Messing with Punctum: Cinemagraphs and the Uncanny

This essay investigates cinemagraphs and their increasing use in social media and advertising. Following on from a brief survey of current commercial practice, the article draws connections with Roland Barthes’ historic reading of photographic production and interpretation, albeit situated within twenty-first-century technology. The essay argues that the principle device of cinemagraphs, the looping of localised motion within an otherwise still image—this juxtaposition of the animate with the inanimate, or the dead with the undead—when executed well, alludes to a now historic framework of photographic interpretation, yet with often uncanny results. When read in the context of an increasingly transhuman culture, of data-base politics, artificial intelligence, and robotics, cinemagraphs perhaps provide a glimpse of “The Uncanny,” in a way that is rather well suited to prevailing Zeitgeist. The text argues that this is not dissimilar to the reception of mechanical automata in the eighteenth and nineteenth centuries, when situated against the then backdrop of the first industrial revolution.



The recent popularity of cinemagraphs in social media and advertising owes as much to a certain novelty factor as to an underlying fascination with the uncanny, particularly in the context of what Brynjolfsson and McAfee (2014) call the “Second Machine Age.”

In Camera Lucida, Barthes (1993, 92) describes “all those young photographers who are at work in the world” as “agents of death.” He posits that punctum represents the unique stigmatum within the photographic image; that which mechanically preserves what “could never be repeated existentially” (Barthes 1993, 92).

But no doubt, as Rubinstein (2015) put so eloquently in his recent essay “What Is Twentieth Century Photography?” our post-photographic culture has put paid to all that.

And yet, I want to argue, cinemagraphs’ principle device of localised, looped motion within an otherwise still image—this juxtaposition of the animate with the inanimate, or the dead with the undead—when executed well, alludes nolens volens to a Barthesian framework, albeit with uncanny results (Royle 2003).

When read in the context of an increasingly transhuman future, of nanotechnology, deep learning, artificial intelligence, and robotics, cinemagraphs perhaps provide glimpses of the uncanny, particularly well suited to our transhuman condition, and much in the same way as mechanical automata did in the eighteenth and nineteenth centuries, and against the backdrop of the first Industrial Revolution.

While it is perhaps fanciful to imbue with epistemological validation a medium that is so clearly bound up in the transactional economies of social media, cinemagraphs nonetheless seem to connect with some long-defunct notions of photographic theory, albeit in a post-photographic context.


What Is a Cinemagraph?

The graphic interchange format (GIF) was introduced in 1987 by internet provider Compuserve with the stated aim to make the most of limited bandwidth, due to primitive modems used for web-access at the time. The GIF standard allows effective, lossless compression, albeit with a reduced gamut of no more than 256 colours. In addition, GIFs support transparent background colours, metadata, and multiple images embedded within one file.

This combination of attributes has made possible a whole range of uses on the web, from ubiquitous memes to banner ads and animations. What came into use, as a work-around for slow transmission speeds, has become one of the most widely supported image file formats on the web due to its versatility.

Even the advent of broadband hasn’t diminished the popularity of GIFs. In a belated acknowledgement of the GIF standard, Facebook started offering GIF support in 2015. Newer formats, such as portable network graphics (PNG), have not yet succeeded in replacing GIFs, mainly due to their lack of animation support.

Cinemagraph was a term invented and trademarked in 2011 by American fashion photographer Jamie Beck and graphic designer Kevin Burg for a particular type of GIF animation.

In Burg’s words (Lin 2014, n.p.), a cinemagraph is a “living photograph. It’s a photograph that has a living moment inside of it.” As to the term cinemagraph, Berg explains: “We were just playing with Greek roots. We were researching how they came up with the term photography” (Lin 2014, n.p.). He states that “calling it a GIF didn’t seem to work either, because a GIF can mean so many things. This is a specific thing with its own criteria. Its own medium,” a nod to a “cinematic moment, married to a photograph” (Fitzpatrick 2013, n.p.).

After Beck and Burg let the initial trademark registration lapse, a rival company, Toronto-based Flixel secured the cinemagraph trademark in 2014 for “compositing the photographs or the video frames into a seamless loop of sequential frames in such a manner that motion in part of the subject between exposures is perceived as a repeating or continued motion, in contrast with the stillness of the rest of the image” (Trademark File 2014, n.p.).

In other words, cinemagraphs are hybrids between full-motion videos and traditional photographs. Assembled with image editing and compositing software, they combine still images with frames of the same subject that show limited movement in a small part of the composition. The moving part of the composition is then masked in such a way that the rest of the image across the whole sequence appears still and motionless—apart from the moving element. The movement within the image will repeat in a more or less obvious manner, depending on the skill of the art-worker, and on subject matter. Saved as a GIF, the cinemagraph can then be uploaded to various web platforms, just like any other GIF. For more complex sequences, which require true colour, or larger files, motion formats can be used as well.

First introduced during a collaboration covering New York Fashion Week, Beck and Burg quickly gained considerable commercial exposure, after publishing their first cinemagraphs on Beck’s Tumblr blog. According to Beck, “the name Cinemagraph went generic overnight. I tweeted what we named it, and it went crazy from there” (Lin 2014, n.p.).

According to Time Magazine, cinemagraphs create “moments that are quiet and contemplative, elevating the humble GIF into something much more refined” (Lin 2014, n.p.). Adweek proclaims that “subtly animated Cinemagraphs are practically hypnotic” and goes on to say that “you’re going to start to see a ton of these on Facebook”(Sloane 2015, n.p.).

As Burg notes, “People can’t stop staring at them, isn’t that what advertisers want?” (Sloane 2015, n.p.). And so it would seem, Beck and Burg have since set up their own studio specialising in cinemagraphs and have produced campaigns for many global fashion-, automotive-, and consumer-brands. Other photographers and marketers have swiftly followed and now routinely offer cinemagraphs as part of their commercial portfolios.

Cinemagraphs have been used in campaigns for brands such as Heineken, Toyota, Mercedes, Nestle, Coca-Cola, Diesel, Chopard, Cartier, Tiffany, Donna Karen, DKNY, and Balenciaga. And the list keeps growing.

According to Rich Tong, head of influencer platform Fohr Card, and former fashion director at Tumblr, a series of six cinemagraphs Beck and Burg produced for Oscar de la Renta in 2011 achieved 55,000 notes and more than 2,000,000 impressions (Indvik 2011). Michael Aaron Flicker, president of marketing agency XenoPsi, points out that cinemagraphs tend “to have more virality, with 71% higher organic reach than still photographs,” and he concludes that “when done well and used effectively, Cinemagraphs are an enchanting and beautiful tool for telling a brand’s story” (Flicker 2016, n.p.) He goes on to say that “auto-play looping” make cinemagraphs “easy to consume, but hard to ignore” (Flicker 2016, n.p.).

According to Mark Homza, cofounder of Cinemagraph provider Flixel, a “Cinemagraph banner ad for Panasonic’s Lumix range of cameras was clicked 60% more than the static version” (Marshall 2015, n.p.). Cinemagraphs, Homza maintains, achieve up to 80 percent better click-throughs than static banner ads (Marshall 2015).

Increasingly gummed up by the entropic goo of social media, yet compelled to stand out from the crowd, brands are naturally drawn to cheap, easily sharable media formats that deliver above-average participation figures. In an online world, at the same time saturated and perpetually hungry for more disposable visual content, on platforms and blogs, across social media and corporate sites, the amount of liking, sharing, and clicking-through matters a great deal. Amalgamated in sophisticated “dashboards,” which show the ebb and flow of customer “engagement,” these data provide marketers with a notional sense of success and validate “share of voice,” albeit within the echo chamber that is social media marketing.

Cinemagraphs have certainly succeeded in complementing the online marketing tool set, yet it remains to be seen to what extend the novelty factor will eventually wear off. Media technology evolves at breakneck speed, and the next big attention-grabbing format is never far away.

But before fully immersive, artificial intelligence enabled artificial reality and virtual reality consign cinemagraphs to the quaint end of the hype-cycle, it may be worth analysing more fully just why the combination of still image with looped animation may be quite so hypnotic as claimed.


Messing with Punctum

Can we speculate that Barthes might have taken an interest in cinemagraphs, especially the obsessive-compulsive repetitions of often-banal detail within them—this juxtaposition of the animate with the inanimate, or the dead with the undead?

Did Barthes not claim in Camera Lucida, his late, elegiac, and mournful text, full of emotional depth and introspection, that a photograph represented a “micro-version of death” (Barthes 1993, 14)? That, to fully understand the traumatic irreversibility of time, one need only look at photographic images?

In Barthes’ view there was always a “defeat of time in them: that is dead and that is going to die” (Barthes 1993, 96). It was this irreducible essence, which provided the connection between otherwise unrelated photographs: the highly personal—a photograph of Barthes’ late mother for example—with the public or historic—a photograph of Queen Victoria or a picture of an empty street in Atget’s Paris. Whichever way you looked, photography was ultimately about one thing and one thing only: time, its passing, and ultimately death. This insight represented a considerable departure from Barthes’ earlier writings on photography, particularly the deft way in which he had previously, and rather more rigorously, one might add, unpicked the semiotic tapestry of photographic codes and signifiers.

Barthes’ decidedly more phenomenological approach in Camera Lucida comprised in the main of the idea of studium, whose constituent parts spoke of cultural, historical, or scientific context, and punctum, a uniquely touching and poignant detail, which connected with the viewer on a more profound and emotional level. Barthes frequently referred to punctum as a stigmatum, or wound, which betrayed with “lacerating emphasis” the passing of time itself (Barthes 1993, 96). In Barthes’ eyes, it was punctum that actually validated the photograph’s existence; it punctures the surface of the image and, in doing so, captured the essential nature of the medium.

Barthes’ narrative proved immensely impactful for a generation of students, theorists, and practitioners. As Geoffrey Batchen argued in his 2009 collection of essays on the subject, Camera Lucida is quite possibly the most widely read book on photography. In a roundabout way it could be said to have influenced artists as diverse as Gerhard Richter, Christian Boltanski, Tacita Dean, Fiona Tan, and more recently Taryn Simon, who all created vast collections of everyday portraits, which owe much of their allure to Barthes (Dillon 2011).

Without doubt, Camera Lucida successfully energised post-1980s photography scholarship; not least in terms of the backlash it generated: in particular the criticism that Barthes had succumbed to a rather sentimental, and essentialist interpretation of a medium clearly capable of considerably more complex subtexts. This, combined with the conspicuous absence of any acknowledgement of potential ontological entanglements brought about by the impending onset of the post-photographic condition, led to the eventual cooling if not wholesale dismissal of Barthes’ ideas. Batchen (2009, 21) asked the obvious question: “Has the photography pursued by Barthes perhaps already gone, transformed into a mere ghost of its former self? Can we any longer feel the affect that so transported Barthes as he looked at certain photographs?”The answer must surely be a regretful but nonetheless resounding “No, we cannot.”

Rubinstein, in his 2015 essay titled “What Is 21st Century Photography?” argued convincingly that the “four horsemen of the photographic apocalypse: Index, Punctum, Document, and Representation can no more account for 21st century digital realities.” He goes on to say that “even if some parts of this form of photography are still visible, they are in a state of advanced decay” (Rubinstein 2015, n.p.).Rubinstein maintains that photographic image making must, in order to stay relevant, move away from an obsession with objects and instead focus on processes, in particular those processes that are largely hidden from view but which nonetheless underscore the information economy of the twenty-first century. He posits that the “demise of the industrial age is curtains for the spectacle of representation: visual surveillance is replaced with predictive policing, industrial processes replaced with trading algorithms, armies replaced with remote controlled killer robots and perspectival geometry replaced with the…topology of the computer screen” (Rubinstein 2015, n.p.). Rubinstein (2015, n.p.) concludes that “these changes do not mean that suddenly, what we see in front of our eyes does not matter, but that many more things that matter are outside our human field of view. The question is, what becomes of photography when the locus of power shifts from the optical nerve to the fibre-optic cable?”

So it is against this dialectical backdrop—with the beautiful ruin of Barthesian punctum on the one hand and Rubinstein’s twenty-first-century photography on the other—that I want to try and situate cinemagraphs and thus attempt a contemporary reading. Despite the fact that Barthes’ late theory of photography is today an all but burnt-out shell orbiting in zero-gravity, like a remnant of a once-exquisite vessel, forever adrift in discursive space, I want to argue that perhaps it can provide a starting point for analysis, albeit one which leads beyond the essentialist Barthesian conclusion. I want to suggest that Barthes’ notion of the centrality of time can still be used to unlock our understanding of cinemagraphs. The trajectory of this argument, however, will inevitably arc toward a reading much closer to Rubinstein’s outlines of a photographic practice fit for the twenty-first century—one which at once represents as well as reveals the anxieties, power-relationships, processes, and transactions that characterise what Brynjolfsson and McAfee (2014) call the “Second Machine Age.”


Cinemagraphs and the Uncanny

As technologist Kevin Kelly (2012, n.p.) argued: “You’ll be paid in the future based on how well you work with robots”—if you have a job at all, one might add.

The Second Machine Age represents a profound shift in the way capital, labour, and profit are organised. Where the first machine age brought about a slow, many decades-long transformation from a largely rural, agricultural society to an urban manufacturing economy, the dawn of the Second Machine Age leads to disruption on an altogether grander scale and in a much shorter time frame. According to Brynjolfsson and McAfee (2014), the benefits of this disruption are shared much more unequally than was the case before. In the “winner-takes-all” markets of global platform-capitalism, larger parts of the population are subject to greater economic uncertainty than ever before. As a consequence, the authors suggest, the Second Machine Age is poised to lead to a breakdown of the societal contract, which stipulates that the opportunities enjoyed by the next generation will always be greater than those of the previous (Brynjolfsson and McAfee 2014).

As the fields of computing, artificial intelligence, bio-tech, robotics, and nanotechnology advance with ever greater speed and sophistication, from natural language processing to fine motor skills, from face-recognition to deep-learning, self-driving cars, and grey goo, the sound of music emanating from Silicon Valley and captivating us so thoroughly, is not the pleasant soundtrack of just another elevator pitch but the relentless drumbeat of the march of the machines: big data, dashboards, metrics, and all. In the wake of this uncanny procession, physicality becomes contingent, a mere mirage of the new underlying lingua franca of zeroes and ones.

Against this collapse of objecthood, Rubinstein’s thesis of the four horsemen of the photographic apocalypse seems entirely on point, his call for twenty-first-century photography legitimate and urgent. I want to suggest that, despite their often banal and commercial nature, the new medium of cinemagraphs can indeed be seen in this context.

And yet, can we really view the repetitive loops of animated detail inside a cinemagraph without also being reminded of Barthes’ preoccupation with punctum and the passage of time itself? In a cinemagraph, time seems stuck in a never-ending cycle of creepy, mechanical monotony. Like an old record stuck in a groove, unable to let the music play, the cinemagraph is a photograph that remains forever unresolved. As with Schrödinger’s cat, at once dead and alive, the cinemagraph forsakes closure; any sense of completion is lost in a state of perpetual, hypnotic suspense. But there is nothing mechanical about cinemagraphs; no cogs, wheels, and springs, no needles jumping in a groove. Instead there is only cold, binary code. Barthes’ sense of punctum, that lacerating wound in an otherwise smooth surface of photographic representation, thus transmogrifies into something more akin to a stuck programming loop.

Whether it is that lock of hair, forever pendulous in an invisible breeze, whether it is the fabric fold of a garment fluttering against an otherwise immobilised, lifeless figure, or some other poignant detail: let us not be fooled. Cinemagraphs do not reference some sort of punctum on steroids, some amplification of Barthes’ ideas by means of animation; punctum on a loop, so to speak, so that the Instagram generation may finally “get it” after sufficient repetition, and thirty years late. No, cinemagraphs merely appropriate punctum in the same manner that Instagram filters appropriate the nostalgic gamut of an allegedly more authentic age of Polaroids and Super 8. In the process, punctum is finished off, its original meaning forever-suspended in endless, uncanny repetition.

In his magisterial survey on the subject, Royle (2003, 1–2) argues that the uncanny “appear(s) to be indissociably bound up with a sense of repetition or ‘coming back’,…the constant or eternal return of the same thing, a compulsion to repeat.” This, he argues, goes hand in hand with “feelings of uncertainty, in particular regarding the reality of who one is…. It is a crisis of the natural, touching upon everything that one might have thought was ‘part of nature’: one’s own nature, human nature, the nature of reality and the world.” Royle attributes the uncanny to a “peculiar commingling of the familiar with the unfamiliar,” something which can be felt “in response to dolls and other lifelike or mechanical objects” (Royle 2003, 1–2).

Just like those ingenious mechanical contraptions made during the golden age of automata in the eighteenth and nineteenth centuries, cinemagraphs, too, evoke the uncanny by dint of their apparent syntheses of robotic otherworldliness with life-like familiarity, of the dead with the undead. And both mechanical automata as well as cinemagraphs can be characterised as “self-operating after they have been set in motion” (Franchi and Güzeldere 2005, 27).

Yet, while mechanical automata are driven by the very constituents of the First Machine Age, the levers, gears, springs, cams, and crankshafts of industrialisation, the cinemagraph is constituted by code, the substrate of the Second Machine Age. Where the former represents an actualised expression of the anxieties and imaginations of the steam-age, its awestruck preoccupations with the magic of clockwork mechanics supposedly underpinning life itself, the latter speaks of a rather more contemporary disposition of foreboding and apprehension, one which only superficially alludes to something mechanical. What unites both is the lingering sense of dread and unease they evoke.



In Camera Lucida, Barthes’ reading of photography invites us to reflect upon who we are; it suggests that aside from all cultural context, punctum can unlock a portal that connects with time itself, thus amplifying our transient sense of humanity. Cinemagraphs on the other hand allude to our new destiny as hybridic cyborgs, undead zombies, whose external Gestalt betrays their internal, immaterial substrate of zeroes and ones. Whatever animated, punctum-esque detail we are looking at, there is nothing human left in these images. Instead we are witnessing an uncanny, post-apocalyptic, and transhuman state of trance, where (artificial) life is eternal, suspended in obsessive-compulsive repetition. The idea of what it is to be human, once upon a time captured alluringly by Barthes’ punctum, is vanquished in the process: our fleeting sense of being in the world, of a beginning and an end. The result of this dissolution within cinemagraphs is a medium, which combines the banal with the profound in the most startling manner. No wonder, the creators of the first cinemagraphs attest to their hypnotic quality.

It would therefore be shortsighted to attribute the recent surge in the use of cinemagraphs to novelty factor alone. On the contrary, I would argue that cinemagraphs have been successful, precisely because they do seem to operate in a discursive space that is cognisant of twentieth-century photo history. But above all, cinemagraphs tap into some fundamental anxieties that underscore our forced march toward the Second Machine Age: anxieties of disruption, upheaval, and alienation. Perhaps, there is, once again, a spectre that is haunting humanity; yet this time, it is the spectre of a disenfranchised, transhuman future.

In the spirit of Rubinstein, and with more than a mere nod toward Barthes, cinemagraphs may just be, in their own, insignificant, banal, and exceedingly commercial ways, expressive of this crisis.



Barthes, Roland. 1993. Camera Lucida. Translated by Richard Howard. London: Vintage Books.

Batchen, Geoffrey. 2009. Photography Degree Zero: Reflections on Roland Barthes’s Camera Lucida. London: MIT Press.

Brynjolfsson, Erik, and Andrew McAfee. 2014. The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies. New York: W. W. Norton & Company.

Dillon, Brian. 2011. “Rereading: Camera Lucida by Roland Barthes.” The Guardian, March 25. Accessed July 13, 2016.

Fitzpatrick, Tommye. 2013. “The Business of Blogging: Jamie Beck and Kevin Burg.” Business of Fashion, June 13. Accessed July 14, 2016.

Flicker, Michael Aaron. 2016. “Why Brands Should Embrace Cinemagraphs for Storytelling: Four Ways Marketers Can Use Cinemagraphs to Bring Their Brands to Life.” Advertising Age, January 5. Accessed July 15, 2016.

Franchi, Stefano, and Güven Güzeldere. 2005. “Machinations of the Mind: Cybernetics and Artificial Intelligence from Automata to Cyborgs.” In Mechanical Bodies, Computational Minds: Artificial Intelligence from Automata to Cyborgs, edited by Stefano Franchi, and Güven Güzeldere, 15–149. Cambridge, MA: MIT Press.

Indvik, Lauren. 2011. “How a Blogging Duo Is Changing Fashion Photography with Animated Cinemagraphs.” Mashable, October 5. Accessed July 14, 2016.

Kelly, Kevin. 2012. “Better Than Human: Why Robots Will – and Must – Take Our Jobs.” Wired, December 24. Accessed 12 July, 2016.

Lin, Joseph C. 2014. “When Photos Come to Life: The Art of the Cinemagraph.” Time Magazine, February 19. Accessed July 14, 2016.

Marshall, Jack. 2015. “Marketers Experiment with Cinemagraph Ads.” The Wall Street Journal, March 23. Accessed July 15, 2016.

Royle, Nicholas. 2003. The Uncanny. Manchester, UK: Manchester University Press.

Rubinstein, Daniel. 2015. “What Is 21st Century Photography?” The Photographers’ Gallery Blog, July 3. Accessed July 14, 2016.

Sloane, Garett. 2015. “Here Is the Captivating Ad Format Facebook Hopes Will Wow Its Users.” Adweek, February 10. Accessed July 14, 2016.

Trademark File. 2014. “Cinemagraph 85703768.” Trademark File, August 15. Accessed July 15, 2016.