Complex Ideas in John Locke’s ‘Essay Concerning Human Understanding’

Portrait of John Locke, by Sir Godfrey Kneller, 1697 / State Hermitage Museum, St. Petersburg, Russia

By Dr. Garth Kemerling / 11.12.2011
Professor of Philosophy
Capella University

Philosophy Pages

Although he granted that some groups of simple ideas naturally occur together in our experience, Locke supposed that most of our complex ideas are manufactured in the human mind by the application of its higher powers. Combining joins several simple ideas together in the formation of a new whole; comparing brings two distinct ideas together without uniting them, giving rise to the idea of a relation between them; and abstracting separates some aspect of an idea from its specific circumstances in order to form a new general idea. Repeated applications of these powers, Locke supposed, give rise to the whole variety of ideas human beings are capable of having. [Essay II xii 1-2] If we’re going to analyze the epistemic origins our complex ideas, it will be helpful to consider the ways in which we gradually build up our supply of them.

On Locke’s view, complex ideas are of three varieties: Modes are invariably conceived as the features of something else, which are never capable of existing independently. Substances, on the other hand, are understood to be the existing things in which modes inhere. Relations are nothing more than mental comparisons in some respect among other ideas. [Essay II xii 3-7] Complex ideas of all three sorts are manufactured by the mind from simpler components.


A simple mode is a complex idea all of whose component parts are variations or combinations of a single simple idea. [Essay II xii 4-5] Consider, for example, the simple idea of space: acquired initially from our senses of sight and touch, this idea provides the sole content for a host of related ideas of our own manufacture. The notion of one-dimensional length can be “folded” upon itself to yield those of area in two dimensions and capacity in three; these notions, in turn, can be modified more subtly to provide our ideas of shapes and figures. Even the notion of place in relation to other bodies within a framework Locke believed to be abstracted from the simple idea of space. [Essay II xiii 2-6] In similar fashion, Locke supposed that the simple idea of unity, repeatedly recombined with itself, provides the entire content for the simple modes of number, including even that of infinity. [Essay II xvi – xvii] In every such case, the complexity of the simple mode arises only from an iteration of a single simple idea, upon whose content the complex idea therefore relies.

Mixed modes, on the other hand, are complex ideas whose components include several distinct simple ideas, often including those derived from different experiential sources. Although such ideas can be acquired through observation of their instances in our experience, Locke supposed that they are much more commonly manufactured by the mind as complex ideas before we first apply them to the world. [Essay II xxii 1-2] This is an important feature of our complex ideas of mixed modes. Among them are included the conceptions we form of human activities, which typically involve not only ideas of sensation describing the overt action but also ideas of reflection regarding the intention with which it is undertaken.

For Action being the great business of Mankind, and the whole matter about which all Laws are conversant, it is no wonder, that the several Modes of Thinking and Motion, should be taken notice of, the Ideas of them observed, and laid up in the memory, and have Names assigned to them; without which, Laws could be but ill made, or Vice and Disorder repressed. Nor could any Communication be well had amongst Men, without such complex Ideas, with Names to them: and therefore Men have setled Names, and supposed setled Ideas in their Minds, of modes of Actions distinguished by their Causes, Means, Objects, Ends, Instruments, Time, Place, and other circumstances; and also of their Powers fitted for those Actions. [Essay II xxii 10]

The idea of stabbing, for example, typically involves a deliberate act of penetrating flesh without any presumption about the instrument employed to do so. In such cases, it may be worthwhile for legislative bodies to conceive of such actions and condemn them as criminal without having to have experiential evidence of their first occurrence. [Essay III v 2-6] The only thing that matters for the formation of a mixed mode, on Locke’s view, is the convenience of its use for us. This clearly varies among distinct cultures, since the frequency with which we have occasion to notice, denominate, and evaluate actions of particular sorts often depends upon their social context. Invented wholly for our own use in matters of moral and legal import, the ideas of mixed modes have only that specific content we choose to give them. [Essay II xxii 5-12] As we’ll see next time, the freedom with which such ideas are manufactured renders problematic our ability to use a common vocabulary in reference to them.


Complex ideas of relations derive from the mental operation of comparing distinct ideas without thereby combining them together into an entirely new whole. [Essay II xii 1, 7] Since this comparison considers each of the ideas in light of its extrinsic conformity with the other, it is commonly expressed by means of one of a pair of reciprocal relative terms—”parent”/”child” or “cause”/”effect,” for example. This is not invariably the case, however, so we must be alert to the possibility that apparently positive terms implicitly signify comparison with something else: “old” or “imperfect,” for example. The tip-off to such cases, on Locke’s view, is that (as in the case of relations of the other sort) removal of the correlative destroys the relation completely, rendering the term inapplicable. [Essay II xxv]

Many of our ideas of relations are acquired through sensory experience of the modes of time, place, and number. The correlative ideas of cause and effect are a special instance, on Locke’s view: observing the changes produced in one thing by the operation of another, we form a sensitive notion of the causal relation even though we have no conception of the underlying mechanism at all. [Essay II xxvi 1-6] Because language is devised for our convenience in satisfying the needs of ordinary life, things that are commonly connected in observation give rise to our complex ideas of natural relations. But we also conceive of instituted or voluntary relations between things that are corrleated with each other only by virtue of our own personal or social agreements. Among these, on Locke’s view, the most important and vital are the moral relations drawn between the complex ideas of specific human actions, together with their circumstances and goals, and the moral rules by reference to which we evaluate them. [Essay II xxviii 1-4] We’ll look at these much more closely later on.


According to Locke, the complex idea of a substance is a collection of simple ideas that is believed capable of existing independently. Observing in experience that several features recur together frequently, we suppose that there must be some common subject that has all of them. Thus, for Locke, the “Notion of pure Substance in general” is nothing but the assumption of an unknown support for a group of qualities that produce simple ideas in us; our only notion of this putative “substratum” in itself is the confused notion of “that which” has these features, or “something, I know not what.” [Essay II xxiii 1-3] Locke consistently ridiculed the use of such scholastic terms as “Inhaerentia” and “Substantia” as pointlessly circular efforts to provide a positive idea of something for which, in fact, “we have no Idea of what it is, but only a confused obscure one of what it does.” From this, he supposed, we can infer little of the metaphysical nature of reality. [Essay II xxiii 17-20] Nevertheless, just as we combine simple ideas of qualities and powers to form the complex ideas of individual substances, we also gather many distinct things under the even more complex ideas of collective substances—army, world, and universe. [Essay II xxiv 1-3]

To the confused idea of substance in general, Locke held, we add either the ideas of regularly observed sets of sensory qualities, in order to form the notion of a bodily substance, or the reflective ideas of various kinds of mental operation, in order to form the notion of a spirit or thinking thing. Since the underlying nature of the substrate is equally obscure in both cases, the “secret and abstract Nature of Substance in general” is equally unknown to us for substances of either sort. [Essay II xxiii 4-6] We do have clear and distinct ideas of the primary qualities of both bodies (solidity and impulse) and mind (thinking and motivity) even though our sensation and reflection offer no insight into “the internal Constitution, and true Nature of things,” so that the relation and/or independence of bodies and minds remains an open question. [Essay II xxiii 29-32] Thus Locke argued that even though we are often more familiar with material objects than with spirits, our ideas of them may be no clearer, and the customary prejudice in favor of bodily substance is philosophically indefensible. [Essay II xxiii 15-16]

With respect to bodily substances, the most accurate complex idea we could form, on Locke’s view, would be one that includes the ideas of those active powers and passive capacities that it exhibits in its interaction with other things—the magnetism of the lodestone and the flammability of wood—along with its characteristic weight, color, or heat. Since he believed the primary qualities of bodies to reside in the unobservable texture of their sub-microscopic parts, Locke found it natural to suppose that we classify distinct kinds of body by reference to the secondary and tertiary qualities these unknown features produce in us or other things. [Essay II xxiii 7-10] To express one of Locke’s favorite examples in our own idiom, we were able to identify genuine pieces of gold for centuries before we had any clue to its atomic structure.

Representational Reliability

Several late chapters of Book II are devoted to a detailed discussion of the success with which ideas of various kinds perform their representative functions. The point here is that since we use ideas as signs, it is vital to be aware of the likelihood that they do actually point beyond themselves to their intended referential objects. The extent to which they serve these functions will determine the reliability of any knowledge we try to acquire about those objects.

Locke explained the clarity of ideas by analogy to visual perception: just as an object is seen clearly when viewed in suitable light, a clear idea is one of which we have a “full and evident perception,” whose content is present before the mind. An idea is distinct, on the other hand, when it is perceived to differ from all others. [Essay II xxix 2-5] Locke had no stake in differentiating sharply between the clarity and distinctness of ideas. In the Essay‘s fourth edition, he made a half-hearted effort to substitute the single adjective “determined” in place of the customary pair “clear and distinct.” His central concern was with the failure of this representational function, in ideas that are confused. This occurs most frequently with respect to complex ideas, whose simple components may be too few or too poorly organized to determine their content precisely. The problem, Locke argued, is that we often use words as if we knew their significance when, in fact, the ideas associated with them are not fully conceived. [Essay II xxix 7-11] We’ll consider this issue more fully next time, but it’s worth noting that Locke believed that many apparently intractable philosophical disputes arise from a failure to employ words to signify clear and distinct ideas. His examples here all arise from the wide-spread failure to hold in mind a correctly-formed, determined idea in association with the word, “infinity.” [Essay II xxix 13-16]

In Locke’s taxonomy, and idea is said to be real (as opposed to fantastical) so long as there is something that it represents. Notice that the accuracy of representation is not at issue here at all. Even if the idea fails to correspond to its object, it is real provided only that the object does exist. Since simple ideas are passively received in the mind, for example, they must be caused by something real, even though—as we’ve already seen—in the case of secondary qualities they fail to resemble their causes. [Essay II xxx 1-2] Because our complex ideas of modes and relations are not supposed to refer to anything else beyond their own archetypal content, they are all real as well. Ideas of substances, however, are intended to refer to existing things and are therefore fantastical if things with the appropriate combinations of features do not in fact exist. [Essay II xxx 4-5]

An idea is further said to be adequate only if it represents its intended object fully and perfectly; inadequate ideas convey the nature of their objects only partially. [Essay II xxxi 1] Locke insisted that all simple ideas are adequate, though doing so required some fancy footwork with respect to our ideas of secondary qualities. Properly understood, he argued, simple ideas represent whatever power it is that produces them in us, whether or not the idea resembles that power—which, in the case of secondary qualities, it does not. As with the reality of ideas, so with their adequacy the vital point for Locke was the causal process by means of which we acquire them; our lack of voluntary control over that process forestalls any possibility of mistake or erroneous judgment. [Essay II xxxi 2, 12] Because complex ideas of modes and relations are assembled by the mind without reference to any external archetype, they are their own archetypes, which they cannot fail to capture adequately. Although our communication with each other about such ideas (especially in the case of mixed modes) may falter because we do not agree about the signification of words we both employ, the ideas themselves are invariably adequate. [Essay II xxxi 3-5, 14] Once again, it is complex ideas of substances that are unreliable; such ideas, according to Locke, have a double intended reference but are inadequate in both respects. If they are supposed to represent the substantial forms of existing things, our ideas of substances are inadequate because (on the corpuscularian theory) these real essences are unknowable. Even when considered more modestly, as collections of properties that co-exist in a common substrate, our complex ideas of substances are merely partial approximations, since it is clear that they include only those features we have most commonly observed; it always remains possible for us to be surprised by the discovery of another property that belongs just as surely in the same being. [Essay II xxxi 6-11] As always, ideas derived from experience can be no more adequate than the experience itself, and with respect to natural things our experience is limited.

Although he maintained that truth and falsity are most properly regarded as characteristic of propositions, Locke granted that ideas are sometimes said to be true or false (better: right or wrong) by virtue of the role they play in the formation or assertion of such propositions. [Essay II xxxii 1-5] It is the relation between ideas and the words used to signify them that matters for this representational criterion. Because of their familiarity in our experience and the frequency of our discourse about them, our ideas of natural substances and their qualities are often truly signified by the common terms we employ to designate them. [Essay II xxxii 6-10] It is the ideas of mixed modes, manufactured independently in the minds of individual thinkers and lacking any external referent, that are most commonly false in this linguistic sense, because we do not agree on the signification of their names. [Essay II xxxii 11-13]

On the whole, then, Locke believed that ideas provide a vital but imperfect foundation for human discourse and knowledge. As we continue our study with an examination of his philosophy of language and of knowledge, we’ll be reminded of the limitations imposed by the reliability of ideas of each type.