The hardships of equality

This post is an attempt to clarify and generalize the main points of:
https://meandiscourse.wordpress.com/2012/11/13/risks_of_sex/

and explain why equality is an excruciatingly hard goal. Note that equality is meant, first and foremost, to be formal equality in front of the law. I am not interested here in questions of economical equality, although extreme economic inequality often has legal and social repercussions.

What does it take to create equality?

Suppose we have two classes of people. People of Class A has higher social and legal status than those of Class B. This leads to privileges. Thus, a person of class A can harm a person of class B without fearing serious repercussions, whereas society strongly enforces against harms done by class B unto class A. In such a case, we say that society enforces inter-class violence stronger than it does intra-class violence.

Risk stems from two possible actions: a person may exert violence on another person, and a person may ask the protection of the Law (or society). This protection can take the form of government violent action and can be applied against innocent individuals (as in the case of false charges, or unintended side effect of enforcement against a guilty party).

We assign probability p(x,y) to person x being harmed by person y, and abuse notation to set:

p(x is harmed by y| x∈A, y∈B)  = p(A,B)

In the case of inequality:

p(A,B) < p(A,A) < p(B,A)                                                                                                               (1.1)

This stems from the assumption of strong inter-class enforcement directed at the lower class, as well as from group B individual’s reduced access to funds, transportation, leisure, weapons, etc. The outcome of those is that the aggregated risk exerted by a high class person on her colleagues is much higher than the risk exerted by a lower caste person. The aggregated risks by class would be:

w(A) = p(A,B)*#B + p(A,A)* #A

w(B) = p(B,B)*#B + p(B,A)* #A

Where, by #A we denote the number of members of class A. So, clearly, given assumptions (1.1):

w(A)<w(B)

Note that we speak hereby of social status and legal inequality – not about economic inequality, although the latter often leads to legal and social inequality.

Suppose a benevolent liberal wishes to convince people of Class A to forfeit their privileges. Would it be rational (in the short term, all else being equal) on their part to accept?

All else being equal, and assuming full and immediate assimilation of the lower class within the higher class, it means that the high class individuals are asked to accept increased access by Group B members to government enforcement and to services and goods for which they previously had limited access. A person of class B could now cause as much harm to class A individuals as other class A members could – both through incurring direct harm (just like members of A could) and through the activation of government enforcement (be it justified or unjustified). Thus, we are asking to replace an upper-class individual’s risk with:

w’(A) = p(A,A) * (#A+#B) = w(A) + #B*[p(A,A) – p(A,B)]

In the case of a large population of lower class people, as in the case of significant inter-class enforcement, we are asking high-class people to accept a significant increase in personal risk. We are asking them to accept a high risk from a larger pool of people. This, clearly, would not be the choice of a free, simple, rational individual. Actually, the rational individual preference would clearly be to demand an increase in the inter-class enforcement, and a preference to reduce as many as her colleagues from class A to class B.

This unfortunate outcome means that, as long as we are speaking about rational evaluations of immediate consequences, high class members of society should be expected to use their power to push on lower classes as hard as they can. It follows, too, that individuals are better off if they can somehow push others into lower castes. In order to increase equality, we must convince the high-class individuals to accept an increase of personal risk – and this is far from being an easy task.

The high-class increase in expected risk will be only more significant once we start to take into consideration real-life outcomes which are bound to cause an increase of p(A,B) far beyond p(A,A) – due to vengeful attitudes, a history of neglect, and poverty. It seems that in actual situations, increased inter-class enforcement results in lower intra-class enforcement. This is, amongst other things, a matter of police attitudes towards the lower classes. If you deem the local policeman to be as dangerous as your fellow thieves, you are less likely to cooperate with him or ask his help when you are being mugged.

We thus need to be aware of the unfortunate facts that there is no such thing as zero risk, and that sometimes, in order to prevent a greater wrong, we must allow smaller ones (allow – not create!).

Since convincing people to accept higher levels of personal risk is very hard, the simplest avenue open to class B would be an increase in inter-class violence, to unbearable levels, at the hope that the cost of inter-class enforcement would force class A to forfeit at least some of its privileges. This is called “terrorism” (when it fails) and “revolution” (when it succeeds).