Reporting
Dragon Skin: The Body Armour Argument That Went Public in 2007
In 2007 a commercially developed body armour called Dragon Skin became the subject of a public argument between its manufacturer, Pinnacle Armor, the American broadcaster NBC, and the US Army. The design used overlapping discs of ceramic composite and not the flat rigid plates of the Army’s issue vest, and was promoted as offering better coverage and flexibility. The Army tested it, reported that it failed, and the dispute over how the testing had been conducted became a larger story than the armour itself.
The two designs
The Army’s issue system at the time, Interceptor, used a soft vest with rigid ceramic plates inserted front and back, plus side plates added during the Iraq war. Protection was concentrated over the torso and the plates were flat, which is straightforward to test and to certify.
Dragon Skin used small overlapping ceramic discs in a flexible carrier, in the manner of scale armour. The claimed advantages were greater flexibility, wider coverage without gaps at the edges of a plate, and the ability to absorb multiple hits across the surface, not degrading a single plate.
What the Army said
The test that settled the Army’s position ran from 16 to 19 May 2006 and was conducted for the service’s programme office for soldier equipment, Program Executive Office Soldier, at the HP White Laboratory, one of the two facilities certified by the National Institute of Justice to run body-armour tests. Naming the venue matters here, because the protocol dispute that followed was argued against that facility’s procedures rather than against the Army’s opinion.
The vests tested were Pinnacle Armor’s SOV 3000 Level IV Dragon Skin, fired against the threat baseline used for the Army’s own Enhanced Small Arms Protective Insert. Testing reported failures, including penetrations, and additionally raised a concern that the adhesive holding the disc array together degraded at high temperature, a serious matter for equipment intended to be used in desert conditions and stored in vehicles. The x-ray plates from the test, later briefed by the same programme office, showed the discs displaced from their intended positions after conditioning.
On that basis the Army prohibited the wear of the armour by soldiers and stated that it did not meet requirements.
What Pinnacle Armor and NBC said
Pinnacle Armor disputed the results and the methodology. Its objections were about test conditions: which samples were used, whether they were current production, how they had been conditioned before firing, and whether the shot placement and sequence reflected realistic use.
NBC ran a news segment presenting the armour favourably and questioning the Army’s handling of the matter, which moved the dispute from a procurement disagreement into a public controversy about whether soldiers were being denied better protection.
Why the argument was hard to resolve publicly
This is the part of the episode that has lasting value, and it has little to do with ceramics.
Ballistic testing results are only comparable when the test protocol is identical: the same threat projectile at the same velocity, the same backing material simulating tissue, the same environmental conditioning, the same number and spacing of shots, and the same criteria for what counts as a failure. Change any of those and two honest laboratories will produce different outcomes for the same armour.
The public argument therefore consisted largely of parties citing results obtained under different protocols, with the manufacturer’s tests and the Army’s tests not being directly comparable. That is not the same thing as either side being dishonest, and it is why the dispute could not be settled by adding more test results to the pile.
The secondary problem was that the detailed protocols and results were not fully public, which left outside observers unable to adjudicate and made the disagreement look like a matter of institutional credibility rather than of measurement.
What it left behind
The armour did not enter service. The episode did, however, become a reference point in two ongoing arguments.
The first is about procurement and outside innovation: whether large military acquisition systems are structurally slow to evaluate designs from outside their established supplier base, and whether the caution is prudence or inertia. Both readings are defensible from this case.
The second is about how technical disputes are reported. Coverage of the affair frequently presented the disagreement as being about whether the armour worked, when the substantive disagreement was about how it should be tested. Those are different questions, and the second is much harder to make into a broadcast segment.
For anyone reading about protective equipment now, the useful takeaway is procedural: a claim that armour stopped a given round means nothing without the protocol attached, and a comparison between two systems tested differently is not a comparison at all.
How body armour is actually specified
Understanding the dispute requires knowing that protective equipment is not specified as “strong” but against named threats. A protection level states which projectile, at which velocity, at which angle, the armour is required to defeat, and how much deformation of the backing material is acceptable when it does. Change the projectile or the velocity and the same plate passes or fails.
Two further variables decide whether laboratory performance survives contact with service. The first is conditioning: armour is exposed to heat, cold, humidity, salt and vibration before firing, because that is what happens to it in a vehicle in a desert. The second is multi-hit behaviour — whether the panel still protects after taking a round, and where.
Both were central to the 2007 argument, and both are the sort of detail that disappears from a summary. That is the durable reason the episode is worth reading: it is a clean case of a technical disagreement being reported as a moral one.
Questions and answers
Why did the dispute become so heated?
Because the disagreement was about the test protocol rather than about the product, and protocol arguments are unusually hard to resolve in public. Each side could point to results obtained under its own conditions. Without agreement on temperature, sequence, round type, backing material and what counts as a failure, two honest tests produce two incompatible answers.
What makes body-armour testing so protocol-dependent?
That the variables interact. Temperature affects material behaviour; the number and spacing of impacts affects whether a plate that stops the first round stops the fourth; the backing material determines what deformation is recorded; and the failure criterion decides whether a stop that transmits significant energy counts as a pass. Change any one and the result changes.
How are claims of this kind assessed now?
By whether the testing followed an agreed protocol with an independent party present, and by whether the full result set was published instead of the favourable subset. Neither condition is exotic and both are frequently absent, which is why a manufacturer's own test result (however genuine) is treated as a claim instead of as a finding.