Reid’s Non-Naïve Direct Realism
Rebecca Copenhaver
Lewis and Clark College
§1. Introduction
It is commonly said of modern philosophy that it introduced a representative theory of perception, a theory that places representative mental items between perceivers and ordinary physical objects.[1] Such a theory, it has been thought, would be a form of indirect realism: we perceive objects only by means of apprehending mental entities that represent them. The moral of the story is that what began with Descartes’ revolution of basing objective truth on subjective certainty ends with Hume’s paroxysms of ambivalence and skepticism in the conclusion of the first book of the Treatise of Human Nature.[2]
Thomas Reid’s criticism of the representative theory of perception, of what he called ‘the theory of ideas’, is well known. “The theory of ideas, like the Trojan horse, had a specious appearance both of innocence and beauty; but if those philosophers had known that it carried in its belly death and destruction to all science and common sense, they would not have broken down their walls to give it admittance.”[3] Many have supposed that Reid’s opposition to this theory is tantamount to opposition to indirect realism and acceptance of direct realism.[4] Recently, however, there has been disagreement over whether Reid is a direct realist. If we take direct realism to be naïve, that is, if we understand it as the claim that there are no entities mediating perception of objects, Reid is not a direct realist. He holds that sensations do mediate perception; he holds that we interpret sensations; and he holds that this interpretation informs us about the external world. The challenge, then, is to locate a version of direct realism that is not naïve, one that illuminates why Reid may have thought of himself as a direct realist. I take up this challenge in this paper first, by examining some definitions of direct realism that have been proposed in the secondary literature on Reid. I argue that these definitions do not capture why Reid thinks he is a direct realist. Second, I propose a definition that I think does capture Reid’s conception of himself as a direct realist.
This definition is useful not only for understanding Reid, but for understanding what was at issue in general for theories of perception in the modern period. The common story about modern philosophy story will need to be rewritten if what is at issue between direct and indirect realists is not the immediacy of perception. In this paper I will argue that direct and indirect realists may reach general agreement that perception is mediated by sensations, causation, and physical processes; as analysis proceeds, however, they will disagree about how such mediating entities represent external objects. Indirect realists hold that it is in virtue of the intrinsic characters of mediating entities and external objects that the former represent the latter, while direct realists hold that it is in virtue of an extrinsic relation that mediating entities represent external objects. And so those modern thinkers who regard perception as a mediated process need not be read as indirect realists.
I begin, however with a general description of Reid’s standard theory of perception.
§2.
Reid’s Theory of Sensation and Perception
Reid describes perception as both physical and psychological. All perceptions have their origin in some physical impression on the brain by an object or some medium between the object and the sensory organs. This physical stimulus leads to a sensation, which Reid calls a mental act. In this paper I will not examine the significance of Reid’s view that sensations are acts, and sometimes I will refer to them as entities, for the sake of clarity. Sensations suggest a conception of an object and a belief about it. Together, this conception and belief constitute a perception.
Reid avoids mention of ‘causes’ when he describes the perceptual process: perceptions are “roused and stimulated” by “corresponding sensations”; sensations are “occasioned” by objects and properties.
When we say that one being acts upon another, we mean that some power or force is exerted by the agent, which produces, or has a tendency to produce, a change in the things acted upon. If this be the meaning of the phrase, as I conceive it is, there appears no reason for asserting that, in perception, either the object acts upon the mind or the mind upon the object.[5]
These claims figure importantly in Reid’s notion of explanation; they imply that Reid’s aim is not to explain perception, sensation, representation or the existence of material objects. On the contrary, one of Reid’s criticisms of the ideal theory is that its explanatory goals could never be met. Reid means his theory not as explanation but as a careful description of knowledge we all possess about the workings of our perceptual systems.
How a sensation should instantly make us conceive and believe the existence of an external thing altogether unlike it, I do not pretend to know; and when I say that the one suggests the other, I mean not to explain the manner of their connection, but to express a fact, which everyone may be conscious of – namely, that, by a law of our nature, such a conception and belief constantly and immediately follow the sensation.[6]
Reid thinks that we just happen to be constituted so that material objects ‘occasion’ mental acts of sensation which then ‘suggest’ conceptions of and beliefs about material objects. That Reid uses these verbs rather than ‘cause’ is a hallmark of his theory, not an accident of his prose.
Finally, according to Reid, sensations are natural signs: they suggest a conception of and belief about material objects and qualities. Sensations suggest them immediately, without inference. The role of sensations as signs is determined in nature by what Reid calls original principles of the mind that assign particular sensations to particular objects and qualities. In this way, Reid claims that sensations ‘signify’ material objects and qualities. In other words, sensations ‘suggest’ perceptions (conceptions and beliefs) but ‘signify’ material objects and qualities.
Sensations alone cannot represent or be about any material object. Sensations suggest conceptions of and beliefs about material objects because our constitution is such that particular perceptions constantly arise in conjunction with particular sensations. A sensation is a sign, then, only insofar as it suggests a perception. Absent the relation of suggestion, sensations themselves signify nothing.
This is a brief overview of Reid’s standard theory of perception. In this paper I will not be considering Reid’s theories concerning non-standard cases of perception; I will not be considering, for example, his geometry of visible figures.
§3. Direct Realism
‘Direct realism’ and ‘indirect realism’ have naïve and non-naïve uses. A theory of perception that denies that direct apprehension of something occurs at some point in the perceptual process, even if that something is an image or idea, would be naïve. Any definition of direct realism that will exclude naïve direct realism must accept that perception is in some sense mediated. Unsurprisingly, Reid is not opposed to mediation as such in perception. “Although there is no reasoning in perception, yet there are certain means and instruments, which, by the appointment of Nature, must intervene between the object and our perception of it; and, by these, our perceptions are limited and regulated”.[7] If Reid’s direct realism is anchored not in opposition to mediation as such, to what sort of mediation is he opposed? In order to answer this question I will examine some definitions of direct realism that have been proposed in the secondary literature on Reid.
In “The Development of Reid’s Realism”, John Immerwahr proposes the following definition of direct realism, and concludes that Reid is not a direct realist in the Inquiry.
By direct realism I mean the theory that we are directly aware of external objects and that we know them without requiring awareness of mental entities which act as cognitive links informing us of an external world.[8]
If we reduce Immerwahr’s definition of direct realism to its premises we arrive at the following:
Direct realism holds that:
1. We are directly aware of external objects, and
2. We are aware of external objects directly insofar as we do not require awareness of mental entities which act as cognitive links informing us of an external world.
The first thing to notice here is that premise two is underspecifed between two ways in which it could be true:
2’) There are mental entities which act as cognitive links informing us of an
external world, but we are unaware of them.
2’’) There are no mental entities which act as cognitive links informing us of an
external world, and thus no such mental entities of which to be aware.
That Immerwahr’s intends premise 2 to be made true by 2’ is made clear by his reason for thinking that Reid is an indirect realist, namely, that because sensations are signs, we must be directly aware of them rather than the objects they signify. I’ll have more to say about Immerwahr’s interpretation of Reid’s theory of signs in a moment. For now, it is important to note that the term ‘awareness’ as it occurs in this definition is also underspecified, and depending upon how one reads this term, one might or might not understand Reid as a direct realist.
If what awareness amounts to is a conscious attending to a mediating mental entity and the formation of beliefs concerning that entity which could serve as a basis for inference, then Reid is a direct realist because according to Reid 1) we need not, and in fact rarely, consciously attend to sensations and 2) the transition from sensations to the external objects which they signify is not inferential and not based upon any belief formed about the sensation. If, however, what ‘awareness’ amounts to is that there is a mediating mental entity which plays a role in informing us about external objects, then Reid is not a direct realist because he does hold that sensations are signs that indicate external objects and that we are informed of external objects in virtue of having sensations. Without specification of what awareness amounts to, Immerwahr’s definition will not aid us in determining whether Reid is a direct realist.
That Immerwahr intends ‘awareness’ to be read as conscious attending becomes clear when we examine why Immerwahr interprets Reid as an indirect realist. Immerwahr points out, rightly, that Reid’s theory of suggestion and sensations as signs is drawn from Berkeley. Immerwahr presents Berkeley’s theory through the following quotation: “In reading a book, what I immediately perceive are the letters; but mediately, or by means of these, are suggested to my mind the notion of God, virtue, truth, etc”.[9] Immerwahr concludes that Reid’s theory of sensations as signs commits him to indirect realism because like Berkeley, Reid must believe that words and sensations are the immediate objects of awareness. “For Reid sensations are the signs and objects the things signified. Thus only sensations are the objects of direct awareness. The sensations play the role of sensa in indirect realism, acting as cognitive links to the external world, which is known only indirectly”.[10] It is true that Reid adopted his theory of sensations as signs from Berkeley, and I will have more to say about Reid’s adoption of Berkeley’s theory towards the end of this paper.
Like Berkeley, Reid illustrates his theory of sensations as signs by using an analogy between words and sensations. It is important to note, however, that Reid regards words and sensations as signifying what they do directly. Reid thinks that when we read a word we attend not to the word, but to it’s sense. Analogously, when we perceive, we attend not to the sensation, but to the object presented in perception. “The sensations of smell, taste, sound and color, are of infinitely more importance as signs or indications, than they are upon their own account; like the words of a language, wherein we do not attend to the sound but the sense”.[11] Reid is making the now familiar point that sensory experience is ‘transparent’ or ‘ephemeral’: in the act of sensation my mind is directed towards the object, not the sensation. According to Reid sensations signify objects regardless of our attention to or awareness of them.
The feelings of touch, which suggest primary qualities, have no names, nor are they ever reflected upon. They pass through the mind instantaneously, and serve only to introduce the notion and belief of external things, which by our constitution, are connected with them. They are natural signs, and the mind immediately passes to the thing signified, without making the least reflection upon the sign, or observing that there was any such thing”.[12]
Nevertheless, I propose that Reid’s insistence that we need not be aware of sensations is not what makes him a direct realist. More is needed: if one were to propose a theory on which perception were mediated by mental entities from which beliefs about the external world were inferred, but inferred unconsciously, we would still be tempted to regard this theory as indirect. Thus, more needs to be said about the sort of transition that is made between the sensation as a sign to the object which the sensation signifies.
In “Sensation and Perception in Reid” George Pappas makes a distinction between perceptual and epistemic direct realism.
Perceptual direct realism (PDR) is best seen as a cluster of distinct theses. One such thesis is existential, for what it comes to is that (a) there are external physical objects which have at least some of the qualities they are typically perceived to have, and (b) these external physical objects and some of their perceived qualities exist independently of perception and percipients. PDR also includes a metaphysical thesis concerning the nature of such objects and their perceived qualities…Finally, PDR incorporates a perceptual thesis according to which external physical objects and some of their qualities are directly or immediately perceived…We can define such a concept as follows: (A) A person S directly perceives an object O at a time t = (1) S perceives O at t; and, (2) it is false that: S would perceive O at t only if S were to perceive R at t, where R ≠ O, and where R is not a part of O nor is O of R, and where R is not a constituent or group of constituents of O, nor is O of R.[13]
Pappas’s definition may be analyzed as follows:
Perceptual direct realism:
PDR1) There are external physical objects which have at least some of the qualities they are typically perceived to have, and
PDR2) These external physical objects and some of their perceived qualities exist independently of perception and percipients.
PDR3) External physical objects and some of their qualities are directly perceived where
(A) A person S directly perceives an object O at a time t =
(1) S perceives O at t ; and,
(2) it is false that: S would perceive O at t only if S were to perceive R at t, where R ≠ O, and where R is not a part of O nor is O of R, where R is not a constituent or group of constituents of O, nor is O of R.
Epistemic Direct Realism:
EDR) Typically perceptual knowledge of physical objects is direct or non-inferential, not being based upon immediately known, or immediately justified beliefs about sensations.
Pappas thinks that Reid holds these premises and so he thinks that Reid is both an epistemic and perceptual direct realist. However, these premises don’t yet illuminate why Reid took himself to be a direct realist. Premises PDR1 and PDR2 are uncontroversial insofar as any direct realist will agree to them. Notice, however, the occurrence of ‘perceive’ in the second sub-premise of PDR3. PDR3 holds that perception of an object O is direct so long as one need not perceive something else in order to perceive O. But there are theories that we would regard as indirect but which meet this condition as well as the other conditions outlines in PDR1-3. For example, one could hold a sense-data theory which accepts the existence of independent external objects (PDR1 and PDR2) and on which we perceive external objects not by perceiving sense-data, but by being acquainted with sense-data. This may seem a trivial terminological point except that what relation we bear to the entities which mediate perception is important – some relations, like ‘perceiving’, will sacrifice directness, yet others, such as ‘having’ may not. If Reid is a direct realist merely because he thinks that we don’t perceive sensations, his direct realism will not be interesting.
EDR specifies which sorts of transitions between the sensations and the things they signify are ruled out in an epistemically direct realist theory. In regards Reid’s theory it rules out that sensations could be signs in virtue of any inference that could be made between the sensation and that which it signifies, and it rules out that perception of objects could be based on any beliefs we form about sensations rather than the objects themselves. Reid does in fact hold that the sign-signified relation is non-inferential and he also holds that our perceptual beliefs are about material objects, not sensations. However, EDR does not rule out the possibility that sensations carry information about the material objects they signify and that our perceptual beliefs are arrived at by extracting this information from the sensations themselves. Such extraction could be non-inferential, it could be by the application of concepts, for example, and such application need not require any belief about the sensation itself. Nevertheless, if sensations were bearers of information one might regard Reid’s theory as indirect.
In fact, in “Thomas Reid’s Account of the Objectivated Character of Perception” Nicholas Wolterstorff does regard Reid’s theory as indirect precisely for this reason. Wolterstorff asks, “when Reid says that in perception we have a conception of the perceived object, does he mean that we apprehend it by way of some singular concept, or does he mean that we apprehend it by way of our having Russellian acquaintance with it”?[14] Wolterstorff contends that Reid’s notion of conception must be by conceptual apprehension, and that therefore he must not be a direct realist. In other words, Wolterstorff contends that Reid’s theory is similar to Kant’s in the sense that both Reid and Kant think that in perception we apply concepts to an intuitive input – intuitions for Kant, sensations for Reid. Wolterstorff reaches this conclusion through his account of Reid’s theory of sensations as signs. This theory, he argues, is evidence that Reid regarded sensations as representing objects and qualities in virtue of information they carry and we interpret. In the next section, I will argue that Wolterstorff draws this conclusion because his definition of direct realism, like that of Immerwahr and Pappas, fails to identify the crucial respect in which Reid regarded himself as a direct realist.
§4. Reid’s Anti-Sensationalism
Wolterstorff is right that Reid speaks of sensations as ‘signs’ or ‘indicators’ of material objects and qualities. Reid also speaks of our ‘external senses’ as providing information. Finally, Reid claims that sensations ‘signify’ material objects and qualities and speaks of signs as being ‘interpreted’. Wolterstorff holds that we cannot make sense of Reid’s talk of interpretation or of signs as indicators unless we allow that Reid thought that sensations carry information and that in perception we apply concepts to sensation in order to extract this information. However, if sensations are the bearers of information which we extract by applying concepts to them, sensations mediate perception in a way that could be troubling for direct realism.
Recall that Pappas’ definition of epistemic direct realism was that perception is direct if we do not infer from a mediating mental entity to a belief about the existence of external objects. But Reid holds a stronger version of this premise. Reid insists not just that sensation are not the basis of any inference, but that sensations by themselves cannot, in principle be the basis of any inference. If signs were carriers of information in their own nature, solely in themselves, then they could (even if by exception) be a basis of inference. I contend that Immerwahr, Pappas and Wolterstorff have not captured in their definitions why Reid thinks this stronger version of EDR is true. Reid’s claim that sensations cannot form the basis of any inference is explained by what Keith DeRose has called Reid’s anti-sensationalism: his denial that the representational relation between sensations and objects is internal, that is, his denial that this relation is fixed by the intrinsic properties of the sensations and the objects.[15]
According to Reid, two contingent features of our constitution are that sensations are occasioned by material objects and qualities and that the sensations thus occasioned have the phenomenal character that they do have.
No man can give a reason, why the vibration of a body might not have given the sensation of smelling, and the effluvia of bodies affected our hearing, if it had so pleased our Maker. In like manner, no man can give a reason, why the sensations of smell, taste, or sound, might not have indicated hardness, as well as that sensation, which by our constitution, does indicate it. Indeed no man can conceive any sensation to resemble any known quality of bodies. Nor can any man show, by any good argument, that all our sensations might not have been as they are, though no body, nor quality of body, had ever existed.[16]
Reid allows not just for the possibility of spectrum inversion, but for the possibility of radical sensory inversion within and between all of the senses. In other words, Reid holds that neither the intrinsic character of our sensations nor the intrinsic character of objects determines whether sensations represent objects. However, because our constitution is such that particular perceptions constantly arise in conjunction with particular sensations, i.e., because sensations suggest conceptions of and beliefs about material objects and qualities, sensations acquire a signifying role in the total perceptual experience. Sensations are signs because God provides for the lawlike regularity of nature and for our minds as natural objects subject to such laws. We conceive of, and believe in, particular objects when presented with particular sensations. Reid is clear that the connection between sensations and the things they suggest arises from our constitution. If sensations were to suggest conceptions and beliefs in virtue of some quality that they possess, some information that they carry, the connection would not arise from our constitution. But Reid is clear that sensations are signs not by their nature, but that by “a law of our nature, such a conception and belief constantly and immediately follow the sensation”.[17]
The constancy of nature’s laws connects the sign with the thing signified.[18]
And in a like manner when certain sensations of my Mind are invariably accompanied with the conception and belief of certain external object, when it can be shewn that this connexion does not arise from Custom or Education, nor can be accounted for by any Law of the human mind hitherto known and received; We ought to hold this Connexion to be itself a Law of the human Mind, until we find some more general Law of which it is the consequence.[19]
Reid thinks of himself as a direct realist despite the fact that he regards sensations as mediating perception, despite the fact that he holds that sensations inform us about the world, despite the fact that he thinks we interpret sensations, and despite the fact that, like Berkeley, he regards sensations as signs of objects. We can understand how Reid can regard himself as a direct realist, despite these claims, if we ask in virtue of what Reid regards sensations as signs. Reid thinks that the representational relation that holds between sensations and objects is an external relation – it is fixed by extrinsic properties of sensations and objects, properties they have in virtue of being connected to one another by God. Reid thinks of himself as a direct realist because he denies that the relation between sensations and objects is internal, that is, he denies that the relation is determined by the intrinsic characters of sensations and objects.
The following definition of non-naïve direct realism clarifies in what sense Reid regarded himself as a direct realist. It consists in Pappas’ first two premises of perceptual direct realism, plus a new premise about what I will call representational direct realism:
PDR1) There are external physical objects which have at least some of the qualities they are typically perceived to have, and
PDR2) These external physical objects and some of their perceived qualities exist independently of perception and percipients.
RDR1) External physical objects and some of their qualities are directly perceived just in case 1) there is a representational relation R that holds between a mediating mental entity S and an object O, and 2) whether R holds is determined not by any intrinsic properties of S and O, but by extrinsic properties of S and O, and 3) S alone cannot, in principle, be the basis of any inference to O.
Premise RDR1 specifies in virtue of what Reid’s sensations are signs – they are signs in virtue of bearing a external relation to objects, a relation not determined by the intrinsic characters of sensations and objects. Notice that Reid and Berkeley agree about in virtue of what sensations are signs – they agree that sensations are signs in virtue of bearing an external rather than internal relation with what they signify. Berkeley does not hold, for example, that words signify in virtue of their shape, size or orientation on the page. Berkeley is not a direct realist, however, because he does not share with Reid the commitment to the existence of external objects and our perceptual knowledge of them made in premises PDR1 and 2.
If we ignore in virtue of what Reid regards sensations as signs we will find it difficult to understand how he views his own direct realism. If, for example, we read Reid as saying that sensations carry information in virtue of their causal relations to objects, we will have to ignore his acrobatic avoidance of causal language. In addition, our notion of direct realism will illuminate Reid’s direct realism, because causal mediation is not sort to which Reid’s direct realism is opposed. Reid opposes the idea that in perception requires an internal relation between mediating mental entities and objects. Similarly, if we treat sensations as signs in virtue of something ‘in their own nature’, something in their intrinsic character, this will conflict with Reid’s insistence that sensations alone cannot be the basis of any inference.
And as the feeling [of touch] hath no similitude to hardness, so neither can our reason perceive the least tie or connection between them; nor will the logician ever be able to show a reason why we should conclude hardness from this feeling, rather than softness, or any other quality whatsoever. But, in reality, all mankind are led by their constitution to conclude hardness from this feeling.[20]
Reid’s examples of other kinds of natural signs (smoke and facial expressions), show that he intends natural signs to be understood as indicators in virtue of the extrinsic relations they bear to what they signify. Smoke undoubtedly represents fire, facial expressions undoubtedly represent emotions, but they do so because they are connected to these things for some group of persons (in this case humans). Aliens without heads from an ice world would have no reason to connect smoke or a grimace with fire or fear. To use Hilary Putnam’s example, an ant’s path in the sand that resembles Winston Churchill does not thereby depict Winston Churchill; it is not the intrinsic character of words or pictures that make them represent objects.[21] In order for Reid’s natural signs to indicate as they do, there must be natural laws connecting them with what they signify and persons subject to those laws.
This
understanding of Reid’s direct realism also explains how he can speak of the
interpretation of sensations and remain a direct realist. Reid describes the interpretation of
sensations as just the movement from the sensation to what it signifies.
“Nature hath established a real connection between the signs and the things
signified; and nature hath also taught us the interpretation of the signs; so that, previous to experience, the sign
suggests the thing signified, and
creates the belief in it”.[22] In other words, the interpretation of the
sign is the very same thing as formation of the conception and belief, and the
formation of the conception and belief is the very same thing as a sign’s
suggesting the conception and belief.
And because the mind passes immediately from the sensation to that conception and belief of the object which we have in perception, in the same manner as it passes from signs to things signified by them, we have therefore called our sensations signs of external objects; finding no word more proper to express the function which Nature hath assigned them in perception, and the relation which they bear to their corresponding objects…Another requisite to our knowing things by signs is, that the appearance of the sign to the mind, be followed by the conception and belief of the things signified. Without this the sign is not understood or interpreted; and therefore is no sign to us, however fit in its own nature for that purpose… [23]
Reid claims that the mind passes from a sensation to a conception of and belief about the external object in the same way as it passes from a sensation as a sign to the object signified. This ‘passing’ just is the interpretation and the interpretation of sensations does not threaten Reid’s direct realism as he understands it. If there is nothing to interpretation over and above having a conception of and belief about an object, then according to this understanding it is not in virtue of interpreting the intrinsic character of sensations that we come to have a conception and belief. Rather, the interpretation, and the conception and belief, come down to the same thing, and all are equally directed towards material objects and qualities.
Reid’s own understanding of direct realism is compatible with his notion of sensations as signs. Thus, the controversy of Reid’s direct realism cannot be settled by deciding whether sensations are signs. Those who understand Reid as a direct realist and those who disagree can agree on this. The controversy may be settled only by understanding in virtue of what sensations are signs in Reid’s theory.
§5.
Conclusion
I have presented a definition of direct realism that helps us understand how Reid understood himself to be a direct realist. Recognizing that Reid’s direct realism is fundamentally grounded in his anti-sensationalism puts us in a better position to decide whether and to what degree his direct realism is coherent with the rest of his work. In addition, this definition is relevant for understanding both modern and contemporary theories in which the directness of perception is a central theme. It remains to be seen how many and which modern thinkers can be regarded as direct realists under this definition, but I suspect that Reid’s version of direct realism has some predecessors.
Finally, contemporary theories of perception that are committed to intermediary phenomenal entities such as qualia may look to Reid for a theory that is committed to analogous entities without sacrificing directness. As regards the institute, Reid’s theory is one in which conscious mental states are all intentional, but the intentionality of at least some of these states or events (sensations) is entirely derivative off of the intentionality of perception, conception or belief.