An Essay Concerning Human Understanding By John Locke Summary and Analysis Book IV: Knowledge and Probability


In the fourth and final book of the Essay, Locke sets forth the major elements included in the theory of knowledge that he has sought to establish by the arguments presented in the first three books. Many of his conclusions can be anticipated by anyone who has followed his line of reasoning up to this point. There are, however, a number of questions which arise when one stops to consider the implications that are involved in many of the statements which he has made. Several of these questions are discussed at some length in this part of the Essay. In fact, the contents of this book may be summarized as the answers that Locke has given to the following list of questions.

What, in general, is the meaning of the term knowledge, or in other words, how is knowledge to be defined? Are there different degrees of knowledge, and if so, how can they be distinguished? What is the scope of human knowledge, and what limitations with reference to it must be recognized? To what extent are the various degrees of knowledge reliable? What is the meaning of the term truth, and how can truth and falsity be distinguished? Can universal propositions, such as we have in the different fields of science, be said to be true or false? Do we have any certain knowledge of our own existence? Is it possible to know anything about the external world? Do we have any knowledge concerning the existence or the non-existence of God? What, if anything, do we know about the principles of morality? Can ideas about what is right or wrong in the area of human conduct be regarded as true or false? What is the role of faith in the achievement of human knowledge? Is there such a thing as "revealed" knowledge? How can one distinguish between what is regarded as revealed truth and blind superstition? What is the relationship between faith and reason? In what areas of experience is it impossible to have anything more than probable knowledge? What are the major sources of error in the judgments that people make concerning the world of their experiences?

All of these questions and many more of a like nature are pertinent to any type of epistemological inquiry. We shall here attempt to state briefly the substance of the answers that Locke has provided in the final portion of his Essay.

Knowledge, according to Locke, is the agreement or disagreement of ideas with one another. To use his own words, it is "the perception of the connection of an agreement, or disagreement and repugnancy, of any of our ideas." This sounds like a purely subjective theory of knowledge, and if one were to follow it with complete consistency, it would lead to solipsism and a denial of any genuine knowledge about anything other than one's own mind.

But Locke does not follow the implications of his definition of knowledge this far. Apparently his main purpose in this connection was to emphasize the fact that all knowledge has its origin in the sensations and reflections which occur within the human mind. Whether the ideas which are formed by these mental processes are adequate to tell us anything about that which is external to one's own mind is something that cannot be inferred from the ideas themselves.

In spite of this fact, Locke never doubts that these ideas do signify something other than themselves, but he suggests very wisely that one should exercise considerable caution in determining just what it is that they do signify. While he admits that they do signify something, he is equally sure that they do not tell us as much as people in general suppose that they do.

The agreement or disagreement of ideas with one another makes possible four different kinds of knowledge. The first one is that of identity or diversity. It consists in that activity of the mind which distinguishes one idea from another. It is what enables one to know that black is not white or that what is round is not square. The second kind is that of the immediate discernment of the relations between ideas. This type of knowledge is illustrated in the field of mathematics when, for example, we perceive that two plus two equals four. The third kind consists in the observation of the coexistence or non-coexistence of certain qualities in the same object. It is in this manner that we come to associate certain sense qualities with a particular substance, as we do in the case of gold, lead, wood, or any other substance. The fourth kind is one's knowledge of real existence or the correspondence between the ideas in one's mind and the objects to which they refer in the outside world. All that can be called knowledge belongs to one or more of these four kinds, but it must be recognized that no one of these types of knowledge is necessarily present in one's consciousness at any given moment. Knowledge may be either actual or habitual. It is actual when one is conscious of it, and it is habitual when it is preserved in one's memory.

Locke maintains that there are different degrees of knowledge, and these can be distinguished by the extent to which they provide certainty in the realm of one's beliefs. They are also distinguished by the way in which the beliefs are obtained. Some ideas are known intuitively. Some are known through a process of demonstration. Still others are known through sense perception.

Of these three ways of knowing, it is intuitive knowledge that provides the highest degree of certainty. Demonstrative knowledge comes next, and sensitive knowledge has the lowest degree of certainty, which amounts to no more than probability. There are, however, different degrees of probability which may be determined on the basis of past experiences, which include that of other persons as well as one's own. When the probability is high enough to meet certain tests which have been established, we may say that the beliefs which are supported by it may rightfully be classified as knowledge.

Among the things which are known intuitively, nothing can be more certain than the perception of agreement or disagreement of ideas with one another. It is in this way that we know with certainty that one object is not identical with another object. In the same way, we know that the whole of anything is greater than any one of its parts. By the same token, we can be certain of our own existence. What is known intuitively is that of which we are immediately aware without the intervention of any proof or the invoking of any outside authority.

Demonstrative knowledge differs from intuitive knowledge inasmuch as it requires a medium by means of which one idea can be compared with another. It is in this connection that we speak of the proof which supports a given conclusion. For example, it is possible to prove that the sum of the angles of a triangle is equal to two right angles. In this case, the connection between premises and conclusion is not something that is evident at first sight or prior to any reasoning process.

In the case of sensitive knowledge, which forms the basis for all of our scientific investigations, we must rely on past experiences for any knowledge about what will happen in the future So far as the logic of the position is concerned, we must admit that what has happened in the past does not tell us anything about what will occur in the future. Nevertheless, Locke maintains that the fact that events have occurred in a certain way so many times in the past makes it highly probable that they will continue with the same sequence in the future. It is this high degree of probability that constitutes the basis for what may be called sensitive knowledge.

The extent to which human knowledge is reliable is determined by the nature and the number of the experiences on which it is based. Only that which has been experienced either directly or else demonstrated by means of inferences drawn from these experiences can be regarded as genuine knowledge. Concerning that which lies beyond experiences, we can know nothing at all. Metaphysicians who write about the nature of the universe as a whole, or who think they can tell us something about the world of spirits, or the character of the substances out of which the universe is composed are only fooling themselves as well as the persons who accept their judgments as being true. Their views can be nothing more than idle speculations so long as there is no sensory experience by means of which their conclusions may be either confirmed or denied.

Because the human mind is limited by the number and kinds of experiences that are possible, the scope of one's knowledge is relatively small. No individual can ever hope to know more than a small fraction of all that exists. Recognition of this fact should have a wholesome effect on one's attitude both with reference to himself and his judgment of those who may happen to disagree with him. It should encourage humility in regard to himself and a spirit of tolerance and charity toward the opinions of others. Although the scope of one's knowledge is necessarily limited, this is no reason for complaint since the possibilities for knowledge that are open to him are sufficient for his needs.

One of the most important items in any epistemological theory is the conceptions of truth and of error that are implied in it. In a strictly deterministic system which allows for no freedom of choice, neither truth nor error is a logical possibility. In a system that does allow for alternative choices, truth and error are possibilities provided there is some fixed standard in comparison with which one's ideas and beliefs may be judged. On the basis of Locke's empiricism, this standard cannot be things in themselves, for there is no way of knowing the nature of things apart from the way in which they affect human minds.

Since Locke does believe that human judgments may be either true or false, he finds it necessary to develop a different standard for determining the validity of one's beliefs. The standard which he proposes is that of the agreement or disagreement of ideas with one another. Truth, as he understands it, is not something that exists eternally, nor is it something that is independent of one's thinking. The truth or the error of which we speak is something that comes into being when we begin to make propositions about things. The truth of a proposition consists in "the joining or separating of signs, as the things signified by them do agree or disagree with one another." The reverse of this statement is what is meant by error or falsehood.

The use of universal propositions in which affirmations are made that go beyond the boundaries of past experience presents something of a problem for an empirical epistemology. When, for example, we say that "all men are mortal," we are asserting not only that people have died in the past but that the same will be true of people for all time to come. Since this statement includes the future as well as the past, we may ask how it is possible for anyone to know that it is a true statement?

Locke's answer to this question is significant because of the way in which it illustrates his main position. He says, in substance, that whether this proposition is true depends on what one means when he speaks about humanity and mortality. If he has in mind some fixed and unchanging essence of humanity that exists independently of our thinking about it, then there is no basis for accepting the proposition as true. Whether an essence of this type really exists or not is something that cannot be known. Even if it does exist, we cannot know what it is like or how it will behave in the future. All that we do know in this connection is what takes place in our own minds. It is true that we may regard our sensations as being the result or effect of some external force, but how or why it acts that way is beyond the limitations of our minds.

This does not imply, as one might think, that universal propositions are meaningless or that we have no way of determining whether they are true. What we need to recognize is that when we use a proposition such as "all men are mortal" in its proper sense, we are referring not to an external essence but to a definition which has been formed in our own mind. The proposition will be true because we have defined man as a mortal being and if he were not mortal he would not be a man. In other words, what this means is that universal propositions are true only when they are tautologies.

This accounts for the fact that in the field of mathematics, the propositions which are used are not only universal but are also true. The multiplication table, for example, is true for all people and for all time. The reason, of course, is the fact that in mathematics we are dealing with ideas and not with things. Propositions of this kind illustrate what Locke meant when he said that truth is the agreement of ideas with one another.

This explanation is sufficient to enable us to understand the validity of universal propositions that occur in the formal sciences such as logic and mathematics or any other area in which the propositions used are of an analytic nature. The situation is somewhat different in the inductive sciences where universal statements are used in reference to the laws of nature which enable us to make predictions about what will happen under a given set of circumstances. How, for example, can we know that the law of gravity will hold true for the movements of falling bodies in the future? Locke's answer is that we really do not know. All that we can say with certainty is that on the basis of past experiences, there is a strong probability that the same sequence of events will occur in the future.

The so-called laws of nature are not fixed and unchanging entities that belong to the external world. They are only generalizations which have been formed by human minds by a process of abstracting from a series of events certain elements that they have had in common. Once this abstraction has been made, it becomes the basis on which predictions are made with reference to future occurrences of the same kind. This is the reason why Locke maintains that in the areas of the physical and the natural sciences, universal statements do not yield certainty but only varying degrees of probability.

Because the knowledge that we obtain by intuition yields the highest degree of certainty, there is nothing of which we can be more sure than the fact of our own existence. Locke tells us that we perceive it so plainly and so certainly that there is no need for proof. With reference to the existence of an external world and of the existence of God, we have a very high degree of certainty, although it is somewhat less than the certainty we have of our own existence. In both of these instances, we have an example of demonstrative rather than intuitive knowledge. This means that the basis for our belief is an inference that has been drawn from ideas that have been established with certainty. We infer that an external world exists since there must be something which causes the sensations we experience.

Likewise, Locke proposes, we infer the existence of God as the cause of the universe and as the author of the moral principles by which human conduct ought to be regulated. The existence of God, he holds, is logically necessary since it is impossible for something to come from nothing. All finite beings exist, and in order to avoid the idea that they came from nothing, we must believe in a being that is eternal rather than temporal and subject to all the characteristics of finitude. Whether God is a material being or only a thinking being is a matter that Locke does not discuss, but he does insist that he is a thinking being since thought cannot be a product of matter.

Inasmuch as our sensory experiences can tell us nothing about the nature of deity or the moral principles which are derived from that source, Locke finds it necessary to rely on revelation for our knowledge about God and the basis for determining what is right or wrong conduct. He recognizes, however, that in so doing there are several precautions which need to be observed. Revealed truth may very easily be confused with blind superstition or wishful thinking. To avoid these errors, some criteria must be established for determining what is a genuine revelation of truth. If revelation is to be relied upon at all as a source of truth, it must be capable of imparting to human beings at least some things that go beyond the limitations of their finite minds.

But while revelation may go beyond human reason, it cannot be contrary to reason. It is quite possible for us to accept by faith truths which are non-rational in the sense that they are not derived by reason, but we cannot accept as true anything that is irrational in the sense that it is either self-contradictory or contradicted by any known facts. The ideas that are obtained by means of faith and revelation do not have the same degree of certainty as intuitive or demonstrative knowledge, but they do go beyond the type of probability that we find in the inductive sciences. Revelation must always be tested by the criterion of reasonableness in relation to the knowledge that has been obtained in other ways. In no case can we accept revelation if it is contradictory to our clear intuitive knowledge.

Does this mean that all belief in miracles must be rejected? It would mean this if we held that the laws of nature are fixed and unchanging patterns which are embedded in the structure of the universe. But if the laws of nature are nothing more than generalizations or abstractions formed in human minds, this conclusion would not follow. We must, according to Locke's view, admit that miracles are possible and that they are supported by the testimony of those who produced the writings that are regarded as revelation. Our only basis for rejecting any of the miracle stories reported in the Bible would be some concrete evidence that the actual events which took place are not described accurately. Since we cannot reproduce the events in order to check the accuracy of the reports, we have no right to deny that they happened. On the other hand, we cannot affirm with certainty that they happened in exactly the way they have been described. In matters of faith which include the acceptance of revelation, we must recognize that reason is, and must be, the final judge with reference to any of our beliefs.

Although Locke is not a rationalist in his philosophical position, he believes that reason has a very important place in the acquisition of human knowledge. Its function is primarily that of testing the validity of ideas which may be derived from different sources. Nothing can be regarded as true unless it meets the demands of reasonableness. Reason is not, however, to be interpreted in the narrow sense that identifies its use with purely deductive processes. It may include the use of the syllogism, but it also involves a great deal more. In this respect, Locke writes, "But God has not been so sparing to men to make them barely two-legged creatures, and left it to Aristotle to make them rational." It is not only possible but a very common occurrence for people to be reasonable without having had any instruction in the use of the syllogism.

The proper use of one's rational faculties will provide a safeguard against the errors that stand in the way of achieving genuine knowledge. Among these errors we may observe such items as the following: assenting to propositions without stopping to examine the evidence on which they are based; making a wrong use of the evidence that we do have by refusing to give careful consideration to what is really implied by it; being swayed by non-rational considerations or, in other words, being guided by our feelings and emotions rather than logical inferences; and refusing to be convinced by unwelcome views which are contrary to our wishes and desires. To be aware of these tendencies toward error is the first step toward overcoming them. We may be confident that in the long run truth will win over error.


The conclusions that Locke reached as the result of his long and painstaking investigations of the knowing process are set forth in this portion of the Essay. Book IV contains twenty-one chapters, and these are devoted to the task of making clear what he believes to be true about the nature and extent of human knowledge. Mainly, he has tried to follow the implications of the method and the presuppositions that were stated in the earlier portions of his work, but when these led to beliefs that he could not accept, he abandoned the ideal of consistency and included in his theory of knowledge a number of ideas that he held to be true. For doing this he has often been taken to task by his critics, who have insisted that any adequate system of thought must at least be consistent with itself.

Without doubt the critics have been right in pointing out these particular defects in Locke's arguments, for it must be admitted that ideas and beliefs can never be entirely correct so long as they are inconsistent either with themselves or with any known facts. On the other hand, it must be recognized that ideas, and indeed whole systems of thought, may be entirely consistent and still be false. It is seldom if ever that human beings are completely consistent in their views about themselves or the world around them, and who is there to say that it is more important to be consistent than it is to be right? Most people who read Locke's Essay will agree that many of his conclusions were right even though they are not consistent with some of the positions on which they were supposed to be based.

It would be a mistake to overlook the greatness of Locke's work because of some of the defects contained in it. The whole truth about human knowledge or any other subject of equal importance is more than can be grasped by any human mind, but this does not mean that it is impossible to gain valuable insights with reference to the topic under consideration. For bringing to light so many of these insights into the nature of the knowing process and the methods by means of which it may be achieved, the modern world has been greatly indebted to the work of this pioneer in the field of epistemology.

By defining knowledge as the agreement or disagreement of ideas with one another, Locke has called attention to at least one important aspect of the knowing process. Sensation and reflection are the sources from which many of our ideas are obtained, and comparison of these ideas with one another is one method of determining whether they are true even though it may not be the only way in which this may be done. It is the method that is used successfully in the field of mathematics or any other science of a purely formal nature. But if this were the only way of determining what is true in any area of investigation, we could never know anything about an outside world. Our knowledge would, under these conditions, be limited to whatever takes place in our own minds. Because this type of conclusion was wholly unacceptable to Locke, he did not follow the implications of his definition of knowledge, but instead he insisted that true ideas are in agreement with the qualities which are present in things.

Thus Locke maintained that two of the four kinds of knowledge which are possible on the basis of one's perception of the agreement or disagreement of ideas imply something more than that which is immediately present in the ideas themselves. Memory is said to be the storehouse for knowledge that is not always present in one's consciousness. It can be seen quite readily that his statement implies a continuing self which does the remembering and which recognizes that it is a memory rather than an experience that is entirely new. There is no sensory evidence for the existence of selfhood which memory seems to imply, although the logic of the situation apparently compels us to believe in it. Again, Locke's belief in the existence of a self was probably correct, but certainly it could not be derived from the sensations themselves nor from any reflections that are based solely upon them.

What is said in the Essay concerning the degrees of knowledge has considerable merit, for nothing is more obvious than the fact that some of our beliefs are more certain than others. Locke maintains, for example, that there are different degrees of certainty in what we believe about our own existence, the existence of God, and the phenomena of nature. These degrees of certainty correspond to the ways in which the knowledge is obtained. He tells us that we have intuitive knowledge of our own existence, demonstrative knowledge of the existence of God, and only probable knowledge of that which the senses reveal to us concerning the phenomena of nature.

Although there appears to be some truth in each of these assertions, it requires only a very simple analysis to point out some of the inadequacies connected with each of them. Intuition does yield the highest degree of certainty that is possible with reference to some things. It is the supreme authority about what is present in our own states of consciousness. It tells us whether we are experiencing pain, or seeing yellow, or red, at any given moment. We can be equally sure about the content of any momentary state of consciousness. But to say, as Locke does say, that we know intuitively of the existence of a self that is continuous over a relatively long period of time is to assert something more than is warranted in the bare facts of the case.

Demonstrative knowledge does yield a high degree of certainty, provided it is based on true premises and one makes no errors in his reasoning. Locke's argument for the existence of God, as well as his argument for the existence of an outside world, is based on the premise that whatever exists must have been caused by something other than itself. Now this is undoubtedly true so far as the world of our experience is concerned, but this does not warrant the belief that the final cause of events is something that lies outside of the realm of human experience. If causality is to mean anything more than a succession of ideas in one's mind — and apparently it does mean something more than that for Locke — there is no sensory experience which provides any basis for believing that things are caused at all.

To be sure, demonstrative knowledge can yield certainty in the field of mathematics, in which all the propositions used are analytic in nature, but it does not yield the same results when applied to the phenomena of nature or to the existence of God. Locke believed in the existence of God, but he was evidently mistaken in his belief that it could be demonstrated or proved by logical deduction from the facts of human experience.

Sense knowledge is what we must rely on for any information we have concerning the phenomena of nature, or what is usually referred to as the area of the physical and natural sciences. It is in this area that Locke is especially concerned to point out the limitations of the human mind. Since all of our beliefs about the external world must be formed on the basis of that which takes place in our own minds, it is obvious that we cannot know the real nature of things-in-themselves. Anyone who recognizes this fact will be more apt to avoid making dogmatic assertions about the character of the universe when viewed as a whole. It is for this reason that Locke is skeptical concerning the validity of any metaphysical speculations. Those who claim to know that reality is composed solely of matter, or that ideas are the stuff out of which all things are made, are in his judgment merely indulging in idle speculations since there is no concrete evidence that will either confirm or deny the truth of what they are saying.

Even though Locke's epistemology appears to exclude all metaphysical assertions, he did not wish to eliminate the belief that we do have some genuine knowledge in the fields of the physical and natural sciences. This was possible on the basis of his theory, provided that we do not claim too much for the knowing processes in these fields of investigation. Scientific knowledge, he maintained, does not reveal the independent nature of external objects. It only reveals the way in which things appear to human minds.

It is true that the scientist believes she can make accurate predictions with reference to the way events will take place in the future under specified conditions. However, the only basis for these predictions is the fact that events have taken place a certain way in the past. This, in itself, provides no guarantee at all concerning the future. Locke admits that this is true, and, accordingly, he insists that, while we do not have any certain knowledge about future events, we can have probable knowledge of when and how they will take place. The degree of probable knowledge will in each case depend on the number and character of the events that have occurred in the past.

It seems appropriate in this connection to inquire just what Locke means by probable knowledge when applied to the predictions that scientists make on the basis of the so-called laws of nature. If probability means something more than an ordinary guess in which the chances of it being correct are equal to those of its being incorrect, then there must be something in the order of nature that governs the way in which events will take place. If we cannot know anything about the order of nature, as Locke has insisted in his former arguments, then there is no basis for asserting that we have even probable knowledge about future events. This is another instance in which Locke's beliefs may be correct even though from a logical point of view they are inconsistent with the views on which they are supposed to be based.

The same type of criticism can be urged with reference to his attitude toward metaphysics. Having excluded all metaphysical speculations as idle and a waste of time, he goes on to affirm his belief in the existence of God, the moral order of the universe, the reality of both mind and matter, the reality of causal relations, and the possibility of probable knowledge in the field of the natural sciences. All of these beliefs are at least to some extent of a metaphysical nature. If what Locke had to say about metaphysical speculations is interpreted to mean that no one can have complete and exact knowledge about the entire universe, it would be impossible to take issue with him, nor would there be any reason to do so. But this is not what Locke meant in his criticism of metaphysics. What he did mean is that it is impossible for anyone to have any knowledge about that which goes beyond the limits of human experience. Furthermore, he held that no one has the right to accept as true any beliefs concerning ultimate reality or the character of the universe as a whole.

Apparently Locke was not aware of the fact that any theory of knowledge, including his own, necessarily presupposes something about the ultimate nature of things. One may be critical of any particular metaphysical theory but cannot avoid having one of his own. The fact that no one's beliefs about the nature of the universe will be complete or exact in every respect does not imply that we can know nothing about it at all. We do have some principles that may serve as guides toward a correct view of things. We know that our beliefs in order to be true must be consistent with themselves, in harmony with all known facts, and they must provide us with a reasonable interpretation of our experiences. When these conditions have been met, it can be said that we have what may rightfully be called some degree of probable knowledge.

Back to Top

Take the Quiz

According to Locke, why can't ideas be present in a soul before it is united with a body?