Ethik-Werkstatt - Volltexte im HTML-Format - kostenlos
-->Übersicht -->Alphabetische Liste aller Texte -->Info zu dieser Website -->Lexikon -->Startseite
The Majority Alternative
(1978)
I.) Voting procedures and the tendency towards the majority alternative
The majority principle states that the collective choice should be that alternative which,
if pairwise compared with each of the other alternatives, in all cases gains a majority of
votes.
This
alternative is called the "majority alternative".
This terminology follows that of Duncan Black, who defined: "That motion, if any,
which is able to obtain a simple majority over all of the other motions
concerned is the majority motion. Similarly in an election, that
candidate, if any, who is able to obtain a simple majority over each of the
others, is the majority candidate." [Cf. D.Black, The Theory of Committees and Elections, Cambridge 1958, p.18]
Because the method of pairwise comparison was invented by the french
philosopher CONDORCET (1743-94), it is also known as "Condorcet winner".
The majority alternative is not one version of the Principle of Majority among
others but it is of special importance. However this special importance is revealed
only, when voters are allowed to build coalitions and develop collective voting strategies in order
to get the best possible outcome for themselves. If these conditions are
fulfilled, then an existing majority alternative will be chosen:
In any kind of voting system, which gives equal weight to individual
preferences, an existing majority alternative will win, if all voters act with
perfect rationality when forming coalitions.
BLACK and most of the studies in collective choice assume "sincere"
voting. Therefore they do not analyse voting as a cooperative game, and the special role of the majority alternative could not be detected by
them.
The theorem about the majority alternative stated above can be easily proved.
If for instance it is not the majority alternative m but some other alternative x
which is chosen by the voting procedure, then those
individuals preferring the majority alternative m in comparison with x could have
formed a winning majority coalition on the basis of m. This result is better for
each member of the coalition than x.
Therefore a result other than an
existing majority alternative is impossible, if everyone knows the preferences
of the other voters and acts rationally.
II.) Problems of the principle of majority
If one excludes the more difficult problem that there is a "paradox of voting" with no majority alternative at all because of circular majorities (x > y > z > x), the question of democratic legitimacy is the normative question: under which conditions may the majority alternative be regarded as an acceptable approximation to the collective interest?
The first problem which may arise with any voting system concerns the
assumption that voters know their own interests sufficiently well. If their
preferences are based on disinformation, affected by logical errors,
influenced by fear of sanctions, determined by subconscious motives etc. then the
voters preferences do
not express their "enlightened" interest. Therefore a majority alternative
determined on the basis of such unqualified preferences may not be acceptable as
approximating the collective interest.
The second normative problem of majority rule arises from the fact that
in determining the majority alternative only an ordinal measurement of
individual interests is necessary. The fact that the same decision may be of
different importance or "salience" for the voters is neglected in
voting procedures, which give each individual equally one vote.
A certain mitigation of the problem may occur where coalitions are formed
over a whole series of decisions or when the voter has to decide between large
bundles of alternatives. If the individuals most heavily affected are not the
same but vary from
decision to decision, the degree of salience of the whole series of decisions
may tend to equalize with respect to the individuals.
But the majority decision
is questionable as an approximation to the collective interest in those cases
where for the loosing minority the decision is of greater salience than for the
winning majority and where this difference is not compensated for by the degree
of quantitative superiority that the majority possesses.
A third problem of the majority system arises from the fact that the set of
voters can never be identical with the set of those individuals whose interests
are affected by the decision - for example if the interests of future
generations are affected. Who takes care of these interests if voters express
only their individual interests?
A fourth problem when applying the majority system arises from the enormous
costs of getting information and making decisions, if everyone has to articulate
his interest with respect to every decision. In most cases the improvement of
the decision in approximating the collective interest will not suffice to
outweigh the costs of the decision process necessary to determine the majority
alternative.
A fifth problem of the majority system stems from an unequal distribution of
social power. Powerful minorities may limit the range of feasible alternatives
by connecting with an alternative consequences, which make the alternative in
question undesirable for a majority of the voters. For instance, capital-owners
will withdraw capital and transfer it to other countries, if a socialist party
gets the majority.
III.) How to deal with the shortcomings of the principle
of majority
Having briefly stated the four major weaknesses of a majority system the
question arises, whether these weaknesses can be removed by modifications of the
majority system.
The most promising of these modifications are the election of
representatives (which shall not be discussed here) and the
decentralization of
decision-making by splitting up the whole set of possible decisions into various
subsets of "domains of decision", so that everyone is no longer involved in
every decision but that each individual is competent for only a restricted
number of domains. If those individuals who are not - or only slightly -
affected by a decision, are eliminated from the corresponding voting body a
certain equalization of the degree of affectedness may be reached.
In addition the problem of sufficiently enlightened interests of the
individuals is much easier to be solved with decentralization, for the number of
decisions an individual is confronted with may be reduced considerably. As the
individual only has to decide those issues by which it is affected to a certain
degree, one may assume that there will be greater motivation to gather the
necessary information. In addition, decision-costs are greatly reduced, because
the number of individuals involved in any decision is much smaller. In the
extreme cases where there are individual domains of decision only one individual
has to decide.
The decentralization of decision-making thus seems to be an effective means
to deal with at least some of the short-comings of the majority system. On the
other hand new problems arise. One has to demarcate domains of decision and to
appoint the individuals belonging to each domain. Problems of coordination
between the different domains have to be solved so that a highly complicated
structure of decision-making may result.
At this point only preliminary ideas concerning the criteria of demarcation
and appointment can be presented. One principle of demarcating different domains
of decision should be to combine those decisions, which are mutually
interrelated. If decision 1 would alter the factual situation relevant for decision 2 and vice versa, both decisions have to be made connectedly. The task
then is to identify those decisions which are mutually interrelated in this way.
This task is complicated by the fact that the domains of decision normally have
to be demarcated in advance without knowing exactly which kind of decision has
to be made in the future.
A guideline for appointing the individuals to the different domains of
decision should be the degree of importance or "salience" that this domain has
for the respective individual. The degree of salience of a decision for an
individual would be measured by the difference of individual utility between the
best and the worst alternative. Thus an interpersonal comparison of utility
differences is needed and all the questions associated with this concept rise
again.
***
Also compare the following similar pages of the Ethics-Workshop:
Das Mehrheitsprinzip *** (349 K)
zum Anfang
Alphabetische Liste aller Texte
Übersicht
Ethik-Werkstatt: Ende der Seite "The Majority Alternative"
Last review 03.06.2008 / Eberhard Wesche
If this website was interesting for you, please inform your friends and colleagues about the Ethics Workshop.