The Sense Data or ‘Qualia’ Position
The first account of a phantasm to be discussed is the ‘sense data position’, which is based upon how Aquinas describes briefly the function of phantasms in his Commentary on Aristotle’s On the Soul: ‘Circa quod sciendum est, quod “phos” in graeco idem est quod lux; et inde venit “phanos” quod est apparitio, vel illuminatio etphantasia’. This passage is rendered into English as follows: ‘Aristotle explains the name phantasia: “Note that phos is the Greek for ‘light. Therefore, the term ‘phanos’ is derived, i.e. ‘appearance’ or ‘enlightening, and phantasia” ’ (Commentary on the Soul, no. 668). Given this textual evidence, it is not difficult to place emphasis on ‘apparitio’, which is translated and obviously can mean ‘appearance. Early on, Kenny suggested this approach: ‘[in] interpreting the notion of phantasm, there are many passages in Aquinas [. . .] where translations such as “sense appearance” or “sense impressions” suggest themselves.’11 Kenny’s later work, however, argues for externalism in Aquinas rejecting a sense datum interpretation. An emphasis placed upon ‘appearance’ or ‘sense impression’ suggests that the phantasm is a sense datum as employed by twentieth- century British and American philosophers. For example, in a general discussion of phantasms, Sellars once used phantasms to elucidate epiphenomenalism: ‘“phantasms” or “sensa” we might call them.’
Furthermore, the above passage from the Commentary on the Soul is not an isolated instance in which Aquinas uses the term ‘appearance’: ‘The word “imagining” [phan- tasia] is itself taken from seeing or appearing’ (Commentary on the Soul, no. 632). ‘Aristotle suggests that phantasia is a sort of movement: that just as the sensing subject is moved by sensible objects, so, in imagining [the use of phantasia], one is moved by certain appearances called phantasms’ (no. 56). In addition to this use of appearance, when considering the formation of the species intelligiblis by means of abstraction with the intellectus agens leading towards the formation of a concept in the intellec- tus possibilis, Aquinas speaks of phantasms as ‘likenesses’ of physical objects: ‘Therefore, material things must be understood according as they are abstracted from matter and from material likenesses, namely, phantasms’ (Summa Theologiae, I q. 85 a. 1).
Given texts like these, it is not terribly difficult to suggest that a phantasm is nothing more than a sense datum, which puts Aquinas under the indirect realism umbrella articulated by several early and mid-twentieth-century analytic philosophers. However, this placement is erroneous textually and structurally. While few philosophers now adopt sense data positions, this interpretation of Aquinas’s account is still moderately lively. Hence, it is useful to indicate the problems with this kind of ascription lest further misreadings and misinterpretations continue. Moreover, rejecting the sense data analysis is a further claim against a tertium quid analysis of Aquinas’s account of sensation and perception.
In order to explicate this position rendering phantasms and sense data coextensive, one must consider passages from the writings of various early and mid-twentieth- century epistemologists providing elucidations of the nature of sense data. The classic exposition of the sense data theory is found Moore’s Some Main Problems of Philosophy:
I hold up this envelope: I look at it [ . . . ] what has happened? We should certainly say that we all saw that envelope. [ . . . ] But now what happened to each of us, when we saw that envelope? I saw a patch of a particular colour, having a certain size, and a certain shape, a shape with rather sharp angles or corners and bound by fairly straight lines. These things: this patch of whitish color and its size and shape I did actually see. And I propose to call these things—the colour and size and shape—sense data, things given or presented by the senses—given in this case, by my sense of sight. [ . . . ] Now all of this seems to me to show very clearly that, if we did all see the same envelope, the envelope which we saw was not identical with the sense data which we saw.
Moore’s philosophical text indicates that a sense datum is the direct object of an act of perception. Moreover, sense data accounts indicate that sense data philosophers adopted a ‘relational’ model of perception. Besides claiming that a sense datum is a term of a private act of direct acquaintance, Moore suggests that a sense datum and the external object are distinct entities. In other words, a sense datum is what is presented to a perceiver when she is in a relational act of awareness; this datum is not identical with the physical object. Moore argued explicitly that sense data are entities wedged between the mental act and the physical object; accordingly, Moore accepted this non-identification of sense data and physical object.
On this perceptual model entailing sense data, which is reducible to a form of representationalism, the important structural question concerns going from the sense datum to the material object. Of course, philosophers have offered several purported resolutions for this problem, ranging from ‘instinctive beliefs’ to ‘constructs’ to ‘convictions from common sense’ to ‘permanent possibilities of sensations’. These various solutions, however, are not the present concern. What is of concern is the difference between a phantasm and a sense datum when used in direct sensation. It is in order to argue effectively for this difference that this discussion of sense data has been provided. Analysing the nature of a sense datum in direct sensation as elucidated by some of the many twentieth-century philosophers who entertained seriously the philosophical import of this epistemological entity enables the discussion to advance, indicating the logical and structural differences between a sense datum and a phantasm.
Two propositions concerning phantasms and sense data require discussion:
- (a) A sense datum approach destroys the direct realism of Aquinas’s theory of perception.
- (b) A phantasm is never connected with the workings of the external sensorium alone, so a fortiori, a phantasm cannot be a sense datum; i.e. the paradigm case for a sense datum is direct sensation.
The structured analysis offered establishes the soundness and textual significance for Proposition-2. One might question this proposition by referring to no. 664 in the Commentary, where Aquinas writes: ‘I mean, at least so long as the sensible object is present and the image-movement is simultaneous with the sense-movement’ (emphasis added). The Latin is ‘quando motusphantasiae estsimul cum motu sensus’. However, this passage is a causal statement about an image being produced in the imagination, not a claim that the image is a sense datum in direct perception. In no. 667, furthermore, Aquinas writes: ‘it seems necessary for there to be a phantasm-producing or imaginative power different from sense.’ Thus, passage no. 664 does not refute the thrust of Proposition-2.
In regard to Proposition-1, however, the following remarks are important. If a phantasm is to be equated with a sense datum and if a sense datum is the direct object of perception and thus distinguished from the material object itself, then it follows that Aquinas faces the same consequences for his theory of perception that any sense datum theory entails. The principal consequence of a sense datum theory is driving a wedge between the object of perception and the physical thing itself. Accordingly, if a phantasm is interpreted as a sense datum, then serious structural difficulties arise because Aquinas argues for epistemological realism. His texts for the external senses indicate repeatedly that the object of sensation is the physical quality itself—either a proper or a common sensible—and not an intermediary entity or a tertium quid. This ‘thing consciousness’ and ‘quality consciousness’ linguistic usage is reiterated over and over in the Aquinian texts. To adopt a sense datum theory entails a drastic modification to this realistic theory of perception. If a phantasm is identified with a sense datum, then Aquinas’s theory of perception is accordingly so modified. This entails that the phantasm is the direct object of sensation, and this in turn entails some form of representationalism.
Approaching Proposition-2 begins with sense datum interpretations of a phantasm as found in contemporary studies of Aquinas. In addition to the text from Kenny’s essay, Hamlyn proposed a sense datum approach. Both Hamlyn and Ayer claimed that a sense datum belongs to the same category of epistemological entities as the ideas and impressions of the British empiricists. Hamlyn suggested that a phantasm is a mental entity needed in any mental act of direct awareness, which follows from a sense datum account of perception. If a sense datum interpretation of Aquinas’s theory of sensation holds, then it follows that a phantasm is involved necessarily with every mental act of sensation using the external sensorium. Often Aquinas is interpreted by analytic philosophers as claiming that phantasms are part of the perceptual process involved with each external sense. Early on, Kenny vacillated on this account:
But it also appears that he [Aquinas] thinks that whenever we see something we have at the same time a phantasm of what we see; and he explains sensory illusions by saying that the senses themselves are not deceived, but only the phantasia on which they act. It seems odd to suggest that whenever we see a horse we have at the same time a mental image of a horse. Perhaps the theory is that if we see accurately our phantasm of a horse is a sense-impression; if we are mistaken about what we see, and there is no horse there at all, then our phantasm is a mental image. This theory seems to be confused in several ways, but it is hard to be sure whether Aquinas held it or not. At all events it seems clear that he did not mean by ‘phantasm’ simply a mental image.
While Kennys example is somewhat confusing, nonetheless it is precisely this necessary connection with the external sensorium that Aquinas denies explicitly. In opposition to the position presented above, Aquinas never uses the term ‘phantasm’ when discussing either direct sensation involving only the external senses or any mental act involving the sensus communis. On the other hand, he mentions explicitly the powers or sense faculties of inner sense in which the phantasm is found, conspicuously omitting any reference to the sensus communis. In considering the relation between phantasms and the external sensorium, one must take seriously the passage noted above, which is one among many found in the Summa Contra Gentiles: ‘the powers in which the phantasms reside [. . . are . . .] namely, the imagination, the sense memory, and the vis cogitativa’ (bk II, ch. 73, no. 11). In the Commentary on the Soul, moreover, Aquinas claims that the locus of the phantasms is the internal sensorium, or what he calls the phantasia: ‘It is by the phantasia that we become conscious of phantasms’ (no. 638). Furthermore, the second quotation given above is qualified immediately by the following words: ‘unless the term “phantasm” is being used metaphorically’ (no. 638). This indicates that above and beyond the functions of the faculties of the internal sensorium, the phantasm does not have a non-metaphorical function in Aquinas’s epistemology and philosophy of mind. Because Aquinas discusses the internal senses where phantasms reside, it would be odd to omit the sensus communis if a phantasm did ‘reside’ with this internal sense. Moreover, it must be emphasized that one looks in vain for passages in which Aquinas posits phantasms as the intentional object of the sensus communis.
-  Wilfrid Sellars, Science, Perception and Reality (London: Humanities Press, 1963), 101. Sellars did notmention Aquinas’s account of phantasms in particular. One gets the impression, however, that he provideda generic notion of phantasms under which Aquinas’s account would be subsumed.
-  A detailed analysis of likeness (similitude) will be offered later in this chapter.
-  G. E. Moore, Some Main Problems of Philosophy (London: George Allen & Unwin, 1953), 30-31.Other arguments for a sense data epistemology are found in e.g. Russell, Price, and Ayer.
-  Klemke once discussed this problem in Moore’s theory of perception, which is similar in structure toAquinas’s worries: ‘Moore’s views on perception, although occupying more pages than any other singleepistemological topic, contain the greatest number of unresolved problems of all. Taking the “problem ofsense data” first, on only one point is Moore consistent: there are sense data. Why so? Because they areobjects of certain sorts of awareness, and one cannot (Moore thinks) have an awareness without an object.But would not physical objects fit the bill? Moore thinks not, for then we could not satisfactorily accountfor the phenomenon of illusion, hallucination, after-images, and perspectival visual sensings. Moore overlooks the view that we could say that certain things (e.g. coins) look certain ways (e.g. elliptical) withoutrequiring that there be sense data as entities’ See E. D. Klemke, The Epistemology of G. E. Moore (Evanston,Ill.: Northwestern University Press, 1969), 190. See also Moore, ‘A Defence of Common Sense’.
-  See D. W. Hamlyn, Sensation and Perception: A History of the Philosophy of Perception (London:Routledge & Kegan Paul, 1969), 30-31.
-  Kenny, Aquinas on Mind, 38.
-  One must recall that in comparing the Summa Theologiae with the Commentary on the Soul, there isa marked difference in the use ofphantasia. In the Summa, ‘phantasia5 refers to the vis imaginativa, whereasin the Commentary, ‘phantasia’ refers to the workings of the entire internal sensorium. In this chapter, theCommentary will be followed and ‘phantasia’ will refer to the entire internal sensorium; i.e. the vis imagi-nativa, the vis cogitativa, and the sense memory.
-  Earlier historians of philosophy like Hamlyn and Weinberg and later students of Aquinas like Kempand Pasnau have possibly confused the ‘sensible species’ with the phantasm. These two epistemologicalconcepts, however, are neither identical nor coextensive.