Dazzle Camouflage: Ships Painted to Be Seen, Not Hidden

In 1917, British ships were being sunk at about 23 a week. The Royal Navy's answer was to paint the survivors in loud black and white zigzags and send them back out.

It sounds like a joke. It was official policy. Wilkinson later estimated that about 4,000 British merchant ships and more than 400 warships ended up in the pattern. The scheme was called dazzle, and it rests on a point most colour advice ignores: sometimes the goal is not to blend in, but to be misread.

The problem was not visibility

By the end of 1916, about one-fifth of British merchant ships had been sunk. From March to December 1917 the total came to 925, as the Smithsonian's account of the programme records. The enemy was the U-boat.

Ordinary camouflage had little to offer against a periscope. And a submarine officer does not need a perfect picture of the target. He needs to judge where it is going. Get the heading wrong and the torpedo misses.

That was the insight of Norman Wilkinson, a marine painter serving as a lieutenant in the Royal Naval Volunteer Reserve. He reportedly had the idea in April 1917 on a train journey, according to Annette Wickham's history of the scheme for the Royal Academy. If a ship could not be made invisible, he reasoned, the opposite might work. Make it highly visible, and break up its form so the observer cannot tell where it is going. Wilkinson called it "disguise of direction."

What the Dazzle Section actually did

The Admiralty took the idea up fast. HMS Industry, a store ship, was the first vessel painted, in May 1917. The Dazzle Section began work at the Royal Academy in London in June. By October the Admiralty asked for 50 troopships to be done.

The work was oddly scientific. The team built a scale model for each of the 37 types of merchant vessel. Each model sat on a turntable against a painted sea and sky and was watched through a periscope, so designers saw the ship the way a U-boat officer would. Most of the staff were art students, and many were women. Wickham notes that sources disagree on how many, from eleven to about twenty. One of them, Eva Mackenzie, later married Wilkinson.

Late in October 1917, the Smithsonian reports, King George V came to watch a demonstration. He studied a model through the periscope and estimated its heading as "South by west." Wilkinson's answer was "East-southeast." The two answers sit roughly 80 degrees apart.

The most important areas for distortion sit around the bridge, where a submariner looks to judge a ship's course, Wickham reports. Sloping lines and curves worked better than verticals. The Smithsonian adds that funnels got angled stripes of their own.

Painters such as the Vorticist Edward Wadsworth supervised the work at the main shipping ports, according to the same Royal Academy history. Picasso, the Smithsonian adds, is said to have claimed the Cubists invented dazzle. The resemblance to avant-garde painting is easy to see. The claim itself is only reported, and the models-and-periscope routine reads more like engineering than art.

The colours that did the work

Here the story becomes useful for anyone who picks colour for a living.

The Dazzle Section did not begin with grey and add stripes. According to Wickham, the team first tried "violently contrasting colours." Those turned out to be less useful, partly because U-boats used colour screens in their periscopes. The Smithsonian lists the early pairings: black and white, green and mauve, orange and blue.

What survived was calmer. Black and white, with shades of blue and green, proved most effective. Wickham also stresses that hues were matched to official colour charts, which she calls essential to the scheme. A pattern only confuses if the black is really black and the white really white on every ship.

These are our own colours, chosen in the spirit of the scheme. They are not matched to the Admiralty's paint. The point is the structure: a hard light and a hard dark carry the pattern, and the blues and greens sit between them.

Why does that structure work? Contrast of lightness does the heavy lifting. If you want to see how strongly the eye reads a luminance difference, run any pair from the palette through the contrast checker. Licorice against Isabelline scores about 16 to 1. Lapis Lazuli against Wintergreen Dream scores under 2 to 1, which is why the blues and greens work as the soft middle of the pattern and not as its edges. The same eye behaviour sits behind simultaneous contrast, where a surrounding colour changes how a neighbour looks.

Built to be seen, on purpose

Dazzle sits in a small family of palettes that chase attention instead of avoiding it. Safety orange is the other famous case: a colour never built to be beautiful, only to be seen. The two share a rule. Visibility is not the enemy of a design goal. It is a tool, and you decide what it is for.

Dazzle uses it for the opposite job. Safety orange wants you to find the object. Dazzle wants you to find it and still get it wrong.

That is a useful check on your own work. Before choosing a palette, ask what a viewer should be unable to misread. A warning label should have no ambiguity. A navigation bar should have no ambiguity. A decorative hero image is allowed to be ambiguous, and sometimes it should be.

Did it work?

This is where the story gets uncomfortable.

An Admiralty analysis in 1918 compared losses among ships that had been dazzled and ships that had not. The Smithsonian summarises the figures. In the first quarter of 1918, 72% of attacked dazzled ships were sunk or damaged, against 62% of undazzled ones. In the second quarter the order reversed: 60% against 68%. Dazzled ships were also attacked slightly more often, 1.47% against 1.2%.

The Admiralty concluded that dazzle probably did no harm but was not clearly helping. It did note that morale was higher on dazzled ships. Crews liked the paint. Insurers, the Smithsonian adds, reportedly lowered premiums on dazzled ships.

A 1919 study at MIT, as reported by the same article, found that some designs caused course errors of up to 58 degrees. Only about 10 degrees was needed for a torpedo to miss. That is a laboratory result, not a battlefield one.

A better test arrived in 2011. Researchers at the University of Bristol showed volunteers two moving patterns on a screen, one plain and one dazzle, and asked which moved faster. At high speed, two high-contrast patterns, zigzags and checks, made the target seem about 7% slower. Low-contrast patterns and slow movement did nothing. That finding, published by Scott-Samuel, Baddeley, Palmer and Cuthill in PLoS ONE, suggests contrast is the active ingredient.

Note what it does not show. The experiment measured speed, not the heading Wilkinson cared about. It used screens, not ships. The authors' own example was a rocket-propelled grenade fired at a fast vehicle, where the error might come to about one metre.

So the honest verdict is mixed. Dazzle did not clearly save ships. It may bend the perception of speed under the right conditions. And the Navy kept painting: the Smithsonian counts about 2,300 British ships dazzled by June 1918 and more than 4,000 by war's end. In the United States, 1,256 ships were dazzled between March and November 1918, and 18 of them were sunk.

What to take from a pattern that confuses on purpose

Most colour advice assumes the viewer wants to understand. Dazzle is a reminder of how much of that understanding is built from a few strong cues, and how easily a few strong cues can be rearranged.

Three lessons carry over to screens:

• Contrast carries the message. Whatever else changes, the hard light and hard dark did the work. Test yours with a palette generator and the contrast checker before you pick the accents.
• Exact values matter. The scheme only worked when the paint matched the chart. A near-black that drifts to grey on a cheap monitor loses its edge.
• Evidence beats a good story. Dazzle looks convincing. The loss tables looked inconclusive. If your colour change feels right but you cannot measure it, say so.

The Dazzle Section ran on a simple bet: that people judge a moving object from a few edges, and that a painter could move those edges. One century later the best evidence says the bet partly paid off, for speed, in a lab, with a margin of about seven per cent. The ships were painted anyway, and the crews were glad of it.

Sources

• When the British Wanted to Camouflage Their Warships, They Made Them Dazzle, Smithsonian Magazine
• Designed to Dazzle, Annette Wickham, Royal Academy of Arts
• World Wars camouflage technique could have benefits in modern warfare, University of Bristol
• Dazzle Camouflage Affects Speed Perception, Scott-Samuel et al., PLoS ONE 6(6): e20233