McCloud's Understanding Comics

The following are excerpts from a discussion of Understanding Comics, a book by Scott McCloud. The discussion was posted in the 'Creating Comics' board of WWW.COMICON.COM. These are only the posts I wrote myself-- there are dozens of insightful posts by others which I haven't sought permission to reproduce, not having the space for them.

I'm very grateful to Steve Conley and Rick Veitch for providing the forum for this discussion, which I hope to continue in the near future.

--Joe Zabel

From 9/19/98

UNDERSTANDING COMICS (THE INVISIBLE ART) by Scott McCloud is a 216-page essay on the art of comics published in 1993. Universally praised for its penetrating insights into the comic art form, the book is a best-seller and a landmark in comics history. Especially notable is that the essay itself was done in the form of a comic by McCloud, a cartoonist previously best known for his series, ZOT.

At the very end of the book, McCloud says, '...I would especially appreciate a PUBLIC discussion of these issues in comics' trade journals, art magazines, computer nets and any other forum. This book is meant to stimulate debate, not settle it.'

Unfortunately, this public discussion has been negligible, at least as far as I can tell. Although the book has been frequently reviewed, and is often mentioned in passing by numerous comics commentators, I've never seen it being actually discussed or debated.

I guess if I searched long enough on the net I'd find something; I heard a rumor that the Comics Journal was going to publish some kind of discussion. But I think a book of this importance is worthy of a little more attention.

Is the discussion board a good place for this kind of discussion? You tell me. If there are enough people watching this board who are familiar with the book and have an opinion, maybe we can have ourselves a nice little chat.

I guess the best place to start would be, 'What do you think of the book? And has it had a specific, tangible influence on your art, your writing, or how you read comics?'

From 9/22/98

I'd like to thank the other folks who've contributed to this discussion so far. It looks like we've got enough interest to sustain an ongoing dialog. I hope everybody who has an opinion will contribute to this thread and the other UC-related threads to come.

In this current note, I'd like to answer for myself the question I posed in the head note of the first thread.

I was impressed and thrilled by UC, and remain so to this day. As a cartoonist, my own work has benefited from many of the ideas and perspectives discussed in the book.

But at the time UC came out, I had been intensely interested in comics for over 15 years. I'd formed a lot of theories and opinions of my own, some quite at variance with McCloud's theories.

So for me, reading UC was like being a passenger in a car that's going a hundred miles an hour-- crying frantically, 'Where're you going?!? Look out!! Turn left here!! Watch out for that tree!!! AAAAAAGGHH!!'

I'm hoping this discussion will give me an opportunity to express a lot of the doubts and differences of opinion I experience when reading UC. If I tend to be a devil's advocate, I hope it will stimulate the discussion and contribute to a deeper appreciation of UC.

From 9/24/98

In an earlier note, Eric Hess made an interesting comment:

'I think one of the reasons UC hasn't sparked much debate is that... well, gosh... there's not much _to_ debate. Scott's just _right_ all the time! (I bet he bowls a strike every frame, too. what a burden it must be...!)'

His sentiments are echoed by a lot of people I've talked to about UC; for most readers, it's a very convincing book. McCloud is extremely competent at marshaling his evidence and presenting it in a clear, compelling, and charmingly poetic style.

Indeed the wealth of evidence he presents is daunting to anyone who wishes to question his conclusions. To refute each point he makes would require a very lengthy and tedious critique.

I don't want to attempt that here. Instead, I'm going to focus on a few key points. I hope that will loosen people's attitudes enough that they will be more inclined to question UC.

In this first note, I want to talk about what I think is the book's most famous and quoted passage. It consists of the last 2 panels on page 36 and the first panel on page 37:

1st panel: McC as cartoonish character. McC: 'That's why I decided to DRAW myself in such a simple STYLE.'

2nd panel: McC as realistic rendering. McC: 'Would you have LISTENED to me if I looked like THIS?'

3rd panel: McC again as cartoonish character. McC: 'I DOUBT it! You would have been far too aware of the MESSENGER to fully receive the MESSAGE.'

Reviewers and fans of UC have often sited this passage as quasi-scientific proof of the book's thesis (Note: I don't think McCloud himself has ever made any such claim.)

But lets see if it stands up as scientific evidence:

1) In a controlled experiment, subjects would be asked to read passages with the cartoonish McC and the realistic McC. There would be no mixing of the styles.

2) Proportionally, the sample of the realistic representation is too brief to provide reliable evidence, even for a quasi-scientific experiment. The face is shown in only one panel, and the dialog is too brief to be a proper test of how listenable the combination was.

3) The sudden transition from one style to another creates a shock, which distracts the reader from the dialog momentarily.

4) The experiment calls for making a choice between the two styles, assuming that the stylistic choice is not neutral. An objective experiment might demonstrate, in fact, that the stylistic choice IS neutral.

5) Most importantly, McC's statement 'I DOUBT it!' in panel 3 invalidates the sequence as a quasi-scientific experiment. Telegraphing the desired reaction to the subject is completely contrary to the methods of science.

A somewhat better method of testing the thesis here would be to read a treatments of some length done in each of these styles. For example, you could read JAPAN INC. as an example of a verbose comic in a cartoonish style, and V FOR VENDETTA, a wordy comic done in a very realistic style. But even this would be a poor experiment, because you would be affected by your personal taste and the relative skillfulness of the individual works (both have received a great deal of critical praise). Also, it would be difficult to avoid being self-conscious of what you were testing.

A real scientific test would require many subjects, many examples of comic art, objective researchers, and a carefully constructed hypothesis that was testable.

Of course, McCloud never claimed to be conducting a scientific experiment; rather, he's trying to demonstrate his point under less than ideal circumstances. And I must say also that he is extremely fair-- his rendering of the realistic McC is extremely skillful and appealing; small as it is, I think it's the best drawing in the book.

In fact, the very first time I read the passage, I remember thinking to myself that yes, I would much rather have had the entire book be narrated by the realistic McC, instead of the cartoonish one.

For those participating in discussion (and I hope that anyone who wants to will), What evidence in UC do you think is the most convincing in support of McCloud's thesis of Iconic images?

Thanks for listening.

From 10/6/98

I'd like to continue the discussion of proof in relation to UC. In the last thread I tried to debunk the evidence in the most well-known of McCloud's arguments in favor of his theory of iconic images. Now I'd like to take on another famous passage from the book.

On page 34 and 35 of UC is a sequence involves the McC character placing a mask literally over the reader's face; then the reader opens his eyes, and McC coaxes the reader into smiling for him.

It's a superb example of the use of first-person perspective, and succeeds in sensitizing the reader to the presence of their own 'non-visual self-awareness.'

McC proceeds to argue that this self-awareness is conceptualized by the mind in a manner that resembles a cartoonish face:

'Good. Now what CHANGED when you smiled? What did you see? Nothing, right.

'Yet you KNOW you smiled! Not just because you felt your cheeks compress or crinkling around your eyes!

'You KNOW you smiled because you trusted this mask called your face to RESPOND. But the face you see in your MIND is not the same as OTHERS see.

'When two people interact, they usually look directly AT one another, seeing their partner's features in VIVID DETAIL.

'Each one ALSO sustains a constant awareness of his or her OWN face, but THIS mind-picture is not nearly so vivid, just a sketchy arrangement... a sense of shape, a sense of GENERAL PLACEMENT.

'Something as SIMPLE and BASIC... as a CARTOON.'

This passage is self-contradictory; first McC states that the reader has seen nothing, then he asserts that the reader sees his own face in his mind. In the confusion, the distinction between SEEING and FEELING is never examined closely; instead, McCloud goes on to describe the 'mind picture' that he implicitly assumes to exist.

McCloud speaks as if all of what he is saying is established fact. But the scientific basis for his statements is highly questionable; it is all based on INTROSPECTION.

Introspection is a very weak source of evidence; in fact, an entire branch of modern psychology, the behaviorist branch, rejects all introspective evidence on principle.

Of course, in this situation, introspection is the only evidence we've got. But if we view this sequence as a quasi-scientific experiment, the introspective evidence is contaminated (as in pt. 3) because McCloud goes on to tell you what you've experienced.

I'd like to suggest another, more elaborate mind experiment. Perform the following actions (or try to recollect when you last performed them) after clearing your mind of any preconceptions of what the results will be:

1) With your eyes closed, cross the first and 2nd fingers of your right hand.

2) With your eyes closed, smile.

3) While engaging in a conversation with another person, smile.

4) Recollect the face of another person while they were smiling.

Think about it, then read on...

Here are my own observations when performing the mind experiment described above:

1) With your eyes closed, cross the first and 2nd fingers of your right hand.

-- I generally perceive the two fingers, and my right hand in general. My attention is not particularly drawn to my left hand or the rest of my body-- it's almost as if my right hand is the only part of me that exists. I experience very faintly the gestural meaning of crossed fingers.

2) With your eyes closed, smile.

-- I perceive my mouth smiling. I don't particularly see or 'picture in my mind' my face, either in detail or as a cartoon; rather, the sensation is like a Cheshire Cat's smile, floating in space. I experience the gestural meaning of a smile more distinctly than the meaning of the crossed fingers, above; it actually affects my mood in a perceptible way.

3) While engaging in a conversation with another person, smile.

-- When in conversation, I am more aware of the social event itself than I am of my particular role in it. The traffic jam of sensory information forces me to edit severely-- instead of seeing my partner's features in 'vivid detail' as McCloud asserts, I experience flashes of perception intermixed with the bombardment of words, gestures, sounds and smells, and all my internal reactions to the conversation. I experience the smile as a signal communicated to the other person; I am much less aware of my mouth as in #2, much more aware of my partner's reaction to the smile, and of its place in the general flow of conversation.

4) Recollect the face of another person while they were smiling.

This is rather hard to do, because in recalling another person, I experience a montage of many impressions and ideas, one constantly overlaying another. This is the one place where my impressions might roughly correspond to the cartoonish image, but only for someone whom I don't know very well.


This appears to be wandering far afield from the subject of cartooning, and that's just my point-- a person's introspection is far too complex to be summarized as a cartoon image or any other kind of simple analogy.

I perceive my conscious self as shooting the rapids of infinitely variable experience. Maybe a stream-of-consciousness novel can capture the essence of that (I'd love to see somebody try a stream-of-consciousness comic book!) but I don't think a simple cartoon image (or, for that matter, a complex photo-montage), is capable of it.

From 10/12/98

In McCloud's theory of comics, the idea of iconic images is a central premise, a kind of building block on which his other discussions depend.

The basic idea is that, in comparing relatively cartoonish styles of drawing with relatively realistic ones, McCloud finds that the cartoonish styles have inherent advantages.

He believes that cartoons resemble our non-visual self awareness, so we inherently identify with them, whereas we react to a more realistically-drawn character as being apart, other from ourselves.

He says that cartoons are conceptually closer to words than realistic portrayals are, and therefore words and cartoons are closer to a 'unified language'.

He also contends that because cartoons exist in the realm of ideas, the transition from one panel to the next flows 'seamlessly,' but realistic drawings are seen more as a series of 'still pictures.'

He doesn't dismiss realism altogether, suggesting its use in portraying 'the beauty of nature.' And he concedes that it is all a matter of taste. But he clearly thinks it represents good intentions that have 'conspired against comics receiving the unified identity it needs.'

There's a lot here I disagree with. But before going further, I'd like to know: 1) Do you think this is a fair summary of McCloud's views? 2) What is your opinion of McCloud's theory of iconic images?

From 10/16/98

In the head of this thread I attempted to summarize UC's theory of iconic images, and I asked for reactions to it. The following is my own basic reaction to the theory.

I think it was a good idea for McCloud to discuss the issue of style in his book. It's important to raise the public's consciousness, so that they will take the cartoon image more seriously as a mode of expression. Some comics fans and publishers have been unwisely biased against the cartoonish approach, overlooking a body of extremely important work.

I also think a lot of Mccloud's statements about cartoons are quite valid; the human tendency to see ourselves in simplified drawings and to identify with them is indeed remarkable.

But I part company most sharply with McCloud when he attempts to denegrate the more realistic approach to comics.

His theories in this regard are not well-founded. His concept of non-visual self awareness is far-fetched and unconvincing. His ideas about comics as a unified language are self-contradictory.

There are plenty of practical considerations regarding how cartoons work in comics as compared to more realistic drawings; each approach has inherent advantages and disadvantages. But none of these are discussed in UC.

I think the realistic approach represents one half or more of the total potential of the comics medium. For McCloud to dismiss it out of hand is unfortunate.

From 10/20/98

There's an important inconsistency in what McCloud's saying regarding iconic images and his concept of comics as a unified language.

McCloud's idea is that for comics to function as a unified language, the words and the pictures should function in a similar way. On page 49 of Understanding Comics he makes the distinction between 'received' and 'perceived' information:

'Pictures are RECEIVED information. We need no formal education to 'GET THE MESSAGE.' The message is INSTANTANEOUS.

'Writing is PERCEIVED information. It takes time and specialized knowledge to decode the abstract symbols of language.'

McCloud goes on to propose that iconic images are more perceived than realistic images: 'When pictures are more abstracted from 'reality', they require greater levels of PERCEPTION, MORE LIKE WORDS.'

And hence, 'Our need for a unified LANGUAGE sends us towards the center [of the realistic image/written word progression] where words and pictures are like two sides of ONE COIN.'

However, earlier in the same chapter, McCloud has demonstrated that a simplified cartoonish face can be instantly recognized in a wall socket, the top of a grated cheese container, the front end of a car, or randomly drawn loops that have a single eye placed in them. Says McCloud (page 31), ' cannot AVOID seeing a face here. Your mind won't LET you!'

These two views cannot be reconciled. If you 'cannot avoid' seeing a face in a cartoon, then it certainly does not require 'time and specialized knowledge.'

Even if we discount this inconsistency, McCloud's theory of a unified language doesn't make sense. Motion pictures and the theater have no problem with the fact that their characters are both seen and heard speaking. The speech is 'perceived,' yet the audience has no trouble with it. So why should it be a problem with comics?

The human mind is quite capable of shifting from one mode of perception to another; in fact it effortlessly makes vastly greater transition than this-- from the recollected past to the anticipated future, from the mundane to the exotic, from the abstract to the vividly real. And it does so millions of times a day.

From 11/2/98

Earlier in this series we discussed the evidence for what McCloud describes as a person's 'non-visual self-awareness.' McCloud asserts that this awareness is in the form of a cartoon-like self-image; I tried to show that self-awareness is far more complex and dependent on the context, the person's current environment.

In discussing McCloud's theory of iconic images, I want to return to the self-awareness concept.

McCloud says that because our self-awareness is cartoon-like in form, we therefore will identify with a cartoon-like image more readily than a realistic image. Of course if we reject McCloud's idea about what self-awareness is like, there's no reason to think that this automatic identification takes place.

But even if we accept McCloud's idea about the nature of self-awareness, there still isn't good reason to believe in this automatic identification. McCloud is making an analogy between a psychological state and an image on the page. But just because something is analogous, that doesn't mean it is truly alike. Automobiles may 'eat' gasoline and 'breath' air, but they aren't living creatures.

If an analogy is called for, a better analogy would between comics and other visually-based storytelling forms, i.e. movies, motion pictures, and the theater. Of course, the use of a cartoon image is relatively rare in these media, but somehow audiences have no problem identifying with Humphrey Bogart in non-cartoon form. How about movies that are a mixture of cartoons and photography? Do you identify with Bob Hoskins or Roger Rabbit? I think for most people it's a little bit of both, and is based on their performances rather than how they're rendered.

Expanding on his theme, McCloud examines Japanese comics (page 44), discussing their use of the 'objectifying power' of realism. 'While MOST characters were designed simply to assist in READER IDENTIFICATION, OTHER characters were drawn more REALISTICALLY in order to OBJECTIFY them, emphasizing their "OTHERNESS" from the reader.'

But this characteristic is certainly not universal in Japanese comics. For example, in Mai The Psychic Girl, we identify with Mai, who is drawn realistically; she has some schoolgirl friends who are drawn in a more cartoonish style, and they are the ones we identify as being 'other' than Mai.

The cartoonish 'other' is evident in Jaime Hernandez's work as well. Often times the more realistically-drawn Maggie and Hopey are featured in street scenes where background characters are drawn in a cartoonish manner. As McCloud's theory would have it, the reader should lock onto the cartoonish characters and identify with them, instead of identifying with Maggie and Hopey. But it doesn't happen.

From 11/3/98

One more comment about non-visual self-awareness--

I think people have a kind of tangible self-image when they are experiencing extreme emotions, especially anger. For instance when I'm feeling really sore about something, and I'm walking down the street, I have an impression of myself that resembles an exaggerated cartoonish character with eyes glaring, steam coming out of his nostrils, and a big frown on his face.

This might aid in a reader's identification with an angry cartoon character. The expression on the character's face is not what someone would literally see (we tend to cloak our negative emotions) but rather, the exaggerated expression suggests what the character is feeling inside. Indeed many of the most popular cartoon characters are angry ones.

Does this lend support to McCloud's theory of iconic images? I think not, for two reasons:

First, McCloud suggests that the reader try smiling (p.35) in order to become aware of their non-visual self-awareness. But otherwise, he doesn't talk about the phenomenon in terms of feeling one's emotions.

Second, McCloud's description of what an iconic image is does not include the idea of exaggeration. Of course most of us, when we think of a cartoonish image, think first of exaggeration. But on p. 46 McCloud describes the progressive onset of iconic tendencies in terms of: 'Complex --> simple; Realistic --> iconic; objective --> subjective; specific --> universal.' As far as I can tell, the quality of exaggeration doesn't enter into it.

From 11/11/98

In past threads about the theory of iconic images, I've criticized McCloud's theories of non-visual self awareness and unified language. I think he's on firmer ground, though, with his discussion of 'amplification through simplification'.

On page 30, he writes, 'When we abstract an image through cartooning, we're not so much eliminating details as we are focusing on specific details. By stripping down an image to its essential 'meaning', an artist can amplify that meaning in a way that realistic art can't.'

He continues on page 31, 'Simplifying characters and images towards a purpose can be an effective storytelling tool for any medium... The ability of cartoons to focus our attention on an idea is, I think, an important part of their special power both in comics and in drawing generally.'

He adds, 'The more cartoony a face is, for instance, the more people it could be said to describe.'

This line of reasoning breaks down into three distinct concepts:

1) Cartoonish images eliminate extraneous detail. I have no argument with this; it's definitely a distraction for readers to view the same complex realistic facial features again and again. But cartoonish comics contain a lot of extraneous drawing as well; motion pictures are loaded with extraneous images, and prose fiction contains an avalanche of extraneous words and letters. Why have two 't's in the word 'letter?'

2) Cartoonish images exist more in the realm of ideas. I agree with this to the extent that cartoonish images are less concrete than realistic drawings. But I don't think this brings them any closer to the other meanings of the term 'idea'. It doesn't by itself allow them to better express an opinion, state a hypothesis, or lay out a plan. Also, some ideas are not well represented by cartoons. The ideas of poverty or human suffering, or conversely the idea of sexual attractiveness, have much more impact when represented in a realistic drawing style. Art Spiegelman has said that he drew his MAUS Holocaust sufferers as mice because he needed to mask their suffering for it to be endurable.

3) Cartoonish images are more universal. This is no doubt true, but what it contributes to the work as a whole depends on the circumstances. For example, I don't think THE GODFATHER would have been more powerful if it had been about Michael Smith, son of a generic mob boss, instead of being about Michael Corleone, son of the Don of a Sicillian crime syndicate. I think that most of the time, a specific, unique, individual character is the better artistic choice. We identify with such a character because we ourselves are specific and unique.

From 11/17/98

I wanted to expand a little on the discussion above on the idea that cartoonish images exist more in the realm of ideas.

The interesting thing about the human face is that it is a template for communication. You don't have to say anything for people to read your face. The same thing goes for a photograph of a face or a drawing representing a face.

So a representation of the human face exists 'in the realm of ideas' regardless of how it's rendered.

Take another look at UC p.36, where McCloud compares a representation of himself in cartoonish style and realistic style. Which version has more ideas associated with it?

To me, the cartoonish face conveys simple cheerfulness and alertness, nothing more. The realistic face, however, conveys a sense of irony, of casualness. One can read a lot into that face, it suggests the person's age, their approximate socio-economic status. The slight mussiness of the hair suggests that the guy's some kind of rebel.

Of course a more cartoonish version of the same face could convey all these things; but you need a certain level of detail to convey them.

I'd go so far as to say that the LESS iconic the face is, the more of these kinds of ideas it will convey.

From 11/20/98

In chapter 6 of UC, titled 'Show and Tell', McCloud discusses the iconic theory but does not further expand upon it. Rather, he attempts to put it into the context of the history of art.

'Traditional thinking,' writes McCloud on p. 140, 'has long held that truly GREAT works of art and literature are only possible when the two [words and pictures] are kept at arm's length. Words and pictures together are at best considered a diversion for the masses, at worse a product of crass commercialism.'

McCloud sets out to refute said 'traditional thinking.' But he doesn't establish in the first place that such views are widely held. In fact, he doesn't quote anyone making a statement to the effect that words and pictures must remain separate. Since we aren't told who he is referring to, what their views actually are, or why they hold these views, McCloud's refutation seems unnecessary.

McCloud sketches in the history of art and literature, from a common ancestry in primitive picture-making to their current states, where the high arts have turned away from representation. He talks about a 'shared instinct' that 'we had reached the end of a long journey and that it was time at last to head for home.'

But there's an important omission from this account. McCloud makes no mention of the invention of photography. While a relatively small number of artists were 'heading for home', the culture as a whole was beginning the exploration of a vast and unknown territory. And it's worth noting that a photographic image is as far from being 'iconic' as you can get.

Also, although McCloud makes this extensive comparison between graphic art and comics, he does not compare comic art to any of its other related artforms. There is no extensive discussion of the history of prose fiction, the theater, or motion pictures. But to truly understand comics, we need to understand storytelling, and the artforms in which storytelling takes place. In my view, graphic art is really the odd man out, since the only storytelling in that artform is of the indirect, implied kind.

This closes out what I have to say (at least for now) about McCloud's specific theories regarding iconic images. Next, I'd like to initiate a more general discussion comparing cartoonish comics with their more realistic brethren.