Vote 2020 graphic
Everything you need to know about and expect during
the most important election of our lifetimes

Are Colorized Photos Rewriting History?

Illustration for article titled Are Colorized Photos Rewriting History?

There's been an explosion in the number of colorized photos lately. People find old black-and-white photos online, and meticulously add color to give us a new perspective on history. But recently one colorized image caught my eye after it was tweeted by the notoriously inaccurate HistoryInPics. It's a stunningly colorful view of the Golden Gate Bridge in 1940. But the photo is a lie.

Advertisement
Illustration for article titled Are Colorized Photos Rewriting History?

HistoryInPics scraped the image from Reddit user kibblenbits. But HistoryInPics never mentions that the photo has been colorized, nor who colorized it. The caption simply reads, "Golden Gate Bridge, 1940" leading us to believe that it's an "authentic" photo from 1940.

Advertisement

Here's where things get weirder. The photo isn't even from 1940, even though the Associated Press lists the original black-and-white photo as having been taken on November 20th of that year. Judging by the makes and models of the cars in the image, it's way off. As first spotted by commenters at The Atlantic, the photo is more recent by at least a decade.

Illustration for article titled Are Colorized Photos Rewriting History?

I asked Jalopnik's own Jason Torchinsky, and he thinks that the photo must be from 1952 or later, given that the maroon car (or, at least the one that's maroon in the colorized photo) looks most like a 1952 Chevy Bel Air.

Jason even made an image to demonstrate this, explaining that you can kind of "make out the extra grille 'teeth' they added in 1952."

Advertisement
Illustration for article titled Are Colorized Photos Rewriting History?

"The bumper guards were options that could have been on any of the years," Jason said. "I personally think it's a '52, but you could argue for '51."

Advertisement

Either way, it's not from 1940. And the photo wasn't taken in color.

Why does any of this matter? Putting aside the Associated Press's error in dating the photo for a moment, I think (and so far, I'm in the minority on this one) that there's a danger that the colorized photos could become more popular in search engine results than the black-and-white ones.

Advertisement
Illustration for article titled Are Colorized Photos Rewriting History?

Take the images above, for example. The top image is a photo by Norman Seeff dating back to around 1985. The photo was recently colorized by Jordan J. Lloyd.

Advertisement

"WTF. I didn't even realize I was on /r/ColorizedHistory. Completely undetectable coloring," one commenter says about Lloyd's colorization of Jobs.

As Lloyd explains in the comments of his Reddit post, he agonized over the color of the sweater and other such details, as the best colorizers do. But who's to say what the color of that sweater really was? And what about the books on the shelf or the art on the wall?

Advertisement

Color images resonate with those of us who grew up during a period when color photography was the norm—and color was obviously commonplace in 1985! But Seeff, the original photographer, made the conscious choice to shoot this particular portrait in black-and-white. And what happens if this colorized version becomes more popular than the black-and-white version? Should we care? Does it matter at all to history?

"My only concern would be that nobody should see color versions as 'better' than black and white originals in any way," Alan Taylor at The Atlantic tells me. "They may seem more relatable, or tickle some part of the brain that itches to fill in the gaps left by monochrome images, but they are definitely alterations—not originals."

Advertisement

"If a colorized image can spark interest in history, great, I'd hope people would dig deep enough to learn about the challenges of early photography, and how some came to master and take full advantage of contrast and tone to aid in storytelling," Taylor said.

Errors happen, as we see with the AP's incorrect date for the Golden Gate Bridge image. But what happens when we start to tinker with old photos in such a way that we're inserting "errors" like the wrong color?

Advertisement
Illustration for article titled Are Colorized Photos Rewriting History?

The colorized photo above originally comes from the Library of Congress and depicts ballet dancers in 1924. But Paul Edwards, the person who did the colorization, doesn't really know the original colors of what those women are wearing.

Advertisement
Illustration for article titled Are Colorized Photos Rewriting History?

There's nothing new about photo colorization. Manual colorization is nearly as old as photography itself. But here in the early 21st century there does appear to be a desire to colorize all the old photos, now that the digital tools that make it easier are so commonplace.

Advertisement

As we've learned time and again, the internet is like that old kids' game of Telephone. And a particular image or fact can start with one person online and inevitably become more distorted as it gets further down the line. Errors are inserted—sometimes intentionally, sometimes by accident—until it winds up completely distorting our understanding of the original message.

The vast majority of photo colorizers probably don't have deception as a goal. But it seems important to have this discussion now. Especially since some of the largest popularizers of history online (I'm looking at you, @HistoryInPics) have a terrible track record when it comes to providing context for these old images. It's one thing to remember our history. It's another to remember it wrong.

Advertisement

Correction: This post mistakenly claimed that the Golden Gate photo had been colorized by Dana Keller. It was actually colorized by kibblenbits.

[Update: You can read my interview with professional colorizer Dana Keller here.]

Advertisement

Images: Photo of the Golden Gate Bridge via the Associated Press with an as-yet undetermined date, colorized by kibblenbits; Black-and-white photo of Steve Jobs circa 1985 by Norman Seeff, colorized by Jordan J. Lloyd; Dancers of the National American Ballet on August 20, 1924 via Library of Congress, colorized by Paul Edwards; Nikola Tesla photo circa 1894, colorized by Dana Keller

Advertisement

Share This Story

Get our newsletter

DISCUSSION

dynamichrome
Jordan Lloyd

Hi Matt,

This is a great article and thank you for taking a critical look at what for some is a hobby, but for a few like myself, is a craft that should be taken seriously. My own caveat in this particular article, is one of personal ethics when it comes to why and whether or not an image should be restored in colour: we now have the tools that allow us a great deal of control, but is it right to do it in the first place?

In this particular instance, you have chosen an image I normally wouldn't touch in any other context; however, I happen to be doing a talk on Colour Restoration at the Apple Store in London tomorrow evening and I thought an image of Jobs would be a fun way to wrap it up - I have been very transparent about this and this is not representative at all of my usual stance towards a colour restoration. Would I colour restore an image when it was intentionally taken in black and white like the Norman Seeff, or an Ansell Adams? No. Because to do so would be to ruin the creator's intent, unless you happen to be doing it for a specified reason (as above), or private practice so you can get better.

However, there is a distinction to be made between approaching images that have been taken in black and white as a limitation of the available camera technology of the era. If this is the case and the image was taken for historical posterity, it's fair game as I'm concerned with the following guideline:

I make no claims whatsoever on historical accuracy. I wasn't there, I didn't time travel and take a snap with my phone. A great deal of research that constitutes the greatest part of any colour restoration eliminates as much guesswork as possible so that the result gives you a flavour of authenticity.

Since you have already linked the Jobs image, then I will also link the references I spent many hours gathering, from which you can make your own decisions: Jobs was wearing that sweater as those images were taken as part of a promotional campaign for Apple. The G1 Transformer Inferno toy is that colour, the cubicle is a blue grey, as evidenced from a colour shot taken in the same session.

There are two elephants in the room as far as I'm concerned:

1. Every article I've read on colourising an image focuses on the result, not the process. The 'dark matter' of the colour restoration process is all the stuff around the image itself. Like any forensic investigation, you start with a few clues and you do whatever it takes to find references in the following priority:

a. Find the exact sign, piece of clothing, car bumper, etc.

b. If not the exact one, then an object in the same group, like a brand or company colour

c. If you can't find this, then you look to period objects of the time. Often following this trail leads you to a. or b.

d. For nature, find out where the image was taken and look at the local fauna, weather conditions etc. I've had debates over the specific colour of mud in a commission before.

e. Use known values: You know the colour of an American flag, or maybe you can dig up the 6 stock colours that were used for the Plymouth Roadrunner in 1969, then go from there.

f. If this fails, then make an educated guess (exhaust a-e as much as you can)

Take all of these, put them in, adjust for lighting, atmosphere and film stock and then you can at least make some claim as to a flavour of authenticity.

2. I call it 'Colour Restoration' for a reason, as often there is an actual restoration involved. I can spend hours cleaning up specks that no one will see unless it's a high quality print because we have utmost reverence for the image, but here's the thing: we restore the image to its original state, devoid of the decades of damage and weathering in less than optimum conditions.

Lastly, and I'm sure all of us who do this would agree, we make absolutely no claims as to replacing a vastly superior black and white image (even if it needs a little love!). The original has been around a long time, and we only seek to uncover what the original photographer may have looked at through their viewfinder. It is a supplement so one can perceive the abstract concept of history in a slightly different way; it may be more 'real' to the beholder, or it simply be a fun way of looking at something that's been around a long time and I'm glad that there is a lot of attention because no one was looking at the originals before. Those of us who take it seriously always credit the creator and supplied detailed information on where to find the original, along with a contextual explanation that has come out through meticulous research; our problem lies with people who take the images out of context, with no attribution and no link to an explanation.

I think this craft (which it most certainly is) needs a critical lens and I think more informed discussion is a great way forward, so thanks for a good discussion and I welcome anyone else to join in.

Thanks,

Jordan Lloyd