This is Signal vs. Noise, a weblog by 37signals about design, business, experience, simplicity, the web, culture, and more. Established 1999 in Chicago. Follow us on Twitter for more information on our products.

Ah, yes, the veil of ignorance.

“The most precisely halved cookie is the one cut by me but with the first piece chosen by you.” ~Me

Thanks for the wiki link. The quote itself made no sense to me in isolation.

That is almost exactly how I explain my concept of fairness and ethics to people.

It’s dead simple to understand. If you had to agree to something, not knowing which side you would be on, would you do it. If so, it’s probably fair to both parties, if not, you should go back to the drawing board. This approach works well for situations ranging from finding a lost wallet, to cutting a cake, to negotiating a complex contract.

Great for simple obvious situations like cake sharing. Impractical to implement for more complex things, like politics. This runs into problems when people have imperfect information or are not able to think through consequences fully. Basically it depends on the assumptions that we always make fully rational/consistent choices and have perfect information. Both are suspect.

Or more importantly this will buckle if those who don’t currently have power hope to gain power. Thus they might create a rule in favor of the elite in hopes of becoming the elite, or vice-versa which is how we get welfare and unemployment benefits.

Great quote.

@Ian – what sort of rule of thumb for fairness would you propose then?

The veil of ignorance is often discredited for practical application – it’s not a feasible human trait to imagine a different set of personal experiences then your own. How you imagine being powerful or powerless depends a lot on how you have experienced power yourself.

@Nathan I’m using that with my two-year old. Nice!

@Ian I think you are misunderstand the concept. The veil of ignorance is remarkable among justice/fairness theories because it not only rejects complete information and consistent rationality but also assumes their opposites. The premise “if they did not know how much power they would have” is a global statement of complete ignorance. A rule constructed under this premise would easily hold up to scrutiny under lesser forms of imperfect information. Likewise, it is exactly from assumption that the powerful act irrationally and inconsistently that the demand arises “everyone agree”. Limits decided under the veil protect the less-powerful from the irrational action of the powerful.

@Rick Don’t you find the practicality criticisms fall into the “doth protest too much” category? All ideals suffer the “hard to imagine” characteristic. Or to put it another way, it’s impossible to imagine a fair world, period. Rawl’s theory is uniquely threatening to privilege because it provides a way to show when a rule is not fair, when proponents of the rule will not agree to strip their privilege in exchange for keeping the rule.

@ David

I got nothing, but then I’m not convinced it’s possible to solve fairness with a rule of thumb or philosophy. To quote Rousseau “It would take gods to make (fair) laws for men.” I guess to go further I’d have to ask in return, what is fairness?

@ Patrick

I understand that in the veil we lose our history with our current roles and don’t know what roles we will have. What I’m arguing is that we also aren’t in a position to assess the potential roles and outcomes accurately and so it likely won’t be fair in the end.

Take a modified cake splitting example. Let’s say, unknown to both kids, that half the cake is filled with nuts and that one happens to be allergic to these. They agree that one cuts and the other choses first, seems fair. However, after the cut and selection the child with the allergy winds up with all the nuts in her half. Now as it stands one child gets half a cake and the other gets none.

This system seemed fair in the veil of ignorance but came out really unfair due to imperfect information. Tell me as a parent you wouldn’t jump in there and make an after the fact change in the interest of fairness?

So for the same reason that it can’t be shown to create fair rules and I’m not sure that it can prove a system is unfair. Only that we think it’s not.