A hundred years of philosophy (2nd ed) London: Penguin

Popkin R and Stroll A (1986) Philosophy made simple (2nd ed) Oxford: Heinemann

Russell B (1958) Portraits from memory London: Allen and Unwin

Russell B (1959) My philosophical development London: Unwin paperbacks