Skip to main content

Primary and Secondary Qualities


The distinction between "primary and secondary qualities," first stated and thus named by Robert Boyle, received its classical formulation in John Locke's Essay. There Locke states that apart from ordinary causal properties or "powers," material objects possess five primary qualitiesextension (size), figure (shape), motion or rest, number, and solidity (impenetrability)and many secondary qualities, such as color, taste, smell, sound, and warmth or cold. This distinction was made in the context of representative realism; that is, it was presupposed that the qualities of objects are quite distinct from, and are in fact causes of, "ideas" (representations or sensa), which are the only immediate objects of sensory awareness. The basis of the distinction was twofold. First, perceived size, shape, motion, number, and solidity are ideas caused by and exactly resembling the corresponding primary qualities of objects; perceived color, taste, smell, sound, and so on are caused by, but do not resemble, the corresponding secondary qualities. Second, the primary qualities are inseparable from matter and are found in every part of it; the secondary qualities are not true qualities of matter but are merely powers in the objects to produce sensory effects in us by means of the primary qualities in their minute parts. Thus, red as experienced (idea or sensum) is the effect of the secondary quality red, which is merely the power possessed by a special texture or surface structure of the object to reflect certain light frequencies and to absorb others.

This formulation is rather clumsy, and since George Berkeley the custom has been to apply the first part of the distinction to the qualities of the ideas or sensa. The primary or spatiotemporal qualities of these data may then be said to characterize the object as well, for instance, the sensum is square and so is the object; but the secondary qualities are said not to characterize the object at all except in a derivative way, for instance, the sensa may be red and fragrant, but the object itself is intrinsically neither colored nor scented; it is red and fragrant only in the secondary sense that it causes the appropriate data of color and smell in the percipient. The doctrine is thus essentially the same as Locke's, but the language is slightly different. This second formulation will be used here.

Though Boyle and Locke invented and popularized the distinction and the terminology of primary and secondary qualities, the distinction dates back in principle to Democritus, who said that sweet and bitter, warm and cold, and color exist only by convention (νόμ ), and in truth there exist only the atoms and the void (Fr. 9, Diels and Kranz). The distinction was revived by Galileo Galilei and accepted by René Descartes, Isaac Newton, and others.

Arguments for the Distinction

relativity and measurement

The relativity argument is the most important one: Secondary qualities are affected by the condition of our sense organs and nervous system, by our distance from the object or its motion relative to us, by the lighting or by such intervening media as fog. Since secondary qualities thus vary according to, and depend for their nature on, factors quite external to the physical object, they cannot be intrinsic properties of it. This point was elaborated by Locke in various examples, two of which follow: (a ) If one takes three bowls of water, one judged hot, one judged cold, and one judged medium, and places one hand in the hot water and the other hand in the cold, and then transfers both hands to the middle bowl, the water in that bowl will feel hot to the hand that has been in the cold water and cold to the hand that has been in the hot water. But since it cannot be both hot and cold, hot and cold are therefore not intrinsic properties of the water. (b ) Marble is not colored in the dark; its color appears only in the light. But presence or absence of light cannot alter its real properties, so that the perceived color cannot be included among them.

If we grant the position of representative realism that hot, cold, and color, as experienced, are qualities of ideas or representations, then it is plausible to suppose on these grounds that they do not also characterize objects or resemble properties of objects. (Locke does not always make it clear that representative realism is to be presupposed). But this claim is apparently open to the insuperable objection, stated by Berkeley, that the primary qualities also vary: The object's apparent shape or size varies just as much as its color or sound. This would mean that shape and size as perceived do not characterize objects or resemble the actual properties of the object, thus subverting the whole basis of the distinction. That Locke did not see this may have been partly because he felt that he had to argue against the commonsense assumption that all sensible qualities characterize objects, and partly because the belief that primary qualities characterize all matter was apparently guaranteed by the physics of his day.

Although this objection is valid against Locke's position, it does not destroy the distinction between the primary and secondary qualities, which it is natural to recast and support by a revised relativity argument. This new point is that, in contrast with the secondary qualities, the main primary qualitiesshape, size, and motioncan all be measured (solidity cannot, but it is dubious anyhow, in that most physical objects, even atoms, are far from solid or impenetrable; number, whether there is one object or two, seems scarcely a quality at all; strictly also in the case of shape, what is measured are various dimensionsdiameters, angles, and so onof the object, and supporters of the distinction must maintain that these are the differentiae of the shape). A plate may look elliptical, but by measuring its diameters and seeing that they are equal, we can establish that it is round; one man may look taller than another, but their relative heights may be settled by measurement, as can the speed of objects relative to the earth. The measured size and shape of a plate may thus be held to characterize it, and the sensible size and shape may agree with and resemble them, so that one can say that size and shape (and motion) are primary. Nevertheless, only in favorable circumstances does a given primary sensible quality also characterize the object (for instance, both object and sensum are round); otherwise, there is only a projective relationship, as between elliptical sensum and round object.

Measurement is objective and does not vary significantly because it is an operation that depends on the coordination of a number of separate perceptions and that may be performed by a number of different persons. Consequently, variations due to the measurer on any particular occasion are compensated for and do not affect the final result, and the various actions confirm that one is not simply establishing the qualities of representations. Measurement also leads to conclusions regarding the dimensions and positions of objects in physical space that can be verified by further activities or operations, such as fitting the objects together, moving one's hand between them, rolling an object to confirm that it is round, and so on. By contrast, the variation found in the sensory qualities seems to be caused by their being simply the content of one single act of perception limited to one person at one time.

If all this is so, the list of primary qualities must be somewhat amended. Shape, size, and motion remain, but one should substitute mass for solidity. Temperature is more difficult: Since it can be measured, it seems at first primary. But what is measured is the property of causing expansion in fluid or metal; this property in no way resembles felt warmth, and in physical theory it is a form of energy. Hence, temperature should not be regarded as a separate primary quality. Material objects do, of course, possess many other propertiescausal and dispositional ones, for exampleas Locke realized by his doctrine of "powers," but part of the distinction is that only the primary ones are intrinsic (that is, possessed without reference to other objects) and that all such powers are ultimately due to patterns of primary qualities. Even so, the distinction would have difficulty in coping with some intrinsic "scientific" properties, such as energy or electric charge.

Apart from this, various objections have been made to the distinction in terms of measurement. First, measured motion and size must be stated in terms of some standard, such as a yard or meter; hence, they are purely relational and are not intrinsic properties of the object. But one can reply that it is only the description or labeling of the measurement that is thus relational; the motion or extension labeled, which is actually measured, seems intrinsic to the object.

Second, since colors and sounds may be measured, are they not also primary? But this objection seems based on a misunderstanding of the processes of measurement, for one way of "measuring" color might be to compare a given shade with a standard on a shade card; but that would be like comparing the sensible size of two objects, not measuring them. Proper measurement goes beyond this kind of sensory experience, and even if one gives the shade a number, one cannot calculate with the results as one can with the dimensions of objects. Normally, however, measurement of colors or sounds is either the measurement of the amplitudes or lengths of light waves or sound waves, or a mixture of wave measurement and the comparison of experiences. If one brings up a decibel meter and says that the sound to be measured is 80 decibels, it is the amplitude of the sound waves that is ultimately responsible for the movement of the pointer to 80. It should be noted, however, that the logarithmic scale is used because of a characteristic of human earsthat experienced loudness is related logarithmically to wave amplitude.

Third, measurement is a perceptual processat least it relies on and largely uses perceptionso it may be only producing various correlations of sensa and never getting through to the supposed properties of material objects at all. This objection is made from the point of view of phenomenalism, however, while the whole primary-secondary quality distinction presupposes representative realism. Supporters of the latter would say that the best explanation of the correlation is that the sense experiences arise in the measurement of actual physical objects.

arguments from science

Science can adequately explain and describe the nature of the physical world solely in terms of primary qualities; hence, while primary qualities must characterize objects, there is no need to suppose that secondary qualities must also. The latter would be otiose, and on the principle of economy, or Ockham's razor (that entities should not be multiplied more than is necessary), it would be unscientific to suppose that they exist as intrinsic properties of objects. The objection to this argument is partly that the science of one's day is not final (thus, Locke was persuaded by seventeenth-century science to include solidity in the list of primary qualities), and mainly that scientific theory and description are not the whole truththey describe only one aspect of the world, being limited by their quantitative approach and their instruments. Secondary qualities may thus be real properties of matter with biological or aesthetic functions; Ockham's razor oversimplifies the facts pertaining to living things.

Investigation of the causal processes on which perception depends shows that the only variables capable of transmitting information about the properties of external objects are spatiotemporal ones, which are associated with primary qualities. Thus, light waves (energy distributed in space and time) pass from the object to the percipient, but nothing resembling experienced color and sound is transmitted. But the main force of this argument, since it applies to all the senses, is neurological. The nerves from the different sense organs to the brain are all similar, and therefore the only variables are the frequencies of the impulses (which convey the intensity of the stimulus), their different neural pathways, and their different destinations in the brain. Indeed, it seems to be the different destinations that primarily govern the type and quality of the sensation. And although one can conceive of primary qualities being transmitted by spatiotemporal variables, it is difficult to conceive of color, warmth, taste, or smell being so transmitted. (It may be objected that radio and television can transmit color and sound by converting them into electrical impulses for transmission and then reconverting them. But, strictly speaking, what is converted is not color or sound but light waves or sound waves; moreover, the radio or television station must use microphones and cameras to effect the conversion, and there is no evidence of such conversion devices at the objects we see or hear.)

Berkeley's Criticisms

Berkeley's formidable criticisms of the distinction between primary and secondary qualities have convinced many people. We have mentioned his objection concerning relativity, which, though valid against Locke, can be avoided by restating the distinction on the basis of measurement. He also has nothing to say on the scientific considerations, which were not explicit in Locke. But he did have some further well-known criticisms. First, he stated, "An idea can be like nothing but an idea." In other words, our sensa, being private, mental, and directly perceivable, cannot resemble properties of material objects that are public, physical, and not objects of direct awareness. But resemblance is claimed only for primary qualities; and though sensa cannot be extended in physical space, it seems reasonable to claim a structural resemblance, a similarity in form, between the spatial relations that they sensibly possess and those attributed to objects by measurement; thus, it can be confirmed by measurement that various relations between the sides of a square sensum hold in the object. A similar resemblance seems plausible in the case of motion. There are, however, some underlying difficulties here. In the older representative realism, sensa were mental; and since the mind was held to be unextended, they could hardly have spatial relations. But newer versions would allow some sensible or subjective space different from physical space; certainly sensa seem spatial, and there seems to be no reason why what is directly perceivable and what is not should be unable to have a similar form or character.

Second, matter consisting only of primary qualitiesfor instance, possessing extension but no color, taste, sound or smellis inconceivable. This objection is beside the point: Admittedly one cannot conceive, in the sense of "imagine" or "picture to oneself" (Berkeley's sense of the word), any such thing, for what we can imagine is limited by past experience and perception. But the range of possible existents need not be confined to this, and there is much in science, particularly in modern physics, that cannot be imagined or pictured.

See also Berkeley, George; Boyle, Robert; Colors; Descartes, René; Galileo Galilei; Leucippus and Democritus; Locke, John; Newton, Isaac; Pain; Perception; Realism; Sensa; Sound.


classical works

Anticipations of the distinction are to be found in Democritus, Fragments 9 and 125, in H. Diels and W. Kranz, Fragmente der Vorsokratiker; Galileo, Il saggiatore (The Assayer; 1623), in his Opere, edited by F. Flora (Milan, 1952), pp. 311314; René Descartes, Principles of Philosophy, Part IV, Nos. 188203, and The World or Essay on Light, Chs. 12. Robert Boyle adumbrated the distinction in his The Origin of Forms and Qualities (London, 1666), especially the section titled "An Excursion about the Relative Nature of Physical Qualities"; John Locke stated it in his Essay concerning Human Understanding (London, 1690), Book II, Ch. 8; Newton's view is best stated in his Opticks, Book I, Part II (paperback, New York, 1952, p. 124), but that was not published until 1704. Berkeley's criticisms are in his Principles of Human Knowledge, Secs. 920. Edwin A. Burtt, The Metaphysical Foundations of Modern Physical Science (London: K. Paul, Trench, Trubner, 1925), contains useful discussion of the early versions.

modern works

No book is devoted to the distinction, and substantial treatments are rare, but helpful contributions may be found in the following:

A more recent scientific comment on the distinction is Russell Brain, Mind, Perception and Science (Oxford: Blackwell, 1951). L. Susan Stebbing, Philosophy and the Physicists (London: Methuen, 1937), attacks scientific versions of it, especially those of Newton and of Arthur S. Eddington (see his The Nature of the Physical World, Cambridge, U.K.: Cambridge University Press, 1928).

Among the general works available are Charlie Dunbar Broad, Scientific Thought (London: Kegan Paul, 1923) and The Mind and Its Place in Nature (London: Kegan Paul, 1925); William Pepperell Montague, The Ways of Knowing (London: Allen and Unwin, 1925); Roderick Chisholm, Perceiving: A Philosophical Study (Ithaca, NY: Cornell University Press, 1957); and more fully, R. J. Hirst, The Problems of Perception (London: Allen and Unwin, 1959). D. M. Armstrong, Perception and the Physical World (New York: Humanities Press, 1961), Chs. 14 and 15, discusses the subject from a direct realist position, and Gilbert Ryle's criticisms of the distinction in his Dilemmas (Cambridge, U.K.: Cambridge University Press, 1954) are well known.

Among articles on the distinction are Reginald Jackson, "Locke's Distinction between Primary and Secondary Qualities," in Mind 38 (149) (1929): 5676; J. J. C. Smart, "Colours," in Philosophy 36 (137) (1961): 128142 (reprinted with additions in his Philosophy and Scientific Realism, New York: Humanities Press, 1963); Colin Strang, "The Perception of Heat," in PAS 61 (19601961): 239252; and for a more general treatment, W. C. Kneale, "Sensation and the Physical World," in Philosophical Quarterly 1 (2) (1951): 109126.

other recommended titles

Averill, Edward Wilson. "The Primary Secondary Quality Distinction." Philosophical Review 91 (1982): 343362.

Ayers, Michael. Locke: Epistemology and Ontology. London: Routledge, 1991.

Bennett, Jonathan. Locke, Berkeley, Hume: Central Themes. Oxford: Oxford University Press, 1973.

Cummins, Robert. "Two Troublesome Claims about Qualities in Locke's Essay." Philosophical Review 84 (1975): 401418.

Dicker, Georges. Perceptual Knowledge. Dordrecht: Reidel, 1980.

Fales, Evan. A Defense of the Given. Lanham, MD: Rowman & Littlefield, 1996.

Fumerton, Richard. Metaphysical and Epistemological Problems of Perception. Lincoln: University of Nebraska Press, 1985.

Jackson, Frank. Perception: A Representative Theory. Cambridge, U.K.: Cambridge University Press, 1977.

Levin, Janet. "Physicalism and the Subjectivity of Secondary Qualities." Australasian Journal of Philosophy 65 (1987): 400411.

Mackie, John. Problems from Locke. Oxford: Clarendon Press, 1976.

McGinn, Colin. The Subjective View. Oxford: Clarendon Press, 1983.

Sellars, Wilfrid. The Metaphysics of Epistemology. Atascadero, CA: Ridgeview Press, 1989.

Smith, A. D. "Of Primary and Secondary Qualities." Philosophical Review 99 (2) (1990): 221254.

Wright, Crispin. "Moral Values, Projection, and Secondary Qualities." PAS, Supp. 62 (1988): 126.

R. J. Hirst (1967)

Bibliography updated by Benjamin Fiedor (2005)

Cite this article
Pick a style below, and copy the text for your bibliography.

  • MLA
  • Chicago
  • APA

"Primary and Secondary Qualities." Encyclopedia of Philosophy. . 20 Feb. 2019 <>.

"Primary and Secondary Qualities." Encyclopedia of Philosophy. . (February 20, 2019).

"Primary and Secondary Qualities." Encyclopedia of Philosophy. . Retrieved February 20, 2019 from

Learn more about citation styles

Citation styles gives you the ability to cite reference entries and articles according to common styles from the Modern Language Association (MLA), The Chicago Manual of Style, and the American Psychological Association (APA).

Within the “Cite this article” tool, pick a style to see how all available information looks when formatted according to that style. Then, copy and paste the text into your bibliography or works cited list.

Because each style has its own formatting nuances that evolve over time and not all information is available for every reference entry or article, cannot guarantee each citation it generates. Therefore, it’s best to use citations as a starting point before checking the style against your school or publication’s requirements and the most-recent information available at these sites:

Modern Language Association

The Chicago Manual of Style

American Psychological Association

  • Most online reference entries and articles do not have page numbers. Therefore, that information is unavailable for most content. However, the date of retrieval is often important. Refer to each style’s convention regarding the best way to format page numbers and retrieval dates.
  • In addition to the MLA, Chicago, and APA styles, your school, university, publication, or institution may have its own requirements for citations. Therefore, be sure to refer to those guidelines when editing your bibliography or works cited list.