The word "idea" is a transliteration of a Greek word of which the root meaning is "see." In classical Greek it never lost the possible meaning "visual aspect"; thus Plato writes of a person as being "very beautiful in idea," meaning "beautiful in visual aspect" or "good-looking" (Protagoras 315e). Very often visual shape is primarily involved, as when Plato refers to the "idea of the earth," meaning "the visible shape of the earth" (Phaedo 108d). The transferred sense of "type" or "kind" springs quite naturally from this use. Thus Thucydides writes of "many ideas [kinds] of warfare" (Histories I, 109).
In Plato's more technical use, the Ideas or Forms are always spoken of as (1) the objects of intelligence, in contrast with the objects of perception; (2) things that truly are, in contrast with changing objects of perception, which are in a state of becoming ; (3) eternal, in contrast with the perishable world of change. But there are at least two irreconcilable strands in Plato's thought about Ideas. Sometimes he seems to have thought of Ideas much as later philosophers have thought of universals, as when he says that "we are accustomed to posit a single form for each group of many things to which we give the same name" (Republic 596a); consistent with this he speaks sometimes of the presence of the form in the particular or of the particulars as participating in the form (Phaedo 100d). But sometimes Plato writes as if his Forms were, rather, perfect exemplars or paradigms of which the sensible world is an imperfect copy or imitation; thus in the Parmenides Socrates says that the Forms are "as it were paradigms" and that "other things are like them and are copies of them" (132d). When the Forms are thus described, we also find Plato insisting that they are "separate," a doctrine in conflict with the language of "presence" and "participation" noted above. It is plausible to suggest that there is here a tension between the theory of universals and the theory of resemblance to standard objects as explanations of common names.
But it is the theory of Ideas as separate and eternal paradigms that appears in the Timaeus, the dialogue that had incomparably the greatest influence on later antiquity and the Middle Ages; there the divine demiurge is depicted as forming the world on the pattern of the eternal Forms. It will therefore be the aspect of Forms as paradigms, perfect exemplars, blueprints, particularly as patterns used by a divine agent in creation, which will be important in the development of the philosophical notion of an idea.
In the Timaeus the Forms, or Ideas, are eternal and independent objects to which the demiurge looks as patterns. But one of the most important and early modifications of this Platonic view is the religious conception of the Ideas as the thoughts of God. This is the view of Plotinus (Ennead III, 9, i), of Philo (De Opificio Mundi 4) and of Augustine (De Diversis Quaestionibus LXXXIII, Question 46). Clement of Alexandria simply defines an idea as a "thought of God" (Stromateis V, iii, 16.3). The ideas are still perfect and eternal exemplars, but now they are in the mind of God.
It is not a very long step to extend the term idea to cover patterns, blueprints, or plans in anybody's mind, not only in God's. Thus we find Thomas Aquinas saying that "the word 'idea' signifies a certain form thought of by an agent in the likeness of which he intends to produce an external work" (Quaestiones Quodlibetales IV, I, lc); similarly Goclenius says that "in general an idea is a form or exemplar of a thing with an eye on which a workman makes what he has planned in his mind" (Lexicon Philosophicum 208a).
When the word idea was taken over into the French and English vernacular by learned men in the sixteenth century, there were thus two elements in the concept of an idea—that it was an exemplar or pattern and that it was a thought in a mind. Using the pattern element alone, François Rabelais could speak of Pantagruel as being the "idea and exemplar of every joyous perfection" (Pantagruel, Book III, Ch. 51); but a pattern and its copy could be easily muddled so that Rabelais also could say, "En leur mariage semble reluire quelque idée et représentation des joyes de paradis" ("In their marriage some idea and representation of the joys of paradise seems to be reflected"; Pantagruel, Book III, Ch. 10). When the other, mental element is introduced, the meaning of "idea" quickly becomes "mental representation"; this is a very common meaning in sixteenth-century French and English, and the phrase of Michel Eyquem de Montaigne, "Ayant par longue conversation planté vivement dans son âme unc générale idée de celle de Plutarque" ("Having by long communion vividly emplanted in his own a general idea of the mind of Plutarch"; Essays, II, 4), could be paralleled many times.
Thus when René Descartes first began to write, the meaning "image or representation," often but not necessarily "in the mind," was already well known in the vernacular. In spite of the fact that Descartes is usually credited with the invention of the non-Platonic use of the term, we find him at first following this vernacular use. In his first Latin work, the Regulae, the word idea appears infrequently, but Descartes always uses it to mean an image or representation; when he first introduces it in the Meditations, he at once says, "Quelques-unes [de mes pensées] sont comme les images des choses, et c'est à celles-là seules que convient proprement le nom d'idée" ("Some [of my thoughts] are like images of things, and it is to these alone that the name 'idea' properly belongs"). It is only under the pressure of philosophical difficulties that he extends the term idea to cover the unimaginable, for which Thomas Hobbes duly reprimanded him: "When I think of a man I represent to myself an idea or image composed of colour and shape. … of God we have no image or idea" (The Third Set of Objections, Objection 5). There is therefore no need for any explanation why the word idea tends to mean "mental image" to seventeenth-century philosophers; this is what the word ordinarily meant in their time.
What does need explanation is why, if Descartes found the word idea to mean "properly" only "an image of a thing," he and other philosophers could use "having an idea" as a proper designation of all thought and could define an idea as the object of a mind when it thinks, in a liberal sense of "think" that includes sense perception. Part of the explanation is to be found in the representative theory of perception, held in some form by all the philosophers of the period; there was no extension of meaning in using "idea" of sense perception because it was believed that what was directly perceived was not things, but images of things—the images caused by and more or less resembling the things themselves. Another part of the explanation is the "image theory" of thinking: To think of something is or includes having either a mental image of that thing or, as some believed, a physical image on that part of the brain termed the "corporeal phantasy." Such a view was in the air at the beginning of the seventeenth century and was accepted by Pierre Gassendi and Hobbes without reservation. Descartes never doubted that many of our thoughts are images of things; his extension of the term arises from his gradual realization of the inadequacy of the image theory to account for all our thought even while he persevered in the use of its terminology. His use of the term to denote any object of thought became the standard one in philosophy, via such influential writings as the Port-Royal Logic and John Locke's Essay. Only a few scholastically trained philosophers, such as Immanuel Kant, have stood out for a more Platonic usage; thus Kant in the Critique of Pure Reason holds to the terminology of the transcendental ideas of reason to which no corresponding object can be perceptually given, as distinct from the concepts of the understanding ("Transcendental Dialectic," I, 2).
Most of the confusions in the "way of ideas" arise at least in part from the use of the term idea to cover both the representative percept and the object of conceptual thought. This can be illustrated in terms of the doctrines of innate ideas, concrete and abstract ideas, and simple and complex ideas.
The mature Gottfried Wilhelm Leibniz always maintained, and Descartes sometimes maintained, that all our ideas are innate. Thus Leibniz said that "all the thoughts and acts of the soul come from its own depths, with no possibility of their being given to it by the senses" (New Essays concerning Human Understanding, Book I, Ch. i, 1). But this is a theory of perception, as is made clear by Descartes in his Notes Directed against a Certain Program, his defense of the view:
Nothing comes from external objects to our mind through the organs of sense save certain corporeal motions … but not even these motions, and the configurations to which they give rise, are conceived by us as they occur in the sense-organs. … Whence it follows that the very ideas of motions and configurations are innate in us. So much more must the ideas of pain, colours, sounds, and the like be innate, so that our mind can, on the occasion of certain corporeal motions, display them to itself; for they have no similarity to the corporeal motions.
There is nothing here from which Locke would dissent, except verbally; no wonder that Leibniz said, in the preface to his New Essays, "I am led to believe that at bottom his [Locke's] view upon this point is not different from mine." The true controversy with Locke is, rather, exhibited by Descartes's view of concepts, expressed in the same terms as and never distinguished from the perceptual theory by philosophers of the time. According to this theory, some ideas are innate—for example, those of God, mind, body; others are adventitious—one's ordinary idea of the sun; still others are made (factae ) or factitious—the ideas of the sun astronomers construct by reasoning. It is those innate and factitious ideas—which Descartes could as little say were occasioned by "corporeal motions" as Locke could say they were caused by "corporeal motions"—which raised a still-pressing difficulty.
Abstract and Concrete Ideas
The distinction between abstract and concrete ideas is virtually the distinction, misleadingly put, of concepts and percepts. The doctrine of abstract ideas was held by the Cartesians, and the best statement of it is to be found in Port-Royal Logic, Book I, Ch. 6. To have an abstract idea is to think of some feature or features of the perceptible without attending to other features that it has and that are as inseparable from it (except in thought) as are the length and breadth of a road. Locke took over the Port-Royal account of what abstraction was without change, even echoing its language, but tried to give a more thorough account of what it involved. He tried to give an account of abstraction in terms of a doctrine of simple and complex ideas, but by failing to distinguish thought and perception, he gives two incompatible accounts of this distinction. In Book III of the Essay he tells us that all ideas save those denoted by proper nouns are abstract. Of these some are indefinable; they are simple ideas. Others are definable; these are complex ideas. "The ideas first in the mind, it is evident, are those of particular things" (Essay, Book IV, Ch. vii, Sec. 9)—that is, we first perceive particular things; in thinking about them, we may come to form some very general ideas by omitting less interesting features and concentrating on those common to a whole group, which taken together form a complex idea; by further abstraction we can get to less and less complex ideas. It is clear that according to this view simple ideas involve the highest degree of abstraction. But in Book II we are told that the simple ideas enter the mind in perception simple and unmixed; they are objects of perception. Thus a theoretical analysis of the construction of concepts is inextricably confused with an atomistic doctrine of perception. If simple ideas are objects of perception and complex ideas are formed from them, then all abstract ideas ought to be imaginable, and George Berkeley's famous sneers about the abstract idea of a triangle have some justification. But neither Berkeley nor David Hume could emancipate himself from the basic confusion; this is true of Hume in spite of his famous distinction between ideas and impressions.
Thus the classical theory of ideas, which had held virtually undisputed sway in the seventeenth and eighteenth centuries among rationalists and empiricists, was based on the theories of representative perception and image-thinking. To continue to use the terminology after these theories had been abandoned as inadequate could lead only to confusion and a skepticism which, consistently developed, would be even more extreme than Hume's.
Reasonably, therefore, outside the empiricist tradition the term idea, as employed in the seventeenth and eighteenth centuries, soon ceased to appear in philosophical writings. Kant's representations have, indeed, some resemblance to ideas of sensation, and the thing-in-itself plays a part somewhat analogous to Locke's substratum. But there are important differences, and his concepts of the understanding are very far from being copies of representations. He does, indeed, use the term idea technically, but with a yet further removed significance. In the Critique of Pure Reason, he says: "I understand by 'Idea' a necessary concept of reason to which no corresponding object can be given in sensation" ("Transcendental Dialectic," I, 2). These ideas, such as that of the absolute unity of the subject, have, Kant holds, a valid regulative employment, but if we try to apply them to experience we become involved in metaphysical paralogisms. Insofar as the term continued to be used in Continental philosophy it was used, as by G. W. F. Hegel, in senses far removed from that in pre-Kantian philosophy.
But in British philosophy the terminology did not die an easy death. The empiricists could not abandon it, especially in their philosophical psychology in which the doctrine of the association of ideas continued to play the dominant role given to it by Hume. It was largely F. H. Bradley's polemic against psychologistic logic that finally led to the abandonment of the "way of ideas." But even Bradley, in the first chapter of his Logic, which is a locus classicus for the attack on psychologism, showed that he had not completely emancipated himself. He could still write that "the idea, in the sense of mental-image, is a sign of the idea in the sense of meaning," and added "without ideas no judgment," though in a note of 1922 he rejected these statements. By 1922 his own work and that of G. E. Moore had led to the elimination of the term idea from British philosophy, except as a part of nontechnical idiom.
In the United States, also, the term idea continued to have considerable currency. It was a key term in the pragmatism of Charles Sanders Peirce, William James, and John Dewey, reflecting the fact that they, too, were heirs to the empiricist tradition though not to Humean skepticism. They avoided this skepticism in part by wholly abandoning the image theory of thinking with which the terminology of ideas was traditionally linked. In Dewey's instrumentalism, ideas became tools for directing our activities, responses to sensation rather than sensations. They were tied to practical transactions. In calling the idea a law of action, Dewey reminds us rather of the definition given by Thomas Aquinas quoted earlier in this article than of the traditional empiricist position. But Peirce could still think, like Bradley, of ideas as psychological entities, as well as in terms of pragmatic epistemology; and in James also the pragmatic doctrine that our ideas of an object have to be explained in terms of the sensations we expect from it and the reactions we make toward it had still not been completely disentangled from a more traditional empiricism.
See also Augustine, St.; Berkeley, George; Bradley, Francis Herbert; Clement of Alexandria; Concepts; Descartes, René; Dewey, John; Empiricism; Gassendi, Pierre; Hegel, Georg Wilhelm Friedrich; Hobbes, Thomas; Hume, David; Innate Ideas; James, William; Kant, Immanuel; Leibniz, Gottfried Wilhelm; Locke, John; Montaigne, Michel Eyquem de; Moore, George Edward; Peirce, Charles Sanders; Philo Judaeus; Plato; Plotinus; Plutarch of Chaeronea; Psychologism; Rabelais, François; Socrates; Thinking; Thomas Aquinas, St.; Thucydides; Universals, A Historical Survey.
For critical and historical discussions, see Albert G. A. Balz, Idea and Essence in the Philosophies of Hobbes and Spinoza (New York: Columbia University Press, 1918); M. H. Carré, Realists and Nominalists (Oxford: Oxford University Press, 1946); E. Garin, La théorie de l'idée suivant l'école thomiste (Paris: Desclèe, de Brouwer, 1932); Paul Natorp, Platons Ideenlehre (Leipzig, 1930); W. D. Ross, Plato's Theory of Ideas (Oxford: Clarendon Press, 1951); G. J. Warnock, Berkeley (Harmondsworth, U.K., 1953); John W. Yolton, John Locke and the Way of Ideas (London: Oxford University Press, 1956).
For the nontechnical use of the term idea (idée ) in the sixteenth and seventeenth centuries, see The New English Dictionary and Huguet's Dictionnaire de la langue française du seizième siècle.
J. O. Urmson (1967)