The problem of “seeing” – Wittgenstein, Köhler, and visual perception


When reading parts of Philosophical Investigations and Remarks on the Philosophy of Psychology, one can observe that the problems concerning visual perception are given in the form of a dialogue between Wittgenstein and an unknown interlocutor. Upon further inspection it becomes evident that this interlocutor is Köhler, and we can recognise him by his ideas, or at least Wittgenstein’s own understanding and interpretation of some important gestalt psychology concepts. Such form of argumentation allows Wittgenstein not only to counter gestalt ideas without interruption, but it also gives him enough room to properly extrapolate the main objections and explain his own concepts which arise from them. There are pages upon pages of seemingly repetitive objections explained time and time again (as Wittgenstein does), but this is not just needless repetition. Wittgenstein tries to break through the language barrier and reach the heart of the problem of gestalt and of visual perception as such. How a few counterarguments can turn into an entirely different interpretation of the problem will be shown in the later part of this article, when we turn to another work of Wittgenstein’s dealing exclusively with “seeing,” Remarks on Colour.

For the first part of this inquiry, it is important to note that all of Wittgenstein’s objections to Köhler can be divided into two main parts. First, Wittgenstein presents Köhler’s ideas and explains why they are wrong. We could say that in this part, Wittgenstein explains how Köhler’s argumentation is wrong. Second, through his own theory of aspect changes, which he presents as an alternative to Köhler’s presentation of organizational change, Wittgenstein transcends the problems themselves and gives concrete reasons for Köhler’s and other gestaltists’ incorrect understanding of human perception. He traces the arguments back to the problem of seeing and using language to describe what we believe to be an objective reality. A very Wittgensteinean turn indeed. In this part, Wittgenstein therefore explains the reasons behind Köhler’s false interpretation and arguments, or in other words, why Köhler is wrong.

With the help of some great researchers on Wittgenstein’s objections to gestalt psychology, we will try to uncover and interpret his thoughts and ideas surrounding the concepts of seeing and perception. In addition, a major point from his late work Remarks on Colour will help us understand the source of his dealing with visual perception from a philosophical perspective, and why there is more to opposing Köhler than a simple disagreement. What this article will try to show, therefore, is not who was right regarding visual perception. Through examining Wittgenstein’s argumentation, we will try to uncover the underlying problems and present opposing views as sources of important insight for future evaluations of these ideas.

Köhler’s ideas

To properly understand Wittgenstein’s objections, we must first acquaint ourselves with Köhler’s main ideas, or at least how Wittgenstein perceives them. “‘Object’ and ‘ground’ ­— Köhler wants to say — are visual concepts like red and round” (Wittgenstein 1980a, 177e). Object and ground, or figure and ground alignment, or organization as Köhler calls it, can be explained most straightforwardly by Wittgenstein’s example of a double cross “illusion”:

(Eilan 2013, 1)

This picture can be interpreted, or seen, as a white cross on a black background or vice versa. We cannot see both of them at the same time, and this represents the phenomenon of organization for gestaltists. When we see one cross or the other, we see a unit, the parts of which seem to naturally belong together. This “belonging together” is explained by Janette Dinishak in her PhD thesis: “In the visual field, for example, certain areas of the field seem to belong together. That is, the visual field organizes into circumscribed units or Gestalten” (2008, 2). Organizational principles are therefore at the heart of a gestalt. Köhler is saying that the change in organization that occurs in the above example, the difference in seeing a white or a black cross, has as much visual impact as changes in colour and shape do. In other words, he is saying that gestalts are just as visual as colour and shape. This means that when a black cross changes to a white cross on a black background in the above example, this change is on the same level or in the same category as if, for example, we were first shown a picture of a green field and then a picture of a blue sea. The object is different, and we see different things in each instance. 

Naomi Eilan, in her article “On the Paradox of Gestalt Switches,” boils down this gestalt idea into two main principles: “According to the Gestaltists, the organizational phenomena are ‘primitive’ and ‘purely visual’” (2013, 5). She calls this “the pure vision claim.” That and “the new object claim,” are the most important foundations for gestalt psychologists’ argument of organization being on the same level as colour and shape. Organizational change is for them a purely visual phenomenon, and as such presents us with a completely new visual object after the change. This is where “the pure vision claim” connects to “the new object claim.” The connection is well summarised by Dinishak when she explains how “Köhler talks this way in his descriptions of the experience of a change in organization. He remarks that a change of organization is a change in visual reality, the coming into and going out of ‘existence’ of visual units or circumscribed wholes, each with its own distinct concrete, real form” (2008, 128). In short, this means that if you treat organization as purely visual, then its change means a change in the actual object of sight.

“Köhler is wrong!”

After explaining Köhler’s main claims, we now turn to Wittgenstein’s counterarguments and his own explanation of the organizational phenomenon. Wittgenstein argues that there is “the categorial difference between the two ‘objects’ of sight,” i.e. between the change in organization, as Köhler calls it, and change in shape or colour (2009, 203e). He reformulates the first as actually being “the change of an aspect.” Presenting another different example of what this aspect means exactly, in paragraph 113 of the second part of Philosophical Investigations he writes, “I observe a face, and then suddenly notice its likeness to another. I see that it has not changed; and yet I see it differently. I call this experience ‘noticing an aspect’” (2009, 203e). We now come to the main clash of opinions: what is a change in organization for gestaltists, belonging to purely visual phenomena and therefore representing a change in the actual object of sight, is for Wittgenstein “only” a change in aspect, more precisely the organizational aspect. For Wittgenstein, changes in colour and shape are simply too different from changes in organization that they could be put in the same category. This is beautifully summarised in what Eilan dubs “the paradox of gestalt switches”: “What is incomprehensible is that nothing, and yet everything has changed, after all. That is the only way to put it” (Wittgenstein 1980b, 86e). Here we can begin to follow Wittgenstein’s arguments focusing mainly on the insurmountable differences between aspect switching and a change in shape and/or colour, which, in his opinion, prevail over the similarities.

For the sake of clarity, we will once again turn to Naomi Eilan and her article, wherein she divides Wittgenstein’s counter arguments into two parts: arguments against the “new object claim,” and arguments against the “pure vision claim.” We have already briefly touched upon the first argument against “the new object claim,” which is simply just “a statement of the paradox” (see above), or, as Eilan also calls it, “the Objection from Phenomenology.” She explains that “[i]f we say we are seeing two different objects, one before and one after the switch, we distort an element of the phenomenology of aspect switches that Wittgenstein regards as essential to it” (Eilan 2013, 6). This means that there is an essential phenomenological component to aspect switches, which we could call “sameness after change.” Wittgenstein emphasizes it many times throughout his critique of Köhler, also in connection with perception. In the second part of Philosophical Investigations he states that“[t]he expression of a change of aspect is an expression of a new perception and, at the same time, an expression of an unchanged perception” (2009, 206e). We can now see that for Wittgenstein there is an essential difference between object change and aspect change, namely that in the second case there is a kind of “sameness” that persists throughout the change, and this cannot be found when objects really change. This phenomenological part that stays the same is what makes the cases of Köhler’s organizational change fundamentally different from the change in colour and shape.

If you disregard this difference and emphasize that there is indeed a new object after aspect change, you are faced with another problem, which brings us to Wittgenstein’s second argument, dealing with the “location” of visual impression: “Someone who puts the ‘organization’ of a visual impression on a level with colours and shapes would be taking it for granted that the visual impression is an inner object. Of course, this makes this object chimerical, a strangely vacillating entity. For the similarity to a picture is now impaired” (Wittgenstein 2009, 206e). The problem, as Wittgenstein perceives it, arises from the previously mentioned “sameness after change.” When the organizational aspect changes, something does indeed change, but we can clearly see that something else stays the same. We cannot grasp what this something is exactly, but the fact is that if you want to disregard that which stays the same, like Köhler does, and focus only on the new object which is completely different after the change, you need to move the object before you from the external world to some kind of an internal world[1]In her article, Naomi Eilan puts much more emphasis on the problem of ‘internalism’ as implied in the New Object Claim. For a detailed explanation of the claim and its consequences for … Continue reading. If you claim that, just as with shape and colour, everything has changed after the organizational change, the subject of the study cannot be the external object before you, because with that external object you can clearly see that after the aspect change, there is that previously mentioned “something” that stayed the same. Therefore, you need a new “playground,” an internal object world in this case, where there really is a complete change in object when aspects switch. As Wittgenstein points out, “this makes object chimerical” and there is no firm grasp of it anymore. Furthermore, taking this course looks as if, instead of trying to solve the problem, you are putting every effort into eluding it. Köhler, in Wittgenstein’s view, disregards the external object world because it does not fit his theory of there really being a new object after the change. The problem is that such argumentation does not explain the phenomenon of “the unchangingness,” which persists as “an essential component of the phenomenology of aspect change” (Eilan 2013, 6). 

“The new object claim,” which was thoroughly discussed above, is obviously linked to “the pure vision claim,” which we have briefly mentioned, but Wittgenstein’s objections to the latter take us in a completely different direction. Let us turn to Wittgenstein’s favoured example of the Rubin-vase:

(Eilan 2013, 1)

Looking at this picture, I say, “I see a vase,” and then, after recognizing two faces, I exclaim, “but now I see two faces!” Similarly, when looking at the painting of a horse, I will say, “I see a horse,” and then, after being shown a picture of a deer, I will say, “I see a deer.” The problem lies in the word ‘see.’ It is indeed the same word used in all four cases, but it is essentially different in the first two. What Wittgenstein claims is that the fact we use the word “see” in all of the above cases does not mean that all of the cases of change are the same. As he explains, there are “[t]wo [in this example, otherwise many more] uses of the word ‘see.’ The one: ‘What do you see there?’ – ‘I see this’ (and then a description, a drawing, a copy). The other: ‘I see a likeness in these two faces’” (Wittgenstein 2009, 203e). Although the same word “see” is used, Wittgenstein claims that it can mean very different things when used with different examples. With this problem, Wittgenstein takes us beyond simply arguing against Köhler – he tries to find reasons behind Köhler’s mistake.

Sources of Köhler’s mistake 

We have already mentioned Wittgenstein’s starting point in explaining many different uses of the word “see.” His main argument against Köhler that follows from this is that Köhler made a mistake because he focused too much on the similarities between what he called organizational changes and changes in shape and colour. Köhler’s mistake can be explained in two ways. First, as Eilan discovers in her article, Wittgenstein believes that Köhler makes a kind of semantic mistake, mixing up two different meanings of the word “see.” There is nothing wrong in saying we see the changes in organization just like we say we see the change of a picture before us. In fact, there is no “better” way of explaining the aspectual change than with the word “see.” The difference, however, is in the meaning. In other words, as Eilan explains it, “Wittgenstein’s diagnosis of Kohler’s New Object Claim is this. What makes Kohler say that we are confronted with a new concrete visual object after the switch is the complete naturalness of saying that we see the new organization, and hence, that it must be borne by a new concrete object” (2013, 9).

The second way of explaining Köhler’s mistake is the way Dinishak does in her PhD thesis. She argues that Wittgenstein believes the main reason for Köhler’s mistake is a false comparison or analogy made by the gestaltist between the two concepts of change. She believes Wittgenstein argues that “the claims that figure in Köhler’s theory express an inclination to favor a comparison between seeing aspects and seeing colour and shape” (2008, 128). Köhler obviously sees there is a kind of difference between changes in colour and shape and changes in organizational aspect, as Wittgenstein calls it, but he neglects those differences in favour of describing the change in aspect through analogy with the change in shape and colour. He believes that, just as in the second case, there is a new visual object that emerges after the change in the first case (we explained this argument in the previous section). In her PhD thesis, Dinishak also quotes paragraph 879 from Wittgenstein’s Remarks on the Philosophy of Psychology Volume I, where he directly mentions analogy and explains what we have just mentioned: “I see two different visual objects, you say, which merely have something in common with one another. In saying that you are only laying stress on some analogies at the expense of others. But now this emphasis needs to be grammatically justified” (Wittgenstein 1980a, 155e). The main problem Wittgenstein sees is this: using an analogy between two concepts often makes us neglect the main differences between these two concepts. In this case, the differences are so important that they negate Köhler’s analogy and overrule the similarities between “seeing” and “seeing aspect” (see Dinishak 2008, 130-131).

Seeing colours

From the concept of organization as explained by Köhler and other gestalt psychologists, Wittgenstein has taken us further, into the problem of seeing itself. But, unlike the psychological or biological treatment of this subject, we have begun uncovering it in a philosophical way. Leaving the opposition between Wittgenstein and Köhler behind, we can ask ourselves this: what more can Wittgenstein’s philosophy tell us about seeing? Or better, where does this philosophically treated concept of visual perception in Wittgenstein’s philosophy come from?

Colours are one of the subjects that occupied Wittgenstein throughout his life and philosophical career. As such, they are directly connected with sight, but for Wittgenstein, colours represent much more. In his hands they become a tool for uncovering all kinds of different philosophical problems that go beyond sight. For example, in Tractatus, colours help him define elementary particles and prepositions (see Wittgenstein 2001, paragraph 6.3751 and beyond), and in Philosophical Investigations coloured squares are used to drive home the point of different contexts of language games (see Wittgenstein 2009, 27e and beyond). However, there exists another text, which focuses almost exclusively on colours and also touches on the problem of seeing — Remarks on Colour.

There are many similarities between Wittgenstein’s oppositions to Köhler and some of the ideas presented in Remarks on Colour, but one of the most important ones is Wittgenstein’s trying to emphasize the importance of a word’s “surroundings.” We have mentioned how one of the reasons he finds behind Köhler’s mistake is that Köhler did not make the right distinction between the use of the word “see” in one context and the other. In Remarks on Colour, Wittgenstein exemplifies the same point in paragraph 57 of the first part: “‘I feel X’ ‘I observe X’  X does not stand for the same concept in the first and the second sentences, even if it may stand for the same verbal expression, e.g. for ‘a pain’” (1950, 9e). This already alludes to one of Wittgenstein’s main points, which is that words have no set or concrete meanings. He emphasizes this by asking the questions about seeing over and over again, like in paragraph 333: “‘And what is ‘seeing’?’ And how should we answer it? By teaching the questioner the use of the word ‘see’?” (Wittgenstein 1950, 61e). If we cannot even decide exactly on what it means to “see,” how can we then use this exact same word as the basis for our arguments like Wittgenstein believes Köhler does? What Wittgenstein is trying to tell us is that things are never easily explained when dealing with language, and we are dealing with language and only with language throughout our lives.

(Salvador Dalí, El Ojo, 1945, source: wikiart)

So, is seeing in connection with colours anything like seeing in connection with aspects? We could say, with Wittgenstein, yes and no. On the one hand, it is completely different, because there is an essential difference between the change in colour in our visual perception and the change in organization. On the other hand, we cannot describe one or the other more precisely than with the word “see.” Or better, we cannot say that one is seeing and the other one is not. When you see a colour change, you do see it, and when you see organization change, you see that too. But you do not see it in the same way. And now, we can ask with Wittgenstein, “But how do you see it?” and “Can the word ‘see’ be explained by comparing these two?” Here, we come back to the problem of differentiating meanings of the same word in different contexts. This is one of Wittgenstein’s key points throughout his late philosophy, i.e. language games, but I also believe this shows us something else: we have to observe every problem from multiple points of view if we ever hope to uncover some truth about it. With what we know of Wittgenstein’s opposition to Köhler, we could say that Köhler’s mistake was he did not look beyond psychology and science when trying to uncover the phenomenon of organizational changes or gestalt switches. However, could we also say that Wittgenstein was too philosophical? That, maybe, not every concept is best explained with philosophical inquiry only?


For Wittgenstein there was much more to the problem of visual objects and visual perception than a simple disagreement with Köhler. What he wanted to achieve and what he showed is that these problems can never be discussed from a singular point of view. What visual objects are and how they are affected by change, for example, which is what this article is concerned with, is examined and explained quite differently when observed from a scientific/psychological point of view, as opposed to a philosophical one.

While both the philosopher and the gestaltist focus on the concept of organization or gestalt in connection with our visual perception, we can, in general,  summarize the main differences between them as follows: Köhler focuses almost exclusively on the differences before and after the organizational change, and this leads him to conclude that organizational changes are, just as changes in colour and shape, a purely visual phenomenon, intertwined with visual objects. Gestaltsare visual and cannot be considered anything but. They are independent of our interpretation. We see one object in one case, and another after the change occurs, but never both at the same time.  Wittgenstein, on the other hand, focuses on the similarities before and after what he calls “aspect switch.” He believes that Köhler is wrong in making an analogy between colour/shape and object/ground. For him, what is essential is that even after the switch “nothing […] has changed” in a sense. He believes that we cannot disregard the fact that there is something different when talking about colour/shape as opposed to object/ground. For Wittgenstein, that “sameness” which persists when organization changes is the key to uncovering a very special phenomenon which he calls “noticing an aspect.”

I believe that the point of showing these two opposing theories was not to decide once and for all who, Wittgenstein or Köhler, was right in this regard, but rather to show that just as with colours, we first need to establish the context and understand the language game we are playing to then properly understand the meaning of anything. Perhaps a better way to put it would be, in Wittgenstein’s terms, “to grasp the words correctly.” There are numerous writings on the topic of science versus philosophy, but Wittgenstein’s opposition to gestalt psychology can teach us that there are no straight paths to truth. Visual perception can be examined from a psychological perspective as well as a philosophical one, and each one offers a unique insight into the problem.


Dinishak, Janette. 2008. “Wittgenstein and Köhler on Seeing and Seeing Aspects: a comparative study.” PhD diss., University of Toronto.

Eilan, Naomi. 2013. “On the Paradox of Gestalt Switches: Wittgenstein’s Response to Kohler.” Journal for the History of Analytical Philosophy 2 (3): 1-21. doi:

Wittgenstein, Ludwig. 1950. Remarks on Colour. Berkeley: University of California Press.

–––. 1980a. Remarks on the Philosophy of Psychology Volume I. Oxford: Basil Blackwell.

–––. 1980b. Remarks on the Philosophy of Psychology Volume II. Oxford: Basil Blackwell.

–––. 2001. Tractatus Logico-Philosophicus. London: Routledge.

–––. 2009. Philosophical Investigations. Chichester: Blackwell Publishing.

*featured image: Salvador Dalí, Los Elefantes, 1948, source: wikiart


1 In her article, Naomi Eilan puts much more emphasis on the problem of ‘internalism’ as implied in the New Object Claim. For a detailed explanation of the claim and its consequences for Wittgenstein, see Eilan 2013, pp. 4 – 5, pp. 10 – 11, and pp. 15 – 16. For why exactly Wittgenstein finds this problematic, see Wittgenstein 2009, pp. 203 – 240.