BMCR 1999.09.24

Aristotle on the Sense-Organs

, Aristotle on the sense-organs. Cambridge classical studies. Cambridge: Cambridge University Press, 1997. xvi, 304 pages ; 22 cm.. $59.95.

Given the amount of attention paid to Aristotle’s psychology over the last few decades it is surprising that we’ve had to wait so long for a monograph exclusively devoted to Aristotle’s account of the sense-organs. With the advent of Johansen’s intriguing and controversial study, one may expect a lively response and reaction that should further the debate on this important subject.

This book contains an introduction, six chapters, a rather limited bibliography, an index locorum, and a general index. The work’s organization is somewhat puzzling at first, placing a chapter on the medium between chapters on hearing and vision, but the author manages to cover the major issues and problems addressed on the subject. The first chapter examines the sense of sight and the anatomy of the eye. The second chapter takes a closer look at the concept of the medium, focusing primarily on the medium of vision. Chapters three through five work through the remaining special senses. In chapter six J. returns to some of the philosophical issues surrounding perception and allows the author to recap how his account of the sense-organs endorses the philosophical positions which he argues throughout.

J. approaches the subject through a philosophical and interpretive controversy. For years scholars have debated what Aristotle means in saying that a sense-organ takes on the form of its object without the matter. J.’s study uses this problem to present Aristotle’s theory of the sense-organs. The fact that J. frames the work around an interpretive issue makes the work engaging and readable. It is difficult to read this book without taking a position and jumping into the fray.

What does it mean for a sensory form to be actualized in the sense-faculty? The book examines two basic models: the first I will call the Literal Model. This position is probably best articulated by Richard Sorabji in his important article Body and Soul in Aristotle 1 and is seen to support a functionalist and hylomorphic reading of Aristotle as articulated by Nussbaum and Putnam. According to this view sense-perception succeeds in actualizing the potential present in the organ by means of a physiological change. There must be a chain of physical events that causally bridges the gap between the sense-object and sense-organ. The make-up of the medium and organ of perception must be constituted in such a way as to bear the physical property. Thus, air is physically the kind of thing that can be moved or affected by the color, since it is transparent and is potentially any color; thus it literally becomes red in the presence of a red sense-object and light. The transparent jelly in the eye likewise can become red; and when it does, we see red.

That physiological changes underlie perception does not imply that the physiological description of the event is a sufficient account of it. The formal description will explain the essence of the phenomenon better. On the hylomorphic view the two accounts supplement each other. A standard example cited in favor of this reading is anger; formally it is a “yearning for retaliation” but physiologically it is described as “a boiling of the blood around the heart” ( De An. I.1 403a-b). Followers of the literal model sometime find it useful to compare Aristotle’s theory with modern functionalism on account of the functional plasticity.

J. also emphasizes that the sense-organs are defined functionally in accordance with a proper sense-object. Therefore, any faculty that receives the sensory form of color is vision, likewise with the other sensibles. The location, appearance, and even (to a limited degree) the composition are incidental. This may suggest that J. is siding with Putnam and Nussbaum who have argued that Aristotle’s theory of perception represents a primitive functionalist theory. On the contrary, J. wishes to refute the idea that Aristotle’s theory is akin to modern functionalism for modern conceptions of functionalism usually hold that functional plasticity is grounded on the physiological mechanism of the organs and medium. Any physiological process that performs the same or similar activity is functionally identified. J. wants to deny that Aristotle’s functionalism is physiologically based or driven as the literal model asserts. He instead sides with Burnyeat’s Phenomenal Model which holds that Aristotle’s functionalism does not entail or necessitate a physiological foundation but is only formally functional. The absence of a physiological basis, argues Burnyeat, makes Aristotle’s theory fundamentally inconsistent with modern functionalism and untenable for modern thinkers.

J. agrees that Aristotle’s theory represents perception as an essentially phenomenal alteration and not a physiological alteration. The sense organ does not literally become or take on the sensory form; the eye does not literally become red. Seeing red is a phenomenal alteration of the physical medium without necessarily a physiological change. It is only affected phenomenally insofar as the sense-object appears through it and only occurs when a perceiver is present. The phenomenal interpretation therefore can claim that matter is required but not material change. The kind of actualization present is like that which follows when a builder builds. The builder is not physiologically altered, only operationally changed; he simply applies the knowledge that he already has in him as a first actuality.

If the phenomenal interpretation is correct then the existence of physical organs seems to require justification. Since J. is committed to the phenomenal interpretation this question dominates much of the subsequent discussion. Although J. clearly adheres to the phenomenal interpretation he believes that Aristotle’s account of the sense-organs is compatible with either model (14). One wonders if this concession is an attempt not to lose readers who have already taken sides on the literal/phenomenal debate.

After identifying the philosophical controversy in his introduction, J. moves on to Aristotle’s argument. He begins by distinguishing Aristotle’s approach from his predecessors and contemporaries, especially Plato and Empedocles (whose theories are sometimes conflated in the De anima), Democritus, and the Pythagoreans. J. stresses that Aristotle’s theory surpasses the competing materialist theories since he takes a “top-down” approach; in contrast the materialists work “bottom-up.” “Top-down” means that Aristotle begins from the function of the eye, seeing, and accounts for the matter on the basis of how it makes this function possible. The top-down approach is identified with what Aristotle’s calls hypothetical necessity ( Phys. II.9). Bottom up approaches entail material necessity. J. needs to separate and isolate these early on. The literal approach wants to bring the two together.

J. points out that Aristotle is very interested in the question of which elements are associated with which sense-faculties. The fact that there are four elements and five senses creates difficulties. The sense-organ of vision (in humans) is constituted primarily of water. Water is essential not on the basis of its wetness or its coolness (Aristotle’s primary attributes of water) but qua its transparency. This leads Aristotle to define the senses in terms of functional properties (transparent, resonant, etc.) instead of the elements. These properties are common to both the medium and the sense-organs and do much of the theoretical work.

The eye and the medium are therefore defined in terms of transparency. Air and water are only actually transparent in the presence of light. Light, not being a body but an activity of the transparent, becomes a necessary condition for vision. Vision is the power to see what is visible, namely colors. These lie on the outer surface of the object of vision. The inquisitive reader may want to know why certain objects are different colors. J. does not want to go there. For if a banana is yellow because of its physical make-up, and turns brown when the physical constitution is altered, we would suspect a physical basis of alteration in the sense-organ.

The eye itself possesses several parts and unfortunately Aristotle is not always clear or consistent in his use of technical language. The membrane ( δέρμα), the κορή, and the πόροι all manifest transparency. J. argues convincingly that for Aristotle the κορή is not the reflection on the eye but the gel within. Finally he offers a detailed analysis of the famous damage to the πόροι passage in De Sen. 438b. The πόροι turn out to be undefined passages that connect the transparency to the seat of perception in the heart.

Surprisingly, the eyes, on J.’s model turn out to be nothing but a complex extension of the transparent medium linking the sense-object with the perceiver. What is the culmination of the transmission? Do we see when the quality becomes actual in the eyes or are the eyes simply conduits to the heart? J. argues that the eyes are but channels of transparency mediating further, eventually to the primary sense-organ in the heart.

There are serious problems that follow from J.’s view that the eyes are nothing but extensions of the transparent medium to the heart. First, there are textual problems. Aristotle clearly holds that vision occurs in the eyes and not in the heart.2 The sense-objects of the primary sense-organ are said to be the common sensibles, not the special sensibles. Secondly, J.’s model creates far greater problems than the simpler model. Mediation of vision requires a route that is both rectilinear and illuminated (for Aristotle the medium is illuminated, not object). This is why we can’t see around corners. There is no transparent and continuous line-of-vision passage from the sense-object to the heart. A different mechanism of mediation (and hence another theory of perception) would be needed — one that can send information through the murky twists and turns of the body.3 Third, and most seriously, if the eyes are nothing more than a continuation of the transparency of the medium, Aristotle’s theory of vision would be an anti-theory. Perception would then not have been explained; instead we would have what seems to be but a rather redundant extension of the outside medium. The explanation of vision in De anima and De sensu would be but a smoke screen for an enormous black box or a Cartesian Theater.

In the second chapter J. expands his analysis of the transparent medium and tries to explain what it means for the medium to be phenomenally altered and not physically altered. He begins by showing why the ‘literal model’ cannot work. If the medium actually takes on the sensory form of the sense-object, the medium itself becomes the sense-object of vision. But Aristotle argues that the eye cannot perceive its object if it is in direct contact with it. Thus if the medium actually becomes red and is in direct contact with the membrane of the eyes, vision should not be possible. J. therefore concludes that the transparent medium cannot bear the sensory quality in the same way as the original sense-object. He proposes that the medium is only phenomenally altered and not physically altered. But what does phenomenally altered mean? It means that it takes on an activity in which sensory forms may be seen through. It acts as a conduit not by taking on the sensory form but by not impeding it.

This is clearly J.’s most convincing argument. If the sensory medium is physically altered then it seems to become the immediate object of vision and is open to the charge that they eye is in direct physical contact with a sense-object. However there is a way out for the literal model. We must remember the role of light. Colors require light to be seen. It may be argued that the reason we cannot see an object if it is pressed against our eye is that the object prevents the light from activating the transparent in the gel. Thus in the case of the transparent medium both the light and the sensory form are given access to the perceiver. The reason that the sense-objects require a medium is that the medium in the case of all the senses contributes something to the actualizing of the sensory form in the sense-organ.

In the third chapter J. moves on to hearing. Sound is the object of hearing. Like vision, hearing requires a medium, either air or water. For sound to be generated in the air the air must not disperse too easily. Air alone is not responsible for and cannot generate sound. Solid objects must strike against one another and strike the air. ( De An. 419b20) Ideally a hard concave object is struck by another hard object. The surface of the object allows the sound occur in the space between the sounding object and the organ of hearing. ( De An. 419b8)

J. identifies the element of the air essential for the production of sound as “the resonant.” J. seems to down-play his phenomenal interpretation of the medium in this chapter for certainly the air is not simply heard through. The sound literally affects the medium through a motion ( κινήσις and φορή), vibrating ( σείεσθαι), and rebounding ( ἀφάλλεσθαι) of air. ( De An. 420a26) For hearing to occur (as opposed to just sounding) the sense-organ in the ear must become like the medium; the calm air in the sense-organ is potentially moved. It seems that the literal model is more at home here than the phenomenal model. The chapter ends with an important discussion on the importance of an animal’s environment to Aristotle’s conception of the plasticity of the sense-organs in different species.

The discussion on the remaining special senses is in some regards the most valuable part of the book. Touch, taste, and smell seldom get the attention that they deserve — especially considering their prominence in the De Sensu. J., however, gives theses senses a quite thorough analysis. Any reconstruction of Aristotle’s theory of perception must reconcile all the senses to be taken seriously. Here J especially shines. Chapter four focuses on the contact senses, touch and taste. Both represent unique challenges to Aristotle’s theory. Touch is problematic since it does not seem to need a medium for perception. The fact that the sense-object of touch has more than a single pair of contraries (hot/cold, wet/dry, hard/soft) raises further problems. J. carefully distinguishes between touch as contact and touch as feeling. He also explains why the organ of touch (located near the heart) is composed of different elements.

The organ of taste is the tongue, and the object of taste is flavor. The tongue as a fleshy organ also possesses the powers of touch (422a8-11). One can feel temperature, resistance, and other tactile qualities with the tongue as with the hand. The similarity of the senses raises special problems. As the organ of taste its proper objects are flavors. Flavor ranges from sweet to bitter (as color ranges from light to dark). As vision requires light and hearing a resonant medium, taste requires moisture. Flavor is dry but the dry by itself is not tasteable. It can only be tasted when mixed with the wet. For salt to be tasted the flavor must be activated by being dissolved or percolated in moisture. Thus moisture is a necessary condition for taste (188). J. compares the process to the production of coffee by means of an espresso machine. The steam does not carry the taste of the coffee in suspension but itself becomes flavored. The same applies to medium in the activity of tasting.

The final special sense examined is smell. Smell is poorly defined in humans. The proper object is the smellable, qualities that usually get their name from the associated tastes. Like taste, the object of smell is also the dry. The organ must therefore also be something that is potentially dry. The dry however is not mixed as in the case of taste, but is mediated by air (and sometimes water). It is not clear what functional property (such as the transparent or the resonant) of the air or water permits the transmission of the sensory form. The medium of smell washes the sensory quality from the sense-object as dye is washed out of colored fabrics. J. is careful that his readers not confuse Aristotle’s theory with the effluence or vapor-theory of smell. No actual particles are transmitted. Nor is there a mixture. The medium must take on the quality as air takes on color. Again, a non-literal interpretation is offered. J. concludes this chapter by examining special cases and problems of smell such as how fish can smell in the water and why some animals have nostril-flaps.

J. ends his work by reiterating his argument for the phenomenal interpretation of Aristotle and how Aristotle’s theory of the sense-organs favors this model.

Aristotle on the Sense-Organs is a valuable contribution to the field. The book is well written and well documented. J. demonstrates an ability to convey difficult positions with great clarity and frequently illustrates the issues with illuminating analogies and examples. In spite of my obvious doubts regarding the philosophical issues mentioned above I found the work to be very stimulating and informative. A safer (less controversial) work on the subject would not have been difficult given J.’s expertise in the field. That he chose to approach the subject through a controversy gives the work a life and energy that it would not otherwise have had. This work represents the most extensive defense of the phenomenal model to date. I would recommend the work to advanced undergraduates, graduate students, and scholars.


1. Richard Sorabji, “Body and Soul in Aristotle”, Philosophy 49, 1974.

2. It is interesting to note that when we experience recollection the motion that starts elsewhere in the soul returns to the special organs. See De. An. I.4 408b18. Also recall that images sometimes remain in the sense-organ after the object is no longer present ( De An. III.2 425b25).

3. Peck in appendix B to Loeb edition of De Gen. An. (1953) has attempted with little success to establish mechanism of transmission for sensory information from the eyes to the heart by means of pneuma. However tantalizing the theory, it does not survive in any tenable form in the surviving texts.