Chapter 14 Chapter 13 Knowledge, Error, and Probable Opinion

The question of the meaning of truth and falsity, which we considered in the last chapter, is much less important than the question of how it is possible to know what is true and what is falsity.This chapter will examine this question fully.Undoubtedly, some of our beliefs are false; we are therefore obliged to ask how certain is it that such and such beliefs are not false.In other words, can we know anything at all?Or is it just a fluke that we believe it to be true?Before answering this question, however, we must first decide what we mean by "cognition," which is not as easy as one might think.

At first glance, we might think that the definition of knowledge is "true belief."We think we have a knowledge of what we believe when what we believe is true.But this would be inconsistent with the common usage of the word "knowledge".To take a trivial example: If a person believes that the late prime minister's surname began with the letter B, what he believes is true.Because the late Prime Minister's name was Sir Henry Campbell Bannerman (Bannerman).But if he believes that Mr. Balfour was the late Prime Minister, and also believes that the late Prime Minister's surname began with the letter B, the belief, though true, cannot be regarded as constituting knowledge.If a newspaper, with its clever foresight, publishes the results of a campaign before the telegrams report the results, it may, by luck, prove afterwards to be correct, to the detriment of less experienced readers. Generate trust.But though their faith was genuine, they cannot be said to have knowledge.It is thus clear that a true belief is not knowledge when it is deduced from a false one.

Likewise, if a true belief is deduced from an erroneous process of reasoning, it cannot be called knowledge, even if the premises on which the deduction is based are true.If I know that all Greeks are men, and that Socrates is a man, then I infer that Socrates is a Greek; then again I cannot be said to know that Socrates is a Greek, since my premises and conclusion Although all are correct, the conclusion is not based on the premises after all. But should we say that nothing is knowledge except what is validly deduced from true premises?Obviously, we cannot say that this definition is too broad and too narrow.First, it is too broad because it is not enough to say that the premises are cognizable if they are true.A person who believes that Mr. Balfour is the late Prime Minister may draw valid deduction from the true premise that the last Prime Minister's surname begins with the letter B, but he cannot be said to know the conclusions reached by the deduction.We must therefore revise our definition to say that knowledge is that which is validly deduced from known premises.Even so, this is a circular definition: it assumes that we already know what "given premises" mean.This definition, therefore, is at best a definition of a kind of knowledge, what we call derived knowledge, as opposed to intuitive knowledge.We can say: "Derived knowledge is something that is effectively deduced from the premises of our intuitive cognition".There is no formal flaw in this statement, but it leaves the problem of definition of intuitive knowledge to be studied.

Let us set aside the question of intuitive knowledge for a moment and examine the definition of derived knowledge suggested above.The main objection to this definition is that it unduly limits knowledge.It often happens that people have a true belief which grows in them because it can be validly deduced from bits of intuitive Logical steps are deduced from direct visual knowledge. For example, let us take the beliefs that arise from reading.If the newspapers had published the news of the king's death, we would be quite justified in believing that the king was dead, because if it were false, the news would not have been published.We have good reason to believe the newsprint's assertion: the king is dead.But here the intuitive knowledge on which our belief is based is the knowledge of the existence of sense-data derived from seeing the printed matter in which the news is published.Such knowledge hardly emerges into consciousness unless one is a poor reader.A small child may know the shape of each word and struggle to read it bit by bit to understand their meaning.But it is not the case with a casual reader who knows the meaning of every word at once, without realizing that his knowledge comes from "seeing the printed words" unless he reflects on it. This kind of feeling material comes from.Thus, although it is possible and possible for the reader to effectively deduce the meaning of each word from them, he does not in fact do so, because he does not actually make any steps that could be called logical inferences. .But it would be absurd to say that readers were unaware that the newsprint carried the news of the king's death.

Therefore, no matter what the result of intuitive knowledge is, even if it is only based on association, as long as there is an effective logical connection, and the person concerned can perceive this connection through introspection, we should admit it as derived knowledge.We can in fact derive from one belief to another by means other than logical reasoning: for example, the passage from print to its meaning illustrates these methods.This method may be called "mental reasoning."As long as there is a discoverable set of logical reasoning alongside mental reasoning, we can think of this mental reasoning as a means of obtaining derived knowledge.Because the word "discoverable" is vague, it makes our definition of derived knowledge less precise than we would like: it doesn't tell us how much reflection is required to make the discovery.But in fact "knowledge" is not a precise concept: in the course of this chapter, as we shall see more fully, it is confounded with "probable opinion."It is therefore unnecessary to search for a very precise definition, since any definition is always subject to misunderstanding.

Nevertheless, when it comes to knowledge, the main difficulty arises not with derived knowledge, but with intuitive knowledge.As long as we study derived knowledge, we fall back on identifying intuitive knowledge.But when it comes to intuitive beliefs, it is by no means easy to discover a criterion for distinguishing what is true from what is false.In this matter it is scarcely possible to arrive at very precise results: all our knowledge of truth is tinged with a degree of doubt, and a theory which ignores this fact is manifestly false.Nevertheless, there are remedies for alleviating the difficulty of this problem.

First, our theory of truth offers the possibility that we can distinguish certain truths as self-evident in the sense that they are guaranteed to be error-free.When a belief is true, we say that there is a corresponding fact in which the several objects of the belief form a single complex.A belief is said to constitute knowledge of a fact in so far as it satisfies the few conditions considered in this chapter which are not quite clear.But of any fact we may have, besides knowledge by belief, a knowledge by perception (the word perception is here used in the widest possible sense).For example, if you know the time of sunset, you can know the fact of sunset at that time: this is the knowledge of the fact through the knowledge of truth; if the sky is clear, you can also look west and see it The setting sun; then you know the same fact through the knowledge of things.

Hence, when it comes to any complex fact, there are always two ways in theory of knowing it: (1) By means of judgments, in which the various parts of the fact are considered as they are in the fact; (2) by knowing complex things in themselves, which may be called perception (in a broad sense), although it is by no means limited to objects of sensation.It may now be noticed that the second method of knowing complex things, the method of knowing, is only possible if there is such a fact; whereas the first method, like all judgments, is subject to error. .The second method presents us with the complex as a whole, and is therefore only possible if there is a relation between the parts of the whole that unites them into the complex whole.The first method is quite different, it presents to me the parts and their relations separately, and requires only the reality of the parts and their relations: the relations may not connect the parts in a judgmental manner. , but such judgments can still be made.

It will be recalled that, at the end of Chapter Eleven, we suggested that there may be two kinds of self-evidence, one which gives us absolute assurance of truth, and one which provides only partial assurance.We can now distinguish the two. We can say that a truth is self-evident (in its first absolute sense) when I know a fact corresponding to it.When Othello believes that Desdemona loves Cassio, if his belief is true, then its corresponding fact is "Desdemona is in love with Cassio".Of this fact, no one but Desdemona can realize it; therefore, in the "self-evident" sense we now consider, Desdemona loves the truth (if it is a truth) of Cassio. , is self-explanatory only to Desdemona.All mental facts, and all facts about sense-data, contain this private element: since only one person can know these mental things, or the sense-data connected with them, in what we now call self-evident, They can only be self-evident to one person.No fact, therefore, can be self-evident to more than one person, so long as it concerns particular beings.Facts, on the other hand, do not have this personal element so long as they are about the universal.Many minds are capable of knowing the same universal; therefore, the relations between universals are known by knowledge to many different people.In any case where we know by cognition a complex fact of several terms in a definite relation, we say that the truth that these terms are so connected has a primary or Absolute self-evidence, in which case the terms so connected must be true.So this self-evidence is an absolute guarantee of truth.

But though this self-evidence is an absolute guarantee of truth, it does not enable us to be absolutely certain that, of any given judgment, the judgment we speak of is true.Suppose we first perceive that the sun is shining, which is a complicated fact, and then judge that "the sun is shining."When passing from perception to judgment, it is necessary to analyze the known fact: we must separate the components of this fact, "the sun" and "shine".Mistakes are possible in this analysis; therefore, even when a fact is primordially or absolutely self-evident, a judgment which is believed to correspond to the fact is not necessarily infallible, because it It may not really correspond to the facts.But if it corresponds to the fact (in the sense of "corresponding" explained in the last chapter), then it must be true.

The second kind of self-evidence is chiefly that of judgment, and does not result from the immediate perception of a fact as a single complex whole.This second self-evidence will vary in degree, and it can be reduced from the highest to merely a tendency to support the belief.For example, a horse hurried past us along a hard road.At first we were no more than sure that we heard hooves; gradually, if we listened carefully, there came a moment when we thought it was fancy, or the shutters upstairs, or the beating of our hearts; at last, We wonder if there is any sound at all, and then we think we don't hear anything anymore.Finally, we know we can't hear anything.In this process there is a succession of self-evident grades, from the highest to the lowest, not in the sense-data themselves, but in the judgments made upon them. Or, let's assume that two colors are compared, one is blue and the other is green.We can say with absolute certainty that they are two different colours; but if the green is allowed to gradually change more and more like blue, then first it should become a turquoise, then a greener blue, and then If it is blue, then there will always be a point where we wonder if there is any difference at all;The same holds true when playing an instrument, or in other situations where successive grades exist.Self-evidence of this kind is thus a matter of rank; it seems obvious that higher ranks are more reliable than lower ranks. In derived knowledge, our underlying premises must have a certain degree of self-evidence, as must the relationship between the premises and the conclusions deduced from them.Take, for example, a piece of reasoning in geometry.It is not enough that the theorems from which we start are self-evident; at each step of reasoning, the relation between premises and conclusion must also be self-evident.In difficult reasonings, the degree of self-evidence of this relation is often very low; and therefore errors of reasoning are not improbable in cases of serious difficulty. From what we have said above, it is evident that when it comes to both intuitive knowledge and derived knowledge, if we consider intuitive knowledge to be reliable in proportion to its degree of self-evidence, then from the existence and logic of salient sense-data There are gradations of certainty in the truths which are physically and mathematically simpler (which may be regarded as quite certain), down to judgments whose probability is not much greater than its opposite.What we firmly believe is called knowledge if it is true, whether it is intuitive or derived logically (logically or psychologically) from intuitive knowledge.What we firmly believe is called error if it is not true.What we firmly believe, if it is neither knowledge nor error, but at the same time what we do not firmly believe, because it does not have the highest self-evidence, or is not derived from the most self-evident things, may be called probabilistic. sexual opinion.Much of what may ordinarily be regarded as knowledge, therefore, is more or less probabilistic opinion. Concerning probabilistic opinions, we can get a great deal of help from consistency, which we reject as a definition of truth, but which can often be used as a criterion.A separate set of probabilistic opinions, if they are consistent with each other, are more likely than any one of them alone.Many hypotheses in science have achieved their probability in this way.They are embedded in a coherent system of probable opinions, which are more probable than they would be in isolation.The same thing applies to general assumptions in philosophy.Often in a single case such assumptions seem highly doubtful, but when we consider the order and coherence they bring to many probable opinions, they become almost sure.This, in particular, applies to things like the distinction between dreams and real life.If our dreams were as consistent as our day-to-day life every night, we would hardly know whether to believe our dreams or our actual life.But in fact, a check of coherence denies the dream and affirms the actual life.Such a check, though it increases the probability where it succeeds, can never provide absolute certainty unless there is already a considerable degree of certainty in the consistent system.The mere organization of probabilistic opinions, therefore, can never, by itself, make probable opinions undoubted knowledge.