in Writing

A working definition of comics

A strategic, spatial assembly of separate images

There you go. I’m picking up on my comics studies reading, currently a good way into Nick Sousani’s Unflattening, and as usual when somebody delineates comics I wince just a little and go “Yeah, except…” – so I foolhardily hacked out the short description above. Just a quick note to say that I’m wholly enjoying Sousani’s book, it was just an unfortunate springboard for me to finally write this.

This is not to say that my version above is more “right” than others, but it’s probably more open to different modes of comics making; based on my own experience experimenting with the form (not genre or medium), and may also unveil some new perspectives for comics studies*.

If we dissect the contents and deliberate omissions of my description, we can start with the choice of “assembly” rather than “juxtaposition”, which is an attempt to dodge any notion that the images must be made by the comics maker. Obviously, this is based on my recent collage work methods where I have worked from a base of found materials. Also, there’s enough mixing up between comics making and the act of drawing already, let’s keep that hornet’s nest out of this equation.
I call it a “spatial assembly” to avoid the hoary comparison to film, as well as embracing both sequence and the segmentivity approach to comics analysis (and reading!). That’s why it’s also a “strategic” assembly because that hopefully covers any execution of intent on the maker’s part, whether the message conveyed is linear narrative, poetic or other. I chose “strategic” over “deliberate” to not only emphasise intention but also the considerations involved in the arrangement of panels. Going with “strategy” you may imagine the maker at their worktable in an admiral’s uniform, brooding over whether the next day’s battle should be in the form of a full page spread, or several smaller panels scattered across the page.

The “images” part, then. In When the last story is told, I have a few panels that are just one solid colour, so we’re talking the very basic level of images. You might say “visual information”, or “visual content”, but a) those sound unnecessarily technical for what is meant to be a short, even glib term, and b) once we open for visual information, writing is right outside the door.
There’s the major point of divergence from other definitions of comics: I don’t buy the bit about comics being a hybrid or compound form of text and images. True, most comics integrate the two, but what people forget or overlook is that like music, comics work perfectly well without words. There are surely things you can only tell, not show, or which are less complicated to say than to illustrate, so you write them – or, perhaps, leave them out because you’re making a comic, not a frigging essay. In any case, the assumption that most comics use text, so by default comics must use text, is a logical fallacy. Comics aren’t by definition hybrid, so I’m leaving it out of the definition (without being blind to the obvious).

What I said about putting it short to the point of glibness? At least I may have succeeded with one of those… I hope readers have winced “Yeah, except…” repeatedly during this transmissions. Please post your objections, corrections,  and demands for satisfaction in the comments.

* Or at least new to me, I claim only a limited perspective of that field.


  • Here’s a thought about Roy Lichtenstein – Allan Haverholm
  • Here’s another thought about strategic, spatial arrangement of images – Allan Haverholm
  • Here’s a thought about strategic, spatial arrangement of images – Allan Haverholm

Write a Comment



  1. This very nearly does it for me. Certainly, as someone who makes non-narrative comics it definitely includes my work, which most definitions exclude. My one problem with it is that it’s a little too broad. I think it’s the ‘seperate images’ thing, because that could be applied to advertising space, or a catalogue? To my mind, it could be improved by maybe extending it to: “A strategic, spatial assembly of separate images, to be understood in relation to eachother.” That still feels clumsy, but it does start to exclude stuff which ISN’T comics?

  2. Thanks Gareth, my aim was indeed to allow for non-narrative comics as well as “straight stories” and anything in between.
    I see your point about advertisements and catalogues, but I often use IKEA manuals as an example of “pure communication” comics. How different is a department store catalogue, really? There is, or should be, a great deal of consideration behind the “strategic spatial assembly” of those, directing the reader’s gaze and giving optimal visibility to the most profitable products. And by spotlighting a product next to a similar, less expensive one on the same page, the store communicates different messages to buyers who want to save money and those who would rather pay for (ostensibly) higher quality.
    As I see it, images on a page relate to each other by default, as do images on a spread or in a book.

    The (intentionally pointed) question is rather, “when is a catalogue not a comic?” The message in this case is not creative or instructional, but commercial, and although the strategy is subtle (as opposed to typing GIVE US YOUR MONEY across every page) it is definitely there.
    I’m being the devil’s advocate of course, and I would rather consider commecial catalogues a distant relative to comics than all the same thing.

    That’s another situation where comics get caught between high and low culture, actually, because both our definition attempts could also fit a gallery exhibition, which I don’t suppose we’d object to? And I understood your comment to mean commercial catalogues, but I realise just now that it might also be exhibition catalogues… Sorry if my reply was way off topic!

  3. More to the point of your question, could we amend the wording to address the relationship between images or panels? My immediate thought is that the word “assembly” is too open, and we could be back with “juxtaposition”, implying an interaction between image elements. I’m cautious with phrasing that might suggest relations between adjacent images only, though.

    “Strategic, spatial assembly of separate, but contextually related images”? That’s clunky, though. Perhaps you or other native English speaking readers can offer alternatives?

  4. My definition of comics is a lot simpler – if it has panels, it’s comics. If it doesn’t, it’s part of the broader spectrum of visual narrative. The panel itself is the defining portion of comics – an artificial frame that you can put just about anything into. There’s a lot that can be done by putting panels next to each other – an entire grammar, in fact. But the panel is the key.

  5. Is it simpler to create a whole supercategory of “visual narrative” for everything that doesn’t fit your definition? 🙂

    Less pithily, I think our definitions are more or less the same, but for one thing you need to define “panels” to the reader. Do you mean in the sense of discrete image units or the actual borders? Because I’ll have to disagree that drawing a frame around an image makes it comics. If you draw another one in relation to the first one, voilà! there’s the thing. So I guess I’ve just outed myself as one of those separatists who don’t consider single cartoons comics.

    I see where you come from with your take on it, and it’s a good casual shorthand for identifying comics. However, as I tried to say in the post, I wanted to formulate a definition that

      a) may be of use in making and researching comics, and
      b) by its wording might emphasize non-traditional strategies (there’s that word again) over conventional ones.

    Although you put it in fewer words, I don’t think your version is necessarily simpler – to understand or work from, that is.
    Does that make any sense, or am I misconstruing your comment?

  6. You are not misconstruing my comment, just forcing me to be more precise.

    I mean it in the sense of actual borders. And yes, there should be more than one to create the sequence. Once you have the panels in sequence, it doesn’t matter what’s inside them – photos, collage, text, cartoons, painting, pixels, etc.

    So my basic definition would be: artificial frames arranged to create a sequence, where the order of the panels implies meaning.

    Like I said, there’s an entire grammar associated with what you can do with the panels once you get past that basic definition, which is what I think you’re aiming for. And there’s an awful lot of complexity that can be spun out of that simple definition.

    Having said that, I’m not a single panel purist. I think that a single panel panel cartoon is a comic in the same way that “Stop” is a full sentence in English – the other elements are understood. A previous and/or subsequent panel are implied.

    (I love discussing stuff like this – preferably over an alcoholic beverage, but this website will have to do.)

  7. Ah, you took the bait, R.M. – allow me to reel you in 🙂

    There is a long history of borderless panels in comics, both as a contrasting element in a series of bordered panels and – as in Will Eisner’s later works – a prevalence of panels that fade out around the edges, but still retain their individuality.

    And the sequence thing – there is an increasing need to expand that concept, or even dethrone it from its form defining character, conflated with linear narrative as it is. See my reading notes on Tamryn Bennett’s segmentivity argument:

    There’s a trap in thinking or talking about comics as a language, present here as you talk about the “grammar” of comics. To formulate that grammar you can only consider existing works within the form, and you end up with a system that will hold valid points, but may unintentionally seal off avenues for new and useful manners of expression – will we make better comics knowing there’s a formal equivalent of the accusative? – like McCloud chucking non-narrative work into his “non sequitur” bin category.

    Compare the schools of art that formed elaborate grammars of symbolism which – viewed a few hundred years later, in a different cultural and historical context – requires background knowledge to be decoded. Shorthand as an exo-system growing on top of the original thing.
    One of my agendas here is to allow for new means of expression – precisely bypassing the intricate system of speech and thought balloons, speed lines, onomatopoeia and emanata that aren’t at all intrinsic to comics. If you take them away, the individual work will/should still function on a formal level.

    My point being, and I think you and Neil Cohn both would agree, that we should be able to consider comics as a language on the same terms as music is a language, communicating emotions and states of mind intuitively, while scholars can identify notes, scales, and motifs. Less a form with a correct sentence structure, and more one that offers a wealth of possible elements to be combined at will.

    You know, I really do set out to just answer briefly and to the point…

  8. I was thinking about this more last night and my point is that comics is defined by the panel and not what’s in the panel. The test being that if you remove the content inside the panel, does the panel layout on the page still convey information? IMO, the answer is yes.

    I have to take a step back to make this into a more coherent argument, though. I’ll send you the paper before I get ready to publish because I’m a big proponent of getting feedback in private instead of being embarrassed in public.

  9. There is certainly information in the relative shape, size and placement of empty panels, but at such a minute level that it would be of little value to others than – well, readers of this comment section. To a large degree I see panel dynamics as an enhancement of the image content in relation to the whole (whether it’s narrative, poetic, or neither). But that’s just how I make comics.

    I look forward to hear your thoughts at a later time, and to reading your paper.

  10. On Twitter, L Nichols comments:

    @haverholm I think that’s a reasonable definition. Although the one place I would differ is the word “separate” since it’s a fuzzy area. source

    @haverholm And by that I just mean like, you can have one image that can be subdivided but still be the same image? If that makes sense? source

    So that’s two votes against the “separate” part, although for different reasons. I can’t say I was perfectly happy with it to begin with, and I do appreciate people chipping in with improvements. Would “discrete” be more appropriate, perhaps even in reference to the whole? As in
    a discrete whole consisting of strategically, spatially assembled individual images? That sentence just veers close to supercompressed gobbledygook for me, though.

    As eager as I am to amend the wording or meaning for accuracy, I’m not sure that saying comics are made up of separate(d) image units contradicts or excludes subdivision as Nichols describes it? Quite the opposite, we could argue that splitting a panorama into a three-panel strip is the most straightforward way of making comics (or comicking an image if you’ll excuse my verbing the noun).

    Thanks for the input, L! It’s definitely food for thought.

  11. Also on Twitter, Mark Badger and I had the following conversation, posted here for context. You will find his tweets indented, below:

    @haverholm doesn’t spatial leave out comics on tablet? source

    @badgetoon Like animated/guided view? Yeah, if it’s just a slideshow of single panels, it doesn’t fit my definition. source

    @haverholm even McClouds infinite canvas is just one space, the screen source

    @badgetoon Exactly, so there’s spatiality. I’m not arguing against virtual space, just basing my point on the early motion comics I’ve seen. source
    @badgetoon …which were, essentially, lo-fi animation. Much like the early Marvel cartoons. But I don’t know how they work now. source

    @haverholm doing mine I decided time replaces space on screen source
    @haverholm so animate panels not figures source

    @badgetoon As, in most comics, space replaces time 🙂 No, unless there are >2 panels on-screen simultaneously, I wouldn’t include that. source
    @badgetoon That’s exactly why I emphasised spatiality, because once the relation between images is primarily in the viewer’s memory >> source
    @badgetoon >> the result is too difficult to separate from film. I figured some smartass would point to frames in a movie reel 😉 source
    @badgetoon Spatial arrangement opens for other things to be interpreted as comics, though. Like gallery shows, which I find pretty funny. source
    @badgetoon And you know, this is entirely based on my own work, so mileage may vary 🙂 source

    Note: For an idea of what Mark means by “Doing mine,” see his made-for-screen comics, Carabella on the run, here

  12. Just saw your discussion here, which looks quite fun! I just want to clarify this statement that was made:

    ” I think you and Neil Cohn both would agree, that we should be able to consider comics as a language on the same terms as music is a language…”

    Actually, I would NEVER EVER EVER “consider comics as a language.” That is actually the opposite of my theories. Comics are not a language. Period. At all. Ever.

    Comics USE a visual language of images (often combined with text) the same way that novels or newspapers USE written language. It would be totally weird to say “novels ARE written language” or “newspapers ARE written language.” That’s what it sounds like to me when someone says “comics are visual language.” No: Comics are written in visual languages.

    This separation is what also allows me to say that comics are “defined” by things that are not the traits of the visual language (i.e., the properties of the drawings). Comics are defined by their context in society, the specific visual languages they use, and other socio-cultural factors. This means that visual languages occur outside of comics (like instruction) and that comics don’t have to use visual languages (like abstract comics with no representational drawings). These are separate things that only overlap in the most canonical cases.

  13. Ah, thanks for setting that straight, Neil! Sorry for attributing my less systematic point of view to you, it must have been a spur of the moment lapsus 🙂

    I do see your point that the general public’s perception of comics (“Joe Sixpack says comics are ______”) equals a definition of the form, or at least of the term “comics” to be taken seriously. However, in this case I’m more interested in setting a basis for discussing comics formally than what genre people as such lump comics in with. My own minority definition if you will, although I hope it will be useful to others as well.

    We’ve had the discussion elsewhere about comics defined by socio-cutural factors, and that conversation is part of what set this post going. Maybe I should put it up on the blog as well if all parties agree to it. The two ideas of comics can then be discussed on their individual terms.


  • This Article was mentioned on

  • This Article was mentioned on

  • This Article was mentioned on