Opposites Attract: Face to Face with Chris Landreth’s Animation

The screen at the front of the auditorium shows two versions of the same computer-generated face. One has skin and the other does not, its muscles revealed. She has been created in the facial animation software produced by the JALI Research company, of which Chris Landreth is Co-founder[1] and Creative Officer. As Landreth manipulates sliders in the software, he explains that we use the orbicularis oculi muscle to narrow our eyes: the woman onscreen and her flayed doppelganger raise their lower eyelids. The expression is ambiguous, but a sense of joy arises with another bit of muscular contraction, this time from the zygomatic major, the puller-up of the lip corners. Landreth teaches us, the audience of his “Making Faces” masterclass at the 2022 Animation Festival of Halifax,[2] that human faces are like alphabets: with 14 primary muscles on each side of the face, or 28 in all, each of these muscular “letters” performs actions in combination with others to spell out the full range of human emotion.[3] The woman onscreen, whom Landreth calls the “Uncanny ValleyGirl”[4] with characteristic paronomastic wit, smiles benevolently at the metaphor.

The muscles that mirror each other on either side of the human face could be a symbol of Landreth’s own work and career, a saga of oppositional pairs: art and science, physiology and psychology, realism and metaphor. A case in point is his current return to software development after many years focussing on writing, directing, visual design, and animation. And, like twinned facial muscles working in tandem, the parts of his career demonstrate the falsity of opposition between oppositional pairs. Software coding is generative and creative; the art of animation is highly technical; psychology is embodied; the body is transformed by thoughts and emotions. The relation of realism and metaphor is perhaps the most difficult pair to parse, but the double helices of art/science and physiology/psychology guide us in this as well.

There are many landmarks in Landreth’s career cited widely and frequently, and with excellent reason. He has played a major role in the evolution of computer-generated animation – and the digital imagination – worldwide. His early films, created while working at Alias Research as an expert tester of the first version of the now-ubiquitous Maya graphics program, did more than contribute substantially to the development of this software: they also showed the artistry and inventiveness that the budding form of digital animation could accomplish. He earned an Oscar nomination for the first of these, the end (1995), and won a Genie for Bingo (1998), the second. The significance of his next film, Ryan (2004), cannot be overstated; Judith Kriger writes that it “very quickly became one of the most celebrated animated short films of all time.”[5] An animated documentary made at a time when the genre was hardly acknowledged to exist,[6] this complex, empathetic sketch of animator and addict Ryan Larkin won an Oscar, a Genie, festival awards at Annecy and Cannes, and countless other accolades. In contrast to the mainstream push towards mere verisimilitude, it also developed a visual language that Landreth calls “psychorealism,” in which metaphors for psychic realities are inscribed on the bodies of otherwise realist characters. The Spine (2009) further explores psychorealism, this time in a piercing drama about a toxic relationship, and Subconscious Password (2013) takes a tour of a psyche with caustic humour. These are the most celebrated moments of his career, and Ryan in particular has been the subject of a wealth of analysis.[7]

Rather than retread this ground, we will look here at some of the less-visited corners of his career – the less-lauded films, the academic publications, the software development – as a supplement to the better-known narratives and as a tribute to his own maverick path.

A Matter of Motion: Scientists Animate

This story from the margins begins with the pressing need, in the formative years of computer-generated animation, for artist-scientist-researchers: today, most digital tools are mature enough that an artist need only be a sophisticated software user, but in the 1980’s, much animated content was created by computer scientists. Although Landreth always created and even exhibited his artwork,[8] his university education began in sciences. After completing a Master of Science in Theoretical and Applied Mechanics in 1986, he continued to work in the field of fluid mechanics with his advisor, Ronald J. Adrian,[9] while also studying computer animation with Donna Cox at the University of Illinois’s National Centre for Supercomputing Applications (NCSA). His work at this time reflects his multifaceted abilities and interests. For example, in the same year that his first film, The Listener (1991), was shown at SIGGRAPH, he was a co-author on two scientific papers on Particle Image Velocimetry,[10] which has since become a key way of measuring the flow of fluids.[11] A common denominator of these two branches of his work at that time is the precise observation of movement, and his animation has certainly proven him to be a master of this. Decades later, while working on Subconscious Password, he described the process of animating Cthulhu’s tentacles in terms of fluid mechanics – airflow and turbulence, mass and air drag – and stated that he used the science all the time despite having moved away from it professionally.[12] Back in the 1990’s, he continued to bridge art and science in his next job visualizing and animating scientific data at the North Carolina Supercomputer Center, where he completed two more animated films. Caustic Sky: A Portrait of Regional Acid Deposition (1992), and Data Driven: The Story of Franz K. (1993) in order to test the software;[13] the latter film includes a character with a full face, hands, and synchronized speech. Both of these films were shown at SIGGRAPH,[14] that nerve centre of artistic-scientific research-creation.

Landreth emphasizes Donna Cox’s importance, and her theories and innovations elucidate core aspects of his own creativity. Cox’s story is the mirror image of Landreth’s: while pursuing her Master of Fine Arts, she was already satisfying her love of science and math by writing digital colour mapping tools in C++.[15] Then, as a new professor of photography at the University of Illinois, she convinced colleagues in sciences “that they needed visually literate people to work with them on teams to visualize their data.”[16] She coined the term and realized “Renaissance Teams”[17] of collaborators from different fields of expertise, an approach that she connects both to women’s socialized strengths in collaboration and to the overt political aims of the Women’s Movement, which influenced her deeply.[18] Landreth flourished and contributed scientifically, artistically, and ethically to this mosaic[19] of science and art. In a 2018 retrospective of women in digital arts, Cox thanks him personally as one of the “exceptionally talented and pioneering Renaissance men” whose “support, enthusiasm, and sharing [of] mutual values [made] the world a better place with collaboration as the foundation of artistic and scientific exploration.”[20]

Landreth’s immersion in this environment illuminates an important thread in his film work that is seldom remarked: the naturalization of and connection with a diverse range of racial, gender, and sexual identities and experiences.[21] As Sobaz Benjamin states, Landreth’s work makes it possible to express stories and emotions “that are incredibly difficult to express for people who have been marginalized by various systems.”[22] Diversity is evident in his films. In the end, for example, the animator figure morphs from a model based on a scan of Landreth’s own head[23] into a Black woman and then a kaleidoscope of faces, bodies, and voices, which visibly disengages the power and privilege of artistic creation from the white male and distributes it widely. The white male never returns as the “real” identity: the last speaker, whose words actively and self-referentially draw the film to a close, is a young Woman of Colour. In The Spine, the audience shares the point of view of a queer white woman whose sexuality is never seen as exceptional. This a far cry from the distinctly white, hetero- and gender-normative universes that many male-identified artist-scientist-researchers created in early computer animation. Landreth’s worlds are often – and justifiably – seen to be perverse and disturbing, but they are also deeply compassionate. For example, when he found that the first draft of The Spine had a very “snide tone,” Landreth was “horrified” and scrapped it in favour of a script that sought the beauty and nobility of a marriage that appeared, from the outside, to be entirely toxic.[24] His most recent film, Be Cool (2017), captures this ethos pithily if not rigorously. A 43-second public service announcement created through the National Film Board of Canada, it shows a computer-generated white man, who looks a great deal like the scan of Landreth in the end, taking selfies in front of difficult sites such as Auschwitz and slums. The winged, floating head of a woman of ambiguous racial identity[25] rises like the sun from the horizon to exhort him pleasantly in song: “Be cool; don’t be an asshole.” The final intertitle tells us that “2/3 of us don’t know we’re assholes.” Landreth can pose as one precisely because he is not one.

“We’re All Twitching Meat”: Physiology and Psychology

One absolutely crucial element that Landreth brought to the table of the NCSA and early computer graphics more broadly is his long-time expertise in faces and their movements. While Cox animated data such as particle flow in a tornado, and the early films of the more commercially oriented John Lasseter anthropomorphize inanimate objects with ever-increasing photorealism, Landreth has always been “obsessed with faces” and portraiture.[26] Self-taught from books of anatomy,[27] Landreth has from the beginning created faces that are expressive by virtue of their realism design and movement but that do not deny their digital artificiality. The eponymous character of The Listener is a case in point. It is nothing more than a slice of a face from nose to beetling brow resting on three piston-like legs and responding to an unnerving, fragmented soundscape that includes a laugh track and a disembodied voice’s repetitive, demanding “Are you comfortable?” Speechless, it reacts with surprise, suspicion, irritation, distress, and fury that are all rendered highly comprehensible by precise, naturalistic physical movements. The proof of this is in audiences’ reactions. The film was shown on MTV[28] and is now available on YouTube (although untitled and uncredited[29]), where user comments attest to powerful reactions of delight, bafflement, fear, and horror. One user states that the film is “distressing, and it’s captivating,” and another says, “I don’t know what the hell, but it’s impressive.” The most verbose captures the emotional and existential disturbance well:

Significantly, none of the users comment that the graphics are dated or the film is emotionally anodyne; the highly expressive face is still works.

This is made possible by a deep knowledge of physiology, which supports psychological realism[30] and emotional legibility. Landreth opens his 1996 article entitled “Faces with Personality” by grounding psychology in physiology: “To create a realistic looking, 3D computer-generated human face that projects a believable and engaging personality, you must understand the anatomy of the face. Humans speak and express emotions using their facial muscles. Knowing which facial muscles are responsible for creating which expressions is an essential key to modeling those expressions in CGI.”[31] Although he does not use the metaphor of the alphabet that he later uses in “Making Faces,” this article includes a list of facial muscles and their actions and a brief discussion of the combinations of muscles to produce both emotional expression and phonemes.

The latter is precisely what his work at JALI Research refines. In the 1996 article, Landreth explains how he animated the faces in the end: he used Maya’s predecessor, PowerAnimator, to create a neutral wireframe face from a scan of his own head and then used control vertexes (CVs) to create a separate version of the face with each individual muscle engaged. A morphing utility allowed him to combine images of muscle engagement in order to accomplish both emotional expression and speech animation. Twenty years later, JALI (Jaw And Lip Integration) is procedural animation software that translates an audio track and speech transcript – in any language[32] – into synchronized lip, tongue, and jaw movements. The detail with which JALI accounts for speech phenomena such as over-enunciation (which uses facial muscles with little jaw movement) and droning vocal tones (which use little lip movement for the same phonemes)[33] is a far cry from Landreth’s problem-solving in PowerAnimator. JALI Research has also allowed Landreth to return to his early work in computer sciences, although his role has evolved yet again to meet the needs of the situation. As an expert tester for Maya at Alias Research during the creation of the end, Landreth refused to write code in order to ensure that the software would work for someone without programming skills. Instead, he wrote a lot of bug reports. at times enlivening this frustrating process with literary experimentation: one notable bug report was written as a pastiche of Tolkien.[34] Now, at JALI Research, he is engaged in the creative process of scripting MEL (Maya Embedded Language) and Python[35] and expanding the software’s ability to realize emotion even beyond jaw and lip synchronization. In their article on the application of JALI in the AAA video game, Cyberpunk 2077 (CD Projekt, 2020), Landreth and his fellow authors discuss the integration of procedural animation of jaw and lip with “paralinguals” such as blinks and brow movements to further refine the representation of emotion and internal thought.[36]

Realist timing of character movements in relation to speech has shaped Landreth’s style from the beginning. Apropos of Ryan, he discusses conventional, tidy correspondence of sentences and keyframe poses: “‘This is something you see a lot in studio animation,’ he says. ‘A character says something and does a gesture, says something else and does another gesture.’” However, Ryan’s hand gestures were often logically and temporally unconnected with what he was saying, so the film was made using straight-ahead animation instead of keyframe animation.[37] But this naturalistic approach, which Landreth sometimes calls “Method Animation” in a reference to Lee Strasberg’s Method Acting,[38] did not start and end with Ryan. In Data Driven: The Story of Franz K., the picture editing is often timed to breaks between phrases or clauses of the narrator’s speech, but many aspects of the design and animation were driven by scientific data rather than verbal phrases, and the movement of the figure onscreen does not punctuate the synchronized speech in a simple way.[39] In the end, the Animator character punctuates speech with a gesture only comic effect, when he makes a quick rictus of a grin after spouting a deliberately pretentious phrase; otherwise, he fidgets naturally. Both in relation to speech and beyond, Landreth emphasizes the twitchiness of human movement: “we’re all twitching meat […] The meat part is that we have mass, momentum, and weight; the twitching part is universal in the way people move. People don’t move in curves. They twitch into motion impulsively, generally with a fast ease in and a slow ease out. It looks like a sawtooth pattern, not a sine wave.”[40] This vision of movement and its psychological motors creates an onscreen image of humanity that is fundamentally off kilter and yet deeply familiar.

Breaking Down Humanity: Realism and Metaphor

The convergence of the off kilter and the familiar is encapsulated in the ethics as well as the aesthetics of psychorealism, which is the clearest method by which Landreth combines realism and metaphor. He explains the relationship in this way:

Landreth uses the “symbolic fantasy stuff” to deliver details about the characters’ realities with visual and narrative compactness. For example, in place of a long, expository backstory, Ryan’s mental illness and addiction has left him with a sliver of a face on little more than an apple core of a head; the protagonist in The Spine literally grows a spine of crystalline beauty over the course of the film before losing it again when a destructive relationship returns. In Wired magazine, Scott Thill writes that this technique is “a far cry from most CGI, where humans look, well, human. Instead, The Spine breaks down its characters’ humanity, displaying their interior dramas with nightmarish skill.”[42] But the physical distortions in all of Landreth’s films only “break down” humanity in the sense of analyzing it, not of annihilating it.

Landreth states that he is attracted to this kind of imagery “not to alienate these characters from us as freaks, but hopefully to bring them closer, to give a means for us to identify with them.”[43] The unspoken assumption here is that “we” – members of the audience – always already carry wounds that enable us to recognize them in others; this is narrativized in Ryan, which introduces us to a mutilated “Chris” before Ryan’s damage becomes apparent. In other words, interpreting these metaphors engages an understanding of our own brokenness. Also apropos of Ryan, Landreth states in conversation[44] that we are all broken in our separate ways but, as in the troubled relationship in The Spine, “it’s the brokenness that makes us beautiful.” This broader ethos throughout Landreth’s work is encapsulated in but not confined to the metaphors of bodily distortion. Both metaphors and other narrative and aesthetic choices are premised on psychological realities within members of the audience as well as the psychological realities of the characters conveyed through physiologically realist – if selective – design and movement.

Mirror images permeate Landreth’s work, from JALI’s Uncanny ValleyGirl and her flayed doppelganger right back to the last shot of The Listener, in which the persecuted half-face looks into the “camera” and its own reflection. The mirror image is perhaps the metaphor par excellence of the tensions within oppositional pairs of art and science, physiology and psychology, realism and metaphor. These tensions are both deceptive and true. Landreth’s body of work is proof that art and science are not fundamentally opposed; however, they are also not the same thing. As Landreth says, his work coding at JALI Research has eclipsed his artistic production, for now. “It’s kind of like […] you go to a gym on Thursdays and you concentrate on your back muscles and you don’t concentrate on your arm muscles, and then next week I’ll do the arm muscles.”[45] All of this work develops muscles; the results are all major contributions to digital animation, whether they be tools or films. Either way, digital animation wins.

Bibliography

Adrian RJ, Offutt PW, Liu ZC, Hanratty TJ, and Landreth CC, “Studies of Liquid Turbulence Using Double-Pulsed Particle Correlation,” in Applications of Laser Techniques to Fluid Mechanics, edited by Adrian RJ et al., 435-50. Heidelberg: Springer Berlin, 1991.

Amsden, Cynthia. “Chris Landreth.” Take 1 8, no. 24 (Summer 1999): 13-15.

Blair, Jennifer. “Animation to Spare in Chris Landreth’s and Ryan Larkin’s Short Films.” Screen 56, no. 1 (2015): 46-63.

Brown, Hannah. “Faces Come to Life in Sderot with Animation Master.” Jerusalem Post October 21, 2021. https://www.jpost.com/israel-news/culture/faces-come-to-life-in-sderot-with-animation-master-682791.

Cox, Donna “Metaphoric Mappings: The Art of Visualization,” in Aesthetic Computing, edited by P Fishwick, 89-114. Cambridge, Mass: MIT Press, 2006.

Cox, Donna. “Using the Supercomputer to Visualize Higher Dimensions: An artist’s Contribution to Scientific Visualization.” Leonardo 21, no.3 (1988): 233-242.

Cox, Donna, Ellen Sandor, and Janine Fron, eds. New Media Futures: The Rise of Women in the Digital Arts. Urbana, Chicago, and Springfield: University of Illinois Press, 2018.

Edwards P, Landreth C, Fiume E, and Singh K. “JALI: An Animator-Centric Viseme Model for Expressive Lip Synchronization.” ACM Transactions on Graphics 35, no. 4 (July 2016): article 127, 11 pages.

Edwards, Pif, et al. “JALI-Driven Expressive Facial Animation and Multilingual Speech in Cyberpunk 2077” SIGGRAPH ’20 Talks, August 17, 2020.

Faigin, Gary. The Artist’s Complete Guide to Facial Expression. New York: Watson-Guptill, 1990.

Fore, Steve. “Reenacting Ryan: The Fantasmatic and the Animated Documentary.” Animation 6, no. 3 (2011): 277-92.

Hannibal, Carmen. “Subjective Perspective as Creative Metaphor in the Animated Film.” Mediaesthetics 2 (2017). https://www.mediaesthetics.org/index.php/mae/article/view/63/138

“In Conversation with Chris Landreth.” Interview with Sobaz Benjamin. Animation Festival of Halifax, May 6, 2022.

Kriger, Judith. Animated Realism: A Behind the Scenes Look at the Animated Documentary Genre. Waltham, MA: Focal Press, 2012.

Landreth, Chris. “Faces with Personality: Modeling Faces That Exude Personality When Animated.” Computer Graphics World 19, no. 2 (February 1996): 58-60.

Liu ZC, Landreth CC, Adrian RJ, Hanratty TJ. “High Resolution Measurement of Turbulent Structure in a Channel with Particle Image Velocimetry.” Experiments in Fluids 10, no.6 (1991): 301-312.

Lorenzo Hernández, María. “Through the Looking Glass.” Animation Studies 5 (2011). https://journal.animationstudies.org/maria-lorenzo-hernandez-through-the-looking-glass/

Moore, Samantha. “Animating Unique Brain States.” Animation Studies 6 (2015). https://journal.animationstudies.org/samantha-moore-animating-unique-brain-states/

Robinson, Chris. Animators Unearthed: A Guide to the Best of Contemporary Animation. New York and London: Continuum, 2010.

Robertson, Barbara. “Psychorealism: Animator Chris Landreth Creates a New Form of Documentary.” Computer Graphics World 27, no. 7 (July 2004): 14-18.

Skoller, Jeffrey. “Making it (Un)real: Contemporary Theories and Practices in Documentary Animation.” Animation 6, no. 3 (2011): 207-214.

Thill, Scott. “Chris Landreth’s Twisted Spine Will Melt Your Mind.” Wired, September 1, 2009. https://www.wired.com/2009/09/chris-landreths-twisted-spine-will-melt-your-mind/.