Jeffrey LewisAdditional Thoughts on Burma

Catherine Dill and I have published a pair of articles concerning allegations that Burma’s generals are building a chemical weapons facility near a place called Pauk.  The article published on the CNS website analyzes satellite images of the site, while the other in Foreign Policy explains why we ought to care.

The short version is that the facility at Pauk is one of a number of sites being built by the Directorate of Defense Industries (DDI), which is believed to be responsible for the arms trade with North Korea.  The Obama Administration has largely neglected nonproliferation concerns, understandably calculating that democratization in Burma would take care of the relationship with North Korea, as well as any illicit weapons programs.  That argument is lot harder to make now, especially given the Burmese government’s detention of the journalists who original reported the story.  Burma’s North Korea and nonproliferation problems show that some Burmese generals remain a law unto themselves, which is really a democratization problem.

In the coming weeks, we’re going to write more about Burma — revisiting some known DDI facilities, revealing some new ones.   The take-away, though, is simple:  Since the Obama Administration began to engage Burma to encourage a transition to democratic , the Directorate of Defense Industries has expanded.    If the Obama Administration wants to sustain its engagement with Burma, it has to make nonproliferation concerns a real priority.

That starts with insisting that Myanmar ratify the Chemical Weapons Convention (instead of making repeated promises to do so).

I always enjoy the comments here at ACW, so I wanted to share the information and see what folks had to say.


  1. P (History)

    The problem with IMINT is often that we can see big buildings and other constructions but have no idea what is inside. Could be crammed with chemical weapons production stuff, could be just empty. Or something in between. I am really curious what Jeffrey and the wonks will dig up on this issue.
    What does seem relevant to me is that Myanmar has been investing quite heavily in new conventional arms in the past ten years. Some of the results could be seen in last month’s big annual parade where new upgraded tanks, rocket launchers, helicopters and other stuff where shown off:

    Claims are floating around that a lot of this new weaponry is assembled, upgraded or even produced in Myanmar, suggesting that the DDI has been busy setting up arms factories. If pictures on the internet are genuine it seems that the DDI likes spacious but rather empty halls, which do look to me not entirely consistent with a “standard ordinance factory” but rather oversized
    Like entry 99 on this page showing Myanmar K-8 light jet aircraft being ‘assembled’ in a large if not huge unfinished hall which suggest major and probably unrealistic future production plans. Some more example here:

    That the Myanmar production sites involve foreigners is without doubt. China, Russia, Ukraine and Serbia are known to have supplied the weapons and technology, including for the local input into the final products. And that some of the foreigners may look like Koreans could be related to the fact that they are Koreans, just not sure if they are from North or South…: (South Korean industry illegally supplying production line for artillery shells 2006) (photo suggesting Myanmar has/ currently builds armoured vehicles with help from South Korea)

    • Jeffrey (History)

      You make a totally fair point. (Although I think I can exclude aircraft production at this site, unless Myanmar plans to truck the airframes out over poor roads.)

      Two thoughts.

      First, we have an instance where locals have made a specific allegation, much of their account checks out, and the response of the government is over-the-top in arresting the journalists and attempting to seize all copies of the paper. What we are asking is that Myanmar make good on its many promises to ratify the CWC and declare the purpose of the facility.

      Second, we have looked at many of the DDI facilities — we are trying to raise funds to analyze them all. Although some include large warehouses such as the ones the images you include (and I am tempted to try some inside-out matching for some facilities) these things are really, really big. The tank factory is close in size, though:

    • P (History)

      The aircraft hall is so big that the aircraft could almost take off from inside it… OK,that is too much James Bond style. But it possibly gives an impression of Myanmarese thinking of how big a production hall should be.

      Here is some more info about Myanmarese arms production which could may be help to understand if the big buildings could be related to conventional weapons. No info on missile production other than Igla manpads (which by the way N. Korea also produces based on a Russian licence)

    • P (History)

      For the record, this P with a bunny face gravatar and myself are two different people.

  2. Allen Thomson (History)

    Well, I’m delighted that you’re back to doing this stuff. If you’re in a fund raising mode, I’d be happy to click a PayPal button if there’s one available. Speaking of money, could you say how much it cost to buy the Astrium image? The price of satellite imagery isn’t always easy to discover.

    A comment on the picture on the cover of Unity: It appears to have been taken from an aircraft, and I suspect that would have been an official aircraft. Which raises the question of how it got out of official channels and onto the cover of Unity. Another defector?

    • Jeffrey (History)

      That’s kind — we were thinking about trying to crowdfund this work. Do you think that might be viable?

      The images are a few hundred dollars a piece. (It’s by square meter with a minimum and pricing plan that varies depending on the use, basically where we publish.) Pauk is very spread out, so it required two images that cost about $400 each.

      The challenge is that we’d like to buy images of like 30 facilities and spend the time to analyze them. Of course, what is really expensive is time to analyze.

      So, perhaps a crowdfunding approach might work. I’d be interested in your opinion, either here or in my inbox.

    • Allen Thomson (History)

      > That’s kind — we were thinking about trying to crowdfund this work. Do you think that might be viable?

      The only way find out is to try.

      My impression, based on some experience, is that not many people care about about this stuff. Just how many people/organizations would be be willing to come out of the woodwork and put money on it is anybody’s guess.

      > The images are a few hundred dollars a piece.

      I, an old retired guy with interest in the matter, would go for an image a year. Like $30 a month. Maybe $100 up front to get things kicked off.

  3. Pete B (History)

    You can see the construction pretty clear in Bing maps.
    You see all the tunnel under construction, and you can se its a pre-fabricated steel and then covered with concrete.
    Then you also can see the points for grounding the supporting beams, and that gives me impression that the buildings are to small for aircraft or tank production.
    But if it was for heavy equipment and manufacturing,there should be much more concreted support plates?

    And if it was for munitions, where are all the storage buildings for explosives, etc?
    The long road to West and North dosnt make any sense?

  4. Cthippo (History)

    One thought that comes to mind, having worked in manufacturing, is that there is at least a 5-1 ratio of warehousing to manufacturing space. What that means is that for every square meter of production line there will be at least 5 meters of warehouse. The 5-1 ratio is for very progressive facilities that take just-in-time and lean manufacturing seriously. In normal manufacturing it can be as high as 20-1.

    I’m also looking at the fences, and the lack of a beaten zone between them and the treeline.

    Beyond that building looks remarkably…generic. For a chemical weapons facility I would expect more specialized liquid handling equipment and piping, as well as air handlers. This facility has a built on spec feel like they put up a big building and expect to find a tenant for it later.

    • Jeffrey (History)

      That’s really helpful as a measure. I am not sure what to make of it at the moment, other than to say I appreciate it and am thinking about it.

  5. P (History)

    Well, I’ve tried to understand this facility using the satellite photograph… Long story short, I’ve not succeeded; however, I’ve found some hints that could be interesting. The few I could make sense of, curiously, did not pointed towards a chemical weapons facility.

    First, a few observations that led me nowhere, but could perhaps be useful to somebody else :

    – The Google Maps image captured a single vehicle on the facility, close to the north-east corner of the northern buildings. I couldn’t identify it;

    – Although the north and east buildings of the north site could be connected, there’s clearly no connection with the western one;

    – the two buildings to the south are actually connected with what looks like a pipe or so;

    – the northern building in the northern site seems very weird. I mean, clearly, the terrain is not level, but what’s weird is that its roof isn’t, either! It looks like the builders only corrected for the terrain’s inequalities on its southern side, which is straight, but not on its southern one, which is curved. In fact, using the little windows and the central line as references, one notices that the building’s curvature gently increases from the south to the north wall. Shoddy construction?

    – There’s a temple (or perhaps the director’s house?) just to the south-west of the south buildings.

    Now, for somewhat more interesting stuff :

    – To the north-east of the south site, the river has just been dammed. The dam is recent : some bushes are still emerging from the water. However, this isn’t for electricity production : firstly, this river is very small, secondly, the electrical substation is located to the north-east of the north site. I guess we can safely exclude recreational, fishing, and agricultural motivations for this construction. Thus, I see only two possible explanations :

    a) flood prevention for the north site (this river flows around it);
    b) a need for guaranteed clear water at one of the sites.

    Hypothesis a) sounds weird, because if this were the case, there would be no need to fill the dam right now, when the water level is low. Unfortunately, I don’t see what we can make of b). Perhaps they just needed water for the construction?

    – The relief is quite pronounced. This looks like the true motivation for the separation of both sites : there just wasn’t enough flat space to put both groups of buildings together. However, the tunnel itself is quite unexpected. It looks like it was built to avoid having to
    build a fence along the road. That’s interesting, because
    it was likely quite expensive to build, yet it was still preferred to a long fence.

    – The road that comes from the west and leads to the north site is perplexing. It was clearly built for this site, as evidenced by the fact that it doesn’t appears in the older photographs. Several bridges were built for it, so this doesn’t looks like a small temporary road. But there’s no gate on it anywhere, not even a hint of a construction site for one. And it clearly leads to the north site, not the south one, although the southern buildings look more like regular warehouses than the northern ones.

    And, now, for the more interesting stuff :

    – The security measures are completely off. For starters, the gate is located *between* the housing and the facility. At Minbu, the gate separates the housing, the facility and the helipads from the rest of the world, whereas here, even the helipads are located on the outside. Huuuuuuuh??

    – Continuing on the security measures, as mentionned by Cthippo, there’s no protection against coming very close to the facility. In fact, we can easily find approximately where the photographs were taken from : the forest just next to the facility.

    – While we’re at it, if it were a chemical weapons facility, this site would get a clear-cut ‘F’ in discretion. There’s not even the beginning of an attempt to conceal its existence.

    – Finally, strategically, the position of this construction is perplexing : It’s located within 200 km from the border with India, 200km from the border with Bangladesh, and 250km from the sea. In other words, close enough to be vulnerable to an air strike from India-Bangladesh-a sea power, but far enough that neither India nor Bangladesh would fear a chemical contamination on their soil if it were destroyed, and far enough from the sea to be unable to benefit from its ease of transportation. Huh??? If Burma wanted to create a chemical weapons factory, they have enough territory to build it in an easier to protect place.

    Now, let me try to risk an explanation. I’m not convinced that this is the truth, only that this is a possibility that should be considered.

    What if this isn’t a military facility at all?

    I mean, the Burmese regime is a highly corrupt one. And, in such states, it is common for the army to control purely economic industries, as a way for its leaders to make more money.

    Also, all the security measures seem designed to protect the facility from its own workers : the gate to search them when they leave the place, and the fences and the tunnel to prevent them from diverting anything from it. On the other hand, nothing seems designed to prevent either communication between the Burmese and foreign workers, or communication between the workers and the locals (even the river that separates them seems unwatched), or even observation of the site.

    So, here’s my hypothesis : what if this is only a high-value economic installation with army participation? Say, a facility intended to separate and cut precious stones, or handle precious woods like teck, or something of the like. This seems to fit with everything on the site : security measures so that none of the precious materials can be diverted, and no security against military attacks or identification, as these are not considered real threats. And Cthippo’s 5-to-1 ratio likely doesn’t applies to things like precious stones, or precious wood, or anything where the value of the production is high and the volume isn’t.

    Now, as I mentionned, this is only an hypothesis, and there are clearly not enough information on this facility to give a definitive opinion on it. But the fact that there seems to be nothing against it should at least lead us to give it some consideration.

    • Jonah Speaks (History)

      Under this alternative hypothesis, Burma’s arrest of journalists is intended to cover up a scandal of army profiteering from precious stones (or whatever). If so, why did the whistle-blowers call it a chemical weapons factory? Why does Myanmar not respond by offering a tour adequate to establish, “See, no chemical weapons here.”

    • P (History)

      > why did the whistle-blowers call it a chemical weapons factory?

      That’s unclear. But remember Jeffrey’s podcast, where he referred to those (undeniably persecuted) Hmong people who claimed they hade been targetted by chemical weapons, and it turned out they likely hadn’t. Since the international community is far more interested into chemical weapons production than corruption and illegal eviction, this can be a way to obtain more attention.

      > Why does Myanmar not respond by offering a tour adequate to establish, “See, no chemical weapons here.”

      Hah! Do you take us for beginners? And why not publish the precise nature of the product that’s handled there, the identity of our foreign buyers, and the number of the bank account of the general in charge, to ease the task of your western anti-corruption sanctions, while we’re at it?

  6. inspectorman (History)

    I have to say that Cthippo and P are probably on the right track. CW production needs a lot more energy than the small electrical substation in the northern zone can provide, and there’s no evidence of any liquid / gas handling; at the very least you’d expect worthwhile quantities of exterior tankage, piping or even bunding. The southern area tankage is not visible in what I can see. As far as the “ventilation” in the article goes, that doesn’t really qualify as ventilation. Just some kind of slatted roof. We need big fans and lots of them. And there’s no storage / movement of a lot of munitions around this site. Whatever is being processed is not making much of an impact on the roads around and within the site, from the Astrium images.

    PS Ordnance not ordinance.

    • Jeffrey (History)

      A few comments.

      First, I am not sure the facility was complete in the January 2014 image — the road to the west was still under construction and some other features seem incomplete.

      Second, there substation at Paul is not obviously smaller to me than the one at the known Rabta chemical weapons facility near Rabta (32°13’8.84″N, 12°53’54.45″E) although I will admit to having no expertise in such a specific area.

      Third, there are large exhaust stacks in the northern area. I do wish we had a larger library of images of known chemical weapons production sites. But my sense is that we’re not going to be able to make a determination based on overhead images alone.

    • P (History)

      > there substation at Paul is not obviously smaller to me than the one at the known Rabta chemical weapons facility near Rabta

      Let’s measure them. The Pauk substation is an approximate square of 60*60m, to which we can add an empty rectangle, for future expansion I guess, of 15*60m. Total area : 4500 square meters. (Since the closest building is separated from the substation by a fence, I guess we can assume its 500 sq. m. aren’t part of it.)

      The Rabta one is an approximate rectangle of 187*73m, to which we have to add a little near-square of 33*30m. Total area : 14641 square meters. That’s more than three times bigger. Also, with its high-power line and its complex elements, the Rabta substation seems me qualitatively different from the Pauk one.

      (Btw, yes, I did take into account the fact that the scale of both photographs is slightly different.)

      > Third, there are large exhaust stacks in the northern area.

      Do you suggest the buildings could be generating their own power? That’s a possibility, but if this were the case, there would be no need for any substation. Also, why not group all the generators together in their own building, as usual?

      > But my sense is that we’re not going to be able to make a determination based on overhead images alone.

      Yes, that’s also what I fear.

    • Jeffrey (History)

      Obviously, I measured them. The question is what to measure. (I also get slightly different measurements, but that’s less important.)

      The Pauk substation is not complete, which may explain the absence of high power lines (although I think I see the towers) and complex elements. We can guess at its final footprint, but only guess. Moreover, the layouts are quite different.

      Making a comparison requires more expertise in electrical substation design than I have.

      Finally, we do not know that Rabta is the minimum size for a chemical factory.

    • Jeffrey (History)

      So, we need another comparison. Apparently the substation at Rabta serves not only the chemical facility, but also serves as a distribution point for two lines running to other facilities.