booth, john

8/11/2019 Booth, John

1/33

Political Participation in Latin America: Levels, Structure, Context, Concentration andRationalityAuthor(s): John A. BoothSource: Latin American Research Review, Vol. 14, No. 3 (1979), pp. 29-60Published by: The Latin American Studies AssociationStable URL: http://www.jstor.org/stable/2502963.

Accessed: 09/04/2014 15:06

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at.http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

.

The Latin American Studies Associationis collaborating with JSTOR to digitize, preserve and extend access to

Latin American Research Review.

http://www.jstor.org

This content downloaded from 200.54.168.42 on Wed, 9 Apr 2014 15:06:16 PMAll use subject to JSTOR Terms and Conditions
http://www.jstor.org/action/showPublisher?publisherCode=lamerhttp://www.jstor.org/stable/2502963?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/2502963?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=lamer


2/33


3/33

LatinAmerican

esearch eview

1.

How

muchpolitical

articipationakes

place, and

what s its nature?

What

re

the

evels f

participation?

2. What s thestructurefpolitical articipation?incethepresence f

political iolence n

theregion s

well

documentedFeierabend nd Feierabend

1966,

eierabend t

al.

1969,Huntington

968,Bwy

1968), hefocushere

willbe

mainly

n nonviolent

ctivity.

o

what

degree

does the

structurefnonviolent

political articipation

n LatinAmerica esemble hat

f

ther reasof he

world?

3. How do

differences

n the nature f

the

sociopolitical

ontext

r

en-

vironmentffect

he

tructurefparticipation?

4. How

concentrated

s political

articipation?o what

xtents

participa-

tion

monopolized

y persons

of

higher

ocioeconomic

tatusor

by political

activists?

5. Is masspolitical articipationnLatinAmerica ational?o what xtent

do

individuals ct

politically n

the basis of

goal-orientationsnd

imagesof

effective

hannels?

While

these are

by no means all

the mportant

uestions

bout political ar-

ticipation

n

Latin

America,2

he

tentative

nswers

now available

have

critical

implications

or

ur

understanding

f

politics

n

the

region

nd

for he

directions

for utureesearch.

POLITICAL

PARTICIPATION DEFINED

Traditionally,

ittle

greement

as

existed

mong tudents f political

articipa-

tion

bout

xactly

owhat he erm hould efer.

ilbrath

1965,

. 1) argues

hat

the

definition ust

notbe

so broad

that t oses

focus ranalytical

tility. ut

others

Verba

nd

Nie

1972,p. 2;

Pateman

970;Euben

1970) aution

hat he

definition

ustnot be too

narrow-and

especially

hat

t

must

ncludemore

than

ust behavior onnected

with

political

arties

nd

elections.

n

search

f a

definitionf

political

articipation

hat

would

nclude

satisfactorily

road

ange

of

political

ction,

with

Mitchell

eligson

have

elsewhere

Booth

nd

Seligson,

1978a)

mployed he

concept fpublicgoods

as a

central heme.

eriving

rom

political conomytudies felectoralehavior e.g., Downs1957,Tullock 968,

Czudnowski

976), his

pproach

s

novelonly s

applied o

thebroader naly-

sis of

political

articipation

n

general

nd to the

rea ofLatin

American

olitics

(Chaffee

976).

Collective

oods

onsist f

goods

that

when

supplied o one

member f a

collectivity

annot

asily

e denied o

others f

he

ame

group.Collective

oods

therefore

iffer

rom

rivate oods

by

how

much

ontrol ver

heir

se rests

n

the

hands

of

their

upplier.While

ccess to private

oods may be

easily

on-

trolled

y

the

supplier,

ccess to

collective oods

may not Olson

1968,pp.

14-15).

Public

oods, he

basis of the

definitionf

political articipation,

onsist

of specialkind f ollectiveoodssupplied ygovernmentsrbycommunities

through

heir ollective

xpenditure.

ational

ecurity

nd

monetary

ystems

provide lassical

xamples

f

public oods;governments

upply

hem,

nd once

they

xist or

ny

citizen

hey

ffectively

xist

or

ll.

But

ommunities

oo

for

30

This content downloaded from 200.54.168.42 on Wed, 9 Apr 2014 15:06:16 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp


4/33

POLITICAL PARTICIPATION IN LATIN AMERICA

example, neighborhoods, villages)

supply public goods, even though they ack

formal overnments.

Towns and

villages

often

provide

themselveswith a

variety

ofpublic goods such as roads, community enters, nd schools through ollec-

tive

expenditure

f such resourcesas

money, abor,

and materials ontributed

y

communitymembers.

An

excellent

xample

of this

type

of

communal supply of

public goods

comes from Adams's

(1959, p. 177) description

of the

Peruvian

village ofMuquiyauyo: Major

communal work projects

.

. are carried n every

few

years.

As is the case with the other towns

in

the

region, Muquiyauyo

has

a

long history of communal works .

. . of two major types: first here are the

upkeep activities, uch as fixing

treets nd bridges

.

.

.

which go on year

n

and

year out.

The

second kind of work

involves the large special projects

n

which

some new

addition or alteration s made to the community'smaterial equip-

ment.

Based on these concepts,

political articipations defined as behaviornflu-

encing r attemptingo nfluencehedistributionfpublic oods.Thus, when a citizen

or

a

communitymembertriesto

affect he distribution f a public good, he/she

has

participatedpolitically.Politicalparticipationdoes not include attemptsto

influence he distribution f

state-controlled rivate goods (such as licenses

or

patents),which

are

controlled

by

their ale.

Some

examples

of the

application of

this

definition

n

the

Latin American context

will

illustrate oth its

implications

and

utility.

A

road provides an

excellentexample

of

a

public good: normally

road is built through ither

governmental r community xpenditure, nd once

there t is difficult o restrictts use. In a rural region inadequately linked to

supply centers nd marketsby road,

attempts y residents ither o build

a road

themselvesor topersuade thegovernment

o do so constitute oliticalparticipa-

tion. Similarly, ctions by citizens

ntended either o bolsteror to undermine

he

stability

f a

particularregime

would constitutepolitical participation.Regime

stability

s

a

public good ( good

taken

n

the

economic,

not

ethico-moral

ense)

because

stable governments an

ensure particular conomic, social,

or

political

arrangements eneficial o certain

ectors

of a

society.Thus, actively upporting

or

opposing

a

regime

involves

attempting

o

influence

the distribution

f

a

public good,

and

hence is

political

participation.3

Several issues that oftenarise in discussions of the nature of political

participation hould be mentionedbriefly n order to clarify his definition nd

its application. First,participation equires ction overt behavior-and not atti-

tudes

or

beliefs.

Since regime

stabilitymay be regarded as a public good, active

support

or

government

oes

constitute

articipation. ome (e.g., Weiner1971,

Woy 1978)

have

suggested

that

political actiondoes not constitute articipation

unless

it

is efficacious-that

is, actuallyinfluences omething. Requiring such

policy

influence

s

a

sine

qua

non of

participation

would result

n

the

logically

flawed

position

of

arguing

that one who

voted or

campaigned

for

a

losing

candidate

in

an

electionhas

not

participated

t all.

Thus,

this

definition

oes

not

requireeffectivenfluence,merelythe attempt o influence the distribution f

public goods.

Nor does

the

definition

equire ntentionality,conscious aware-

ness of

taking political

action.

In

Latin

America,

where

the

involvement

of

the

state

n

labor relations s

commonly

xtensive,

he

strike-basically

an

economic

31



5/33

LatinAmerican esearch

eview

act-often

has extensive

mplications

or he distributionf publicgoods, re-

gardless f he

workers'

ntent. olitical articipationeed notoccur

nlywithin

formal overnmentrenas, since many public goods exist and are distributed

outsideformal overnment.inally, olitical articipation

eed

not be conven-

tionial,hat s, eitheregalor acceptable

o

a

regime;

ction f nykind-violent

or not, egal

or

illegal-aimed

at

influencing

he distributionf publicgoods

entails olitical articipation.

One advantage f his efinition

s thebreadth f hepolitical

ehaviorst

encompasses.

his breadth acilitates

n approach o citizen olitical

ctivity

f

muchgreater

cope thanthe

traditionalocusupon elections nd

parties hat

dominated articipation

esearch n LatinAmerica

rior o

1970

nd that on-

tinues

s a

major

rend

oday.4

second dvantage f this pproachies n ts

theoreticalmplications. yexaminingndividual ehavior s itrelates o the

distribution

f publicgoods within ocial systems,

ne

must

necessarily

on-

sider

hepolitical

ctorwithin he ontext f arger

ocial, olitical,

nd economic

structuresnd processes. he analystmust herefore

o beyond

he ndividual

to examine the reciprocal ffects

f the nteractionetween he

state nd the

political articipant

Booth nd Seligson 978a).

LEVELS OF

PARTICIPATION

The

first

uery

sks how muchnonviolent itizen articipation

akesplace

n

LatinAmericanocieties. he answerwill ikelyurprise hosewho still elieve

that

oliticalnactivity

haracterizes ass

publics

n the

region.

Breakdowns

f

the

evels of several

ypes

f

activitymong

Costa Rican

family

eads

reveal

participation

t

rates ather

igher

han

one

might

ave

expected

Booth 976).

Well ver

half f the

population

elonged

o

at leastone

voluntaryssociation,

was

moderately

ctive

n

both eadership

nd attendance

n

suchorganizations,

and

had taken

art

n

at east ne

communitymprovementroject

f ome ort.

In

fact,

4

percent eported

oting

n the 1970national

lection, figure

hat

closely

matches he

fficiallyeported

urnout f83.3percentTribunal

upremo

de

Elecciones

970).Over 30 percent

f theCosta Ricanrespondents

iscussed

national oliticsccasionally5ndhad contactedt east ne public fficial. ver

20 percent

iscussed ocalpolitics

t east

ccasionally

nd had been

members f

a

political arty,

hile10

percent

r more

had contacted

legislative

ssembly

delegate,

municipal ouncilman,

r

a

municipal

xecutive.

verall,

he

verage

respondent

ad participated

n 3.6 of the seventeen ypes

f

activity

xamined

(median

=

2.8),

and

only

ne

n

sevenhad never

ngaged

n

any

of hem.

One

might easonably rgue

thatCosta Ricans ould

drastically

xceed

the

LatinAmerican

orm or

articipation

ue to thatnation's iberal onstitu-

tional

radition

nd active

party

ystem,

ut data from thernations

uickly

dispel

his

hypothesis.

ooking

irstt

voting

we

find

hat atin

Americans

re

active oters,whereveregimes ermitlections.nthedecade1961-70, osta

Rica

was

slightly

bove

average

38 percent)

n

LatinAmerica

n thenumber

f

registered

oters

xpressed

s a

percent

f the

totalpopulation,

which anged

from

low

of

1'9 ercent

n

Guatemala o

a

high

f58

percent

n

Uruguay.

osta

32



6/33

POLITICAL

PARTICIPATION IN

LATIN AMERICA

Ricanswere

among

the most active

voters

n

Latin

America,

with

over

80 per-

-cent f

the registered

oterscasting

ballots

compared

to

a

range

from

3

percent

in Colombia in 1970 to 95 percent n Peru in 1963 Willems 1975,pp. 288-91). In

the1973

presidential lection,

more than 90percent f

Venezuelans voted Martz

and Baloyra 1976).

For further

omparativereference,

ational

election turnout

in the

U.S.

is about 55-60

percent Campbell

et

al.

1964, p. 49),

about 95

percent

in

Austria, nd

near

60 percent

n India

(Verba

et al.

1971, p. 36).

One

cautionary

note

must be added. Voting

statisticsfrom Latin

America must

be compared

with extreme

are

for wo reasons. First,

he accuracyof the

figures

hemselves

is

suspect due toelectoral

fraud n certain

ettings.Second, voting

s

mandatory

in

several countries, such as

Mexico,

Costa Rica, and

Venezuela, a fact

that

clearly boosts turnout

see, forexample,

Booth

1975a, p. 91,

for

data on

the

Costa Rica turnout ncrease when votingbecame mandatory).However, even

when

voting

is

required,

it does not

become

universal, leaving

some 10

to 15

percentvariance

n

turnout nthe Costa

Rican and

Venezuelandemocracies.

Campaigning

and

partisan

activismcomprise

a

second mode of

political

activitysee

below for

discussion

of the modal

structure fparticipation). or

comparison,

Verba

et al.

(1971, p. 36)

present data that permit n

estimatethat

roughlyone fifth

f the votingage

population of

Austria,Japan, ndia, and

the

U.S.

engage

in

some

sort of

campaigning.

Biles 1978) reports hat

n 1970

some

6

percent

of

urban

Uruguayans

belonged

to a

political

club or

party,

nd

that

some 13 percent

had

campaigned

or contributed

money

to

a

candidate.

Baloyra

and Martz 1978) reportmuchgreater ctivityn Venezuela in 1973:nearly hree-

fourths f

the

electorate

exposed themselves

to

campaign

stimuli,

and

nearly

half

took an active

campaigning role

(volunteerwork,

electioneering,

ttending

meetings, tc.).

The levels of

contacting

public

officials eported mong

Costa Ricans fall

within the

range reported by

Cornelius (1974, p.

1135)

for several other

Latin

American

nations.

Contacting mong

residentsofseveral

ower-class communi-

ties in

Mexico,

Peru, Brazil,

and

Chile varied from6 to

42

percent.

Some

11

percent of the

residents of Montevideo,

Uruguay,

admittedhaving received

some favorfrom

contact

with a

publicofficial Biles

1978). For nations outside

LatinAmerica,Verba et al. (1971) reportranges ofdifferentypesofcontacting

of

from

to

10

percent n

Nigeria

thelowest level) to

between 3

and 16

percent

inAustria the

highest evel).

For

communal

activism-collective problem

solving-there are few

analyses that

provide quantitativedata

fromLatinAmerica.

However, numer-

ous

studies

(for

example, Adams 1959,

Roberts 1'973,Fagen and

Tuohy

1972,

Fishel

forthcoming, astillo

1964, Doughty

1968,

Dobyns 1964) report

n

detail

on

the existenceof

communal

activism

n a wide

variety

f

community

ettings,

both

urban

and rural. Dietz

and Moore

(1977, p. 27) report

hat

73

percent

f

the

residents of six of

Lima's squatter

settlementshad cooperated with

other

resi-

dents in some community mprovement ffort. ooth and Seligson (1979), em-

ploying

two

independent data

sets, confirm

hat

around two-thirds f

Costa

Rican peasants

had

done some communal

problem

olving.

One finaloverall

comparison

will

place

Latin America

n

a

cross-regional

33



7/33

LatinAmerican

esearch

eview

and

cross-cultural erspective.Elsewhere

I

have

shown thatCosta Ricans' mean

level

of activity

n

voting,

partymembership,

ttendance at

political

meetings,

communitymprovement ctivism, nd contacting ublicofficialsBooth 1975a,

pp. 109-10;

1976, p. 629) ranks

them

as

slightly

more

active than

Indians

and

Nigerians,

but

only slightly

ess active than

Austrians,

Americans,

nd

Japanese

(Kim et al.

1974). Since these evels forCosta

Rica are roughly

imilar o those for

the

rest of

Latin

America,

we

must therefore onclude that

evels of mass par-

ticipation

n Latin America differ

ather

ittle rom hose

elsewhere.

THE

STRUCTURE

OF PARTICIPATION

What s the

structure f

political

participation

n

Latin

America, and how does

it

compare withotherregions of theworld? Referring rimarily o the U.S. and

Western

Europe,

Berelson

et al.

(1954),

Lane

(1959),

and Milbrath

1965)

once

depicted political

participation

s

a

unidimensional

phenomenon

consistingpri-

marily

f

electoral ctivities.

However,

recent

research n seven societies

nclud-

ing the United

States,

three

European nations,Japan, ndia,

and Nigeria

Verba

et

al.

1971,

Verba et al.

1973),

has revealed that nonviolent

participation

n

each

of these societies

usually

consists

of

several

ndependent

factors

r

modes;

these

include voting,

campaigning, ontacting ublic

officials, nd cooperative

ctivity

in

the community.But what

of Latin

America? Do such clearly distinctnon-

violentmodes

of political

behavior appear insocieties n this

cultural egion?

My (1975a, 1976) factor nalysis of seventeenparticipation ariables iso-

lated six factors f

politicalparticipation

mong Costa Rican

familyheads: vot-

ing, political

party ctivity likeVerba and

company's campaigning),

contacting

public officials,

oliticalcommunication,

nterestgroup activity like Verba

et

al.'s

community ctivity), nd

community

mprovement ctivism.Thus, four

of

the

dimensions

of

mass political

activitymong a national

probability ample

of

Costa

Ricans resemble those found

in other

regions

of the world. Other

Latin

American tudies

report

everal nonviolent

dimensions

of

political

participation,

and

confirmVerba et

al.'s

(1973)

contention

hat the

modal structure

may

alter

according to

the particular nvironment see

below).

Biles's

(1978) survey

of a

broad range ofpolitical ctivities mong urbanUruguayansrevealed fourmodes

of

participation

n

1970: voting,

political ommunication,

ommunal activity,nd

campaigning-particularized

ontacting.Baloyra and

Martz (1978)

demonstrate

that

n

Venezuela in 1973,

campaign activism, lsewhere

thought o be a

single

dimension,actually

consisted of two

separate

subdimensions-exposure

to the

campaign

and

involvement n

the

campaign. Studies conducted in

Ecuador

(Moore 1977, forthcoming)nd Peru

(Dietz

1977,

Dietz and

Moore 1977) have

also found several

dimensions

of

political

participation

n

these authoritarian

contexts.

Despite

the

absence of

voting,

Dietz and

Moore's detailed look

at

participation

mong

Lima's urban

squatters

reveals a communal

problem

solv-

ing mode, an organizational activity-contacting ocal officialsmode, and a

contacting

national officials

mode. Seligson and Booth,

(forthcoming-c)

how

that

the modal structure f

political

participationmong

a

cross-section f Costa

Rican

urban

dwellers (not just the poor) differs rom

hat of

peasants

in

ways

34



8/33

POLITICAL PARTICIPATION IN LATIN

AMERICA

linked to demographicand organizational

contrastsbetween city nd country-

side.

In sum, then, political participation

n Latin American

nations, as else-

where, exhibits multidimensional tructure.

While too few

surveys of whole

Latin American societies

have been conductedto permitfull

blown comparison

of

the

region

with the

participatory

tructures ound

by Verba, Nie,

and their

associates (1971,

1972, 1973), the studies done to date

have

produced

no

really

surprisingdeviations from the expectation

of multiple modes. The modes of

citizen ction are not always the same,

however. As the following ection hows,

the political, economic,

and social contexts may alter participatory

tructure

from ne society o another, r within societyfrom ne sector o another.

CONTEXTUAL DETERMINANTS OF PARTICIPATION

To what extent nd

in what ways does

context, r the socioeconomicand politi-

cal

environment,

nfluence

political participation?

tudies by Verba, Nie, and

Kim (1971) and

Verba et al. (1973) compare

modes of citizenactivity n seven

nations and conclude

that constitution

nd regime type account for structural

differences iscovered between Nigeria

and Yugoslavia and

five other nations.

Leeds and Leeds

(1976, p. 193) go so

far s to claim that virtually ll the differ-

ences

in

the

political

behaviorofthe urban poor

in

threeLatin

Americannations

stem from contextual

factors. Are

contextual phenomena

so

important?

The

followingdiscussion divides evidence of the environmental ffects pon politi-

cal

activity

nto studiescomparing oliticalregimes and studies

within articular

regimes.6

Turning

first o comparisons ofregimes' impacts upon

participation,we

find that

data take

two basic forms:

omparisons

of different

ational

political

systems,

and

analyses

of

regime

changes

within

particular

ocieties.

First,

as

indicated

n

the previous sections, both

structures nd levels of politicalpartici-

pation vary

with national

context.

Most

notably,

f

course,

certain uthoritarian

regimes sharply

curtail

or

completely uppress voting

and much

political party

activity.

Biles's

(1978)

Uruguayan study

observes another

type

of contextual

effect; ampaigning and personalized contactingmodes (separate elsewhere)

combine

into

a

single

mode. Biles

speculates

that this

derives from

Uruguay's

intense and

long-lived

pattern f patronagepolitics Weinstein

975),

which

has

transformed

olitical

clubs

and

parties

nto

almost

ubiquitous

brokers

between

citizens

with demands and

the

state.

Leeds

and Leeds

(1976) carefully

nalyze the impactof regimetype upon

the

politicalactivity

f the urban

poor

of

Brazil, Peru,

and Chile.

They

find

that

differences

n

the number of

political parties

with a mass base within

each

societydramatically

lterthe

tactics,

trategies,

nd success of the urbanpoor in

their ursuitof the public goods.

In

Brazil,

where

party

lites

have

avoided

mass-

based organizing, aveladwellershave resorted o interest xchanges (trocas e

interesses),

nformal

nd

temporary

ut

sophisticated

arrangements

o

bargain

potential support

for

competing

candidates

in

exchange

for

specific mprove-

ments and

services.

In

Peru,

where only

the APRA

has

had a

mass

base,

other

35



9/33

LatinAmerican esearch

eview

parties ave attemptedo counteracthis dvantage hrough actics fmass co-

optation. hefrequenthanges n ruling liteshave producedmassive ureau-

cratic roliferations successive overnmentsreate ew agencies nd programs

for

heurban

poor.

The

typical

arriada

esponse

n

Peru

has

been: 1)

to

direct

demands o bureaucracies, hich

ast

onger hangovernmentsnd are more

dependable

han

parties;

nd

(2)

to

manipulate

he

multipleeven competing)

national nd nternationalssistance rograms

o

maximumdvantage.

n

Chile,

where

rior

o

1973

everal

olitical arties

ad broad

mass

bases,party ompe-

tition or

electoral

upport mong

the

urban

poor

has

been intense.

n

the

callamipasformednformallyy gradual

ccretion

f

residents)

nd the

pobla-

cionzesgovernmentlanned ettlements)

heresidents ave

cautiouslymanipu-

lated ompeting arty gencies or elp.But n the

ampanienitos

typicallyormed

by party-organizednvasion)residents ave sought public goods primarily

through he bureaucraticinkagesprovidedby

the

party

hat

organized

he

invasion.

Walton's1976,pp. 134-35) omparison ftwoMexican nd twoColom-

biancities etails ow politicalnd economic haracteristicsfurban lites ffect

the nature nd success of demand making y lower-level roups. He finds

similaritiesetween ali and Monterrey,here lites rovide ewer ublic oods

to

lower-sector

roups

han

n

Medellin

nd

Guadalajara.

Greater

olicy

on-

cessions, n response o greaterower-sectoremandmaking n the atter air,

occur

ecauseMedellin's conomic lite

ependsupon

ower-sector

ooperation

for steady abor force nd becauseGuadalajara'sbureaucraticeadersneed

successful

evelopment rojects

o

promote

areer dvancement ithin hena-

tional

RI.

Elite

vulnerabilityPortes nd

Walton

976, p. 172-75),

which

he

urban

oormay xploit n Medellin nd Guadalajara, s at ower evels

n

Mon-

terrey,

here he lite s less

pressed ythe

PRI

due to political iversity,nd n

Cali,

where

business eaders

find

n

ample

and

steady

abor

upply.

he find-

ings by Walton

nd

by Leeds

and

Leeds suggest hatone critical ontextual

variable

nvolves

he

nature f he lite.

A

combinationf ntra-eliteompetition

and dependency pon ower-sectorroups eemsto promote lternativehan-

nelsformassdemandmaking

nd

toencourage uchdemands.

Recent tudiesofsocialsecurity ystemsn LatinAmericaMesa-Lago

forthcoming,alloy1977,Rosenberg

nd

Malloy1978)detailhow increasing

participationy

uccessive

waves

of

newly mergingnterestectors ave reated

unworkable

nd

inequitable ystems

hat

have severely vertaxed ational co-

nomic

nd

political

esources. uccessful

ffortso

reformuch programs ave

so

far

ucceeded

only

where uthoritarian

egimes

ave

seized

power

nd re-

pressed nterest-grouparticipation

n

social

ecurityolicymaking.hus,

where

vulnerableivilian egimes ave dependedupon ower-sectorupport, ressure

politics

as

been

encouraged. egrettably,t has been military

nstitutions

hat

have manifestedhe nternal olidaritynd power to reformuccessfullyhe

resultant eformed ublicpolicies, ut this t the cost of severely epressing

participationRosenberg

nd

Malloy '978).

Studies fregime hangeswithin ations rovide ther xamples f

con-

textual

ffects

pon participation.or nstance, ollierdescribes n detailhow

36



10/33

POLITICAL

PARTICIPATION IN LATIN AMERICA

policychanges by

successive

Peruvian

regimes since the -19,50save

altered the

government's

reatment f the urban

poor,

and how the

poor's politicalbehavior

has shifted n response: Squatter settlements nd public policytoward settle-

ments have been used as

a

means of

inking

he urban

poor

to

the

state

n

a

way

that s

viewed as

constructive, ather han

disruptive,by

those

in

power

(1975,

p.

37).

Thus, as

successive elites have tried to control

participation, olicy has

variouslypromoted

mobilization of

squatters to supportOdria

('1948-56),

self-

help activityunder Prado

(19'56-60),

massive electoral nvolvement-followed

by

frustration-in the party

governmentperiod (1961-68), and

self-help

and

extraparty

mobilizationunder military

ule 1968-75).

(See also Woy 1978,Dietz

and

Palmer1978, Palmer

1973, Palmerand

Middlebrook 1976, and

Palmer and

Rodriguez

1972.)

Shepard Formanargues that the nature of [the Brazilianelite's] power

struggle

t

any given

time has

defined

the

nature of

peasant politicalparticipa-

tion as well

(1975, pp.

'143-44).

He

traceshow these

struggleshave transformed

peasant political

participation: s local elites

have

gradually

been

suppressed by

centralizing

ational

elites,

peasants' political

ctivity

as

not become

indepen-

dent,

but

has shifted rom

patron-dependency upon

a

single

ocal landlord

to

patron-clientship withplural national

nstitutions.

ince 1964, ofcourse, mili-

tary

ule has

dramatically

estricted uch

activity.

rian

Loveman's

(1976,

forth-

coming)

analysis

of abor

organization mong

Chilean

peasants

since

191'9

hows

that

opportunisticmanipulation

by

urban based

parties

of the left

n

pursuit

of

theirown goals has provoked repeated waves of ruralstrikes, ncouraged by

the

opposition

to

weaken a

government

n

power.

Once

a

regime

change oc-

curred,

however,

the

parties

have

typically upported

renewed

repression

of

the

peasant

unions.

Loveman

also notes

how the

Unidad

Popular

coalition's

rise

to

power

in

1970

brought

about a

dramatic

growth

n

peasant activism s the

left

was

finally

free to

compete

more

openly for mass support and

eschewed its

previous

vacillation.Adams (1970) traces

the evolutionof political

power struc-

tures

n

Guatemala from1944to 1966,

highlighting he initial

xpansion ('1944-

54)

and

subsequent

contraction1'954-66)

of politicalparticipation

mong various

social sectors as

regimes

and

their

deologies

have

shifted

fromrevolution to

counterrevolution,witha corresponding ncrease in rulingelite solidarity see

also Boothforthcoming).

The

numerous studies that treat

ontextual ffects pon

political partici-

pation

within

egimes ubdivide

into twogroups:those

comparing he

activity

f

differentocial

sectors,

nd

those

that

examine

differences

ithin

singlesocial

stratum. n

the

first

group, Fagen and

Tuohy (1972)

compare

the

activity f

upper-,

middle-,

and

lower-class residents of the Mexican

provincialcapital

of

Jalapa. They

report

hat

contactingpublicofficials, olitical

group

membership,

party

membership, ttempts

o influence

public

decisions,

and

campaigning

re

highestamong

upper-sectorgroups,

intermediate

orthe middle-class

sample,

and lowest among the poor. They further ind this inequalityaccentuated by

educational

attainment,

ltimately eporting surprise

at the disproportionate

extent to which the

upper

class .

. .

enjoys

certain

political

advantages

in an

institutional

etting

that is

ostensibly geared

to the

representation

f middle-

37



11/33

LatinAmerican

esearch eview

and lower-class interests

1972, p. 87). Clearly,

the

structure f the

Mexican

political ystemprovides significant arriers o

participation mong lower

status

individuals.

Several

Costa Rican studies also report

substantial

participation

differ-

ences across social strata.

Seligson

and Booth

(forthcoming-c)ompare political

activism among

Costa Rican urbanites and

peasants. They report only slight

structuraldifferences

n

modes of

activity,

ut

they

find that urban

dwellers

exceed peasants

in

political participation

within

the

national arena

(voting,

con-

tactingofficials),

while

peasants engage

more in

communal

improvement

nd

group activity.Other

Costa Rican studies

that examine the impact of region

(Booth et al. 1973, Booth 1974) and

community ize (Booth

et al.

1973;

Booth

1975a, 1975b)

reportgreaternational arena

activity nd less communalparticipa-

tion among those in large and centrally ocated communities,with the reverse

true

n

small or solated towns and villages.

The

common thread running hrough

hese reports s a striking ifferen-

tial

in

public

services

isolated

by controlling

or region, community ize, and

urban-rural onditions. Each

studyconcludes that a

major cause forrural,

pe-

ripheral area, and village residents' intense communal

problem solving

is

the

lack of basic

services

which

they

confront.

Urban,

central

region,

and

larger-

town residents

enjoy many basic services

provided by government,

hus reduc-

ing the impetus

for communalism. Another

contextual effect tems fromthe

costs and benefitsof

participation

n

a

highlycentralized

political system.

The

institutions fgovernment nd the servicestheydistribute ie within asy reach

of residents of

Costa Rica's

metropolitan enter,

but

remain remote

and

costly

for hose

isolated

from he center.These differentialostsof access

and

unequal

distributions

f

payoffs f national arena

participation ccount formuch ofthe

observed

variation

n

voting, contactingpublic officials, nd interest-groupc-

tivity mong

different ommunity izes, between rural and urban strata,

and

between central

nd peripheral egionsofCosta Rica.

A

final

group of studies delves intocontextual ffects pon citizen

action

within a

particular ocial stratum.

For

example, Seligson (1978) scrutinizes he

effect f economic

development

evels

upon

four

forms f citizen

ctivity

mong

peasants. He discovers that ow levels of ruraleconomic development increase

levels of group

activism, contact with local

government,communityproject

activity,

nd

voting. He concludes: In

areas where infra-structures

poorly

developed

and

government ervices are

minimal, ndividuals are

compelled to

participate olitically

f

heyhope

to

see

some

improvement 1978, pp.

151-52).

Bourque

and

Warren

forthcoming)

xamine the

differences

n

participa-

tion between the

women

of two Peruvian

rural communities

nd

report

sig-

nificant

ontextual

ffect. hey note that the

women of

the

traditional

ighland

village,

where the

males

controlkey economicprocesses, take ess part

politically

than the

women

of the

nearby

new commercial town.

Bourque

and Warren

attributethe difference o the commercial town women's greater economic

power-many

operate

businesses

independently

of males. Handelman

(forth-

coming)

notes that workers

n

Mexico's PRI affiliated

abor

unions exhibit de-

38



12/33

POLITICAL PARTICIPATION IN LATIN

AMERICA

pressed political activity, ut

that members of

radical electricalworkersunions

take part at a

notably higherrate. He traces

this within-classvariation to the

ideologicaldifferences etweenthe two types ofunions.

Cornelius

(1973; 1975, pp. 110-34)

examines the impact of

socioeconomic

contextupon the political

activity f the poor of

Mexico City.He demonstrates

that different

eighborhoods

have distinctpolitical cultures as

measured with

aggregated scores

on

indices of civic

mindedness, dispositionto conform o

communitynorms, perception

of external hreat, nd strength

f self-helpori-

entation.

These

community

ttributes

roduce significant

ariationsfrom arrio

to

barrio

n

an overallparticipation

ndex,

even with

controlsfor

ocioeconomic

status SES), length of

residence,

and

psychological

nvolvement

1975, p. 114).

Cornelius also

itemizes 1975,pp.

124-32-with

many

usefulreferences

oother

research) ten other aspects of the social and material contextthat influence

participation

mong the urban poor. While

internal cleavage,

low population

density,

nd

large

size reduce

articipation

within barrio,

everal things ncrease

activity: ES homogeneity,

tability

f

residence, boundedness, establishment

by invasion,

external

threats,

repression,

success with

previous petitions

to

authority,ocally

oriented

eadership

and

voluntary ssociations,

and such

de-

velopmental

needs as lack of land titles and

basic public

services. (For other

studies thatcite

similar

ontextual ffects pon participation y the urban

poor,

see Goldrich et al.

1967; Perlman1975, 1976;

Portes 1969; Roberts

1973;

Dietz

1974; Moore

1977, forthcoming;

s well as thebibliography n

Cornelius 1973.)

To sum up, politicalparticipation n LatinAmericamanifests iverse con-

textual nfluences.The

(evolving) structure f

sociopolitical nstitutions rovides

the framework

within which

citizens attempt to

influence

the distribution f

public goods.

Constitutionalstructures, egime

types, elite

interactions, nd

institutional

onfigurations

eterminethe opportunity, hannels

for, nd costs

of

participation,

hereby

nfluencing he levels, structure,

trategy, actics, nd

policy

influence of

political activity.

hifts n

such

structural henomena may

produce dramatic lterations

n

patterns

f citizens'

efforts o

influence he dis-

tribution f national

public goods.

At lower levels

within social

systems, phe-

nomena as varied as the

availabilityof public

services, local politicalculture,

community ize, and residencepatternsprovidesimilarframeworks hat deter-

mine the

motives, tactics,modes, and intensity

f political

ctivity.What

Portes

has

said of Latin

America's urban

poor seems appropriate for

extension to in-

clude most

other social

sectors-political

behavior

is rational

adaptation

to

what structural

ircumstances

ermit

nd

encourage (1976, p. 108).

THE

CONCENTRATION

OF

PARTICIPATION

To

what extent

s


n Latin

America

concentrated,

hat

s,

monopolized by

militant

olitical

activists r

by

persons high

in

socioeconomic

status? n theUnitedStatespoliticalparticipation endsto be somewhat concen-

trated;

ndividuals

who perform

he

most

difficultnd

time-consuming

cts

also

often

perform

most

of the easier ones.

Furthermore, articipation

s

unevenly

39



13/33

LatinAmerican

esearch eview

distributed,

ithmanypeople

taking

art t fairlyow levels,

nd farfewer

engaging

n argenumbers

f ctivitiesVerba

nd Nie

1972,

p. 25-35;

Milbrath

1965, p. 16-21;Matthews nd Prothro966, p. 52-58;Lane 1959, p. 93-94).

My

research n

a national

robabilityample

f1,442Costa

Rican amily

heads has revealed

hat evels

of ctivityary ubstantially

Booth

975a).While

but

14 percent ad engaged

n none ofthemeasured cts,only

1

percent

ad

takenpart

n

six

or more,

with

median

corefor he

sample

of 2.8.

Going

further,he study

focused n

the eastcommon

ctivitiesthose

done

by

30

percent

r

ess ofthe ample),which

hepreviously

ited

uthority

ould

sug-

gest

to

be

highly

oncentratedmongpolitical

ctivists.

nly half f all

Costa

Ricanfamily eads

had done

anyoftheserarer cts,

nd only bout15 percent

had engaged

n

three r

more.Thus, there s, indeed,

some

concentrationf

political articipationn Costa Rica, as observed n other ocieties.A small

minority

f

participants

s

disproportionately

ctive.

However,

he

evidence

revealed

nother ide of the

coin-extreme

nactivitys also

rare, nd

almost

half

f

therespondents

ad engaged

n

between ne and five

f

even

the

rarer

forms

fparticipation.

The question

f how muchconcentration

r overlapping

ctually

xists

may

be examined

tillfurther.

arly

peculation Milbrath

965, Lane 1959)

suggested

hat

participation

s so concentratedhat

lmost

very erson

who

performs

rare ct

engages

n

themore ommon

nes as well.

However,

im t

al.

(1974)

and Verba

nd

Nie

(1972)

have shown that his

assumption

reaks

downfor he seven Verba-Nie tudy ocieties,withparticipation orewide-

spread nd much

ess concentratedhan

xpected.

heyfurtheruggest

hat

he

concentration

f

participation

ncreases

s the evelofeconomic

evelopmentf

the ociety

ncreases.

he CostaRican

data Booth 975a, p.

107-14)parallelled

the

Kim t

al. findings.

comparison

f heoverlapping

fparticipation

mong

behaviors epresentative

f he ixmodesrevealed hat he oncentration

s only

about

half s

great

s

its

theoreticallyossible

maximum.

Many

more

people

takepart

n

themoredifficultnd

time-consuming

ess

common ctsthanone

would

expect

f but

a

few

political

ctivists

ominated ll

arenas of activity.

Unfortunately,

o similar

nalyses

f he

oncentration

f

participation

ave

yet

come o ight or ther atinAmericanations.

Another

way

to

explore

he degree

of

concentration

s

to examine

he

extent

o

whichSES affects articipation.

he

correlation

etween

high

tatus

and more ntense

olitical

ctivity as been widely

bserved Salisbury

975, .

326;

Milbrath nd Goel 1977,

pp. 86-106).

s it

true hat hose

of higher ES

dominate r monopolize olitical

ctivity?uch a participation

onopoly ould

be said

to exist

f

either 1)

no

activity

ccurred

t

all

amongpoorer

itizens,

(2) the

richwere

greatlymorepolitically

ctive

han ower tatus

roups,

r

3)

nonelites

ad virtuallyo

nfluencever

hedistribution

f

public

oods.

Data

from

hroughout

atinAmerica ave

completely

ndermined

he

propositionhat hepoor are politicallynactive. he size of thegroup fcom-

pletely

nactive

osta Rican

citizens 14 percent)

s tiny n comparison

o the

ranks f

the nation's

ow SES

groups Booth

975a).

Furthermore,esearch

n

Costa Rican

peasants,

most

quite mpoverished

y any

standards

Booth

nd

40



14/33

POLITICAL

PARTICIPATION IN

LATIN AMERICA

Seligson 1979; Seligson 1978, forthcoming)

eveals substantial

evels of partici-

pation. Additionally, tudies from

lsewhere in Latin America

report mportant

amounts of politicalactivity mong peasants in Mexico (Landsbergerand Gie-

risch forthcoming), hile (Loveman 1976,

forthcoming), nd

Brazil (Forman

1976), and among the urbanpoor in several

countries Ray

1969;

Cornelius

1974,

1975; Dietz 1974, 1977;

Portes1969, 1971,

1972; Moore 1977;Peattie1968; Perlman

1975, 1976; Roberts

973). Thus, lower-status roups take an

activerole n politics

in

many Latin American nations. This

role involves not only

communal and

organizational ctivity, ut party

participation nd contacting

public officials s

well.

Thus, theupper-status ectorsdo not

dominate

participation

n

this ense.

The second condition mentioned

above concernsdisproportionate oliti-

cal

participation y

wealthier itizens. s thepolitical ctivism f

the upper

strata

extraordinarilyreat n comparison to the ower strata? f so, towhat extent nd

inwhich arenas of

activity? he answer to

the first uestion is a qualified yes-

individuals

high

on the status

hierarchy re more active in

certain political

arenas. Forexample,

Fagen

and

Tuohy

1972, pp. 81-106) report

hat

both

social

class

and educational

attainmenthave a strongpositive

relationship,

s noted

above. Elsewhere, Costa Ricans at large

exhibit ow-to-moderatepositive cor-

relationsbetween SES

and five of sixmodes of political

participation-voting,

partisan ctivism,

political ommunication, rganizational

ctivism, nd contact-

ing

public officialsBooth 1975a, 1978).

Among Costa Rican

peasants, landown-

ership,

an

important

ndex

of

social

and economic

standing

n

the

countryside,

contributeso a significantncrease nparticipation norganizations, ommunity

betterment

fforts,

nd

contacting

ocal

public

officials

Booth

and

Seligson 1979).

Cornelius

(1975, p. 94) reports

a

weak but

significant ositive

SES correlation

with

voting

nd

campaigning mong

his

Mexico

Citymigrant oor.

Data

on

other

Latin

American nations

also reveal

positive

correlations

between

participation

ndicators

nd

socioeconomic

status.

Biles

(1978) reports

weak

positive

correlation etween

SES and

organizational

membership

nurban

Uruguay. McClintock's 1976a, 1976b) data on the

Peruviancountryside

n

1969

and 1974

revealweak-to-moderate ositive

associations between

SES

and

politi-

cal

participation evels

in

a variety foccupational

contexts,

despite

the relative

homogeneity of her population samples. Bourque and Warren forthcoming)

show that

among

women

in

two rural Peruvian

villages, exceptionally high

political

ctivity

ccurs

among

those

with

the

greater

han

normal

status

derived

from conomic

independence frommen.

Fishel (1976) reports

that

in

another

Peruvian district he

socioeconomic elite

takes a greaterpoliticalrole (especially

in

contacting public

officials)than the

majority of citizens. It seems safe to

conclude, therefore, hat

higherSES

individuals, n both thenational arena and

a

variety

of

local

contexts,

tend

to

engage

more

in

politics

than

lower

SES

citizens.

However, not all evidence reveals such

positiverelationships.

Several

of

the studies just citedreportverydifferentindingsforcertain ypesofactivity.

For

example,Cornelius's Mexico

City

urban

poor

are

significantly

ore

ikely

o

takepart

n

communal

problem solving

if

they

are

low in

socioeconomic

status

(1975, p. 94). Similarly,

mong

Costa

Ricans,

communal activism

occurs most

41



15/33

LatinAmerican

esearch eview

frequently mong persons

low on

the

SES

hierarchy, egardless

of

whether he

population is consideredas a whole (Booth

1975a), or ruralresidents

re

singled

out (Boothand Seligson 1975, Seligson and Booth forthcoming-c,eligson 1978).

Likewise, Biles (1978) encountered no

correlationbetween SES and voting or

political

communication

among

his urban

Urguayans,

and found a

negative

association

for

campaigning/contacting.

aloyra

and Martz's

(1978) study

of

participation

n

the

1973 Venezuelan

election

reports

no association

between

voting and SES. They attribute

his

phenomenon

to the

mandatory

nature of

voting and to the intensive mobilization of

the Venezuelan campaign. Lands-

berger

and Gierisch

(forthcoming) eport

no

statistically ignificant egree

of

association

between

objective

status ndices

such as the amount of and

owned,

education,

and

income and

the voting and partisan activism

of their Mexican

peasant sample. And finally,Dietz and Moore (1977, pp. 37-41) detected no

significant orrelationsbetween SES and community mprovement fforts y

Lima's

poor squatters

nd

migrants; hey lso report hat ctivists

re

systemati-

cally slightly oorer

than

nactiveresidents.

These findings reveal patterns

n

political participation hat hold con-

siderable nterest. ersons of

higherSES evidentlypredominate

n

participation

within

the national government rena. This

is likely o

first ecause

the greater

one's wealth and

standing, the more

likely

one

is

to have interestsfor the

defense

of which

nationalpoliticshas

importance; econd, those of higher tatus

are more

likely

to have

the educational and economic resources that facilitate

participationwithinthe complex,centralizednationalpolitical systemsofLatin

America. On the

other

hand, persons

of ower SES

commonly

ake

more active

politicalroles

than

elites

in

communal

improvement

ctivism.

As

noted

in

the

previous section, the great centralizationof most public services within the

region's metropolimilitates n favor of

communal improvement fforts mong

those

marginalized

from

uch services

by living

n

poor urban barrios or in the

countryside:

Where

national

policies

distribute

public goods unevenly

and

offer cant

prospect

for

ocal

improvement,neighbors

often

turn

to each other

for he creationof

public goods through

ommunal effort

Booth

and

Seligson

1978a;

see

also

the

studies

n

Seligson

and

Booth forthcoming-b).

The final issue under the rubric of upper-SES domination of political

participation

oncerns

the

question

of

the

relative

degree

of

influence upon

public policy. While the definitionof political participation hosen explicitly

rejects

the

notion

that

policy nfluence s

required forparticipation o occur, the

substantive

significanceof the whole issue

depends

to an

importantdegree

upon

the

public's policy mpact.

That

is,

if

nonelites do takepart

n

politics

but

have

no

influence n policy, hen theparticipation s irrelevant nd unworthy f

anythingmore

than

passing note. How much policy influence do lower- and

middle-sector

groups exercise

n Latin

America? There is evidence that certain

segments

of mass

publics sometimes influence national

public policy

in

both

democratic nd authoritarian ontexts.Butsince particular olicies seldom affect

all

social strata

simultaneously,we must

isolate specificgroups

in

order to ex-

amine

this

nfluence.

One such

social sector onsists of the

urban poor.

The now

almost

Brobdingnagian

iterature n

Latin

America's urban

migrants nd squat-

42



16/33

POLITICAL

PARTICIPATION IN LATIN

AMERICA

ters defies

comprehensivetreatment

ere

due to

the

broader

scope

of

this

re-

view. The

citations hatfollow,

herefore,

re illustrative;

or

further

nalysis

and

citations,the reader should consult Cornelius's (1973, 1974, 1975) works on

Mexico

City,whichalso cite

much of the

relevant ross-national

material, s well

as Handelman's

(1

975a) reviewof research

on the urban

poor in Santiago,

Chile,

and thebibliography

n Portes nd Walton

1976).

Peattie's case

study of the poor

Caracas barrio of

La

Laja

describes

the

developmentof a

demand-makingprocess

that

eventually

securedwater

lines,

a baseball field,

and

a nutrition

enter.

While

observing

that such

communities

may oftenfail

n

policy

nfluencedue to

lack of government

esourcesor

official

recalcitrance, eattie

notes that

success

in

these cases

resulted

frommobilizing

enough group pressure

to establish new channels

of connection

with thecen-

tersofpower outside the community 1968, p. 69). Similarly, orneliusreports

that conventional

(nonprotest)

demand

making had been partially

ffectiven

all six ofhis low

income MexicoCitycommunities:

Most of thebenefits

ited by

community

eaders

would not have been

provided in the absence

of overt

de-

mand making efforts

1974,

p. 1140). He attributes

governmental

failureto

meet other

needs

as sometimesdue to

interrupted

r inept demand making,

and at

other imesto lack of official

esponsiveness.

Studies that

document

similar impact upon

public policy

abound (for

example,

see

Ray

1969, Fagen

and Tuohy 1972,

Roberts1973,

Dietz

1974,

Leeds

and

Leeds

1976,

Portes and Walton

1976, Moore

1977), but one should

not

conclude that the policy mpactoftheurban poor is extensive,or that the poor

often

force

governments

to reorder their

budget

allocation

priorities.

Moore

(forthcoming)

ffirms hat the

capacity

to

.

.

.

extract small-scale

(but

real)

benefits

fromthe

system by manipulating

t, does

not belie the fact that

this

participation

ill

not alter

he basic structural onditions of] ystemicnequality.

The generally

o-optativenature

of concessions to urban

squatters

nd migrants

has

been

demonstrated

byEckstein 1977)

forMexico, Perlman

1976) for

Brazil,

and

Collier 1975)

forPeru,among others.

Governments

often make

policy

concessions

not

on

a broad basis

that

affects

arge segments

of the urban poor,

but rather

distributively,s in the

case

ofconceding lots and titles o squatters,greatlydisaggregating enefitsby dis-

pensing

themon an

individual basis

(Collier

1975, pp.

46-47). Thus,

rather

han

pursue

costly

redistributive

olicies

such

as

government

housing,

the

state ac-

quires

land

relatively heaply

and doles it out

piecemeal,

maximizing

control

while minimizing

he

likelihood of serious

challenges

to the

overall allocation

of

public

resources.

The ruralpoor also can

influencenational government

olicy

to

a limited

degree. Even small

and isolated

rural communities

n several

nations obtain

pork-barrel

ppropriations

from heirnational

governments.

Doughty

and

Ne-

gr6n's

1964, pp. 56-59)

study

of

the

Peruvian

Andean

village

of

Pararin

reports

thaton several occasions the governmenthas reacted to communityrequests

with

assistance.

In

response

to delegations

toLima,

the

government

has

helped

pave

some streets,provided

materials

for

ne of the

schools,

donated

technical

aid

for

he installation f

a

generator,

nd

supplied

earthquake

relief.

Similarly,

43



17/33

Latin

American esearch

eview

the Costa Rican government

had

provided

assistance

in

some form r another o

two-thirds fthe communal

improvementprojectssurveyed

in

109 towns and

villages. This aid took several forms:direct ash appropriationsbythe National

Assembly, material

and machinery donations by the Public Works Ministry,

technical assistance

from

the

National Water and Sewer Service, provision of

teachers nd books

for ommunity-builtchools, and so forthBooth et al. 1973).

Further xamples

of

such

government esponse

to

peasant demand makingmay

be found

in

Dobyns (1964), Whyte 1969),

Castillo

(1964), Gonzalez

and Ham-

mock 1973),

and

Fishel

forthcoming).

One phenomenon thatclearly facilitates

olicy influenceby peasants

is

organization.

The

peasant leagues of Bolivia

(McEwen 1975, Malloy 1970) have

proven instrumental n securing assistance to

rural communities,resulting n

organized villages receivingfargreater ttentionfrom he centralgovernment.

Powell's (1971) study

of

the Venezuelan

peasant

movement

reveals how

campe-

sino syndicate

eaders

function

s

brokers

with the

government

n

seeking

and

reform nd servicedevelopment.He concludes that

governmental grarianpoli-

cies

are

responsive

to

the

inputs generated by

the

peasant Federation,

even

though

rational

planning

considerations

have

usually

modified

the

state's

out-

puts p. 171).

Despite such

evidence,

one

should not assume

that

Latin American tates'

responsiveness to

demands from he ruralpoor

have been far-reaching. s

with

the urban poor,co-optationand the use of highly

distributive olicies minimize

the cost of necessary concessions to campesinos. Even in Mexico, where the

ruralpoor have

violently tated theirdemands for

eform, olicies of the revolu-

tionary

governments

have

continued to favor

urban/industrial evelopmental

strategies.Land redistribution as been

relatively

ow

cost

(especially since it

has undermined

portions

of a

hostile

former

ligarchy),

nd

has followeda very

disaggregated or distributive pattern

somewhat analogous to that Collier

(1975) described for

Peruvian urban squatters. Land

has been seized at declared

tax value

and

parcelled

out to

ejidos

in

fits nd

starts oincidingwith unrest n

the

countryside.

The

government

nd the

PRI

leadership have also co-opted

peasant

leaders

in

order to

manipulate

their

dvocacy

of

peasant

interests

Han-

sen 1971, pp. 87-107). Although Costa Rican peasants have so far argely es-

chewed

violence, demand for

and there has been satisfied n a similarly isag-

gregatedfashion,first hroughpermitting

ndividual colonization of public and;

when

public

land

was

exhausted,

the

government

urned

to

compensating pri-

vate argeholdersfor

osses to invaders Seligson

1977, Booth and Seligson 1979).

Turningto middle-sector

groups, Malloy (1977), Rosenberg and Malloy

(1978), and Mesa-Lago (forthcoming) resent an

overview for he whole region

of

the success of

pressure

groups (especially

middle-sector

lements)

n

gaining

concessions and

special

treatment

n

thedevelopment

of

social

security ystems.

Special

interest

groups have,

in

fact,

been so successful that efforts

t social

securityreformhave failed in all but a few military egimes with sufficient

coercive

power

to

repress

interest

group protestsagainst the loss

of

special

privilege.Within few years after eru's 1968

military oup, the governmentally

established

and

manipulated organizations of

SINAMOS managed to attain an

44



18/33

POLITICAL PARTICIPATION IN LATIN AMERICA

independent nterest epresentation ole sometimesrepresenting he proscribed

parties). This phenomenon led the military eform overnment o reevaluate ts

commitment o mass participation n defense of its own policymaking Woy

1978, Dietz and Palmer1978).

Overall, then, the evidence indicates that various social sectors do, in-

deed, influence public policy, though elites seek to manipulate the political

activity f other groups to their advantage (Adams forthcoming, aylis

1978).

Thus, throughoutmost of Latin America, as Eckstein 1977) so graphically ocu-

ments for Mexico, policy concessions to the poor tend to be manipulative and

co-optative. Government responses to middle- and especially to upper-sector

pressures are

more

extensive, belying greater nfluence for these groups than

thatof the poor.

In

sum, the concentration f political participation

n

Latin

America appears no more pronounced than in otherculturalregions. However,

although nonelites engage in significant mounts of political ctivityn a variety

of

contexts,

heir

policy influence,while undeniable,

remains

slight

withinthe

context fnationalgovernment nstitutions.

PARTICIPANT RATIONALITY

The finalresearch question explores the issue of the politicalrationality

f

mass

publics

in

Latin America. A test of citizen rationalitymust evaluate primarily

whether

mass

politicalbehavior exhibitsgoal-oriented haracteristics. ne test-

able model of goal-oriented political behavior builds on the premise that to

behave rationally

ne must

be

sufficiently

riented toward the future o

have

goals,

must

perceive some political

nstitution

s

salient

relevant)

for

realizing

those goals, and then must act in congruence with those goals and perceptions

(Lane 1959, pp. 103-362; Czudnowski 1968; Touraine and

Pecaut

1970; Portes

1972, pp. 271-72; Almond and Verba 1965, p. 219). Thus, rationalmass political

behavior would consist of a pattern f activity n political nstitutionsn propor-

tion to

the

intensity f relevantgoals and to

the

perception

of

those

institutions

as

useful for pursuing these goals. Nonrational participationwould be either

random, ntense n the absence of goals and institutional aliency, r low

in

the

presence of high goal orientation nd salience.

One

study

based

upon

a

national probability ample of Costa

Ricans

(Booth 1977, 1978) tested the goal orientation-saliencymodel within hreepoliti-

cal

contexts.

Behavior

conformed o the means-ends

model of

political ationality

in

each case. Participationwithin the communal, local governmental, nd na-

tional arenas

occurred east

among

those

with ow

goals

and

who also failed to

perceive

the

nstitution

s

salient,

more

for hose with

eitherno

saliency percep-

tion

and

high goals

or vice

versa,

and

most

among

those

with both the

more

intensegoals and perceiving aliency.

Evidence

fromother

Latin American

nations

also sheds

light upon

the

questionofcitizen rationality. haffee 1975, 1976, 1977, forthcoming)heorizes

that

both

participation nd nonparticipation

n

politics result from

ndividuals'

calculation

of the relative

costs

and rewards of action.

He

argues

that a rational

individual will only expend an effort o influence the distribution f public

45



19/33

Latin

American esearch eview

goods

if

the

potential

rewards

outweigh

his/her

otential

osses. Chaffee

cites

several examples

of both

conventional

and

unconventional political

action to

supportthishypothesis.Neuse's (1'978) nalysisofthevotingpatterns fChilean

women

from1952

to 1973 reports rational action. Neuse attributes he rise

in

lower-class female participation fter 970 to a response to the determinedmo-

bilizationefforts f theUnidad Popular coalition. This shift pparentlyreflects

women's decisions to pursue theirpersonal goals throughpolitical nstitutions

because

the

parties

had

persuaded women

that

politics ould be of greater se to

them

than

before.

Other

studies, these fromauthoritarian ettings, support the apparent

rationality f patterns of mass political activity.Dietz (1974, 1977), Cornelius

(1974), and Cornelius and Dietz (1976) have examined demand making among

urbanmigrants nd squatters n Lima and Mexico City using a model similar o

the one

I

have

tested

see

Booth and

Seligson 1978a, p. 15, for discussion of the

Cornelius-Dietz model).

Their

findings

etail the

complex

tactics

hroughwhich

the

urban

poor

seek

government

ssistance

for he

provision

of needed

services.

These

rational behavior

patterns significantly

esemble those observed

by

Mathiason

(1972) among the urban poor of democratic

Venezuela.

Finally,

Por-

tes's (1972) surveysof numerous studies of thepoliticalbehavior of urban slum

dwellers

in

several cities (including Santiago, Guatemala City, Lima, Bogota,

and

Barranquilla)

ed him to conclude such

participation

s rational

see

also

Portes1976).

In summary, vidence from everal Latin American countriesrevealsra-

tional

patterns

f

politicalbehavior

n

many types of political ctivity. lthough

these findingsdo notnecessarilygainsay contentionsfrom he politicalculture

literature hat Latin Americans may be emotional, dogmatic, personalistic,or

paternalistic n their political behavior,7 they do cast some doubt upon the

common

conclusion that their participation n politics is thereforenecessarily

irrational.A more reasonable nterpretation ould recognizehow the complexity

of

human

motivation, ulture,

nd

the individual's social and physical environ-

ment

must

affectdecision making (Seligson and Booth 1976). Much further

research

s

requiredto illuminate nd explicate he effects f such factors n goal-

orientedbehavior and decision making. Further tudy should dwell upon im-

proving empirical tests of these models, taking nto account the multiplicity f

influences

n

the social environment.

DISCUSSION

To

sum

up,

the

available

evidence

provides roughly

he

followingpicture:

non-

violent political activity mong

citizens is

quite

extensive.

Though varying

n

-type

nd

degree

both

within nd

between

societies, participation

eems

neither

exceptionally infrequent

nor

widespread, given

the overall level of socio-

economic developmentof Latin America. There are several distinctmodes of

nonviolent

politicalparticipation, ery

imilar

o those observed

in

other ultural

regions, ncluding voting, partisan activism,contactingpublic officials,

iscus-

sing politics,organizational ctivity,

nd

community mprovement

work.

Varia-

46



20/33

POLITICAL

PARTICIPATION

IN LATIN

AMERICA

tions

n the socioeconomic

and

pQlitical

nvironment

eem to accountfor

major

differences

n styles

of

participation:

marginal sectors

such

as the urban

and

rural poor engage more frequentlyn collectiveproblem solving within the

communal

arena,

while the

urbanmiddle

and upper sectors,

ntegrated

nto

and

served

by the national

politicalsystem, take

a more active

part

in the national

arena.

Elitesmanipulate

mass

participation

o serve

elite goals.

Butdespite

the

fact

hatsome concentration

f


n the

hands of

socioeco-

nomic

elites

and political

activists ccurs,

and though

elites exercise

dispropor-

tionate

policy

nfluence, large proportion

f Latin Americancitizens

takes

part

actively

n

politics.

And finally,

atin

Americanmass publicsparticipate

npat-

terns

that suggest

rationality

f the

means-ends

type,

pursuing

their

political

goals

in

accordance

with theirperceptions

of the

usefulness

of

certainpolitical

institutionsor hosegoals.

Reassessing

odernitynd

Political

Mobilization

Politicalmobilization,

ong

portrayed

s a transition

romow

levelsof

participa-

tion to higher

levels,

has

been

centralto

the theory

of

modernization

Portes

1973,1974).

The first

ften

dimensions

of

modernity ha

booth, john

Documents