Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site watmath.UUCP
Path: utzoo!watmath!cmgiuliani
From: cmgiuliani@watmath.UUCP (cmgiuliani)
Newsgroups: net.women
Subject: Generalizations & Individuals -- Insurance & Discrimination
Message-ID: <7867@watmath.UUCP>
Date: Sun, 27-May-84 00:27:33 EDT
Article-I.D.: watmath.7867
Posted: Sun May 27 00:27:33 1984
Date-Received: Thu, 31-May-84 20:34:25 EDT
References: <21@decwrl.UUCP>, <6500007@uokvax.UUCP>
Organization: U of Waterloo, Ontario
Lines: 26

newbug snack.

You cannot attack the practice of charging women less for insurance on the
basis that it would be unfair to extend such statistical practices to such
matters as how much to pay for physical labour -- should women be paid less
because they are not as strong.

Insurance use Statistical generalizations because they have no other
information available to them.  It is too expensive/difficult to obtain
other relevant info.  In any such case, "discrimination" is justifiable.
If I have to hire either person A or person B for a technical job, knowing
only that person A is male and person B is female, I will hire the male. 
This is based on the *true* generalization that men are better at (or if you
like more commonly educated in) technical fields.  If I want such a person
for a receptionist, I will choose the female, since women are generally better
at social functions. 

This is a lousy way to hire people -- I would rather know more about the 
candidates.  It is a lousy way to handle insurance, but no other way is
practical.  What we call "discrimination" in the perjorative sense is the
incorrect application of generalities when more significant info is available.
But sometimes discrimination is the only logical course.

                              Carlo @ the U of Waterloo