John Rawls and Ethelo

Ethelo draws its philosophical roots from the work of John Rawls, a 20th-century philosophy and social contract theorist. 

Rawls wrote about the importance of fairness in social decision-making. He proposed a framework for analyzing democracy and justifying its existence as the basis of our society. He argued that democratic principles, even in the purest ideal of democracy, would not prevent inequality nor should they. However, he proposed that a properly constructed democracy would put a limit on the extent of that inequality. 

Rawls was a social contract theorist, and followed a tradition of Jean Jacques Rousseau and others. The core idea of social contract theory is that the laws and rules that society imposes on individuals can only be justified if we can assume everyone in society agreed to those rules - or, rather, they would agree, if they were asked. Such an assumption is philosophically handy. However, whether it's useful to imagine such a fiction, or whether people would agree to the current society is obviously very much in question. In time the idea lost out to ideas such as utilitarianism of John Stuart Mill, who proposed a more pragmatic approach of “the greatest happiness for the greatest number.”

Rawls’ “great idea” caused social contract theory to be pulled out of the dustbin of history, brushed off, and put on a pedestal once again. He proposed a unique new approach which created a solid ground for assuming collective agreement around a theoretical social contract - and a way to strike the balance between personal satisfaction and inequality. This was fairness, or in his words justice. His two most famous books - A Theory of Justice and Justice as Fairness, explained and expanded the idea. His approach, and centering of fairness as the key component of a valid social contract reinvigorated social contract theory and cemented Rawls’ reputation as one of the greatest philosophers of the 20th century. How did he do it?

Rawls proposed a now famous test called the  original position to justify the social contract as a frame of reference that can allow us to determine where the line between personal satisfaction and inequality should be drawn.

The Veil of Ignorance

Rawls proposed that we imagine that everyone in society is situated as “impartial equals” behind a "veil of ignorance" as they negotiate the social contract. More concretely, we should imagine all members of a society are seated at a vast conference table, as they negotiate the terms of the social contract - the laws, regulations, policies and even customs of the society. But there is a catch; no one at the table knows who they will be born as when the negotiation is done. The “veil of ignorance” prevents them from knowing what life they will live in the new society, and this enforces a risk-management approach to the negotiations. 

Where does the logic of this veil of ignorance lead us? As a matter of first principle, he argued, everyone will accept some degree of inequality if it means our chance of doing better personally (even in a random draw) improves as a result. There is a benefit to all of us, he argued, when highly productive people have more resources to play with; a rising tide raises all boats. However, he said that while some level of inequality may be justified, there is a limit. A rational actor will only accept so much risk, if they don’t know who they will be born as. For example, a society that has 21% children living in poverty while an elite 1% own more than half the wealth would not survive Rawls’ analysis. 

Rawls’ “veil of ignorance” test caused great debate. Certainly he did not have the last word; his “original position” has been criticized for example, for ignoring the role of society in shaping identity, preferences and relationships. There are logical inconsistencies which have been pointed out. However, his strategy for tackling the issue of fairness in social contracts evoked a powerful, intuitive resonance and became the center of debate which continues to this day. His ideas had an enormous impact on the field of social contract theory and political philosophy generally.

Rawl’s philosophical intuition on the importance of fairness has found evidentiary support in other disciplines. It has been validated empirically in the social sciences, under the general heading of “social preferences” and “inequity aversion.” Studies in game theory and behaviour economics show that people will reject unfair outcomes even when they would otherwise benefit. Likewise, they will support outcomes they dislike if the process was seen as fair. This powerful phenomenon is not limited to humans either - it has been documented in animal experiments involving monkeys, birds and other social animals. 

Fairness, it seems, is a fundamental social instinct, and although we can come up with words to describe it, it precedes language itself. Certainly it is deeply embedded in our expectations of the rules of society.

Ethelo and Rawls

Ethelo is a group decision methodology based on the idea of fairness, and draws directly from Rawl’s theory and his “veil of ignorance” approach.

Imagine the negotiating table, at which the social contract - or indeed, any group decision - is being negotiated. On the table, is a stack of all the potential social contracts. There may be a very large number - we would include every variation and permutation of possible agreements. As we go through the stack of contracts together, everyone turns a colour which indicates their satisfaction with the contract in front of them. We cannot see their face - but we can see the colours of everyone in the room turns as they go through the stack of contracts.

We can turn the idea of the veil of ignorance on its head, and solve the problem in reverse by imagining a special twist. Instead of an imaginary negotiation where we are sitting at the table, not knowing what body we will be born into, instead, imagine a room of anonymous people changing colour as they work through the stack of contracts. That is, each possible social contract is represented as a distribution of profiles across the colour spectrum. Our search is not for the ideal social contract - but for the optimal distribution of colour. We are looking for an ideal distribution of colour.

This twist is an equivalent situation to Rawl’s problem, but one that is easier to solve. By imagining different contracts as different colour distributions, we can more easily compare them - and use mathematical and computational tools for that comparison.

For example, imagine two social contracts with exactly the same AVERAGE satisfaction; 

Contract One

Contract Two

Why? Two reasons;

1. Personal risk: The odds of being quite unhappy are higher in the first contract. The first contract has a higher number of very happy people, to be true - but the chance of being one of those very happy people is not high. 
2. Collective effectiveness; The odds of being able to execute on the decision effectively, and achieve a good outcome is much lower in the first outcome than in the second. This is key. In the first outcome, there will be division and polarization. Those who did oppose the outcome will resist it - and they won’t go away. This will act as a drag on the energy of the group. On the other hand, the second contract will have much greater unity - a sense that everyone is roughly in agreement, and that the outcome is fair. This might also be framed in terms of risk; the risk of failure or of problems not recognized by the majority, is higher in the first.

Scenario Analysis

In a normal voting scenario, there are only a limited number of discrete potential outcomes put in front of a group. Imagine Brexit for example, or a US election. Voters have little influence over the choices they are faced with; their only role is to select a favourite. This naturally leads to highly differentiated options, and in the results, the voting group splits. No matter what the outcome, there will be winners and losers. In such a situation, majority vote will govern and minorities will lose. However democracy doesn’t mean majority vote. It means one person, one vote - that people should have equal influence. And, implicitly - equal satisfaction with the outcome - or as equal as possible.

Majority vote is simply an aggregation methodology - and there are other approaches to the aggregation of participant preferences that can lead to better outcomes. When group size is small, many organizations prefer consensus achieved through discussion. However, such an approach does not scale easily. The key breakthrough of Ethelo, which is scalable, is to lift the hood on the generation of choices, and the logical structure of the decision itself. 

At the heart of the Ethelo approach is the idea of scenario analysis. Instead of just a few potential outcomes, Ethelo conceives of a group decision as consisting of a set of decision parameters. That is, a decision can be broken down into a set of smaller decisions revolving around key ingredients such as options, issues, criteria and constraints which can be recombined in a multitude of ways to simulate all possible outcomes. In this approach, rather than just one or two potential outcomes, there can be a vast number - all slight variations of each other, covering the whole spectrum of possibility.

The Ethelo platform is designed to enable custom workflows for different decision types. It enables participants to collaborate on solving all these smaller problems, discussing them as they go. Each participant comes up with their own favourite solution - but in doing so, provides the Ethelo platform with enough information to model their preferences across the whole range of possible scenarios. Then, it searches through all those scenarios to find outcomes that will have a fair distribution of support, that that maximizes support while minimizing polarization.

The Importance of Fairness

The observant reader may have noticed; we have left one question unanswered. That is, what is the “ideal” distribution of colour across the group, or in other words, satisfaction? How can we know in advance which possible outcome would strike the right balance of fairness and personal satisfaction for participants? Some people may be more risk tolerant than others for example - more willing to accept outcomes with sharp divisions between winners and losers. Indeed, this was a criticism of Rawls. But this challenge can also be solved. We can simply ask people directly.

When Ethelo undertakes its scenario analysis it looks at each potential outcome as having two key characteristics; average satisfaction, which it seeks to maximize, and a second characteristic which it seeks to minimize. That second characteristic is inequality, or in mathematical terms variance. Variance describes how polarized a distribution is. For example, if 50% of voters completely oppose an outcome, and 50% completely support it, then variance is maximized. However, if everyone is left feeling exactly equally happy, variance is minimized.

In the decision language of Ethelo, the relative weighting between average satisfaction and variance in satisfaction is called fairness. As Rawls would have observed - a typical participant will accept some variance if it increases the average level of satisfaction (and their chance of personal satisfaction). And just as participants can use Ethelo’s custom tools to vote on options and weigh the importance of issues, they can also use sliders or other graphical interfaces to indicate how important fairness is to them. 

It is therefore possible to treat fairness itself as one of the decision parameters that participants give feedback on, alongside the options, issues etc. We can find the ideal fairness for a group for a given decision. This also allows an enormous amount of flexibility. Rather than assuming that all groups will warrant the same amount of fairness in all decisions - fairness levels can be customized on a case-by-case basis.

Rawls provided a test for the validity of society and the rules by which it operates. That test showed how far we had to go - and also a framework for how to get there. The answer can be found in democracy - but a better designed democracy than we have now.

Utilitarianism

Utilitarianism is perhaps the most successful theory of modern political philosophy. Its great benefit is its simplicity; “the greatest good for the greatest number.” However, this simple maxim hides some mathematical ambiguities. Should we be aiming for the maximum TOTAL amount of good, which means maximizing both the average good and the number of people? Or should we maximize the average good - in which case a small number of very happy people would be better than a large but slightly unhappier group? Or should we aim to improve the lot of the least well-off people - maximizing the minimal utility?

The beauty of utilitarianism is how nicely it conforms to rules of mathematics, and of measurement. It can be easily applied by large, modern bureaucracies to measures of income, health or productivity. However, it subtly fails to capture the idea of fairness. The author Ursula LeGuis captured this in her short story,  "The Ones Who Walk Away from Omelas, where a very happy village had a dark secret; all the unhappiness was carried by a sick, imprisoned child. The village of Omelas had maximized its utility; but at a great moral cost. Totalitarianism, which punishes dissent to enforce a maximum happiness, highlights a similar utilitarian dilemma.

Rawls' Theory of Fairness successfully responded to these challenges facing utilitarianism. However, it nonetheless lacked the easy mathematical applicability of utilitarianism. But that is because Rawls was not a mathematician. By using a two-part analysis of utility - one that measures both average utility AND variance in utility - in combination with scenario analysis, we can bring the rigour of utilitarianism to social contract theory and systematize fairness.

Ethelo and the mathematical framework underlying it is in fact an ethical framework. It does not look merely at outcomes, like utilitarianism. It does not look at an internal application of a universal moral imperative like Kant’s deontological framework. Rather, it understands the good as the fair expression of collective intention, that is, an expression that minimizes dissonance and maximizes happiness. It is a kind of utilitarianism, but it doesn’t look directly at what is “happiness”. Rather, it focuses on providing a fair process for people to analyze decisions collectively and select an optimal outcome. So it is a form of procedural ethics, the rules of a deliberative negotiation around a social contract.

• Variance in satisfaction: 

. What would make an ideal distribution of colour? Well, this is a matter of taste. Different people have different tolerance for risk. However, we can agree on certain rules for example; 

• The value of a distribution is not merely about how many people support an outcome, or even the average level of support. We also care about the distribution of that support.