Choice Architecture and the Bradley Manning Verdict

How behavioral economics is used to shift perception away from Manning’s espionage convictions

Bradley Manning’s verdict was handed down today. I did a cursory search for “Bradley Manning news” and out of the 13 headlines on the first page of results 11 start with “Bradley Manning acquitted of Aiding the Enemy….” The fact that he was convicted on five espionage charges is not mentioned in the headline or is cut off.

At first glance, this verdict would appear to be great news. Manning’s been acquitted of the most serious charge! As various experts have pointed out, a conviction on aiding the enemy would have set a dangerous legal precedent. Harvard Law Professor Yochai Benkler testified that that argument “essentially means that any leak to a media organization that can be read by any enemy anywhere in the world becomes automatically aiding the enemy.” Manning has been acquitted of this most egregious charge, and does not face the lifetime jail sentence that goes with it. That is good news indeed.

On closer inspection however, the intense focus on the acquittal of that one charge shifts focus away from the other serious charges that Manning has been convicted of, five counts of espionage under the Espionage Act, and five counts of theft. This is in addition to the 10 lesser charges he has already pled guilty to. Altogether, Manning is facing up to 136 years in jail. Both the ACLU and Amnesty International stated that this verdict’s purpose is for the US government to intimidate any potential whistleblower into staying silent.

In behavioral economics there’s a concept called “Choice Architecture,” which describes how people change their behavior based on what options are given to them, or in other words, how the available choices can be “architected” to influence decision. An example of this phenomenon can be seen when the Economist changed their subscription choices. At first, the Economist offered two options:

  • Web Subscription: $59
  • Print and Web Subscription: $125

Under this scheme, most people chose the lower priced web-only for $59. The people at the Economist are clever, though, and they decided to add a third option in the middle.

  • Print Subscription: $125

You would be absolutely correct in thinking no one would choose print-only for the same price as web and print! But that wasn’t the point. By adding this seemingly nonsensical third choice, the Economist shifted how buyers perceived the other choices. Under this new scheme, most people now chose to pay more than double for print-and-web, because they were now comparing against the third choice, which made print-and-web looks like a deal compared against print-only for the same price.

The same pattern can be observed in the Manning case. The prosecution pushed for the aiding the enemy” charge, which Manning supporters rightfully denounced, and Amnesty International called “ludicrous.” Even people who want to see Manning in jail for a long time conceded that charge was a bit much. By focusing on the one charge that’s way out there, though, prosecutors modulated people’s perception about the reasonableness of the other charges.

If Judge Lind did not have the aiding the enemy charge to strike down, would her verdict of finding Manning guilty on 20 of the remaining 21 charges seem more extreme? Would her own perception of what would be a just sentence be shifted?

Without the aiding the enemy charge, might the general response have been different? Perhaps it’d have been more along the lines of:

Holy $#@%, Manning has been convicted under the Espionage Act, which is the first time a leaker of official secrets has been convicted under a law that was meant for actual spies. This sets a dangerous precedent and will squelch any leakers from coming forward about government wrongdoing in the future.

Rather than the collective sigh that occurred:

Whew, Manning has been acquitted of aiding the enemy.

I expect the prosecution had this effect in mind when they charged Manning with aiding the enemy. And it worked.