Further Reflections on Visual Illustrations of Philosophy: A Response to Freeland, Schellekens, and Strayer

Donate to CA

The free access to this article was made possible by support from readers like you. Please consider donating any amount to help defray the cost of our operation.


Further Reflections on Visual Illustrations of Philosophy: A Response to Freeland, Schellekens, and Strayer

Thomas E. Wartenberg


The incisive commentaries about Thoughtful Images by Cynthia Freeland, Elisabeth Schellekens, and Jeff Strayer provide the occasion for a rethinking of some of the fundamental claims about visual illustrations of philosophy that I made in the book. The essay begins with a discussion of the “logic” of the term ‘illustration’ and then discusses the relationship between illustrations and technology. A variety of different themes are then addressed, such as the possibility of musical illustrations of philosophy; differences in the processing of written texts and visual images; additions to fidelity and felicity as norms applying to successful illustrations of philosophy; and revisions in the theory of illustration put forward in the book.

Key Words
Aristotle; art; Mel Bochner; cognition; illustration; felicity; fidelity; Michel Foucault; Martin Heidegger; Joseph Kosuth; images; Robert Motherwell; music;  Friedrich Nietzsche; Plato; Jean-Jacques Rousseau


1. Thoughtful Images reconsidered

My primary goal in writing Thoughtful Images was to draw attention to a very rich and varied tradition of illustrations of philosophy that had not received adequate attention from either philosophers of art or art historians. To achieve this goal, I developed a theoretical taxonomy of different types of illustration. I included both in the book itself: a typology of forms of illustration, and an historical survey of illustrations of philosophy from the Ancient Greeks to modern graphic novels.

In developing my account of illustrations of philosophy, two themes emerged that I would like to emphasize. The first has to do with the concept of illustration itself. As I will discuss in a moment, there is widespread confusion about the logic of this concept. Many take illustration to be an artform on a par with painting, sculpture, engraving, and so on. There is some justification for doing so. After all, there are journals devoted to illustration and institutions for the education of illustrators. However, a philosophical account of illustration demonstrates that, in its primary philosophical meaning, ‘illustration’ does not refer to a specific artform.

Rather, to call an artwork an illustration is to posit something external to the work that serves as its source. There are many uncontroversial examples of this. One is John Tenniel’s illustrations for Alices Adventures in Wonderland. Choose your favorite of Tenniel’s illustrations and you will see that it is linked to a passage from Lewis Carroll’s book that it renders visually. A visual illustration—the type I am concerned with in the book—is a transposition of a work from one artistic medium, here written language, into another, here a visual image.[1] An alternative way of characterizing an illustration is as a visualization, though that term is also too restrictive.

A number of consequences follow from this general point about the logic or grammar of the concept of illustration. The most basic is that two norms apply to text-based illustrations: fidelity and felicity. Since an illustration is a visualization of something else that is normally not visual—be it a linguistic text, a concept, or a theory—it needs to be an accurate enough transposition of the source for the illustration to be recognizable as its visualization. Hence, the norm of fidelity. But illustrations do not need to slavishly visualize their sources. They can, and often do, alter the source material to create an artwork that manifests its integrity in its own medium. I characterize this norm as that of felicity, a term I borrow from translation theory.

A second theme I’d like to highlight is the relationship between illustrations and technology. What my investigation into the different illustrations of philosophy revealed is that how philosophy is illustrated at a given time is dependent upon the technology available to artists at that time. The complex Broadsides that illustrated Aristotle’s philosophy in the sixteenth century required the existence of print technology. Without that, they could not have been executed in the manner that they were. A similar point applies to contemporary works of what I call “graphic philosophy.” Although comics were traditionally printed using Ben-Day dots to create shading and even colors, recent graphic novels have eschewed them. The images printed in works of graphic literature such as Logicomix use a more complex printing technology that allows more colors to be directly printed on a page without employing Ben Day dots. Although not every work of graphic literature includes color images, the ones that do are only possible with modern printing techniques.

I now turn to the discussion of my book by Cynthia Freeland, Elisabeth Schellekens, and Jeff Strayer. In so doing, I am less interested in defending some of the claims I made in the book than in acknowledging the ways their discussion has helped me see the need to modify or supplement things that I said in the book. I am grateful to them for assisting me in developing my account of illustrations of philosophy.

2. Freeland and the processing of images versus texts

Freeland raises important issues about the account of illustration I put forward in Thoughtful Images. She questions some of the basic claims I made about the nature of images and their relationship the relationship they bear to their source.

Before turning to those issues, I want to clarify a possible misunderstanding evident in her remarks. The term ‘image’ is polysemous. Among other meanings, it encompasses literary images as well as visual images. These two types of images need to be clearly distinguished from one another. When I speak of the text-image dichotomy in the book, it is a shorthand for a contrast between “written linguistic text” and “visual image.” I don’t discuss literary images at all from a theoretical point of view.

This explains why illustrations of Plato’s Cave are text-based, even though that section of the Republic employs an extended literary image. The ubiquity of illustrations of the Cave in introductory philosophy textbooks suggests to me that many readers have difficulty envisioning the set-up Plato describes pace Freeland. That’s also true of the Divided Line section of the Republic. I can testify that, as a student, I had to make my own drawing of the latter image in order to understand the different parts of the line and their relationships. I believe this justifies my contention that, at least in some cases, visual images are easier to grasp than extended textual descriptions.

Freeland asks me to clarify the relationship that holds between texts and images. The central point that needs to be made is that, despite some of the things I say in the book, the relationship between text and images in works of art changes throughout history and between individual works.

To begin to address this issue it is helpful to consider the psychological research on the topic. There are two studies that are relevant. Both restrict their analysis to “simple” images. What this means becomes clear by looking at what the authors say about the visual works they consider. Since he is addressing marketers, Matthew Dunn specifies that he is only concerned with “intentionally-designed visuals that fit on a screen.”[2] The authors of the second study specifically limit themselves to what they call “normal representational pictures” and not all types of visual images. Works of fine art are explicitly excluded from consideration. The authors also limit their consideration to a specific type of written text, as becomes clear in the following passage:

We also limit our concern to written text (not oral prose or other auditorially presented verbal material); to meaningful, connected discourse (not word lists, single sentences, or deliberately ambiguous passages); to experimenter-provided illustrations (not learner-made drawings or mental imagery); to illustrations that can be generally classified as “representational pictures” (not charts, graphs, or diagrams); and to comparisons of learning from illustrated text vs. learning from text.[3]

The first article just cited is focused on the speed at which information is processed. Summarizing the empirical research on that issue, Dunn states that an 8-word sentence takes .8 to 1.6 seconds to be processed while a visual image takes only between 13 and 150 milliseconds. That means an image is processed 8-92 times faster than an 8-word sentence and 92-577 times faster than a 25-word one. The implication is that there is a significant difference in the speed with which images and text are processed by the brain. This difference can be accounted for by the difference between visual images and written text. Language, both written and oral, must be accessed sequentially, with the words read and understood one after the other. Simple visual images are not experienced in this way. The stick figure in figure 1 can be seen to represent a man immediately and doesn’t need to be taken in sequentially by first looking at the head and then at the body and finally at the legs. (Size is also relevant here, but we need not worry about that.) Simple images can be taken in, as I put it in one passage, “all at once.”

Figure 1: Stick figure of a man


Of course, this does not mean that complex works of visual art share this property. If we look at Mel Bochner’s Range (1979), for example, while we can notice the overall pattern of colored V’s and W’s in black and red, the intricate array of numbers is something that we have to inspect carefully and try to interpret. One might even claim the print to be more perplexing than most philosophical texts because it is not at all obvious what it is meant to convey. Even when I claim that the drawing on which the print is based illustrates one of Wittgenstein’s claims in On Certainty, I am not saying that this work of visual art can be apprehended more quickly than Wittgenstein’s text.

The second article I referenced also summarizes empirical psychological research. W. Howard Levie and Richard Lentz claim that the presence of pictures along with written text makes it easier for students to learn and remember the information presented in the text. One study of children’s picture books involves forty-six comparisons between learning from illustrated texts and texts without visual illustrations. In all but one case, the authors report, “the group mean for those reading illustrated texts was superior to that of the group reading text alone.”[4] The authors also report that the children learned an estimated 33% more from illustrated texts than from texts without illustrations. These results were replicated in other studies as well.

Although these results were obtained from studies of elementary school children, we can also see its significance by considering some of the illustrations of philosophy I discuss in Thoughtful Images. The beautiful illustrations of the French translations of Aristotle are examples of artworks that have a clear pedagogical function. Charles V commissioned these works to assist his courtiers in understanding the terminology employed by Aristotle in his ethical and political works. Because French lacked words for such concepts as virtue, the illustrations were a tool used to clarify Aristotle’s terminology for Charles’ advisors who were not fluent in Latin, the language used for previous contemporary translations of his works. Similarly, the Broadsides that Martin Meurisse created, with the help of the philosophy professor Leonard Gaultier, were intended as heuristic aids for Gaultier’s students who had to demonstrate detailed knowledge of Aristotle’s philosophy in their formal examinations. Here, the imagery of a garden was a way of assisting students in memorizing and retaining Aristotle’s ideas.

The works that I have discussed in relation to the psychological research are all text-based. But there are also other types of illustrations of philosophy—concept-, theory-, and quotation-based. These illustrations are not explicitly linked to a specific textual passage, so it’s not obvious how the empirical research on images compared to text applies to them.

What I do claim, however, is that all illustrations “illuminate or shed light on” their sources. Let me give an example of how such illumination can happen. I argue that Bochner’s Range works illustrate Wittgenstein’s claim that mistakes, and hence doubt, only make sense in the context of a rule, a claim that is central to Wittgenstein’s contention in On Certainty that, if one accepts the ideas of the philosophical skeptic, “One gives oneself a false picture of doubt.”[5] By seeing how Bochner creates a numerical system in which a mistake can only be perceived by viewing it as violating one of the rules of the system he used to create the work, I came to have a better understanding of the rationale for Wittgenstein’s claim. In discovering the meaning of Bochner’s work by developing an interpretation of it, I had the experience of coming to perceive an error that violated a rule used in the work. This made me more certain of the validity of the claim than I was from simply reading Wittgenstein’s text. Bochner’s illustration illuminated the philosophical claim for me.

Freeland suggests that it would be more appropriate to consider films as vehicles for illustrating philosophy since they contain multiple images and not just a single one. I limited the scope of my investigation in Thoughtful Images to static works and so ruled out a consideration of film as a vehicle for philosophical ideas or theories. But in Thinking on Screen: Film as Philosophy, I did devote a chapter to films that illustrate philosophy.[6] The chapter had two parts. The first was a theoretical discussion of illustration that addressed themes I addressed in the second chapter of Thoughtful Images in a more adequate manner. But the second part of the chapter presented an interpretation of Charlie Chaplin’s Modern Times according to which the film was an illustration of Karl Marx’s theory of alienation and exploitation. I agree with Freeland that films can illustrate philosophy and maybe even present more fully worked out illustrations of philosophical theories because they contain sequences of images, but that does not mean that static works, such as the ones I consider in this book, are not also significant illustrations of philosophy.

In the book, I point out that art historians, unlike philosophers, have focused on how philosophy has been illustrated in visual works, and even make reference to their analyses in my discussion of both the illustrations of Aristotle’s ethics and Broadsides of his metaphysics. While Svetlana Alper’s study of Vermeer’s The Art of Painting is fascinating, Vermeer’s painting does not qualify as an illustration of philosophy in the sense I discuss, because there is no specific philosophical theory that the painting illustrates. Unlike Foucault in his analysis of Les Meninas, Alpers does not invoke a theory about the relationship between maps and paintings in terms of which it would make sense to see the painting as an illustration.

In this context, it is worth pointing out that paintings and other works of art may very well be a way of doing philosophy. I don’t discuss this point in general in the book, though it is related to the concern upon which the book is focused: visual artworks that illustrate philosophy. In discussing some works that illustrate philosophy, such as Joseph Kosuth’s One and Three Chairs, I also made innovative philosophical claims, but such claims are independent of their illustrating philosophy.

The larger question Freeland raises as to how to justify different interpretations of works of art is, as she points out, a thorny one. She is dubious that artistic intentions can do all the work required; she may well be right about that. I will come back to this issue in my comments on Strayer’s contribution.

Finally, I address a question Freeland poses for me: Do I think music can illustrate philosophy? That’s a very interesting question. In fact, the entire question of what music can illustrate is intriguing. Recently, I attended the premier of a work by the late Lewis Spratlan, entitled August Landscape, that was intended as a musical illustration of a painting having the same title that was painted by Lorna Ritz. Whether he succeeded was something hotly contested in the discussion that followed the performance of the work. (Similar questions can be posed about Mussorgsky’s Pictures at an Exhibition, which was intended to illustrate paintings by Viktor Hartmann.)

Music can certainly provide auditory images of certain phenomena. Vivaldi’s Four Seasons illustrates a season and included, in addition to poems that the work was intended to illustrate, instructions such as “The barking dog” (in the second movement of “Spring”), “Languor caused by the heat” (in the first movement of “Summer”), and “the drunkards have fallen asleep” (in the second movement of “Autumn”).[7] Beethoven’s Sixth Symphony is another attempt to present musical images of natural occurrences. As the entry on the work by the Eastman School of Music states: “As the subtitle ‘Gewitter, Sturm’ suggests, the fourth movement Allegro depicts a storm, eventually giving way to sunshine after rain.”[8] But even though these examples show that music can illustrate some things, it does not answer the question concerning the possibility of musical illustrations of philosophy.

I would like to end by proposing a tentatively affirmative answer to Freeland’s question. Central to Nietzsche’s philosophy is the distinction between the orderly Apollonian worldview and the chaotic Dionysiac one. A piece of baroque music such as Handel’s Water Music can be taken to illustrate the Apollonian, while a late work by John Coltrane such as Ogunde plausibly illustrates the Dionysiac. A more contemporary example is https://soundcloud.com/dionysian-music.

Even more to the point are musical works that are intended as illustrations of philosophy. A prominent example is Leonard Bernstein’s Serenade (after Plato’s Symposium) (1954).[9] The various sections of the musical work correspond to the different speeches from Plato’s dialogue, although some of the sections switch the order of the speeches. All the movements are labeled not only with their tempi but also with the speeches from the Symposium they illustrate. The first movement refers to Phaedrus and Pausanias’s speeches; the second to Aristophanes’s; the third to Eryximachus’s; the fourth to Agathon’s; and the fifth, Socrates’s speech and Alcibiades’s interruption of the evening.

In describing the final movement of the piece, Bernstein says, “The famous interruption by Alcibiades and his band of drunken revelers ushers in the Allegro, which is an extended rondo ranging in spirit from agitation through jig-like dance music to joyful celebration.” His own analyses show how Bernstein intended his composition to reflect the Platonic dialogue.

But the work is not just an illustration of the dialogue but also Bernstein’s individual interpretation of it. As the website of the Houston Symphony puts it:

Bernstein seems to have preferred Agathon’s poetic vision of love to Socrates’ philosophical one, noting that “his seniority adds to the feeling of didactic soberness in an otherwise pleasant and convivial after-dinner discussion.”[10]

In composing a musical illustration of the Symposium, Bernstein did not simply follow the dictates of the norm of fidelity; he also allowed the norm of felicity to guide him.

This points to the fact that visual illustration is only one form taken by that the transposition of a written philosophical work into an artistic medium. I have begun to think about what I call the “re-envisioning” of works between different mediums. I think this could be a fruitful area for future analysis.

3. Schellekens and the theory of illustrations of philosophy

Elisabeth Schellekens’ comments provide me with the opportunity to supplement the theoretical framework I developed for thinking about illustrations generally and illustrations of philosophy more specifically.

Schellekens appears to misunderstand the role I took the fidelity-felicity distinction to play in my account of illustration. This is understandable because she is looking for a fully developed account of illustrations of philosophy, something that the book does not provide. As a result, she treats the distinction between the norms of fidelity and felicity as the basis for a full-blown theory of illustration and notes that it is inadequate. But she also points the way to supplementing my account in a useful manner.

I introduced the distinction between the norms of fidelity and felicity as a way of understanding the nature of text-based illustrations. The use of this dichotomy explained how an illustration could depart from some aspects of a text, despite being faithful to others, and still successfully illustrate that text. A clear example of the role of these two often competing norms can be found in Rousseau’s illustrations for Emile. In these illustrations, while it is apparent what classical myth is being illustrated, there are features of the illustrations that depart from the text being illustrated. For example, the illustration for book I, Thetis, adds the figure of Charon towards the rear of the image, departing from the literary description of the scene.[11] And similar points can be made for others of the Emile illustrations.

When we turn to illustrations of philosophical concepts and theories, the notions of fidelity and felicity do not apply as clearly as they do to an illustration of a section of text. Still, they do have some purchase, particularly in the case of fidelity. Still, by themselves, they do not provide an account of illustration that explains how philosophical concepts and theories can be illustrated.

In discussing Michael Craig-Martin’s An Oak Tree, Schellekens points out that what she takes to be the success of this work as an illustration of the concept of transubstantiation (a concept I would characterize as theological rather than philosophical, but that’s a small point) is not easily explained simply by appealing to the norms of fidelity and felicity. She argues that the work successfully illustrates “transubstantiation” because of the aptness or appropriateness of Craig-Martin’s choice of a shelf and glass of water to illustrate the concept. Had he chosen other objects such as a flowerpot on a ledge, Schellenkens asserts, “the piece might be neither a successful illustration of transubstantiation, nor an illustration of it at all.” The work succeeds because of the aptness of Craig-Martin’s choice of objects to illustrate the concept.

If this is right, then aptness should be added to the concepts an adequate theory of illustration of philosophy needs to include. It reflects the creativity employed by an artist in illustrating a philosophical concept. Illustration of such concepts is not automatic matter, but the result of artistic choice.

It should be noted that this work does not consist simply of the two physical objects just noted. There is a verbal text displayed next to the objects that appears to be an interview with the artist. In it, he explains the significance of the work. The work consists of the physical component along with the text. Even with the text, it is the aptness of the choice of objects to illustrate the concept of transubstantiation that accounts for the work’s success.

I actually wonder whether the work successfully illustrates the concept of transubstantiation or whether it shows the puzzles inherent in it. These problems come to the fore in the interviewer’s questions, for they betray his puzzlement with the work’s title, An Oak Tree. Since what we see when we look at the work appears to be an ordinary glass of water on a shelf, in what sense is it really an oak tree?

Schellekens turns her attention to the question of what allows a visual work to illustrate a philosophical theory. Again, she correctly points out that it must be more than the work’s accord with the norms of fidelity and felicity. Since a philosophical theory is a complex entity having a variety of different but related component, perhaps there needs to a corresponding complexity in an artwork in order for it to successfully illustrate a complex philosophical theory.

I think that this is the case. Consider Joseph Kosuth’s One and Three Chairs, a work I discussed in the book. It consists of three “chairs”: a rather ordinary folding wooden chair, a photograph of that chair displayed on the wall above and to the left of the wooden chair, and, on the right, a dictionary definition of the word “chair.”

The three “chairs” in the piece each correspond to one of the grades of reality in Plato’s metaphysics: the wooden chair corresponds to appearances; the photograph of the chair to the semblances that Plato takes to be the province of art, representations of representations, as it were; and the dictionary definition corresponds to the form of the chair, that aspect of reality that for Plato is fully real. Unlike An Oak Tree, there is no explanatory text, so the viewer has to grasp the analogy with Plato’s metaphysics without any assistance.

The feature of the work that is most interesting from my perspective is that it functions as a critique of Plato’s view of visual art. Although Plato’s view of art is complex, one important aspect is his rejection of art as a way of attaining truth. Unlike philosophy, which has truth as its goal, art fails to reach the truth since it represents only appearances. One and Three Chairs denies the adequacy of Plato’s view of art by presenting Plato’s metaphysics visually. It thus achieves that which is the aim of philosophy, the presentation of truth.

One and Three Chairs is a successful illustration of Plato’s metaphysics. It’s three “objects” faithfully present analogues of the three categories of Plato’s theory. It has a structure that reflects that of Plato’s view. But the work does more than that.

As Schellekens points out, a successful illustration of a philosophical theory should also have a cognitive component, something that supplements the theory or, at least, our understanding of it. The work accomplishes this by challenging Plato’s contention that art can only represent appearances. Kosuth’s work not only satisfies the norms of fidelity and aptness by having objects that can easily be seen to stand for elements in Plato’s metaphysics, it also departs from that theory by having a cognitive component that shows the limitations of Plato’s view. Even a work as visual as Kosuth’s has a cognitive element that Plato failed to recognize as integral to art.

It’s not clear to me whether Schellekens’ suggestion that works of art that illustrate philosophical theories need to have complexity fully accounts for Kosuth’s work. The norm of felicity is helpful to a degree in that it points towards the work’s embodiment of elements that depart from its source. Perhaps the most that can be said is that a work that illustrates a philosophical concept or theory has to have a cognitive component that transcends the source from which it is derived.

4. Strayer and the typology of illustration

Jeffrey Strayer’s comments on Thoughtful Images are studded with fascinating interpretations of the history of Modern Art. For example, his discussion of the import of J.M.W. Turner’s paintings for the history of Modern Art challenges Clement Greenberg’s claim that Eduard Manet is the “father” of Impressionism. Such discussions show Strayer’s intimate acquaintance with the history of art.

The main theoretical innovation in Strayer’s comments is the development of a novel typology of illustration derived from my own discussion of theory-based illustrations. Before I discuss Strayer’s typology, I respond to a number of the more specific criticisms he makes of my book.

Strayer claims that an illustration can illustrate more than one theory and uses one of Bochner’s Range works as an example that both illustrates Wittgenstein’s claims in On Certainty and Popper’s thesis that new properties of mathematical objects can be discovered.[12] The reason I don’t think this example succeeds is that the relationships Strayer claims to be novel and thus illustrate Popper’s view are ones that follow from properties of the numeric sequences and size of the drawing Bochner created. Let’s just consider his first example. Strayer claims that there is a new relationship between the numerals that was not the one Bochner intended. But this “new” relationship is just one derived from the rules used in this block of the print. The basic sequence consists of the numerals 0 to 9 in numeric order, and there are 26 numerals in a single column. The rules for this block specify that, at the end of a column, the sequence will continue in the opposite direction. This means that after 24 more numerals, the basic sequence begins again. As a result, the top two positions in the second column will be occupied by a “4” and a “3”—and similarly for the rest of Strayer’s example. So, what Strayer claims is a novel discovery is simply a feature of the structure of the work that can be accounted for by the rules used to generate the work. For this reason, I don’t think that the work can support Strayer’s interpretation of it as supporting the claim that an illustration can illustrate more than one theory.

Strayer raises the question of whether there are series of works whose seriality is essential to their being illustrations. In so doing, he poses the question of what makes a sequence of works a series. (I think here of Kant’s famous claim that a sequence of perceptions is not the same thing as a perception of a sequence.) Not any sequence of works bearing the same of similar titles constitutes a series. For a sequence of works to constitute a series, viewing the works as a series needs to make an important contribution to their meaning. Monet’s Haystack paintings are one example a series. While one can view a single work in that series, seeing the entire series allows a viewer to understand Monet’s interest in depicting the alterations in the appearances of objects at different times of day because of the differences in how light illuminates them. Being part of a series is essential to understanding these works.

Strayer goes on to deny that Bochner’s Wittgenstein Illustrations constitute a series. But there are many of features of these drawings that are not fully intelligible without understanding and viewing them as part of a series. For example, the fact that there is a matrix that is invariant until the final drawing and thus represents the fixed background for our empirical beliefs, an idea Wittgenstein develops in On Certainty, cannot be recognized simply by looking at one drawing. Similarly, a viewer cannot understand the final “Diamond Branch” without seeing it juxtaposed with the other works in the series, for it illustrates Wittgenstein’s claim that the foundations of our thought can change, one of the ideas developed in On Certainty. The Wittgenstein Illustrations are a series of works whose seriality is essential to their illustrating Wittgenstein’s claims in On Certainty.

Strayer cites Roman Opalka’s 1965/1-∞ as an example of a series. Opalka himself called each canvas a “detail,” suggesting that he viewed the individual canvases as parts of a single work and not as discrete works. If this is correct, then 1965/1-∞ is not an example of a series. Strayer suggests that the relationship of this work to illustration is not straightforward, and I agree with him.

Do Robert Motherwell’s sequence of seven In Platos Cave paintings constitute a series as Strayer claims? I don’t really see how. Strayer mostly unpacks their meaning in a uniform way by saying that they position the viewer in the position of the Cave dwellers. This doesn’t seem right. When viewing one of those works, one can stand close to it in order to see the details of the painting and then move farther back in order to take in the entire work. A viewer is not trapped in a single position as are the inhabitants of Plato’s Cave. What the works do convey is the atmosphere of the Cave, for the fire that projects the images on the Cave’s wall would have generated smoke and, in that enclosed environment, created something like the “scene” depicted in Motherwell’s works. This is an innovative departure from typical illustrations of the Cave that fail to account for the atmosphere a fire would have created. Despite Strayer’s attempt to argue that the seven versions of In Platos Cave constitute a series by virtue of representing Plato’s theory of knowledge as recollection, I don’t see that the meaning of an individual work is enhanced by placing it in the context of the others, despite the illuminating comments Strayer makes about how In Platos Cave illustrates the text of the Allegory of the Cave.

In discussing my account of Abstract Expressionism and Greenbergian Modernism, Strayer raises a number of important questions. As I mention in the book, and he emphasizes in his comments, there are alternative accounts of what the works of painters like Jackson Pollock demonstrate to be the essence of painting. Although Greenberg took them to be demonstrating that flatness is the essence of painting it is also plausible to take them to have been emphasizing the presence of paint, which they place on their canvases in a manner to emphasize its physicality and thickness, as the crucial element in painting.

What exactly constitutes the flatness Greenberg sees as the essence of painting? It’s not just that the support on which paint is applies is a flat surface; it’s also that a painting should not attempt to use the surface to represent a three-dimensional space. Pace Strayer, plays are not flat in this sense since, despite usually taking place on a flat surface, the space in which the action takes place is clearly three-dimensional. And only certain experimental films and photographs—I think here of Stan Brakhage—use the flatness of the picture plane in a manner that eschews the representation of a three-dimensional world.

But what if the Abstract Expressionists were wrong about the essence of painting? Does that vitiate their attempt to do philosophy in paint? I think not. Many theories put forward by philosophers have been revealed to be false without that making their efforts not count as philosophy. The same for these artists. Even if flatness or paint has been shown not to be the essence of painting, they attempted to demonstrate their view of the essence of painting and, as such, did do philosophy in paint.

Strayer also claims that Monet’s late Water Lilies works show that unconscious intentions are a legitimate category of analysis. I don’t see how they show that. Even if Strayer’s analysis of those paintings is correct, why think that Monet unconsciously intended the effects Strayer posits. Wouldn’t it make more sense to say that the works achieved an effect that was not one that Monet intended for them to have but which we can see? I think a lot more would have to be said to justify the assertion that these works demonstrate the existence of unconscious intentions on Monet’s part.

I now turn to Strayer’s typology of technique-based, problem-based, and history-based illustration. With this typology, Strayer extends my notion of theory-based illustration into a new typology of illustration. After reflecting on his efforts, I have decided that my own account of theory-based illustration in Thoughtful Images needs to be amended by distinguishing between a painting being an illustration of philosophy and the use of a painting to illustrate philosophy.[13]

I used the idea of theory-based illustration to explain illustrations of philosophy that are not tied to a specific text or concept but rather to an overall theory. Once more, Bochner’s Wittgenstein Illustrations can help me make the point. These works are theory-based because, although based on On Certainty, they do not illustrate specific passages from that work, despite quotations from that book being placed at the bottom of the drawings. Instead, they provide a visual analogue to the structure of human language and knowledge. This supports Wittgenstein’s claim that doubt only occurs in specific contexts, with the result that it is not possible to raise a generalized doubt in the manner Descartes does to support methodological skepticism. In so doing, they illustrate the fundamental thesis of On Certainty.

However, in chapter 5 of Thoughtful Images I extended the applicability of the idea of theory-based illustrations to three works of art that were not intended by their creators to be illustrations or at least not illustrations of the philosophical theories I took them to illustrate. I did this by examining the uses to which Nietzsche, Heidegger, and Foucault put those works, for they used them to illustrate their respective theories.

I now think I should have distinguished between the claims made by these philosophers. In the case of Foucault, I stated that his interpretation gained plausibility because it makes sense to see Velasquez as addressing the question of representation by means of his painting. In this sense, the validity of Foucault’s interpretation was based upon positing an intention of Velasquez’s part that he may have had. While this won’t justify the claim that Velasquez’s painting illustrates an incoherence at the heart of representation as understood at that time, it will support Foucault’s claim that the painting illustrates the nature of representation.

The case of Heidegger is different. For one thing, Heidegger may have misidentified the shoes as those of a peasant woman. Meyer Shapiro argues that they are Heidegger’s and provides evidence to support his view.[14] Nonetheless, as Freeland also points out, we may learn a great deal about Heidegger’s philosophy by means of his interpretation of the painting as illustrating his views. But that’s not the same thing as saying that the painting is an illustration of his philosophy rather than it can serve as such. The latter does not make a claim about the painting itself being an illustration, but only about its illustrative use.

Nietzsche’s case is similar. I imagine he was struck by how Raphael’s painting can serve as a visual illustration of his distinction between the Apollonian and the Dionysiac. But Nietzsche never mentions the fact that the painting clearly illustrates two Biblical texts. It seems obvious that Raphael intended to illustrate these passages and to make a point via their juxtaposition. But, once again, the painting has an illustrative use, that is, is a useful way of illuminating Nietzsche’s theoretical distinction.

Part of the problem here is that we need to pay attention to one of the basic assumptions I make about an illustration, namely that the work’s creator intends it to illustrate the text, concept, or theory in question. As I say in the book, “there standardly is an intention on the part of the creator of an illustration to elucidate its source is some way or other,” though in a footnote I refer to my chapter 5 discussion of works not intended by their creators to be illustrations.[15]

I would now say that both Heidegger and Nietzsche use an artwork to illustrate their respective theories without categorizing those works as being illustrations of those theories. Although this may seem a trivial semantic distinction, it is not. It leaves the status of works inviolate when they are used illustratively. I don’t believe, as Strayer does, that a later theoretical development can change the status of a work, making it an illustration of a theory that the creator of the work could not have known of.

Analogously, Strayer’s innovative interpretations of works of art do not make those works illustrations of, for example, concepts in the history of art, even though he can use those works to illustrate concepts like fragmentation. Perhaps, he was misled by my discussion of theory-based illustration into making a more universal claim about illustration than I believe to be warranted. I did intend to extend the scope of works considered illustrations by characterizing as illustrations works like The Rape of Europa that are not traditionally thought of as illustrations by art historians. However, I think the concept of illustration borders on incoherent if all works in the history of art are treated as illustrations.

It is important here to bear in mind a point I mentioned earlier in my remarks: that the term ‘illustration’ functions at a different level than those of specific artistic genres like painting, etching, drawing, and so on. Something is an ‘illustration’ if and only if it is the result of an artist’s intentional re-envisioning of a source.[16] Thus, The Rape of Europa is an illustration because Titian re-envisioned a passage from Ovid in the painting, a claim I justify in chapter 2 of the book. The same work of art can be a painting—or an etching, a drawing, and so on—and also an illustration since the logic of the two concepts are different.

I am thankful to Strayer for giving me the opportunity to rethink one of the claims I made in Thoughtful Images. His erudite commentary on my book has many insightful things to say, despite the disagreements I have just registered.

5. Conclusion

Thoughtful Images was an attempt to develop a theory and history of illustrations of philosophy. As a first attempt at such a project, I knew that it would have a variety of shortcomings. Although my discussants have praised the book, they have also pointed out ways in which it falls short of an adequate account of the matters it discusses. I have tried to indicate the ways in which their comments point towards developing a less flawed account of the theory and history of illustrations of philosophy. I am grateful for their probing questions and suggestions that have helped me begin to develop that account.


Thomas E. Wartenberg

Thomas E. Wartenberg is Professor Emeritus of Philosophy at Mount Holyoke College. His main areas of focus are aesthetics, the philosophy of film, and philosophy for children. In addition to his latest book Thoughtful Images: Illustrating Philosophy Through Art, his publications include Unlikely Couples: Movie Romance as Social Criticism, Thinking on Screen: Film as Philosophy, Existentialism: A Beginner’s Guide, Big Ideas for Little Kids: Teaching Philosophy Through Children’s Literature, A Sneetch Is a Sneetch and Other Philosophical Discoveries, and Thinking Through Stories: Children, Philosophy, and Picture Books. His philosophy for children website, teachingchildrenphilosophy.org, was awarded the 2011 APA/PDC Prize for Excellence and Innovations in Philosophy Programs. He received the 2013 Merritt Prize for his contributions to the philosophy of education and has held Fulbright and National Endowment for the Humanities Fellowships. He is Film Editor for Philosophy Now.

Published on November 25, 2023.

Cite this article: Thomas E, Wartenberg, “Further Reflections on Visual Illustrations of Philosophy: A Response to Freeland, Schellekens, and Strayer,” Contemporary Aesthetics, Volume 21 (2023), accessed date.


End Notes

[1] There is no term that adequately characterizes the relationship between an illustration and its source. Here, I use ‘transposition’ and ‘visualization.’ ‘Translation’ is another option. But each of these has associations that are not general enough to characterize the illustration-source relationship.

[2] Matthew Dunn, “Research: Is a Picture Worth 1,000 Words or 60,000 Words in Marketing?” https://www.emailaudience.com/research-picture-worth-1000-words-marketing, 2023, accessed on March 15, 2023.

[3] W. Howard Levie and Richard Lentz, “Effects of Text Illustrations: A Review of Research,” Educational Communication and Technology, 30, No. 4 (1982), 195. Levie and Lentz’s summary of many empirical studies shows that “with a few exceptions, the results showed that illustrations had a significant positive effect on learning illustrated text information and no effect on learning non-illustrated text information” (198). They state that “subjects reading illustrated text learn an average of one-third more” (203).

[4] Levie and Lentz, 198.

[5] Ludwig Wittgenstein, Ludwig, On Certainty, G. E. M. Anscombe, G. H. von Wright, and Denis Paul, eds. and trans. (New York: Harper and Row, 1972), ¶249.

[6] Thomas E. Wartenberg, Thinking on Screen: Film as Philosophy. (Abingdon and New York: Routledge, 2007).

[7] https://en.wikipedia.org/wiki/The_Four_Seasons_(Vivaldi).

[8] https://www.esm.rochester.edu/beethoven/symphony-no-6/. The same article states, “there are obvious mimetic moments in the music such as the murmuring of the brook, calls of the nightingale and cuckoo.”

[9] This work was brought to my attention by one of the attendees at session of the 2023 meetings of the American Society for Aesthetics where these papers were first presented. Unfortunately, I do not recall who it was.

[10] The Houston Symphony, “A Greek Riddle: Bernstein’s Serenade (after Plato’s Symposium),” 2018, https://houstonsymphony.org/bernstein-serenade/, accessed on April 28, 2023.

[11] Cf. Thomas E. Wartenberg, Thoughtful Images: Illustrating Philosophy Through Art (Oxford and New York: Oxford University Press, 2023), 87-88

[12] The work Strayer discusses is one that was published as the endpaper of Ludwig Wittgenstein, On Certainty, G.E.M. Anscombe and G.H. Wright ed., Denis Paul and Anscombe, trans., Mel Bochner, illus. (San Francisco, Arion Press, 1991). The Range work at issue is the endpaper for the cover of the book. It is less complex than the original drawings I discuss in the book. These do not have the features Strayer mentions. I discuss this work in “Illustrating philosophy:  Mel Bochner’s Wittgenstein drawings,” Word and Image, 31, No. 3 (2015), 233-248.

[13] In his review of Thoughtful Images, “Philosophy’s Illustrious Companions” (Art Journal, forthcoming), Michael Corris criticizes my extension of the term ‘illustration’ to cover works that were not so intended.

[14] Meyer Schapiro, “The Still Life as a Personal Object—A Note on Heidegger and van Gogh” in Simmel, M.L. (ed.) The Reach of Mind: Essays in Memory of Kurt Goldstein (Berlin and Heidelberg: Springer, 1968) 203-09.

[15] Wartenberg, 21.

[16] I discuss the notion of re-envisioning in an unpublished paper presented at the 2023 Meetings of the American Society for Aesthetics. The source of an illustration can be a text, concept, or theory, as I argue in the book.