Thursday, March 4, 2010

Eye-Mouth; Mouth-Eye

So a while back, as you most likely no longer remember, I said that I had something interesting to say about the face of sadly-deceased* character actor Randall Tex Cobb:

and its relationship with the paintings of Francis Bacon (also sadly deceased):

*Randall Tex Cobb, it turns out, is alive and well. I must have been thinking of some other guy. Bacon remains dead.

Well, I've been meaning to let you know my thinking on the resemblance. It is this: in both Bacon's version of Pope Innocent X, and Cobb, the mouth is much more prominent than the eyes - in fact, the eyes are nearly eliminated. This effect is more striking in Cobb's most famous role, in Raising Arizona:
His eyes, squinting, are nearly hidden in the folds of his face, while his mouth is framed by the beard and moustache and enlarged by the omnipresent cigar.

There is something horrific about both figures - Cobb's apocalyptic bounty hunter, and Bacon's pope. Let me introduce a different thought to try to explain this quality of the horrific.

As you most likely have also forgotten by now, I wrote at one point that I suspected that Margaret Livingstone's research would eventually reveal a prioritization of neural recognition of facial features, with eyes ranking above mouths. Score! She kindly sent me a paper that included this fascinating graph, derived from studies of facial recognition neurons in macaques:

You will notice that eyes get a lot more neural firepower than mouths. My contention, therefore, is this: since when we "see" a face, a large proportion of our processing capability goes into understanding the eyes, and a much smaller proportion into understanding the mouth, there is something horrific about a condition where the eyes are minimized or eliminated. Let's go hunting for examples - it's not hard, once you think about it:

Yes, there's Giger's immortal design for the alien in Alien. Giger, an artist possessed, like Bacon, of a natural sense of the horrific, seems to have single-handedly spawned a tradition of eyeless monsters with his design. This hideous thing turns up again in Lord of the Rings:

Disclaimer: I haven't read the books, so I don't know if the Mouth of Sauron is described that way there too. But now that Peter Jackson has gotten the ball rolling again, it's a short leap to the latest deployment, in Clash of the Titans:

Look, you might say, a face that's all mouth - be it Jaws or the Sarlacc - is always horrific, because it's going to bite you. I'm not arguing that that's not true; I'm arguing that there is a neurological component to the fearfulness of the image. After all, a rabid dog is going to bite you too, but why is this the cover of Cujo?

The biting is scary; the face without eyes is horrific. When we stumble across stimuli that provoke a non-linear response, it is always a good idea to check the machinery receiving the stimuli, which is to say, ourselves. So this is the basis of my contention that there's a processing issue involved in the horrific property of the eyeless face.

And I've thought of a tricky and exciting bit of evidence in favor of this proposition! Consider the reverse instance: a mouthless face. The first instance I became aware of was the cover of an old edition of Harlan Ellison's I Have No Mouth and I Must Scream:

I am not the only one to note the thematic and visual resemblance of this image to a memorable scene in The Matrix:

The only other instance of this image that I can think of in pop culture is the poster for the underrated movie of Silent Hill (a film lousy with eyeless mouth-monsters, by the way):

Now, I would not describe these images as horrific - would you? The term uncanny comes more to mind. Thinking about it this way, I think the topic has something interesting to teach us about the always-elusive distinction between the horrific and the uncanny. Consider this:

When we see the eyeless face, we immediately know something is wrong, because our brains are looking for eyes first, and mouth second.

When we see the mouthless face, there is a tiny, tiny delay before we realize something is wrong, because our brains go through the following steps: 1. Recognize outline of head 2. Search for eyes 3. Search for mouth.

So what this teaches us is that the horrific corresponds with an immediately apparent breaking of the ordinary laws of nature. The uncanny corresponds with a subtle breaking of the same laws. The uncanny inspires terror because we first achieve comfort with the image, and then suddenly we realize the image is, in fact, unnatural. The horrific inspires immediate revulsion, but the uncanny inspires deep fear: fear that either the reality we thought we could trust, or our own rationality that we thought we could trust, are in fact untrustworthy.

Let me give you a really good example of this difference which comes to mind. It has to do with two nightmares a friend of mine once had:

1. She found herself beside a highway where there had been a massive multi-car pileup. Everybody involved in the accident had been mutilated and killed. There were bodies, blood, and gore everywhere.

2. Two weeks later, she dreamt that she came to the same place. But now it smelled terrible, because the bodies hadn't been removed, and they were rotting on the ground.

The first nightmare is horrifying: it is a scene of ghastly violence.

The second nightmare is uncanny: certainly, it is natural for a dead body left in the open to rot. But a tiny fraction of a moment later, one realizes it is profoundly unnatural for time to pass in a continuous dream-world during the period that the dreamer is not dreaming it. Two weeks passed in the dream - but where was this dream during that time? What parallel world did the dream inhabit? This second nightmare, this profoundly unnerving nightmare, breaks the rules of nature as we understand them. Either nature is mad, or we are.

This proposition about the nature of the uncanny also helps to explain why we have so many more uncanny experiences as children than we do as adults. When we are children, we really do encounter many more subtle, delayed-awareness mistakes about the laws of reality - because we're still learning those laws. The sudden revelation that something we thought happened one way, in fact is happening another way, is more frequent in childhood, and it is often accompanied by a twinge of fear: fear of the uncanny.

I'm not sure what, if anything, this little essay has to do with art, but I hope it has at least provided some interesting material to mull over.


  1. First, thanks for the Tex Cobb scare. I had to check online to make sure he was still alive before I noticed your little footnote.

    Second, some random thoughts. Maybe because I'm a long-time fight fan, but what strikes me first about Cobb is not his mouth but his mashed up pug's nose, which took a lot of punishment during his pro career. I think it's that, more than his mouth, that makes his face so striking. The nose is often like a target, and if it is too delicate looking, a person loses any aura of toughness he may have, and instead comes across as urbane or civilized.

    Consider Bob Hope and Woody Allen. Each has a prominent schnoz: Bob's thin and Woody's long. This is perhaps one reason why both visages are anything but horrific. (and in reality, it is interesting to note, both men were fairly skilled amateur boxers in their youths).

    Back to Cobb. His flattened proboscis tells me one thing. You can hit this guy with the kitchen sink and he'll still come at ya. Formidable indeed. That, more than his mouth IMHO, made him so effective in Raising Arizona.

    Also, perhaps another reason why an eyeless face is so horrific is because this is how corpses can appear: eyes sunken or gone altogether, jaw hanging loose. That was my first impression of the Bacon painting: a dead pope.

    Finally, the idea of the mouthless face being uncanny and hence terrifying. Not sure if I agree. Reason is because mouthless faces don't appear in nature whereas ones that are effectively eyeless can. When one sees an eyeless face, a fight or flight reaction can take place. Take any look at an angry gorilla to see what I mean ( But presented with a face without a mouth, an adult of average intelligent will know it's fake or a trick of photography. Uncanny, weird, eerie, maybe. Terfifying, not so much. This is because a face sans mouth requires some suspension of belief. A raging gorilla coming at you at full speed most certainly does not.

  2. REPLY PART 1:

    Chris -

    First of all, I'm sorry I gave you a scare on the Cobb front. From my research on his vitality/mortality status, it looks like he's one of those guys, like Peter Boyle, that for some reason everyone thinks is dead every few years, before it turns out he isn't. Go figure. I was pretty sure he was dead for a long while, and was rather surprised and pleased to find out he isn't. Unfortunately, J.T. Walsh is actually dead, although for consolation, M. Emmett Walsh was alive last time I checked.

    To your main point. I think we are, once again, talking past one another. I am prepared to concede all of your main points, which are very interesting, well thought through, a clearly written. And I don't think they undermine my proposition at all. Why? Because we are treating different parts of the cognitive process.

    You are talking about socialized or acquired cognition. Which is to say, nothing that you describe can be known unless one has lived in the world a little bit and run into things like prize fighters, gorillas, and rotten corpses. In the case of the bared-teeth, there may be an element of instinct involved, but there's no way to understand the significance of, say, the mashed nose, without a good deal of integrated knowledge of things as they are. The case of instinct, and the case of socialized cognition, are way past the stage of cognition that interests me here.

    Consider cognition this way:

    1. Brute data entry
    2. Neurological processing
    3. Instinctual and pre-conscious socialized cognition
    4. Emotional cognition
    5. Rational and conscious cognition

    Step 1 feeds steps 2 and 3.
    Steps 2 and 3 feed steps 4 and 5 in a quasi-unrelated process. After that, steps 4 and 5 interact with one another.

  3. REPLY PART 2:

    Now, step 2 has been altered and conditioned by the same factors that feed into step 3. But as a free-standing cognitive process, they don't directly have anything to do with step 3. Which is to say, there may be reasons, like the angry gorilla, that we process mouths a certain way. But our processing of mouths at the level of 2 does not, on a case-by-case basis, have anything to do with gorilla rage.

    My interest is the mechanics of step 2, the neurological processing. Your analysis treats step 3, the socialized pre-conscious cognition of the face.

    As with all cases of analysis of subjective processes, these analyses depend in part on what science can give us - in this case, Livingstone's neuron studies, and in part on deep introspection of minute responses. The latter case is not provable or even really reducable to analytic components. You just have to train yourself as best you can to catch the preliminary flickers of your own cognitive mechanics, and see what you can learn about it. In defense of this method, I offer many of the pre-modern accomplishments of the philosophy of mind, which depends on people like Descartes and Barkley sitting in dark rooms thinking it over at length and trying to figure out *how* they're thinking what they're thinking.

    So when I say the eyeless face is horrific, what I mean is that I have tried to surprise myself with eyeless faces, and catch the very transitory texture of my initial grasp of them. In this light, for instance, Cobb's imposing nose visually functions as part of his mouth. Why? Its flattened structures reduces contrast on the nose proper, making it effectively invisible in the first instant of cognition. The only visible part of it is the long dark horizontal shadow of its underside. In the first instance of cognition, this is perceived as part of the exaggerated mouth structure that includes the mouth, beard, and moustache - my reference here being "Raising Arizona."

    Similarly, my conclusions on the mouthless face derive from the same method, including also the quality of the fear I remember when I saw that Ellison book cover as a child - it was not the sharp fear of attack or physical threat, but fear of a kind of decay of the dependable, of strangeness walking the world.

    There is no place in this kind of analysis for "monkey might bite me" or "heavyweight champ cannot be stopped." It does not rise to the level of such rationalistic thinking. I am looking for the traces of the very beginning of the process of understanding a face. I have, you might argue, offered hyper-rational descriptions of the implications of the emotions called horror and fear-of-the-uncanny. But the implications are not the emotions; the emotions themselves are indivisible emotional quantities.

    Why do I even care? Because a good artist should be sensitive to those very things - the indivisible emotional quantities - and learn, at least intuitively, to manipulate and control them. One artwork can present rational reasons for fear. A second can present those same reasons, but if it includes the irrational reasons as well, then it really takes flight. Whatever you think of the success of "The Shining," this seems to me to be the reason that Kubrick spent so much time reading Jung before shooting it - he wanted to push your buttons at an elemental level, and primarily with regard to the uncanny, not the horrifying (which is probably why nobody thought it was scary).

    I hope this explanation helps to clear up what seems to me a simple misunderstanding of intent.

  4. I understand that as an artist you must think this way regarding visual stimuli. I'm sure poets turn phrases in their mind looking for that scintilla of unlearned response that makes their sensitive natures react.

    "I should have been a pair of ragged claws
    Scuttling across the floors of silent seas."

    As unforgettable as this image is, it probably wasn't the image it evokes that attracted Eliot to it, but the viceral response it produced as language, as words and sound.

    And I think this viceral response is really what you're getting at.

    Your flowchart really should go like this:
    Step 1 - data entry
    Step 2 - how you react to the data before your frontal lobes get a hold of it.
    Step 3 - how you react to the data after your frontal lobes get a hold of it.

    I agree that my earlier interpretation concerns step 3. I also recognize that your analysis is an attempt at verbalizing step 2, if that can be done at all. My guess is that every artist has something differnt going on at step 2, and it is just as easy to talk about.

    As for my step 2, check this out.

    Horrific indeed.

  5. I'm glad you see what I'm getting at. My playwright friend Mac once wrote a character in a state of panic shouting, "My hands! My hands!" Made no sense at all, but it worked perfectly. Somebody asked him how he came up with that line. He shrugged and said, "Right number of syllables." This seems to go to your point on the textual end of the same phenomenon.

    And yeah, that's a pretty freaking nasty video - thanks for reminding me!

  6. Well, you scared me with Tex Cobb. I scared you with that YouTube video.


  7. Hi Daniel,
    I have stumbled to your blog through You are very smart guy, plus love your art, I have learned a lot from the posts about eye-mouth theory, line I’m self taught (still in process-rather self teaching artist) your blog entries are gold,