In 1940, the Nazi's communications encryption had been broken by the British. Military Intelligence were able to decrypt a signal which indicated that the city of Coventry was to be bombed. The military chiefs took this information to the Prime Minister, Winston Churchill. If he ordered the evacuation of the city, he would save hundreds of thousands of lives - but the enemy would know that their encryption was compromised.

What, gentle reader, would you do? Is it worth sacrificing lives now in the hope of saving more in the future? Is revealing your hidden knowledge a price worth paying if it means saving a city?

This utilitarianist problem was probably not actually placed in front of Winston Churchill - but it is an interesting thought experiment.

This brings us to the recent revelation that the security services of the world are tapping our communications, listening to our phone calls, and tracking our every move.

So why is there still crime?

If "Prism" (which I'll use as a shorthand for all digital surveillance) is so good and effective, why is there still any form of crime or terrorism?

There are, I think, several possibilities.

• It doesn't work - either it can't capture the data or it's impossible to sift.
• Criminals don't communicate via electronic means.
• The Government don't want Prism revealed.

To go down the conspiracy theory route, is it acceptable to allow a major criminal attack in order not to reveal how intelligence is captured?

Or, to look at it through a different lens, is Prism stopping 99% of crimes, and the ones which do occur are statistical outliers?

If Prism is as amazing as claimed - and assuming revelations don't push criminals to secure crypto and unmonitored communications - why can't it be used to ticket speeding cars, foil bank robberies, or stop traders manipulating LIBOR?

Are minor and major criminal acts the price that governments are prepared to pay in order to keep Prism secret?

The leaders of several huge corporations have issued statements saying that their companies do not allow the US Government to illegally spy on their users.

I'm sure they believe that. I'd even go so far as to say that I'm sure the entire board and top management genuinely have no knowledge of any malfeasance.

Why would they? We're talking about spies - experts in the art of subterfuge and espionage. Why would a spy agency do anything as crude as ask permission?

Consider the Greek wiretapping scandal. Apparently, no one in the senior corporate structure at Vodafone Greece had authorised the tapping of hundreds of politicians' phones. Despite the high profile nature of the victims and the apparent suicide of Vodafone Greece's Network Planning Manager, there has never been a conclusive answer to how such interception took place and who - if anyone - authorised it.

Someone - an employee, a vendor, an intern, a janitor, a remote hacker - under the direction of someone else - terrorists, spies, rivals - flicked a switch which enabled wholesale surveillance of the Greek political establishment.

That doesn't require a CEO to know anything about it.

Let's examine another possibility.

Silicon Valley companies are well known for recruiting the best and the brightest from top flight American universities. The same students who would often be asked to work for the security services.

Is it beyond belief that such "patriotic" students have recruited into Google, Facebook, Yahoo, etc? And, from there, have found ways to provide direct access to the Government?

It doesn't matter what a CEO or PR department says. I'm not accusing them of lying - I'm saying that it's possible that they've been completely bypassed. Just as the French engaged in industrial espionage against American electronics companies, and the Chinese apparently bypassed Google's security, I don't think it is unreasonable that the American Military Industrial Complex would compel employees to engage in internal espionage.

Plausible deniability - it's not just for Presidents any more.