The CIA – in a declassified document -has stated that the work de Mesquita has done for them enjoys a success rate of 90%.

His work includes analysis and predictions on: achieving an Arab-Israeli peace, containing the nuclear ambitions of Iran and North Korea, anticipating and preventing corporate fraud and responding to international terrorism and climate change.

RT caught up with de Mesquita before his talk to promote his new book: "The Predictioneer's Game" at Trinity College Dublin, Ireland.

Game theory

RT: Your computer model is based upon the study of Game Theory; can you explain it and its relevance to your work?

BBdM: Game Theory is mathematics for dealing with how people interact strategically with each other, i.e. what they do when they are taking into account what other people will do in response to their actions.

Game Theory starts from the premise that people, like bacteria for example, are self-interested and so they always look for the course of action that is likely to give them the best result from their point of view. It assumes that everybody has values, things they want and beliefs. As people are self-interested, we can estimate their values and beliefs and their probabilities of being successful or unsuccessful in their actions, we can predict their behaviour. This is the foundation on which I can create a mathematical computer model to predict outcomes.

RT: Isn’t that a very cold view of the world?

BBdM: Yes, absolutely. I'm not encouraging people to be self- interested, I'm just trying to explain how the world appears to work.

RT: How did you start with your predictions?

BBdM: Back in the late 1970's I was doing research on Indian politics while also developing a model about diplomacy in war. I was asked by the US Department of State who I thought would be the next Prime Minister of India after the Indira Gandhi government fell. I had this war model and it occurred to me that the programs of negotiating governments and the problem of negotiating in diplomacy were exactly the same. I fed some data about India into the war model. I had my own personal opinion as to who would be the next PM of India, but the program came up with a completely different name. The computer turned out to be right. I thought that I was either lucky or accurate. So, I began to develop it more seriously and now here I am.

90% success rate

RT: The CIA gave you a success rate of 90% with your predictions. Can you give an idea of the high profile predictions you got right, and also where did you go wrong?

BBdM: Among the high profile predictions – that are at least in the public domain and I'm allowed to talk about are: I developed a strategy that helped to get Ferdinand Marcos to leave the Philippines to clear the way for Cory Aquino to be elected.

Also, there is one major organisation in the world that has both China and Taiwan in it – The Asian Development Bank. When China applied for membership, they demanded that Taiwan be kicked out. I designed the strategy that would get China to join and keep Taiwan in, and, 25 years later, that’s still the case.

On the failure side, I did a study in 1993 on health care reform in the US. I got everything wrong, but I learned something important in that study. What happened was a major player in that game (a Congressman) had dropped out because he was going to prison. I could not have anticipated that, but I have since amended the model to factor in such events.

On Iran

RT: What about more recently, and your involvement with predictions about Iran?

BBdM: I did a study on Iran's nuclear program that played a part in helping to persuade senior government officials that Iran is probably not trying to build a nuclear bomb. This contributed to the 2007 New National Intelligence estimate that came at a time when many believed the Bush administration was preparing to bomb Iran. My work may have contributed to preventing that from happening, which I think was a pretty big deal.

RT: You are adamant that Iran will not build a nuclear bomb?

BBdM: I've been reading the papers today and I've seen that Iran has negotiated a nuclear deal. This fits in completely with my predictions. I say in my book that the motivators for Iran to have a nuclear program are as follows: to deter the US and Israel, to gain pride of place in competition with Al Qaeda for exporting fundamentalist Islam, domestic pride and to regain a leadership position in the Middle East. Having demonstrated the ability to enrich uranium at a level that produces weapons grade fuel, Iran sends a message on all of those dimensions. They know how to build a nuclear weapon; however, building it will bring upon them a lot of grief, like economic sanctions. Developing enough weapons grade fuel to show that they can make a nuclear weapon without actually doing it achieves all of their goals without all of the risks of foreign military intervention and domestic political pressure.

On climate talks in Copenhagen

RT: What about your predictions on climate change, you say that the upcoming talks in Copenhagen will fail?

BBdM: The difficulty with universal treaties – like the one proposed for Copenhagen – is that you get everyone to sign-on, either by not asking them to change behaviour or by asking them to change but having no punishment if they violate their agreement. Universal treaties almost never do anything useful; they are just feel-good exercises. They deflect attention from the failure at home to take tough political decisions that will actually make the difference.

Take Kyoto for example, 175 countries signed up but 137 of them signed up to do nothing! All they have to do is report what they are doing, i.e. nothing at all. Of the 38 who had to reduce green house emissions, almost all of them lined up after Kyoto and said ‘we can’t meet the standards’. Even host nation Japan said they could not honour the agreement. I predict the same will happen in Copenhagen. They may come out with a feel-good treaty that nearly everyone will sign, but those that pause to read the fine print will see there will be no punishments for non-compliance. There will be no teeth to it.

The US and EU have shown no willingness to reach into their own pockets to compensate poor, high polluting countries like India or China, to help them use cleaner technology. It’s not popular with voters to ask them to pay other people in other countries to change their behaviour.

RT: So what do you predict will happen after that?

BBdM: Although Climate Change is an important issue, I predict technology will provide the answer eventually. Not Copenhagen.

On peace in the Middle East

RT: Any predictions for the proposed peace deal in the Middle East?

BBdM: It’s more of a tentative prediction, but the early indicators in my analysis led me to be optimistic that there will be a meaningful agreement between Israel and Palestine by the end of Obama's first term. The isolation in negotiations with Hamas over the last couple of years has been effective in creating wedges between Hamas’ more moderate and more hard-line wings. The moderates see they need to put more on the table in order not to be left out. The peace agreement will require a change in Israel’s government, but not many governments last for three years there, so I'm fairly confident of that. However, the quality of the data I have on this is not as high as the other predictions.

RT: Finally, is there any danger in putting these predictions out there, you may actually change the results?

BBdM: There is no danger of that, the model assumes that people are doing what is in their best interest, so for them to thwart the predictions just for the purpose of thwarting them would be against their self interests.

Ciaran Walsh for RT