Photo

What is a photo

I either went too deep, or not deep enough

Fred

27 Aug 2024 — 8 min read

“This is the Google Pixel 9, and this is another family photo, without the person taking the photo in it. Sure would be amazing if the people in the photo could leave the photo, so the person taking the photo could pop in, and make one big happy photo. I'll stop saying photo now. Pretty cool huh?”

“This is the Google Pixel 9 with the Magic Editor, which lets you start with a photo, and then make a better photo by adding more photo. Oh yeah, definitely post that one.”

Pics or it didn’t happen

Questioning the authenticity of photography is as old as photography itself. Roland Barthes theorised that images can never be neutral, and always carry meaning through context. Walter Benjamin questioned the relationship between original and reproductions of works of art. McLuhan believed that images cannot be dissociated from the medium they were displayed on. Susan Sonntag asked how photography could be used to manipulate opinions. Jean Baudrillard developed the concept of hyperreality, and I believe his thoughts contains some of the keys necessary to understand what is going on.

Four consecutive stages of the image:

reflecting a a profound reality;masking and denaturing a profound reality;masking the absence of a profound reality;having no relation to any reality whatsoever; being its own pure simulacrum.

In the first case, the image is a good appearance - representation is of the sacramental order. In the second, it is an evil appearance - it is of the order of maleficence. In the third, it plays at being an appearance - it is of the order of sorcery. In the fourth, it is no longer of the order of appearances, but of simulation.

(We’re taking an outrageous shortcut here, but) any sign or image may remain at a certain stage forever, or evolve, from the first stage to the last one, depending on an infinity of external influences. The images resulting from using features like “Add Me” and “Magic Editor” feel like they fall somewhere between the second and third stage.

Using “Add Me” feels like masking and denaturing a profound reality (second stage). I think of the man on the right side as actual light, that was only added after the initial capture. He wasn’t faked, but his addition was. We can safely assume that the three family members were all present in this location, at the same time, but the image generated simulates something that never really took place. There’s a chance the three people did actually pose at the same time, off camera, but strictly based on the information we can get from the edited image, there’s absolutely no way to confirm that. The final image is not a complete simulation, but it’s not natural either.

Using “Magic Editor” (as demonstrated in this video) feels in between the two. The initial framing and the final output are aligned, but the added information (that wasn’t initially in the frame) is somehow synthetic. None of the light that was there actually reached the sensor. This synthetic light was absent from the frame, and derived from what was in the initial capture. The center of the image was present, but what’s added afterwards was absent.

None of these features go as far as the fourth stage. Both still have a relatively clear relationship to a reality.

Dissimulating and simulating

[…] To dissimulate is to pretend not to have what one has. To simulate is to feign to have what one doesn't have. One implies a presence, the other an absence. […] pretending, or dissimulating, leaves the principle of reality intact: the difference is always clear, it is simply masked, whereas simulation threatens the difference between the "true" and the "false," the "real" and the "imaginary”[…]

According to this definition, both “Add Me” and “Magic Editor” simulate, as they threaten the difference between the “true” and the “false”, each in their own way. “Magic Editor” can be used for dissimulating (removing people in the background, clouds or trees could equal to dissimulating), but here, both ads focus on simulation instead of dissimulation. The context of “the three family members standing up next to each other in front of the camera” may have happened, but what we’re seeing does not leave the principle of reality intact: it contains real elements, laid out in a way that isn’t real. The background sky and the foreground water added to the surfer picture were allegedly there, but the final image is not an accurate reflection of their existence. Even if they were absolutely perfectly rendered and indistinguishable from what they really looked like, they would only be realistic simulations at best, and never reflections.

In this sense, both final images are partial simulations of a reality.

Why it feels different this time

Skills

My career as a designer started twenty years ago, somewhere on the seventh floor of a Parisian advertising agency. As a junior freshly added to a small team, I was given the most mundane tasks, which I gladly accepted, seeing them as opportunities to learn and gain experience quickly.

While I didn’t learn everything from scratch - some tools, like the Magic Wand or Content-Aware fill, were already advanced enough to be useful - I actually learned image editing “the old school way”, relying on the pen tool, the feather tool, the sponge, the clone stamp, layer masks, blending modes… And beyond technique, I also had to learn and understand perspective, volumes, light and shadows, and even human anatomy to make sure my edits were invisible to the untrained eye. “Add a patch of grass over there”, “remove a few creases and wrinkles”, “make the sky more blue” or “lift the subject’s right arm a tiny bit” are some of the many tasks I was asked to complete, and I know firsthand how much effort goes into making an image look unedited.

Discarding the potential danger of these new features by saying that’s nothing new, or that “everybody’s been doing it forever” fells a little lazy to me. This makes it sound like absolutely everybody has always had the ability, time and knowledge to fake images. While Photoshop has been there for decades, it has always needed someone skilled enough to make the edits, and this someone was able to make subjective and creative decisions (and mistakes) on their edits. Without anyone operating it, the software wasn’t capable of doing much on its own. What’s new is that these edits can now truly be made without anyone, better and faster than any graphic designer out there could.

Capturing memories / crafting memories

“When you define a memory as [what you’re remembering], there is a fallibility to it: You could have a true and perfect representation of a moment that felt completely fake and completely wrong. What some of these edits do is help you create the moment that is the way you remember it, that's authentic to your memory and to the greater context, but maybe isn't authentic to a particular millisecond.”

Isaac Reynolds, in For Google’s Pixel Camera Team, It’s All About The Memories - Wired, Aug 13, 2024

Let’s get back to the family picture ad. While the final picture is sold to us as ‘the moment the way you remember it’, or ‘authentic to the memory and to the greater context’ (not my words, but words from the people who made these tools), I believe that the initial image (before the addition of third person) is actually how this moment should be remembered. Most people will probably perceive the final edit as a “better” picture, but when looked at from an authenticity or memory point of view, I think they don’t even compare. The initial one is the one to keep (and the same is true for the surfer picture).

I feel these new features blur the border between capturing a memory and crafting a memory. And I don’t see it as a good thing.

When reflecting on my own experience as a photographer, I believe my intention has always been (and still is) to capture memories. You can probably find photographers out there who would define their photography as a crafting game, but I personally don’t. Some of my favourite pictures don’t even look crafted or polished. I like them because they accurately reflect what I saw and what I wanted to capture. I genuinely think you capture a memory by looking around, by framing a specific subject, and by pushing a button. You can make a lot of deliberate and creative choices about framing or exposure, but none of these choices feel like cheating. At the most basic level, a photo should be a record of “a reality” (which is very different from “the reality”).

When you start using tools such as “Add Me” or “Magic Editor”, you’re definitely past the capture level, and stepping into “crafted memories” territory. You go further than just adjusting colors, saturation or contrast with your standard tools. You don’t play around with something that was there, but simulate something that wasn’t. You’re crafting an “idealised” version of reality, that uses a reality as a base, but can’t be truly labelled as a reality itself anymore.

Synthetic memories

These new tools generate partial simulations (still reality-based, with some simulated elements) faster and better than before, and that’s pretty much it. I’d say there’s nothing to panic about. I’m aware of the issues that will arise from their increased circulation, but we’re still far from the complete apocalypse some people like to predict. I’m sure many saw the rise of Photoshop as the beginning of the end, when it comes to trusting what you see. The trust in photographic integrity has undeniably eroded over time, but I’m hoping there’s still some of it left, here and there. I’m still a little optimistic.

My problem with these tools is perhaps the way they promote their own version of “better”. As if images could only be made better (and never degraded) through editing and simulating, without ever losing anything in the process. This mindset also validates sharing these partial simulations to the world (“definitely post that one”). You should be proud of this better photo you’ve just crafted. It is better than what you were able to capture without our tools, so go ahead and show it to the world. No one cares if it’s not authentic or accurate. Just look at it. That’s how you want to remember it. You know it’s better. Trust us.

“Better” is a subjective measure of quality, that falls apart as soon as someone has a different interpretation of what better means. Someone like me, for example, who believes that fewer edits or alterations is what makes images better, and that adding simulated elements can result in losing authenticity.

While some skepticism is always healthy, I believe we should still begin with the assumption that a photo is a record of reality, even in an age where editing tools become ubiquitous and can fake almost anything. I know this might sound counter-intuitive, but this assumption is crucial, and needs to remain intact for as long as possible. It can and will fail at times, but that’s part of what makes it so precious. Because the day we default to believing that every photo is “fake until proven otherwise” is the day we truly lose something fundamental: a collective understanding of authenticity. At that point, we would risk descending into the same kind of madness and denial as flat earthers, who reject photographic evidence of the Earth's shape, leading us down a path where “nothing is real”, and where truth itself is constantly in question.

If you’ve made it this far, you’ve probably already guessed I won’t be using any of these new features. At a personal level, I don’t feel like I can do much more than scratching the surface of this complex topic. I’m just someone who likes to take pictures, who’s weirdly attached to the idea of integrity in photography, who can actually manipulate pixels, and who briefly studied the subject, a long time ago. You shouldn’t need to go back to philosophers from the previous century to find answers (I tried, and failed), but you can give it a shot and have fun in the process like I did.

In the end, it comes down to critically assessing the integrity of all parties, from the companies building these tools to anyone using these tools, in equal parts. There’s a lot of value in photographic integrity, that we shouldn’t trade for “better” or “synthetic” memories. I understand what the companies get by creating such tools, but I just don’t see what we receive in this trade, in exchange for our collective trust. And if we even get anything of value, is it worth what we’re trading it for?

Sources