Wittgenstein on Seeing (1999)
Travis J. Denneson
The basic evil of Russell’s logic, as also of mine in the Tractatus, is that what a proposition is is illustrated by a few commonplace examples, and then pre-supposed as understood in full generality.
–§ 38, Remarks on the Philosophy of Psychology, Volume I
You need new conceptual glasses.
–§ 525, Remarks on the Philosophy of Psychology, Volume II
I. Introduction
Upon a careful and painstaking reading of the rather cryptic and difficult passages within Ludwig Wittgenstein’s Remarks on the Philosophy of Psychology, any reasonably intelligent person might still be left wondering what this obscure Austrian thinker might possibly be attempting to convey. Such a response to a work of Wittgenstein’s is not at all uncommon. Even some prominent thinkers misunderstand Wittgenstein’s ideas, as evidenced by the fact that many perceive of him as subscribing to philosophical schools of thought with which he would want no affiliation. He has been referred to as a behaviorist, a skeptic, a verificationist, and is even thought by some to be a practitioner of a sort of a priori anti-science. In my opinion, it is not very fair nor does it demonstrate a grasp of the sheer complexity of Wittgenstein’s thought to label him as a proponent of any of the above intellectual camps. In fact, Wittgenstein’s thought does not fit neatly into any type of established philosophical outlook or movement, but rather it seems to stand alone. It is known that he even repudiated the schools of thought which he himself had influenced, such as logical positivism and the “Oxford School” of linguistic philosophy. Since Wittgenstein’s ideas seem to elude classification so thoroughly, it is difficult to refer to them as anything but ‘Wittgensteinian’.
So, one might now ask, what exactly is ‘Wittgensteinian’ thought? What is it to maintain a ‘Wittgensteinian’ position on an issue? The answer is not altogether clear, because, as mentioned above, his work is of the utmost complexity. In fact, the question itself does not even seem to be applicable in Wittgenstein’s case. To organize all of Wittgenstein’s arguments and ideas into a neat, coherent philosophical system or program, in my opinion, would seem to do his work a great injustice. The reasoning behind such an opinion is that Wittgenstein does not subscribe to or advance any form of positive philosophical theory per se, for he does not believe that it is the purpose of philosophy to do so.Instead, Wittgenstein’s “method”, if we want to call it that, tends to be characterized by examining the various philosophical issues of his day in ways that no one before has. He tries to actually look at how things are, rather than think about how things must be according to various a priori philosophical principles. As Warren Goldfarb puts it, Wittgenstein is “a philosopher whose major concern is to fight against a priorism, to demolish pictures of how things must be, to expose ‘preconceived ideas to which reality must correspond'”.(1)
Therefore, it can be said that one of the most important things to keep in mind when reading Wittgenstein’s work is that he is concerned with freeing us from traditional, a priori philosophical presuppositions and is attempting to push us to look at philosophical issues in new and different ways. To put it simply, Wittgenstein’s aim is to dissolve the conceptual confusions in philosophy which lead us astray and compel us to impose certainties upon the world that do not really exist. In regard to such confusions concerning psychological concepts in particular,in can be said that Wittgenstein quite literally has volumes to say on the subject. In the Remarks on thePhilosophy of Psychology, Volumes I & II, Wittgenstein provides his readers with a wealth of counterexamples to our traditional philosophical accounts of various psychological phenomenon, all of which are designed to help demonstrate how such accounts seem to be misguided and mired in confusion. One of the most interesting of Wittgenstein’s challenges can be gleaned from the large number of passages devoted to the discussion of what he believes are the differences between ‘seeing’ and ‘interpreting’ that which one sees.
In §1 of the Remarks, for instance, Wittgenstein begins by presenting us with the above figure which can be seen in two different ways. One can see it either as an ‘F’, or as the mirror-image of an ‘F’. In both instances the physical appearance of the object does not seem at all to change, but the way that we see it somehow does. Wittgenstein then goes on to ask: “what does seeing the figure now this way and now that consist in?” (my italics) (2)
Here, Wittgenstein appears to be addressing our traditional philosophical inclination to search for some essential fact which constitutes our seeing the one way and seeing it another. He continues in this manner by asking whether we are actually seeing something different in each instance or whether we are seeing the same thing and merely interpreting it one way or the other. His response to this is not the latter, as our traditional philosophical inclinations would have us expect, but rather it is the former, namely that to see the figure one way and then another is to really see something different in each instance. What could Wittgenstein mean by this assertion? The answer lies somewhere in how the words ‘to see’ and ‘to interpret’ are conceptualized. As Wittgenstein puts it, interpreting is an action. When we see the figure one way instead of the other, we are not actively producing an interpretation of it, but rather our seeing it one way or another is an expression of our visual experience.
In more specific terms, we become passive observers to the different aspects that the object seems to take on as we view it. Therefore, according to Wittgenstein, the way that we actually see the image changes in this particular instance, not the way that we interpret it. Wittgenstein’s claims, it should be noted, have the same implications for our other senses as well. The purpose of this essay is to attempt to elucidate, assess, and defend Wittgenstein’s contentions that there are certain damning conceptual confusions in the traditional philosophical account of what it is to ‘see’ or to have any other sensory experience for that matter.
II. Down a Crooked Path
In order to help make sense of Wittgenstein’s remarks, it becomes imperative to further and more thoroughly explicate the traditional philosophical views to which he seems to be objecting. Philosophers have always wrestled with the problems of sense and perception. Is there really an external world? If there is, can its ultimate reality be known if all that we have to rely on is our perceptions of it derived from our senses? What is it to see? What does it mean to see? Modern science, particularly psychology, tries to shed light on the question of how we can be said to see, hear, taste, smell, and feel in terms of theories which explain how sensations become perceptions. Such theories are based upon empirical observations of the workings of the human body.
What science has discovered about the human body has led traditional philosophy, in its attempt to conform to the findings of science, to accept a number of presuppositions. One such presupposition is that the brain in itself is the seat of consciousness, and the rest of the body is a sort of mechanistic, organic vehicle in which the brain resides and by which it maneuvers through the world. The sensory apparati, namely the eyes, nose, ears, tongue, and nerve-endings on the skin, are treated as mere tools which enable us to absorb sensory ‘data’ from the outside world. Relative to ourselves, it would seem, the essence of the universe around us is to all effects and purposes mere data: lines, shapes, colors, light emissions, textures, etc. The data that the senses acquire and deliver to the brain via electrical impulses along the body’s nervous system, as current psychological theories dictate, is manifested in the brain in the form of unorganized percepts (which can be considered another form of data). It is then the job of the brain to somehow organize this perceptual data, (there is still no scientific consensus as to how the brain is said to perform this function) into a recognizable perception. This type of analysis is infused in Bertrand Russell’s treatment of what we can be said to be doing when we ‘see’:
In our environment it frequently happens that events occur together in bundles–such bundles as distinguish a cat from another kind of object. . . Then physics allows us to infer that light of certain frequencies is proceeding from the object to our eyes. Induction allows us to infer that this pattern of light, which, we will suppose, looks like a cat, probably proceeds from a region in which the other properties of cats are also present. Up to a point, we can test this hypothesis by experiment: we can touch the cat, and pick it up by the tail to see if it mews. Usually the experiment succeeds; when it does not, its failure is easily accounted for without modifying the laws of physics. (It is in this respect that physics is superior to ignorant common sense.) But all this elaborate work of induction, in so far as it belongs to common sense rather than science, is performed spontaneously by habit, which transforms the mere sensation into a perceptive experience. Broadly speaking, a perceptive experience is a dogmatic belief in what physics and induction show to be probable; it is wrong in its dogmatism, but usually right in its content. (3)
As I understand Russell, to ‘see’ a cat, in a nutshell, is to infer that our eyes are being affected by a bundle of certain frequencies of light. We in turn infer that this pattern of light comes from an area before us which beholds an object that resembles a cat. Based upon past experience with similar bundles and through force of habit, we infer or interpret that the resulting perception is in fact that of a cat (he even refers to ‘seeing’ the cat as a hypothesis, and further suggests a method by which to test it!). What seems disturbing about this account to me, and I’m inclined to think Wittgenstein would agree, is that there are an awful lot of “inferences” or “hypotheses” being made for a process which is described as an entirely “spontaneous” one. He tries to compensate for this oddity by declaring that all of this inference-making is done by habit, implying that it therefore goes unnoticed by conscious thought. It is not very clear what he means by “habit” in this account, and his use of the word will be discussed in detail later in this essay. For now, however, we are concerned with one simple question. What compels certain philosophers to conceptualize ‘seeing’ in such a manner?
Modern philosophers of the traditional vein, in their attempts to align the study of philosophy with the methodological commitments of science, have come under this aforementioned presupposition that the brain is really the “I” and the eyes do the seeing for it. This is because the scientific, empirical account of how we see–that is, strictly speaking, how the respective parts of the body work together–is based solely upon observations of the workings of the human body, vis-à-vis, how the eyes, nerves, and brain function in relation to the laws of physics. Philosophers of psychology, in their efforts to determine theoretically what it is to ‘see’, or to provide a theoretical account of what it means to speak of ‘seeing’ something, have become tied up in this empirical, scientific picture. The meaning of what it is to ‘see’ is entangled in scientific explanations of how the respective parts of our bodies function. Thus to ‘see’ is merely a job performed by the eyes: to take in raw visual data, since that is what the eyes apparently do. And since the eyes are the only things doing the actual ‘seeing’, all that is left for us to do is to infer or interpret what the eyes ‘see’, and where this interpretation occurs, of course, is in the brain. For such theoretical reductions regarding psychological concepts, traditional philosophy is indebted at least in part to the influence of the verificationist movement in the early twentieth century, out of which came programs such as logical positivism (or logical empiricism) and Russell’s logical atomism. Such schools of thought, which Wittgenstein somewhat influenced in his early years, are characterized by an attempt to rid philosophy of speculative metaphysics and align it with the methodological commitments of science.
Logical positivists, for example, maintain that the meaning of a proposition is its mode of verification. In other words, if the truth or falsity of statement cannot be empirically verified, then that statement is devoid of meaning and is thus nonsensical as an utterance. For instance, the statement, “God exists,” simply has no meaning since it cannot be empirically verified to be true or false. In the case of seeing, however, the statement, “I see a cat,” is a meaningful one since its truth or falsity can, according to the proponents of such schools of thought, be empirically verified. And since the meaning of this statement, according to proponents of this movement in philosophy, is the mode of empirically verifying its truth or falsity, such meaning must be put in terms of the method in which statement’s truth or falsity is determined. In this case, it involves observing the physical structure and mechanics of the eyes, optic nerves, and the region of the brain to which they attach, as well as the physical laws regarding the behavior of light.
So the meaning of the statement, “I see a cat,” will lie in a purely physical account of the process that is played out by our sensory apparati and our brain, which can be dryly provided as something like the following (yes, why not hash it over one more time?): (1) a bundle of specific light frequencies is entering my eyes and is being refracted through the lenses, registering on my rods and cones; (2) the eyes are now sending this data by electrochemical signals along the optic nerve to my brain; (3) such data is now spontaneously interpreted, based upon inductive familiarity with similar bundles of data in the past, and determined to indicate the presence of a cat.
Russell’s similar account above, along with his being the proponent of a closely related program called “logical atomism,” shows that he accepts a similar enough analysis. Is ‘seeing’ something which is empirically verifiable, which can be reduced to a theoretical account? A verificationist is committed to this type of theoretical conceptualization of ‘seeing’, because conceptualizing it in any other way would render such statements meaningless. And if they were to be deemed meaningless, then how could any form of empirical verification be meaningful when empirical verification in itself is in fact wholly dependent upon statements which declare sensory observations (i.e., “I see where the optic nerve attaches to the brain”)? In other words, if a verificationist cannot provide an empirically verifiable, theoretical account of what it is to ‘see’, then the entire verificationist project is dead because the means of verification itself will be rendered unverifiable.
The chief motive of logical positivism is to purge philosophy of the futile metaphysical quibbles that have kept it from advancing in the way that other disciplines have. Thus, it is unacceptable to a positivist to allow that ‘seeing’ can be conceptualized as simply something which we do, as something which can stand on its own two feet, without need of verification by a further supporting account. For to do so might give it that ghostly, metaphysical aspect which positivists are so intent on eradicating. Furthermore, it is surely true that we can ‘see’, as there are very few things that could seem more obvious. So if a verificationist cannot provide an account of ‘seeing’, what can he give an account of? Either way we wish to look at it, a verificationist is forced to give some kind of theoretical account like the one above, or else abandon his program altogether.
Philosophers who have allowed these elements in the philosophical tradition to influence them have thus created a sharp divide between what one sees and what one infers from what one sees, namely that what one sees is raw sensory data, and all else is interpretation. Again, we can rely on Russell to lend his support to this idea:
There are in fact no illusions of the senses, but only mistakes in interpreting sensational data as signs of things other than themselves. Or to speak more exactly, there is no evidence that there are illusions of the senses. Every sensation which is of a familiar kind brings with it various associated beliefs and expectations. When, say, we see and hear an airplane, we do not merely have the visual sensation and the auditory sensation of a whirring noise; spontaneously and without conscious thought we interpret what we see and hear and fill it out with customary adjuncts. To what extent we do this becomes obvious when we make a mistake–for example, when what we thought was an airplane turns out to be a bird. (4)
From this passage, it is apparent that Russell, in light of this rather simple example, maintains that there must be some constituting or essential object being perceived, and whether it is perceived to be one thing or another is determined solely by the brain’s interpretation, be it mistaken or not, of what that object is. The idea has come to be a fundamental presupposition of many modern philosophers of mind and psychology. In the case of the aforementioned figure ‘F’, therefore, this traditional analysis has instilled in many modern philosophers the conviction that there must be some common, essential object of perception between the ‘F’ and the mirror-image of the ‘F’, which is interpreted differently in each instance. According to this traditional picture, my seeing it one way and then another is due solely to whether I interpret it to be one way or the other, since my eyes have apparently done the ‘seeing’ for me beforehand.
Wittgenstein is particularly troubled by this sort of theoretical reduction of what we can be said to be doing when we say we see something. The above figure is meant to show, as are the ones soon to be discussed, that there are in fact illusions of the senses and thus to conceptualize seeing (as well as any other sensory experience) simply as a process of absorbing and interpreting ‘data’ is to terribly confuse the idea of what it actually means to ‘see’. Wittgenstein’s aim is to steer us off of this crooked path of theorizing based on such a priori presuppositions. In the following sections we will examine where Wittgenstein seems to believe that the philosophy of psychology, in regard to the senses, has gone astray.
III. Interpretation
In §7-8 of Volume I of the Remarks on the Philosophy of Psychology, Wittgenstein writes:
7. It is easy to describe the cases in which we are right to say we interpret what we see, as such-and-such.
8. When we interpret, we make a conjecture, we express a hypothesis, which may subsequently turn out false. If we say ‘I see this figure as an F’, there isn’t any verification or falsification for that, just as there isn’t for ‘I see a luminous red’. This is the kind of similarity that we must look for, in order to justify the use of the word ‘see’ in that context. If someone says that he knows by introspection that it is a case of ‘seeing’, the answer is: ‘And how do I know what you are calling introspection? You explain one mystery to me by another.’ (5)
Wittgenstein wants to begin by attempting to clarify what can properly be called a case of interpreting. An examination of the way in which we conceptualize ‘interpretation’ will do much to shed light on the way in which we conceptualize ‘seeing’. As Wittgenstein writes in the above sections, he takes ‘interpretation’ to be an action in which we make a conjecture or an inference, which may end up being false. This is something which is done, to at least some degree, consciously and deliberately.
When we look at the above figure, we can interpret it to be any number of things, such as a brick, a lidless box, a glass cube, a wire frame, or even a fallen monolith if we like. The point is that we consciously and deliberately drum up ideas concerning what this figure may be meant to represent. Ideas such as these can properly be called interpretations. The simplicity of the above figure demonstrates quite well that an interpretation seems to require, in my opinion, the use of one’s imagination. In order to make an interpretation, we seem to have to imagine the figure as being embellished a little this way or a little that, and/or as existing in one context or another.
For example, to interpret the figure as a brick, we might imagine it with an earthen color and a rough texture. To interpret it as a wire frame, we imagine that the sides of the figure are not solid, and that the lines are made out of thin metal wire. We could also interpret the figure to be a fallen monolith by imagining it composed of solid bedrock and lying on the ground at some ancient archaeological site, such as the Sphinx Temple on the Giza necropolis in Egypt. The point is that an interpretation is something which is not immediately seen, but is actively applied to that which is seen. We mentally embellish the object in a way which conforms to what we believe the object is or may be meant to represent. It would almost seem that interpretations for the above figure in particular are limited only by the limits of the observer’s imagination, since it appears completely solitary with virtually no contextual clues, aside from its shape, which might rule various interpretations out.
Of course, we can offer rather outlandish interpretations which will obviously turn out to be mistaken, such as if we were to interpret the figure to be a pyramid or a jack o’ lantern. Most interpretations of this figure, however, are going to be made in terms of what it actually looks like, and suffice it to say, there are a great many things in the world which share the appearance of this figure. In ordinary everyday life, however, there are many things for which only one interpretation is correct or the most plausible, such as when a person infers that there is a UFO hovering in the night sky and it turns out that it is only the planet Venus. In this case, it is also safe to say that a hypothesis is consciously made which subsequently turns out to be false.
Such a person is still, in some degree, making a conscious inference. What this person spontaneously ‘sees’ is a bright light in the sky, and then consciously interprets it to be a UFO. It might also be apt of us to say in this case, depending on the level of our skeptical sensibilities, that such a person’s imagination has run wild, so to speak. Is this not the way that we have come to use and understand the concept of interpretation in our everyday language game, namely that an interpretation is at least to some degree a conscious and deliberate inference which may or may not turn out to be correct? In my opinion the answer is yes. I also believe that Wittgenstein would agree, and it seems that he wants to argue that traditional philosophy, in its attempt to theoretically reduce what it means to ‘see’, has stretched the meaning of interpretation far beyond the boundaries of its customary usage.
In Russell’s account of what it is to ‘see’ a cat, he claims that through induction, we “infer” that the light patterns before us proceed from a cat. (‘Inference’ and ‘interpretation’ are interchangeable terms as far as such issues are concerned.) Moreover, Russell claims, as also earlier quoted, that “spontaneously and without conscious thought we interpret what we see and hear and fill it out with customary adjuncts.” This claim is especially troubling. How on earth does one make an interpretation without conscious thought? How is that possible? We could say, as I understand Russell in his account of ‘seeing’ a cat, that these inferences are made out of habit, and therefore occur undetected by conscious thought. In other words, we are constantly inferring from what our eyes ‘see’ without even thinking about it. However, this seems like a rather untenable position, no matter how one looks at it. If, by the word “habit,” Russell means that we have come to make these inferences so often and so routinely that such a practice has now become seemingly automatic to us, would he not be implying that at one time in our lives, before we developed such habits, our perceptions actually did not occur to us spontaneously as they do now? Personally, I do not recall ever having had to develop the habit of ‘seeing’ things properly; of going through a process of hypothesizing about whether I see a cat or something else. If Russell means this by “habit,” he is then treating ‘seeing’ as if it were a conscious process which we have developed to the point of needing to think little or nothing of while doing and have come to take for granted, like walking, riding a bicycle, or driving a five-speed clutch. Is it so that we at some time in our early lives found it necessary to make conscious inferences regarding everything in our environment? This is simply unfathomable, because without the spontaneity of our alleged inferences, we would have been plagued by the constant awareness that we could be mistaken about everything that we see. If such were the case, we would all be born into a state of severe skeptical doubt. My doubt, in particular, is that Russell would actually mean such silliness by his use of the word “habit.” I find it to be much more plausible that by the word “habit,” Russell is referring to a natural disposition or inclination of the human brain which works at the unconscious level. However, even though this use may have fewer ridiculous ramifications, his account still remains problematic.
If to ‘interpret’ is a habit in this sense of the term, then Russell seems to be conceptualizing the word ‘interpretation’ in a manner which includes unconscious or subconscious processes in the brain as part of its meaning. If it is so that Russell wishes to call the brain’s unconscious processing of sensory data ‘interpretation’, then I suppose he may. But in doing so he has to recognize that he is now utilizing two different meanings or uses for the word ‘interpretation’: (1) the unconscious processing and organizing of sensory data by the brain; and (2) the conscious and deliberate act of conjecturing or expressing a hypothesis. However, Russell does not seem to get himself clear about how he is using the term, and tends to slip seamlessly from one usage to the other, incorporating both senses of the term into one conflicted concept. Therefore, a theoretical account of ‘seeing’, such as Russell’s, gets a lot of mileage out of this confused conception of ‘interpretation’, since it seems to use two different applications of the word at the same time. It can account for both why we are sometimes mistaken in our perceptions as well as why it is that our perceptions seem to occur spontaneously and without conscious thought. On the one hand, if we use the word ‘interpretation’ to mean the brain’s unconscious processing of data acquired by the senses, we can account for the fact that many of our perceptions seem to occur spontaneously and without conscious thought. The customary usage of the word cannot account for this consideration, because it would seem obvious to us that seeing does not always require conscious inferences. On the other hand, the customary conceptualization of the word can account for why we might think we see an airplane in the sky and it turns out to be a bird, to use Russell’s example–that our conscious inferences based on our perceptions can sometimes be mistaken. And unless we wish to say that the unconscious, mechanistic processing of sensory ‘data’ in the brain can sometimes be “mistaken” in the way that a hypothesis can, we seem to have to admit that this usage of the word does not account for mistakes or ambiguities in perception. (Let us rule out cases such as mental illness and ingestion of mood-altering and hallucinogenic substances, since such instances seem more akin to an overall disfunctioning or alteration of the functioning of the brain or certain regions of the brain, rather than they do to isolated and random “mistakes”.) Therefore, it seems that in the case of such theoretical reductions of ‘seeing’, the usage of the term ‘interpretation’ is terribly confused in that it is characterized by two apparently incompatible elements somehow entangled together into one distorted concept.
With such a confused conception of ‘interpretation’, Russell and other philosophers who argue along a traditional line are trying to have it both ways, so to speak. The meaning of the word is stretched so far as to include that which it seems to contradict. The problem, of course, lies in using the word ‘interpretation’ to denote the unconscious processing of sensory data in the brain. However the brain may organize and process information (even the terms ‘process’ or ‘organize’ may not be fitting, for we know relatively very little about how the brain functions in this regard) from the sensory apparati, it is not a case of ‘interpretation’ as the term is customarily used.
So why call it ‘interpreting’? How did this term come to be used in this seemingly improper manner? As mentioned earlier in this essay, in the traditional philosophical picture, there is a tendency to think of the brain itself as the “I”, or as the seat of consciousness. And if all consciousness resides in the brain itself, the conscious act of ‘interpreting’ can also be ascribed to the brain. Therefore, when modern psychology or neuroscience provides us with an empirical account of ‘seeing’, and tells us that the brain somehow ‘organizes’ visual data into recognizable perceptions, we tend to associate ‘organizing’ with ‘interpreting’, and say that it simply happens spontaneously and without conscious thought. This sort of conceptualization of how the brain works unconsciously by way of a leap of association is the result of the personification of the brain as a conscious, sentient entity in itself. It leads us to ascribe qualities to the brain that are, as I wish to argue, those of a human as a whole, such as inference-making. Therefore the brain is thought of as making interpretations on both the conscious and the unconscious level. We thus end up with an application of the word ‘interpretation’ which seems to go against its customary usage, namely that we take a word which is used to denote a conscious activity and use it to denote an unconscious one. This is the trap into which traditional philosophy has fallen: to maintain that the eyes ‘see’ and the ears ‘hear’, and that we, as brains, consciously as well as unconsciously interpret the information that we receive from the sensory apparati that are positioned throughout the bodies in which we reside. In other words, we each exist as a brain in a vat–and in our case our bodies are the vats. How can a brain by itself exhibit consciousness?
Since the aforementioned purpose of this paper is to deal with how human beings are said to experience the world through their senses, I wish to argue that this way of perceiving ourselves is wrong-headed, and this confused conceptualization of the term ‘interpretation’, as explained above, is symptomatic of such thinking. Again, to ‘interpret’ is to perform the act of making a conjecture, or to express a hypothesis, which may or may not turn out to be correct. To this regard it is an action that is at least to some degree performed consciously and deliberately. This is the way that we have come to use and understand the concept in our language game. To use the term to denote unconscious processes (such as synaptic or neural functions, chemical balances, etc., as explained by neuroscience) performed by the brain is to confuse the concept of interpretation with something that it is not. With such confusions concerning the concept of ‘interpretation’ hopefully now behind us, we can more readily turn our attention to the discussion of what it is to ‘see’.
IV. Seeing
What do we see when we observe the above figure? What we see in the above figure, of course, is dependent upon that with which we are familiar. Those who are not acquainted with the shape and form of a rabbit but are with that of a duck will see only a duck–and vice versa. To see the above image as a duck, and then to see it as a rabbit, is to see two different aspects of the image, just as it is to see the F-figure as an ‘F’ or a mirror-image of an ‘F’ (though these can be considered to be two different kinds of aspect perception). When we normally speak of seeing in our everyday language-game, we are not inclined to say, “I see the picture as a duck,” but rather we simply say, “I see a duck.”
In §515-517 of Volume II of the Remarks on the Philosophy of Psychology, Wittgenstein puts it this way:
515. When I’m looking at the photograph, I don’t tell myself ‘That could be seen as a human being’. Nor when looking at an F do I say: ‘That could be seen as an F’.
516. If somebody showed me the figure and asked me ‘What is that?’, I could answer him only that way. –I couldn’t answer: ‘I take that to be a . . .’ or ‘Probably that is a . . .’ Any more than I take letters to be this or that when I’m reading a book.
517. ‘I see it as a . . .’ is connected with ‘I’m trying to see it as . . .’, or ‘I can’t see it as . . . yet’. But you cannot try to see the regular F as a regular F. (6)
When we look at the duck-rabbit, without any awareness that it can be seen two different ways, we only see either a duck or a rabbit. Let’s say that we can only see the duck, for we are entirely unfamiliar with rabbits. We would never then say, or it would at least seem very peculiar of us to say, “I see it as a duck,” just as it would seem utterly strange to hear someone who is looking up toward the sky at a distant airplane say, “I see it as an airplane!” One would simply say, “I see an airplane.” We speak this way normally because no alternatives are relevant in such a case, since to our knowledge there are no alternative aspects to the object in question. In other words, when we are observing a singular object that is quite familiar and seemingly unmistakable to us, we simply see it, without any need for conjecture or inference. When we say that we see something, we are expressing a belief that a specific perception is apparent to us, wherein no alternative perceptions are relevant. For to say that we see an object as something holds the implication that it can be perceived in more than one way, but does this mean that we are making different interpretations of one essential object or image, or are we actually seeing different aspects? Now, let us say that we are familiar with both ducks and rabbits, and can therefore recognize both aspects of the duck-rabbit image. In other words, we see the image as a duck and then we see the image as a rabbit (in whichever order it may occur). Are we ‘seeing’ one essential object of perception, and merely interpreting it differently, or are we genuinely ‘seeing’ a duck and then ‘seeing’ a rabbit? The traditional stance on this issue would, of course, be of the former persuasion.
However, there is something about the nature of this picture which tells against the traditional, theoretical account of what it is to ‘see’, namely that it appears to have the effect of an illusion. We see it as two entirely different, alternating images, despite the fact that the drawing itself does not at all change. The duck and the rabbit both seem to impress themselves upon us in some eery, inexplicable manner. Remember that Russell, as quoted earlier, maintains that “there are in fact no illusions of the senses, but only mistakes in interpreting sensational data as things other than themselves.” Where is the interpretation in this case? What are we interpreting? The complete ambiguity of this figure makes it extremely difficult, if not impossible, to find something constituting about its appearance. It therefore causes problems for the traditional account. For an account of ‘seeing’ such as Russell’s to hold water, there must be something which we can point to that constitutes both the duck and the rabbit as the proper object of perception. There does not seem to be, however, anything that we can point to in this regard.
In Russell’s defense, we could say that it is a drawing of the same shape, a specific conglomeration of lines and curves, or something to that effect, but this seems trivial and unsatisfying to us–almost as if we were to say that the proper object of sense in this case is a “thing”. It might seem logical or common-sensical to someone like Russell that the duck-rabbit figure is one and the same picture, and that we simply interpret it differently, but as it appears, we simply cannot escape the experience of seeing two entirely unique pictures. In respect to Russell’s claim, vis-à-vis, that there are no illusions of the senses, only mistakes in interpreting sensational data, how would seeing the ambiguous figure one way or other be a mistake? What is the correct way to see it? Wittgenstein puts it this way:
75. Should I say: “The picture-rabbit and the picture-duck look just the same”?! Something militates against that–But can’t I say: they look just the same, namely like this–and now I produce the ambiguous drawing. (The draft of water, the draft of a treaty.) But if I now wanted to offer reasons against this way of putting things–what would I have to say? That one sees the picture differently each time, if it is now a duck and now a rabbit–or, that what is the beak in the duck is the ears in the rabbit, etc.? (7)
It does in fact seem wrong to say that the picture-duck and the picture-rabbit look the same, because they are two completely different pictures. Nor would it be possible to say that they can both be viewed at the same time. To view the two pictures at the same time, to use an example of Wittgenstein’s regarding ‘meaning’ (8), would be like trying to use the word “bank” in a manner that meant both of its meanings at the same time (financial bank, river bank). Even if it were so that the financial bank happens to be built on a river bank, we would still only be able to mean one designation or the other at one time if we were to say, for example, “I am going to the bank.” Sharing the same physical location in space does not make it so that the word to denote the objects can be used to mean both concepts at once. The concepts themselves are entirely alien to each other. The physical, symbolical appearance and phonetic sound of word may be the same, but the meaning remains ambiguous, just as in the duck-rabbit picture, wherein the basic physical structure and shape of the drawing is the same, but the apparent picture is ambiguous. This is what it is to see something as. Wittgenstein wants to draw a fine line between ‘seeing’ and ‘interpretation’ by showing that there are cases in which we can see things as without making inferences. An interpretation, as we have already established, is a conscious, deliberate act. In the duck-rabbit picture, there are no active inferences being made, no conjecture as to what we see. We simply alternate between passively seeing the ambiguous picture as a duck and seeing it as a rabbit. In Part II of the Philosophical Investigations, Wittgenstein elaborates:
Imagine a physiological explanation of the experience. Let it be this: When we look at the figure, our eyes scan it repeatedly, always following a particular path. The path corresponds to a particular pattern of oscillation of the eyeballs in the act of looking. It is possible to jump from one such pattern to another and for the two to alternate. (Aspects A.) Certain patterns of movement are physiologically impossible; hence, for example, I cannot see the schematic cube as two interpenetrating prisms. And so on. (9)
Therefore, ‘seeing’, or ‘seeing as’ is simply an experience which neither has nor needs any kind of theoretical verification. The meaning of the concept lies in this experience. To say that we ‘see’ something can stand on its own two feet, without the need for verification by any further supporting account. Let us put it another way. What can we say of those who have what is called “perfect pitch”? Those who are gifted in such a way have the ability to recognize the pitch of a sound as spontaneously and readily as the rest of us can recognize colors (of those of us who are not colorblind, anyway). If we were to play a specific note, for instance, say E flat, on the guitar for such a person, he will immediately recognize it as E flat, in just the same way that we can immediately recognize the color blue when it is presented to us. Such people have no need for pitch-pipes, since they can tune an instrument by simply listening to it and ‘hearing’ whether or not it is in tune. Where is the inference in this case? The best that the rest of us can do is to try and remember certain pitches and make an inference by attempting to match what we hear with what we remember having heard in the past. It would seem that a person with perfect pitch experiences a genuine ‘hearing’ of one pitch or another. A theoretical account of ‘hearing’ that is along the same lines as the account of ‘seeing’ would thus not seem appropriate, particularly in light of this phenomenon of perfect pitch. There is no verification for ‘hearing’ in this case, and there are no interpretations being made.
Seeing the duck-rabbit figure as a duck, or as a rabbit, is therefore much like having perfect pitch, in that there are no active inferences being made. It is a passive experience, just as it is a passive experience for a person with perfect pitch to ‘hear’ E flat. Furthermore, to try to give a theoretical account of what it is to ‘see’–to put it in terms, as discussed earlier in this essay, which describe the physical processes undergone by the respective areas in the body–is akin to giving a theoretical account of what it is to walk. It is like describing how messages from the brain tell one foot to place itself in front of the other in such and such a fashion, in what manner the knees, ankles, and toes bend, etc., etc., and claiming that such an account of the mechanics of human bipedal locomotion is what it is to walk. Is that really what it means to walk? Or is that just the way that science attempts to explain how we walk? When we tell someone to ‘walk’ to the store, is this just short for telling them to undergo the above process? What if we say, “He is going to go walking,” that is to say, to go on a ‘walk’? In our language game, we use intentional or action concepts to describe what we do. Science and traditional philosophy try to provide a physiological account of these concepts. Wittgenstein hints at the difficulties that are involved in such a practice:
76. The facts of human natural history that throw light on our problem, are difficult for us to find out, for our talk passes them by, it is occupied with other things. (In the same way we tell someone: “Go into the shop and buy . . .” –not: “Put your left foot in front of your right foot etc. etc., then put coins down on the counter, etc. etc.”) (10)
There is a distinction to be made here concerning this issue. Some have made the claim, as mentioned in the introduction to this essay, that Wittgenstein is practicing a kind of philosophical anti-science, in that his arguments regarding mind and psychology are seen as an attack on neuroscience and psychology. Wittgenstein does not have any quarrels with legitimate scientific inquiry or its findings. His aim is to prevent us from adopting a ‘scientistic’ view of things, a view that every linguistic concept we use to describe what we do, such as ‘seeing’, ‘believing’, and ‘understanding’, point to factual, physical things in the world or in ourselves, and can thus be scientifically investigated and expounded. He wants to show us that some concepts are in need of clarification before they can be properly examined or determined to be worth examining at all:
For Wittgenstein, it is characteristic of the notions that figure in philosophical problems–prominently, mental concepts and linguistic concepts like meaning–that a structure is imposed on them, without grounding in the ordinary use of these notions and without being noticed, when they are taken to be amenable to certain explanatory projects. Hence, only through clarification of what the legitimate questions are can proper sense be made of the applicability of science. A scientistic viewpoint ignores this need for clarification. As a result, for Wittgenstein scientism is just as misguidedly metaphysical as traditional, more transparently a prioristic, approaches. (11)
In the case of ‘seeing’, therefore, Wittgenstein is trying to clarify the concept so as to show where scientific examination would and would not be applicable. A scientific explanation of ‘seeing’ will only be a physiological one, which in turn cannot explain, cannot penetrate into the brute fact of the experience of ‘seeing’ the figure as a duck and then as a rabbit, for example. Scientific examination is simply not applicable in such a case. Science can tell us how the eyes and nerves work, what kind of chemicals are released in the brain, how much electrical activity is occurring in what part of the brain under what stimuli, and so on. But to be able to give an account of what it is to ‘believe’, ‘desire’, ‘understand’, or ‘see’ (in the sense it has been discussed in this essay) is an entirely different matter. And for philosophy to fall into the trap of working from a priori presuppositions that are grounded in such misapplied scientific projects is a great mistake.
V. Conclusion
The purpose of this paper has been to discuss and defend Wittgenstein’s views concerning the conceptual confusions that are rampant in the traditional philosophical picture of what it is to ‘see’, or to have any other sensory experience for that matter. The chief confusions lie in the prevailing and allegedly common-sensical conceptions of the terms ‘interpretation’ and ‘seeing’. An interpretation or inference is a conscious action which is performed over and above what happens when we ‘see’ something. It is not tenable to use the concept to denote unconscious, mechanistic processes in the brain. The fact that there are illusions of the senses, contrary to what Russell claims, somehow revolts against the idea that every object of sense contains root data that the brain merely interprets in different ways. Such illusions, through their ambiguity, show that there are in fact cases where we cansee as without need for interpretation, and thus ‘seeing’ is an experience which seems to come as a brute fact, neither having nor requiring verification in the form of a physicalistic account.
Wittgenstein wants to break us out of the scientistic habit of presupposing that mental and psychological attributions like beliefs, desires, understanding, and sensory experiences of every kind can be reduced to theoretical, physiological accounts. He seems to be showing us, contrary to those who mistakenly take him to be a behaviorist, that there are internal, inexplicable things going on within us, that the things we do and experience cannot all be explained or accounted for by pointing to some physical origin or process.
We do not each exist as a brain in a vat. Our eyes are not simply tools used by the brain which do the ‘seeing’ for it. Our eyes do not ‘see’, we do. Our brains are not conscious, but rather we are, and we make the ‘interpretations’ concerning what we see when we find it necessary to do so. Our ears do not ‘hear’, we do, and so on. These are the simple brute facts of our existence. The brain is merely another organ in the body, the purpose of which is to facilitate the various things that human bodies do, such as thinking, walking, seeing, desiring, and interpreting. The scientistic inclination to search for a physical account of mental and psychological notions is an expression of the mental discomfort we feel at the thought of being unable to provide reasons for why we are the way we are and why we do the things we do. Each of us exists as a whole creature, as a conundrum made of physical material that is, in some enigmatic and marvelous way, much more than the sum of its parts.
Endnotes
1. Goldfarb, Warren. “Wittgenstein on Understanding”. Midwest Studies in Philosophy, XVII. 1992. pp. 110. Goldfarb quotes Wittgenstein from the Philosophical Investigations, §131.
2. Wittgenstein, Ludwig. Remarks on the Philosophy of Psychology. Volume I. University of Chicago Press. 1980. pp. 2e
3. Denonn, Lester E. Ed. The Bertrand Russell Dictionary of Mind, Matter, and Morals. Citadel Press. New York. 1952. pp. 176. Excerpted from Russell’s An Inquiry into Meaning and Truth.
4. Russell, Bertrand. Human Knowledge: Its Scope and Limits. Simon & Schuster. New York. 1948. pp. 167
5. Wittgenstein, Ludwig. Remarks on the Philosophy of Psychology. Volume I. University of Chicago Press. 1980. pp. 3e
6. Wittgenstein, Ludwig. Remarks on the Philosophy of Psychology. Volume II. University of Chicago Press. 1980. pp. 93e
7. Wittgenstein, Ludwig. Remarks on the Philosophy of Psychology. Volume I. University of Chicago Press. 1980. pp. 16e
8. See §94 and §184, for example, in Ibid.
9. Wittgenstein, Ludwig. Philosophical Investigations. Prentice Hall. New Jersey. 1958. pp. 212e
10. Wittgenstein, Ludwig. Remarks on the Philosophy of Psychology. Vol. I. University of Chicago Press. 1980. pp. 17e
11. Goldfarb, Warren. “Wittgenstein on Understanding”. Midwest Studies in Philosophy, XVII. 1992. pp. 112