ALMEIDA, Mauro de - Rubber Tappers of The Upper Jurua River
ALMEIDA, Mauro de - Rubber Tappers of The Upper Jurua River
ALMEIDA, Mauro de - Rubber Tappers of The Upper Jurua River
1992
SUMMARY
This thesis studies the forest labour process of seringueiros (rubber tappers) in
the contemporary Amazon. It investigates labour processes from a Marxist
anthropological perspective, focusing on value and exploitation on the capitalist
periphery. The analysis is supported by an ethnographic description of contemporary
seringais (rubber estates) in the State of Acre, where I was born.
This work is organised in three independent parts. Chapters 1 to 4 constitute a
study of the local history of rubber estates and their interface with world and national
history. They deal with the the cycle of expansion and decline of the rubber trade on
the Upper Jurua region of Acre (1870-1943), the renewed prosperity of the extractive
economy in the post-war period (1943-1980) and the conflicts between rubber patrons
(patr6es) and tappers during the last decade (1980-1990). I conclude that the
contemporary rubber estate system was a product of regional Brazilian politics rather
than a response to the imperatives of the world economy. It developed into its present
form as a result of Brazilian State economic policies, which favoured and subsidised a
technologically stagnant regional elite in an area marginal to the world market. Another
conclusion holds that a forest peasantry with a highly-diversified local economy
developed on the contemporary estates. This forest peasantry possesses its own stakes
in the forest economy. It is not simply a proletariat forced to remain in the forest and
supply the world or national market by virtue of debts.
, Chapters 5 and 6 describe in detail the trade-post system and the debt system on
the basis of field work done on the Tejo River Valley. I describe the trade-post
institution as based on the monopoly of natural resources and of trade, supported by
state agencies, extracting rents and mercantile profits from a population of rubber
tappers operating independent economic units in the heart of the forest. I argue that .
system is unable to control the forest labour process. I also interpret debt relations as a
consequence of the extractive character of the forest economy and not as an imposition
of trade-posts.
.
Chapters 7 through 10 proposes the model of a forest house economy, including
Its social groups, its use of the natural resources, its labour process and its overall
working. Far from specialised rubber producers, the rubber tappers' forest house
economy is characterised in technical terms by the amplitude of forest niches they
occupy (including hunting, collecting and cultivation). The technological and social
patterns of this economy possesses ecological and technological characteristics that are
essentially different from non-forest peasant economies (" settler" economies in the
Amazon), and also from the large-scale productive units (fazendas). My argument
favours the inclusion of the tappers' extensive economic strategies in the forest as part
of a wider development policy.
.
To Manuela
TABLE OF CONTENTS
List of Maps, vi
List of Tables, vi
List of Diagrams, vii
Acknowledgements, ix
Glossary, xi
Acronyms, xii
Introduction, 1
1. The Rubber Boom in the Jurua Valley (1850-1912), 9
2. Crisis in the Rubber Market (1912-1943), 31
3. The State and the Rubber Trad.e (1945-1990), 45
4. Peasants and Patrons on the Tejo River (1980-1990), 69
5. Trade Posts, 95
6. Debts, Rubber and Merchandise, 131
7. Forest Houses and Persons, 169
8. Forest Houses and Nature, 199
9. Forest Houses and Labour: the Making of Rubber, 223
10. House Action Plans, 261
11. Conclusion, 289
Tables, 319
Diagrams, 357
Bibliography, 377
LIST OF MAPS
)_._1
~
W
-J
1.1
2.1
2.2
5.1
5.2
5.2a
5.3
5.4
5.5
5.6
5.7
5.8
5.8a
5.9
5.10
5.11
5.12
6.1
6.2a
6.2b
6.3
6.4
6.4b
6.5
6.6
7.1
7.2
7.3
7.4
7.5
7.5a
7.6
7.7
7.7a
7.8
7.9
8.1
8.1b
8.2
8.3
8.3a
8.3b
VI
Vll
JLe':
fLe.:
ACKNOWLEDGEMENTS
I give my thanks to Dr. Stephen Hugh-Jones who believed that I would finish
this thesis some day. My studies at the University of Cambridge and the initial field
research (1981-1984) were financed by the Brazilian National Research Council
(CNPq) and by the State University of Campinas (UNICAMP). A second research trip
(June-August 1987) was funded by UNICAMP while CNPq financed a second stay in
Cambridge (September 1990-February 1991). Campinas State University's ViceChancellor Dr. Renato de Sousa. signed a convention for technical cooperation with the
Conselho Nacional de Seringueiros; the Instituto Brasileiro de Meio Ambiente
(IBAMA) funded research under contract with the Upper Jurua Extractive Reserve's
Association (1991). Marisa Rodrigues, both as a friend and as head of faculty in my
university, was a great support.
I thank anthropologist Carlos Alberto Ricardo, a friend and one of the
Documenta~o
Informa~ao
(CEDI) , which
made mnumerable services and several airline tickets available to me. I thank
anthropologist Mary Allegretti for introducing me to the rubber tapper's political
movement in 1985. Thanks are given to the Conselho Nacional de Seringueiros and to
the
Associa~o
leaders and friends I only mention Chico Mendes. I should stress that the views
presented in this dissertation do not claim to express' the Conselho' s position on any
issue. Particular thanks are given to my Upper Jurua friends, among which I mention
Chico Ginu on behalf of all of them. Macedo became a colleague, a friend and a
teacher for me from 1988. During my visits to Rio Branco, I stayed at my relatives'
house, as well as with my friends anthropologist Terri Aqui and Vera Paiva; and in
Cruzeiro do SuI with Dona Alice, the Messias, the Dantas family, Antonio Macedo,
and Raimundo Cardoso. I thank Iram Rodrigues for his unfailing friendship and help.
ix
To Elizabeth Spalding I have a great debt for her help and friendship. I must
thank in particular Dr. John Monteiro for his invaluable help with the English, although
all remaining errors are of my responsability. Alicia Rolla kindly draw the fine maps of
the Upper Jurua.
This research and dissertation owe their existence to Manuela Carneiro da
Cunha, companheira. I thank my beloved daughter Luana, deprived of most of her
father's time for so long, always supportive. Final and deep thanks are given to my
father Guilherme (now deceased) and to my mother Maristela, the original sources of
my knowledge on the seringal.
This dissertation is the result of my own work and includes nothing which is
the outcome of work done in collaboration.
GLOSSARY
Aviamento
- Advanced supplies; also manioc-flour equipment
Aviar - To supply in advance or flado (aviador;aviado)
BalTaciio - Trade-post
Barranqueiro - River bank (barranco) cultivator
Batelao - A covered boat; as used by a regatiio or patrao
Baleeira - A flat-bottomed, covered boat
Caboclo - Indian or descendant of Indians
Capoeira - Abandoned clearing in a forest
Casa-de-farinha - Manioc processin~ house,
,
Colocac;iio - A settlement; set of trails and Its terrItory
Colocar - To settle someone in a colocac;iio
Embiara - Small game
.
Empanema~ - To make (so!Deone, something) panema
Estiva - BaSIC, bulky, supplIes for a rubber tapper
Estrada - Rubber trail; more generally, a road
Faca - Tapper's knife, and by extension a tapper
Farinha - Manioc-flour
Fiado - On credit; in confidence; credit system
Fregues - Someone who buys on credit; a customer; client
Jgarape - Small river
Lamparina - House la~p using kerosene
,
Madeira - Wood; speCIfIcally a rubber tree urnt
M an'eteiro .- A petty retail trader in a seringal
M ata bruta - Virgin forest
Mae-da-cac;a - Mother-of-the-game; a forest entity
,
Miie-da-seringueira - Mother-of-the-rubber tree; a forest entIty
Novena - Annual nine-days of rosary; associated fairs
Panema - Condition preventing success in hunting
Paranii - Small river with flooded banks
Patrao - Trader who advances merchandise; a boss; owner
Pela - Rubber ball of 50 kgs or more
Poronga - Forest lamp
Prancha - A rubber plank of 10-12 kgs
Principio - Rubber bundle of small size as sold to regatao
Rancho - Food, especially game
Regatiio - Itinerant river trader
Renda - Payment for rubber trails; rent
RoC;ado - Manioc garden plot; any cultivated area
Seringa - Rubber tree
Seringal - Rubber estate; zone under an owner or trader
Seringueiro - Rubber tapper
Saldo - Positive credit balance ("borracha de saldo")
Terra firme - River bank which is never flooded; upland
Terreiro - Back-yard
Tingui - To fish by the use of poisoning
Ubci - A flat-bottomed one-trunk canoe
Vci17ea - River bank subject to flooding; floodplain
Vereda - Trail habitually used by a wild animal
Vizinhar - To share game with a neighbour (vizinho)
Xl
ACRONYMS
ASAREAJ - Associa~ao de Serin~eiros e Agricultores da Reserva Extrativista do
Alto Jurua. Upper Jurua Extractive Reserve Association of Rubber Tappers and
Agriculturalists.
r AP 1 - BRA Z ILI AN
AMAZ ON
,-
"
(
(
I
I
.I
---/
I
I
I
I
\
J
I
PE RU
I
".I
I
,J
- I
--
Xli
JLe.:
I
i
MAP 2 - UPPER
--~--------~~---
....
..... .....
-- --
/'i-:::'::=-=
~-:::./
BR-364 -
.. :,.V
'l' ...
... /-.
.. ' 'ir
..!:?
...... .......... .
--
---
-:;tv
..
-'- ~ '
....... ..'
.".-~;:::-
~-,..-
;::::::./':
".
.. .. ..... .......
. .....
.' '
". ' . .
..".
.'. ... .. ". ' .. ..
e
.... ".
". ' .
'.
'"
"
:.;
....:
--~------
"................. ~ ....
-- -.
LEGENDS
Boundaries of main
Ru bber States
.. ......
..
I
"...
........ ..
.
".
..
I
~~~---.- ~.~~~~.--~--~~
'.
""..
""..
"".. ....
Town
Unpaved rood
'"
Planned rood
.L.____
".
""..
'.
- .-. - - .
'. ".
.-..
. ..... "y"
"1,
s
C")
SAGE
S.
MANTEl .. "
FORTALEZA
-'.
LEGENDS
IRACEMA
1ft
'ti!'
tt
I g . M onft!igo
rt'
Stream (Igarape)
Settlement (Colocacoo)
s. CAM"lEAO
S. BO" HOR"
1 \
1 1 I.
JJ
LI L I~ I~' I ~ I~
MANTEIGA
\'",
\
\
,,
"
,,
\
\
\
~-.(
I
I
I
,,
,,
\
,,
,"
\\
"
" ,,
", "-
" a
" 0
't:I/~
Trade Post
, '"
,.
LEGENDS
/;.'
\
TACARATU
SETTLE MENT
......
,
\
---
MATO GROSSO
\'\
---,,""5
\
cHouse
Po t h
,,
,.,-"
~ Rubber Trail
--- -- --/
I
I
........
",""
"
,.. "
.......
\ \ J 1
\ I
I I~ L' I~ \~
:':.
'b.
:A
:A
'.A
..
....
A:.
.... A
..
A'
A..
/:. ........'
'.
.'
"........ :.
'.A
..
:A
: . A
.........~... .
....... AA
.A
.'
A
/i.':
..
A ..... ,:
1.
..''
"'A
'. ......... -
e'
"
:A.
6.
I>
.'
....:
,A
: to.
A.
: A.
: ..
:A.
'.
...';..... :,:
'
.....
A.
A
A
...................
.. '
.... .......
...... .. -:.
......
[A
'-.1>
COLOCAcAO TACARATU
:1:>.
.:A
.... A
A.
~
-='::":i to.
i..
b..:'
A
.:-.A.
b ....
.~... .'
i,: ."l.
.'.:'/:..
... :
..
1:>.':"
fA
'
.... ....
A'
..:'A,
Seringueira
lOOm
:-A
'
."':
b."
'b..
.. ;.: ..A:
:
A.
.. .....
.......
A.'.: A
A' .'.
~:
b.,'
'.
A.' ...
A
A ..
' A ;
A
:
...~"'.
A A
.b
.: A ',/:.:
:.1>
.0.
;A
A (
'"
;." A"i,.
.:"'..'.e.I:>.
.... A
... A
"A.
INTRODUCTION
Introduction
This thesis has the objective of studying the forest labour process of
seringueiros (rubber tappers) in the contemporary Amazon. The motivation for this
study was my interest in investigating labour processes from a Marxist
anthropological perspective, focusing on value and exploitation on the capitalist
periphery. Following the decisive incentive of my supervisor, Dr. Hugh-Jones, I
combined this interest with an ethnographic description of contemporary seringais
(rubber estates) in the State of Acre, where I was born.
The continued existence of these rubber estates constituted something of an
enigma. Why has a native forest extractive economy, based on so-called debt slavery,
been maintained on the periphery of the Brazilian Amazon for over a century? How
come seringais still exist on the eve of the twenty-first century? Between the original
research and the final write-up of this thesis, other enigmatic issues emerged.
Following Chico Mendes' murder, the rubber tappers became known world-wide,
and for some time after that event were promoted as heroes who defended the
- I
rLe.!
2
3
process itself had been described as brutalising, leading more than one writer to the
interpretations for the. discussion of these issues. Thus, the analytical focus on
conclusion that the true function of debt slavery was to force rubber tappers to
labour processes and social relations in the forest economy, combined with a
accept a labour form from which they spontaneously would flee if given the chance -
fleeing not to become independent forest workers, but to abandon the forest
understanding of the alternatives for the future development of the forest economy.
Summary
the residues of the extractive economy, which was seen essentially as a large-scale
predatory system whose whole raison d'etre resides in the transfer of information
a study of the local history of rubber estates and their interface with world and
and energy from the periphery to the metropolis (Bunker 1985; Almeida 1990a,
national history. They deal with the the cycle of expansion and decline of the rubber
1992; Hornborg 1992; Alier & Schlupmann 1991:34). The peasantisation of the.
trade on the Upper Jurua region of Acre (1870-1943), the renewed prosperity of the
rubber tappers was treated only as a recent and transitional phenomenon (Bakx
extractive economy in the post-war period (1943-1980), the conflicts between rubber
1986).
patrons (patroes) and tappers during the last decade (1980-1990). I conclude that
The rubber tappers' political movement in the 1980s introduced a new
'
the contemporary rubber estate system was a product of regional Brazilian politics
element to this picture. Many tappers not only remained ' in the forest immediately
rather than a response to the imperatives of the world economy. It developed into .
following the patroes (rubber estate owners and traders) exodus from the area, but
its present form as a result of Brazilian State economic policies, which favoured and
also planned to stay there in the future - composing a scenario marked by the
continuity of the forest economy in areas of the Amazon occupied by rubber tappers
world market. Another conclusion holds that a forest peasantry with a highly-
crisis, as rubber tappers who were expelled or under threat of expulsion from
peasantry possesses its own stakes in the forest economy. It is not simply a
extractive areas expressed their rights to forest territory (Bakx 1988, 1990).
proletariat forced to remain in the forest and supply the world or national market by
}Iowever, the
begi~ng
virtue of debts.
subsidies, which in the past had been pocketed by estate owners and rubber
Chapters 5 and 6 describe in detail the trade-post system and the debt
system on the basis of field work done on the Tejo River valley. I describe the trade-
population of r~bber tappers operating independent economic units in the heart of the
sustainable development (Hecht & Cockburn 1991). This thesis provides data and
forest.
I argue that this system is unable to control the forest labour process. I also
including its social groups, its use of the natural resources, labour process and its
overall working. Far from specialised rubber producers, the rubber tappers' forest
~haracterised
they occupy (including hunting, collecting and cultivation). The technological and
social
patterns
of this
economy possesses
ecological
and
technological
"Le.
This thesis is based especially on three kinds of sources. The first is direct
barracao (trade post), and on many occasions studied accounts and debts in detail
observation during field work in the municipal district of Cnizeiro do SuI, Acre,
with the house heads. In my short stays at the trade post I tried to obtain additional
accounting material and observed the transactions at the shop. These data are used
1983; and June..;August 1987. The second involves my volunteer advisory activities -
in Chapters 5 and 6.
"
for the National Council of Rubber Tappers (from 1985 to 1988) and for the Tejo
River Rubber Tappers' Association (from 1989 to 1992). Third, I have relied on my
do SuI, the Center for Historical Documentation of the Federal University of Acre
(in Rio Branco), the Cambridge University Library and the British Library. Other
- I
Cruzeiro do SuI, and particularly on the Tejo River, resulted in two household
surveys covering the Riozinho area of the Restaura~ao Estate (a sub-estate of the
missionaries had frequented the Upper Jurmi region since the late nineteenth
Tejo River's major rubber estate). During the period of research, the Riozinho
century. This material was the basis for Chapters 1 and 2. Chapter 3
national policies relies on different sources such as official statistics and reports,
November 1982, and during those same months a year later, applying a
laws and newspapers. Chapter 4 includes information obtained in my 1987 trip and
in several visits every year since then (ranging from weeks to months).
on
I could not use all of the material obtained. I kept a diary on labour and
~onsumption habit~,
hosts on their work outings. I had to draw maps, make measurements and take photographs. I employed published statistical data from the mGE (the government
statistics agency) as well as data from the RADAMBRASIL project, covering the
Brazilian part of the Upper Jurua region. Early research (1982-3) focused entirely
on the daily life of rubber tappers. One basic difficulty lay in the fact that tappers'
houses form groups of two or three on a settlement, linked to other settlements by
paths involving a two or three-hour walk through the forest. Since I adopted the
practice of studying work activities by participating in them, it became difficult to
make casual visits to other settlements, except on weekends. Settlements were quite
different from one another in their basic economic orientation. It became apparent
that survey data based on questionnaires afforded only a normative and idealised
picture of activities~ which appeared far more varied and complex from the vantage point of direct, personal observation. Survey data revealed patterns and plans.
Observation as a participant revealed actions and processes whereby plans
interacted with circumstance, and varied from house to house. Thus I faced a
dilemma. To remain only a short time on each settlement meant a rupture in the
-'
continuity of work patterns whose minimum dimension should be at least a year; but
to remain too long on a single settlement meant accepting a single pattern as
general, within the many possibilities manifested by different houses and
settlements. Instead of one location viewed by a single observer, there was a wider
group of 25 micro-locations forming a network covering around 23,000 hectares.
Under these circumstances, where the focus was a settlement where I became
"employed" and had to assist the head in all the activities he undertook, the finalresult was a discontinuous sequence of observations and flashes of different styles.
But the data could not be aggregated as a statistic for a "typical" house. The
~egions
0f
projects and
reports (usually helping out Antonio Macedo); research for the Tappers'
[Le.
researched in 1982, 1983 and part of 1987 and then visited by me every year became
the first Extractive Reserve, by the Federal Decree of 22 January 1990, and its land
was officially expropriated on 15 January 1992.
Introduction
The formation of a popUlation of rubber tappers or seringueiros, which
emerged with th"! institution of the seringais or rubber estates, passed through
several phases in the Jurua Valley. The first phase, prior to 1870, may be
characterised by the commercial extraction of rubber by the Amazonian Indian
popUlation; in the second phase (1870 and 1912), known as the rubber boom, the
extractive frontier advanced rapidly in the direction of the Upper Jurua River,
A-
through the territorial expansion of the extraction zones and the migration of rubber
~ .l,
tappers under a system not unlike indentured labour; in a third phase (1912 to
1943), the rubber estates survived as enterprises oriented towards a regional market
J)
vJ
-J
and a stable population; in a fourth phase (1943 to 1985) Amazonian patr6es (patrons, estate owners and commercial agents) became the political and economic
clients of the national state; finally, the period beginning in 1985 and extending to
the present day has been marked by the exit of the patrons from the Amazonian
scene and by the concurrent emergence of rubber tappers' organisations fighting for
land rights and proposing alternatives for the development of the forest economy
(Table 1.1).
river was visited by river traders from Ega sought to acquire especially turtle wax,
sarsaparilla and Indian slaves in their expeditions up the Jurua (Castelnau,
1851:87,128-9; Osculati, 1854:239; Herdon 1853:190,284), relying on indian labour.
10
11
Thus, by 1850 traders already had reached the Middle Jurmi all the way to the
river banks to acquire turtle wax and manatees, and pirarucli fish; and they go to the
Tarauaca River, exchanging commodities with the Indians for cocoa, sarsaparilla,
forest to extract sarsaparilla, copaiba oil, cloves, etc.; which in effect Inight be called
vanilla and copaiba oil. The dry season also was the period for rubber extraction,
and this product found its way back to Tefe between November and December,
when the rains filled the river floodplains (Castelo Branco 1922; Tastevin c.191922)r However, the labour supply was constrained by the limited size of indigenous .
populations, which became smaller as a result of resistance, flight and disease
(Oliveira Filho 1979:126; Ribeiro 1970:42-7; Cardoso de Oliveira; Lazarini 1981:v),
while the demand for rubber grew steadily between 1850 and 1870. In 1853, dried
pirarucli fish led regional exports in value (7,000), followed by sarsaparilla. Rubber
occupied a modest fifth place, with a total value of less than a thousand pounds
sterling, less significant than tobacco, copaiba oil, Brazil nuts and mixira (turtle wax)
(Tavares Bastos 1975 (1886):134). But by 1855, latex already had climbed to second
place, surpassed only by pirarucli, and by 1857 was the ~ost important article in
terms of export earnings, generating over 13,000 per year (Bastos 1975 [1886]:134).
In 1863, rubber accounted for almost a third of the total value of regional exports
(half a metric ton, worth 50,000), growing to over one thousand metric tons (worth .
130,000) by 1865 (Tavares Bastos 1975 [1886]:134-5, 136,229,233-4).
However, the Indian population 1 was insufficient or unwilling to supply the
required number of workers, and the caboclo frequently was described as averse to
disciplined work, preferring seasonal collecting activities and the sale of forest
products to itinerant peddlers (Moran, 1974; Parker, 1985). For example, in 1852
Captain Araujo e Amazonas complained that the inhabitants of Barra (today
Manaus) "spend the summer in what they call work; which they do by going to the
. 1. The Cata~ of the Juru.a we~e a~ong the indigenous peoples who
supplIed rubber to flver traders dunng thIS penod. For the trade in Indian quasislav~s on the Purus see Polak 1894:iv; The South American Missionary Society, 1876
paSSIm, esp. Vol.1O, 1976:234-8; Clough, 1872; Chandless, 1887:299-301" .
12
13
Amazon (Bastos 1975 [1886]:passim; Coelho 1982:24). This contributed to the mass
influx of peasants from the northeastern province of Ceara, whose backlands were
Pano-speaking groups inhabiting the upper course of the Jurua Basin represented
punished by repeated drought crises between 1877 and 1880. In 1878 alone, 54,000
peasants moved on to Amazonia, while some 50,000 others perished in misery in the
were removed from the rubber extractors' path, often exterminated by killers who
deposit~
wheat flour and everything that may be found in a grocery store; "everything that a
civilised man might need" (Tastevin 1920:136; cf. 1925a:413-22; 1926:46-49;
1928:213-15).
These descriptions suggest that by 1911, just before the collapse of the
market for natural rubber, the Tejo river had become a centralised trade area with
investment in transportation, supplies, opening of trails and supervision of tappers5.
The fact that the Tejo was one of the most distant rubber zones suggests the high
profits generated by the rubber business. It should be noted, however, that the Tejo
river also attracted rubber tappers because of its high natural productivity.
According to a fl!:...bber tapper who penned his memoirs, a single tapper could collect
anywhere from six to ten kgs. of rubber per day in the Manteiga and Riozinho, while _
in 120 days of annual work could produce a ton, which was reason enough to
overcome his fear of Cashinahua and Catuquina Indians (Cabral, 1949:35-42)6.
These figures are still valid this area, which is were I did field research. Cabral's
memoirs describe how In 1899, after working as a tapper in a middle Jurua estate,
network. During the rubber cycle, the Tejo area was cut by many mule trails, thirtyfive of which still existed in 1920. The Tejo rubber estates were supplied and
managed from their headquarters at the mouth of the Tejo, described in 1905 by
Mendonc;a as having "great warehouses, an excellent office and around twenty
houses, along with cattle" (Mendonc;a 1907:110;27;194). According to Tastevin, the
Tejo headquarters had houses for the employees, . namely the fishermen, the
mateiros (which he first says are "in charge of supervising the rubber tappers" and
he followed his brother with five men to work in the unoccupied Riozinho and
Manteiga estates, authorized by the local owner. They were on their own to build up
their barracks, open up new trails and fight the neighbouring Indians (Cabral
1949:43-49). I proceed to add more pieces of information of the characteristics of
the upper Jurua rubber zone.
Limitations in Land Ownership and Political Power
In 1903, the eastern Acre territory was formally incorporated into Brazilian
Mello & Co. personnel boat, at the place later known as vila Thaumaturgo. Thewhole western Acre became then the Upper Jurua Department, directly under
federal rule.
The federal status of the Upper Jurua Department clashed with the interests
both of Manaus and of local merchants. The new and wealthy rubber zones, linked
to Manaus and BeIem by steamship navigation, were claimed by the state
government of Amazo~as.
~or
sovereignty (Stokes 1974). At the same time, in the western Acre area which was
install an independent unit of the Brazilian Federation, separate from the state of
claimed independently by Peru, caucheros entered -in conflict with the rubber
Amazonas, but endowed with its own political and administrative order. Instead, the
tappers in the Am6nia river, just as few miles downstream of the Tejo river. The
federal department system, modeled after the federal lands of the United States,
caucheros were nomadic extractors who covered a determined area in search of the
meant that neither the local patron elite nor the regional elite of Manaus, who until
Castilloa elastica tree, which they struck down in order to obtain forty to fifty kgs. of
1904 ruled de facto over the upper Jurua area, were to control decisions regarding _
caucho latex all at once, producing in this fashion a total output of between two and
four tons per season (Pennano 1988:53). Caucheros never settled permanently,
This situation stood in stark contrast to the images of greatness that the
disappearing once and for all from the Upper Jurua after the boom years. 7 The
prospective rubber tappers, it was a city greater than Manaus or Belem, a place
yearly output around half a ton rubber, received regular supplies and constituted a
where any and everyone could strike a fortune. Exaggeration apart, the area did
fixed population whose boundaries were limited to the natural distribution of Hevea
brasiliensis.
town of 3,000 inhabitants in 1912 (Azevedo 1905; Tastevin 1913). In the urban area
of Cruzeiro do SuI, a monumental, thirty metre wide boulevard was projected. The
do SuI which he founded at the mouth of the Moa river. In the same year, the
Peruvian garrison at the Amonia river was defeated with the participation of the
lodge and a chapel, and from 1906 had a newspaper which defended political
autonomy for the region (Revista do Instituto Historico e Geographico Brazileiro,
1908, 1:48). In addition, there were a steam-driven sawmill, two brick and tile
factories, electrical power, an ice factory and about 150 commercial establishments
belonging to Brazilian and "Oriental" (Lebanese, Greek and Jewish) merchants. The
3,000 tons of rubber that left the Acrean Jurua annually allowed for the importation
18
of foodstuffs, bottled drinks, clothing, tools, medicine and firearms, though with a
deficit. No wonder the Cruzeiro do SuI elite became irritated by those of Rio de
Janeiro who treated the town as if it were a mere camp in the forest.
During the boom years, in fact, rubber exports produced revenues
comparable to those generated by coffee exports and Acre was the richest source ofrubber. Local patrons complained that federal taxes were excessive. They also
alleged that the distance separating them from the federal capital of Rio de Janeiro
made it impossible for efficient federal administration to exist. In 1910, the patrons
rebelled, pleading independent state status within the federation. But the federal
government responded by sending troops and stifling the uprising that same year, as
reported by one of its leaders in a book aptly named "The End of the Epopee"
19
for a small fee, but since what really mattered in the early days was the effective
occupation of rubber . trails and not the legal ownership of the areas, the estates
expanded with no concern for boundaries or areas (Magalhaes 1977; Araujo 1910).
After 1904, the federal government did not create mechanisms for the formal
acquisition of land by patrons, or to validate titles issued by the Amazonas State at a
time when the region officially belonged to Peru. Thus, according to Antonio Jose
Araujo, a lawyer who worked in Cruzeiro do SuI after 1909:
All the land in Acre is devoluta (public), it is not legitimate property, nor can
it be legalised for lack of laws to this effect, and it is kept under the power ofa few individuals only under the status of posse (claim) ... while the private
individual does not own land, which belongs to the Union, or rather to the
(Costa 1973[1924]).
Territory (of Acre), on the other hand there is not one square foot of land
The absence of local political institutions had important effects on the land
Issue. The land, in practice forests that were rich in rubber trees, became
monopolised by the patrons; in a legal sense, however, this monopoly was not
bolstered by legal titles, with the exception of deeds issued by the State ofAmazonas before 1904. In the early years of the boom, in order to take control of a
rubber zone, one needed simply "to occupy the area, to erect whatever building one
wishes, and to exploit it": as long as no one challenged this form of appropriation,
the occupant might remain decades in the area, according to a contemporary
(Libermann 1897). Only the frontage areas along river banks could become disputed
areas. A rubber estate, in effect, was a simple "assemblage of rubber tree trails", the
8. In 1913 the federal government set a deadline. of thre~ years for the
urchase of land - a solution refused by the patrons then III the mIddl~ of a d~ep
~conomic crisis (Rego Barros 1981:153). Additional sources on t?e l~nd Issue ~atIllg
from 1907 are summarised in Gon~alves (1991:35-36), confirmIng the pIcture
proposed here.
20
21
lawYer Araujo) ~as' that a category of independent rubber tappers could emerge.
Parissier:
Another effect of the federal rule was that the patron's capacity to coerce the
".. .in order to pay his often fictitious debt, the tapper is forced to make yet
rubber tappers by violent means was apparently restricted at least at some moments
another season at his patron and extend thus indefinitely his slavery, and if,
during the boom. One of the issues was the trade monopoly sought by patrons,
"owners" of rubber territories. In fact, during the rubber boom period a group of
were a robber, and he will not be able to find any employment elsewhere."
small, itinerant merchants operated throughout the area with river craft of one to
(Parissier 1898c:83).
five tons, propelled by oars or poles. These reg~t6es (peddlers) made up for a
Parissier asserted that, in theory, the rubber tappers could payoff their initial
shortage of capital through high turnover rates and through their specialisation in
expenses and their annual purchases at current prices, but most tappers remained in
patrons certainly must have sought to undermine these itinerant peddlers. One
indication of this emerges in Thaumaturgo's second decree of 1904, his "River Law",
in the lower Jurua, because "nearly all the able men go to the seringais, that is to
inspired by the "constant complaints against the restraint exercised by owners who
say, to collect rubber, because this is better paid than whatever other job" (Parissier
judge themselves endowed with exclusive rights over the [water courses of publi y
1898c: 168). There is thus an apparent contradiction between the fact that debts
were the means to impose abusive prices and thus exploitation, and the fact that -
'
the itinerant peddlers rights to navigation and commerce, subject however to ' the
Alfredo Lustosa Cabral, who worked on the Tejo headwaters in 1989, writes
proviso that itinerant traders proved to the patrons that they had not sold any
in his memoirs that in the rubber estates in the middle Jurua and in the Amazonas
in general, debts were impossible to payoff, the rubber trees's yield was 150 to 200
This point suggests that there could be rubber tappers able to pay their debts
kgs annually (2-3 kgs. a day), and the trade was strictly monopolistic. This was called
and still purchase from itinerant traders or to itinerant priests. 9 The generic image
the "toco" system, . were all rubber belonged to the patron to payoff debts. In
contrast, in the "upper river rubber zones" such as the Tejo headwaters, "the rubber
9. Father Parissi er wrote that the best time to receive contributions to the
chur~h.rell between October and Janu~ry, after the tappers settled their accounts
(Panssler 1922:113), and Father Tastevlll confirms this opinion indirectly when he
asserts that the J?atrons did not approve of visits to the church during this period of
the year (Tastevlll 1914).
iLC._._
tapper was free, he had a right to extract the rubber on his own account" (Cabral
10. On the possibility of positive balances and the consumption of
"superfluous" items, see Walle (c.1911:65, 148, 173); Wagley (1964:87, 173).
22
23
1949: 118). According to Cabral, in these places owners leased a pair of trails for 66
remained in debt or enjoyed a positive credit with his supplier. Back in Acre, the
kgs rubber and the tapper purchased his merchandise to the patron or to any other
patron then assessed the value of each tapper's rubber (at the current market price)
seller (Cabral 1949:118). This suggest that there were different trade arrangements
and deducted the tapper's debts from this total. The sale was thus "at the risk of the'
which were applied in different zones. Indeed, according to Woodrofe, in one such
system, "all rubber must be sold to the proprietor who makes it a condition that he
may pay in goods or cash"; in the other system, "the rubber must be delivered to the
direct supervision was nearly impossible, implied the possibility of positive balances.
proprietor who sends it to the local market, paying all the freight and other charges,
Even a small debt was sufficient to force a tapper to remain on a rubber estate until
deducting for himself 15 per cent as a commission and paying the remaining 85 per
he settled it, but a small positive balance never was enough to buy the year's
cent of the net proceeds of the sale to the seringueiro" (Woodrofe 1916:56-7). This
supplies, from either the same patron or another, without contracting new debts.
Nonetheless, a positive balance allowed for the tapper to buy at least part of his
supplies from an itinerant peddler or even from a different patron. On the other
lead
. .
The extractor leased two trails for which he paid the rent of 60 kgs rubber.
The surplus was destined either to the payment of his debt towards the
the patron. There were frequent complains on the part of patrons against such
evasions and in particular on the lack of action on the part of federal administration .
parentheses mine).
the rubber estates, at times spanning entire rivers, facilitated the control over
The point in this seemingly too subtle distinction was that in the second case
exchange, making it difficult for tappers in debt to leave the estates or sell their
the rubber, while "delivered to the proprietor", remained the tapper's property, a
system known as the "embarked rubber" system. Older tappers and trade-post
employees describe the workings of the latter system as follows. At the beginning of
movement on foot along the trails deep within the estates could hardly be controlled
each year, individual patrons received merchandise on credit from suppliers in '
in the same manner, and the backwaters were out of reach of the trade post
Mahaus or Belem; in turn, they supplied the rubber tappers at the outset of the
personnel. 11
Mendon~a
1989 [1907]:227,
However, the
season. At the end of the season, the tappers turned over their rubber to the
The contemporary sources confirm the common knowledge that patrons used
patrons, who shipped the product once a year to Belem or Manaus. There, the
routinely of violence against tappers and agree in describing the rubber patrons as
rubber (which carried the initials of each tapper who had extracted it) was classified
and priced accordingly. The patron then settled in rubber (at current prices) the
merchandise he had received in advance the year before. At that point, he either
11. On flight and other forms of resistance, see Reis (1953:95); Levi-Strauss
(1978 [1955]:484); Wagley (1964:92),; Allegretti (1983:49 ff).
Castelo Branco wrote that during the first ten years of occupation (presumably
Oral traditi~n records stories of estate overseers who ambushed and killed
1895-1905), the rubber tapper was a "true slave, subject to severe punishment,
rubber tappers who had come out ahead in their accounts and left the estate with
including whippings" (Castelo Branco 1922). However, Thaumaturgo says in his first
money, and other brutal practices which are also part of oral tradition include the
report as a federal authority in Cruzeiro do SuI that he arrested some of the "most
torture and murder of tappers, particularly on "lower" rivers such as the Gregorio.
Owners could prohibit the cultivation of garden plots, forcing rubber tappers to
purchase foodst~ffs as part of their debt, and did not allow the tappers to burn and
may well be that Thaumaturgo's "Labour Law" prohibiting the use of violence on the-
part of patrons against tappers and peddlers (Decree number 15, December 1904;
Thaumaturgo de Azevedo 1905:188-194) did not remain only a dead letter during
-,
some of the administrations which followed. Thus, writing in 1910, the lawyer
in a country where no policy is possible and the law of the mighty rules
absolute, the caboclo never goes into the forest, or even steps outside of his
house, without his rifle ... when they aim at the patron, they may encircle the
request the government to force tappers "who desert the estates" to "return to their
trade post, kill whomever they can and put fire to the shed. This is what
posts until they settle their accounts" with the argument that the patrons had
happened to Mr. Bonifacio ... who has escaped but for a miracle, but who
invested capital in each rubber tapper (Araujo,191O:103-4), being at the same time
(Parissier 1889c:68).
On the other hand, the rubber tappers themselves were capable of reacting
13. "Does the patron not su~fer a setback every time t?ey; (the w?rkers),
following bad counsels, abandon theIr shanty and the rubber traIls m the mIddle of
the night in quest of other places where they might escape from the debt they owe?
The '44,' regime has been substituted by the regime of 'the law' (federal
administration). Instead of tracking down fugitives, the patron must track down the
authorities. But the authorities tell him that in our country and under the
government under which we liye, one may not coerce .a citizen to :work ag~inst ~s
will . . . Thus one can explam one of the most senous accusatIons agamst HIS
Excellency Dr. Bueno, present prefect of Alto Jurua". Araujo, 1910:104.
14. "The hands represent ... an appreciable inversion of capital ... certainly
it is a loss both in the patron's labour and in the patron's capital, the fact of (a
rubber tapper) falling ill, escaping or dying" (Andrade 1937:105). ".,' . it is the owner
who introduces the workers. He makes large advances and takes nsks to lose them
by the constant escapes, diseases and deaths" (Araujo 1910:103).
As mentioned above, "Mr. Bonifacio" was an estate owner on the Tejo River
and boss at the main trade-post. The evidence from other reports suggests that such
violent uprisings continued to occur. In 1908, a report sent to the federal
administration by Bueno
was widespread and many conflicts occurred between patrons and rubber tappers as
a consequence of a fall in the price of rubber"(apud Gon~alves, 1991:28). Writing in
1913, when the tappers were again "revolted" with the fall of rubber prices, Father -
15. Fishing and hunting were tolerated, be~~use they did not necessarily
compete with extraction activities. Guns, ammurutIon and salt were advanced
(Panssier 1898b).
26
III
1914).16
27
settling their accounts, but also with violence. The patron-client relationship was not
that of the all-powerful estate owner and the passive rubber tapper (Hecht and
Cockburn 1989:165; Almeida 1984, 1988). The conclusion is that the working of an
engendered by world capitalism, not only transferring surplus value to the centre but
also transferring raw material and information, degrading nature in the process
(Wolf 1982:326-9; Bunker 1985; cf. Kahn 1980:204). However, it can be argued that
extraction activities in the Amazon did not create enclaves as in the case of guano in
throughout Amazoiria during the boom phase. As the previous sections should have -
Peru, nor were they vertically controlled from the centre -- but rather by trade
suggested, the Hevea brasiliensis rubber estates of the Upper Jurua differed
companies as
& Co. --; and that workers resisted spoliation and monopoly
significantly from areas like the Putumayo, in ecological, political and social terms.
with varying degrees of success (Weinstein, 1983:14-23). The existing evidence for
Ecologically, the Putumayo was a Castilloa elastica zone (rather than a Hevea
the Upper Jurua, the westernmost frontier of Brazilian expansion led by rubber
brasiliensis zone); the Hevea brasiliensis of the up river areas could be sapped
Melo l
During the boom years, the rubber estate system required vigilance and
was recruited from the local indigenous population, which was treated as a natural
coercion for three reasons: to ensure the patrons' commercial monopoly; to impose
. resource much in the same fashion as the Castilloa elastica, while in the upper Jurua
zone labour was imported with high costs and required skills to sustain productivity.
restricting the free movement of rubber tappers. But in the upper Jurua patrons did
Furthermore, unlike other areas, the northeastern migrants and the patrons in this
not have a full support on the part of federal authorities, who stayed for short
periods each and were sent directly from Rio de Janeiro. Estate owners faced the
disputed by Peru and Ecuador, a sort of no-man's-land where the patrons wielded
over frontier resources, and the difficulty of controlling the movement and activities
absolute control, a fact that placed it alongside the Belgian Congo as terror zones
of workers within the isolated forest, both due to the dispersion of persons over
due to denunciations of atrocities against labourers (Taussig 1987). The upper Jurua
huge areas and due to limitations imposed by federal government. On the other
was from 1904 within direct reach of the Brazilian federal administration which
hand, rubber tappers reacted sporadically, not only leaving the estate without
stifled local uprisings both in western and in eastern Acre during the boom.
Economically, while areas like the Putumayo developed as enclaves of
British capitalism, the rubber estates of the Upper Jurua and Purus were financed
by merchants who were not directly subjected to British capital (as emphasized by
28
Weinstein 1983). indeed, the main reason behind the "Acre Revolution", when -
hypotheses suggest that the point was not necessarily maximizing net profits at the
eastern Acre rubber bosses seceded from Bolivia, financed by Belem and Manaus
local level (as a bala~ce between imports and exports), but obtaining a maximum
merchants and backed by the governor of Amazonas, was the Bolivian project of
turning the area over to American interests, which would assume direct
administrative or military control over the area (Stokes 1974; Tocantins 1961). The
patrons remained themselves in constant debt -- thus under the obligation to sell to -
upshot of the "revolution", however, was the installation of federal and military
the Belem trader only. Patrons faced no great pressure to cut back on expenses or to
administration from 1904, against the wishes of the local patron elite, whose leader
increase efficiency. Local patrons remained constantly in debt, but debt meant that
in eastern Acre, Placido de Castro, himself a rubber patron, was assassinated during
the prolonged conflicts with federal administrators (Prefeitura do Alto Acre, 1907;
Costa 1973 [1924]); AssembIeia Nacional Constituinte, 1946:passim).
Flight, attacks on trading posts, the consumption of items not reduced to
--.I
"subsistence", the pervasive presence of river peddlers and the possibility of positive -
demand continued to rise, prices were pressed upward, because supply was never
balances all suggest that the system was characterised by a struggle over distribution,
great enough -- neither the accessible natural areas nor the labour supply proved
and not simply by the automatic pumping of a maximum surplus. The patrons'
monopoly was based on the fact that each tapper hypothetically was specialised in
at least at the
. investments, after deducting expenses. Anything less than that would entice the _
activities and for consumption. Under these ideal conditions, a perfect monopoly
tapper to abandon the rubber estate, with or without debt. The most productive
could establish virtually arbitrary prices, provided that patrons could had complete
rubber estates such as the Tejo river would generate extraordinary profits. In order
to appropriate these returns, local patrons themselves increased supplies and thus
On the other hand, such system would not necessarily increase productivity,
We might imagine the following model for the boom period. Rising prices
which was historically low (less than the current average of 400kgs/year) in the
Amazon. A system based at least in part in incentive could more easily exploit the
differences in natural productivity or individual ability, given the difficulty ofmonitor directly workers dispersed at the rate of one per 300 hectares of tropical
forest. The locus of violence would thus be the monopoly itself, trade process not
the productive process; nor real incomes at the "minimum subsistence" level were
required, provided that in the average most of tappers were in debt. These
17 "The aviador patron needs, say in June, 500 million reis. Th.e exporter
advances 'him the sum ... The patron... must turn over ~b~er ... at a pnce already
established by exclusive buyers. The shipment is t;Iot sufficIent to pay the de~r Th~
e orter
forces him to sign another contract WIth the rubber harvest as co atera
xp And this situation is repeated and remains the same form many years, because
th~ relation of the rubber tapper to the patron is roughly the sa!lle as the f~~~~~5~~
the exporter" (Parliamentary Papers, Defesada Borracha, 1913 m Carone
.
8) . .
30
31
workers to capitalist control, transforming the labour process 18 --, patrons could use
Chapter 2
a combination of sheer violence against free trade and incentives to increase the
total output. Indeed, the labour process remained in the hands of the rubber tappers
themselves throughout the boom period. This meant that the tappers, in controlling
the labour process, developed a base for their subsequent permanence in the forest
even after the collapse of world wild-rubber prices.
18. "A wage labour regime, which might prove a solution to the problem, was
not practicable on the rubber estates due to the physical impossibility of controlling
workers on those Mesopotamian latifundia" (Rego Barros 1981:199).
though boom-period statistics vary, they suggest productivity rates similar or inferior
to present-day ones, with 350 kgs. per year on the Lower Amazon and 700 kgs. per
year in upland Acre (Chaves, 1913:48), or a 400 kgs. average output and 800 kgs.
maximum (Walle 1911:167). Other say that 230 kgs was the average (Santos
1980:235).
output per tapper and in terms of rubber output per area, boom-period expansion
developed with the horizontal expansion of extractive activities to all Hevea
brasiliensis (as well as other similar species) frontiers.
Only significant technical innovations could reverse the natural limitations to
exp~nsion
boundaries of forest areas with natural Hevea brasiliensis - and at the same time
subordinate workers and improve their individual efficiency. This revolution,
however, did not occur in the Amazon, but in Malaysia, Indonesia, Ceylon and
Indochina, primarily through the development of plantations based on indentured
labour (Coates, 1987;Dean, 1987;Tocantins, 1961). Beginning in 1912, the rubber
extracted from domesticated Hevea brasiliensis trees planted under the plantation
system in Malaysia -- with seeds previously smuggled from Amazonia -- flooded the
world market. Under the plantation system, each cultivated hectare included
- ,
Amazonia became superfluous on the world market from one day to the next.
The main cQnsequence if this was the disruption of the commercial structure _
Given its low and limited productivity both as measured in terms of rubber
the
between 200 and 400 rubber trees, which were tapped by a single worker; in the
of the rubber trade, initiated by the insolvency of import-export firms in Belem and
Manaus (Weinstein 1983). The principal characteristic of the rubber estates of
!
Amazonia, however, was their resiliency. Between the peak year of the rubber fever
(1912) and the worst year of the recession (1932), production in Acre State as a
whole fell from 12,000 tons to 3,000 tons, while prices plummeted at a much greater
rate, from a maximum of 26,000 reis to a minimum of only 4,000 reis (Table 2.1).
Within the municipal district of Cruzeiro do -SuI, output fell from 1,300 tons
during the boom to 500 tons already in 1923. But the local estates continued in
existence in the upper Jurua, although transformed.
Local population rose from 13,000 to 15,000 between 1913 and 1920.
extractive enterprise, 300 hectares of forest were needed to occupy a tapper with the
1922:615).
same number of trees. Individual annual output on the plantations was between one
and two tons, while the average output in Amazonia remained at about 400 kgs. per
-I
the '
domestication process increased the productivity by area three hundred times, while
.
..J
Current observations suggest processes that help explain the resiliency of the
rubber estates during the crisis between wars. From 1982 to 1991 rubber prices fell
from US$1.80 in 1982 to 40 cents in 1992 as paid to tappers. At the Riozinho Estate
in the Tejo headwaters where I did field research, the number of families fell from a
.maximum of 68 to 53 between 1982 and 1991, a decline to 78% of the previous level
production, peaking at 42,000 tons in the Amazon in 1912, reached over 100,000
tons in Asia already by 1915 and, in the decades that followed, levelled off at
something over 1,000,000 tons, though occupying an area ten times smaller than the
1. Population statistics for 1913 come from a municipal census, while data for
1920 is from the national census, subject to methodological reservations. See also
Castelo Branco (1923:721).
35
34
that hardly may be considered mass flight, while the overal Tejo river population did
not drop signicantly. Now in the Thaumaturgo area as a whole only 65 percent of
the houses extracted rubber, while 35 percent cultivated riverside plots producing
1922:721).
tobacco, sugar and beans, while others sought employment on small cattle ranches
near the river. Thus, the main result of the acute fall in prices in the 80s was a
transference of many families from the hinterland extractive zones to riverside
agricultural activities (Almeida, 1991).
Both in the 1980s' crisis in rubber prices and in that of the period of 1912 to
1943 local response appears to have been the same: the conversion of labour and
And a decade later the change was confirmed in reference to the middle
Jurua river:
... the labour regime has changed. Thus today there are many rubber estates
leased to the seringueiros themselves, who pay rent to the owners, varying
from 50 to 100 kgs., for every pair of rubber trails: under this hypothesis,
tappers enjoy the liberty to buy and sell freely, which they do preferentially
,I
monopoly. During the involution years, patrons who no longer could operate by
advancing merchandise to the tappers chose to lease rubber trails to tappers without
These late concessions to old tapper demands may also be seen as reactions to the
social upheaval that reigned on the rubber estates in the aftermath of the crisis, or
-,
Any number of trails in this well-preserved estate for lease, at 25 kgs. of fine
rubber for each trail, and the lessees will remain free to buy merchandise for
emerged side-by-side with accounts of revolted rubber tappers who fled from the
their consumption and sell the rubber they produce to whomever suits them
estates without settling their debts. 2 The shock felt by the Tejo River rubber tappers
36
37
thus was described by Father Tastevin in 1913, who in this year visited his entire
course:
In 1918" the empire constructed by Melo & Co., the Para firm that had_
bought up the Tejo River on the eve of the crisis between 1910 and 1912, collapsed
The entire country suffered from extreme scarcity. All one had to eat was
in bankruptcy. The property was acquired by Nicolau & Co., which in turn went
maniac and beans, and even this threatened to disappear. In some places
under in 1942. Raimundo Quirino Nobre, a former manager of Nicolau & Co.,
there was no sugar or coffee ... One must remember that this is a rubber
became sole proprietor. 3 Nicolau & Co. maintained a fleet of twelve steamboats of
region par excellence, where all hands are busy with the extraction of this
the gaiola ("birdcages" propelled by a stern paddle) type, but even these gave way to
precious gum. All food and all beverages are imported by river in ste'a mboats
belonging t~ the large trading companies of Manaus and BeIem. Therefore,_
the canoes of itinerant traders and to locally-built craft of ten to twenty tons, which
after the 1940s were propelled by three to eighteen horsepower outboard motors.
when the steamboats fail to arrive at the expected time for one reason pr
The trails covered by mule trains, which allowed for bi-monthly visits to the rubber
another, the spectre of dearth immediately appears (Tastevin 1914).
tappers' homes (quinzenas), disappeared. Tappers in the most productive areas of
In a report written that same year, a public official of Cruzeiro do SuI
the Tejo headwaters, a place that cannot 'be reached even by canoe, now had to
described how the Tejo River rubber tappers were reacting to the insolvency of
Melo & Co.:
carry their product along forest trails. During the rainy season, rubber was
transported by the tappers themselves to the warehouses, from where tons were
at this time the tappers of Melo and Co., 500 or 600 strong, threaten to
floated downriver on rafts to Cruzeiro do SuI.
Restaura~ao
abandon collectively the rubber tree trails, only because they have heard that
estate on the Tejo River, saw three schools closed in 1923 (Castelo Branco
this company, based in the capital of Para, now faces a most difficult
1923:645). The estate's church, which prided itself for possessing the region's largest
situation (Prefeitura do Alto Jurua 1914-15).
-1,
bell and for holding novena ceremonies, also disappeared. In short, the rubber
This same account asserted that the "older rubber tappers" not only fled but
estates had de capitalised.
also attacked the trading posts:
Also the Indians who live on the Breu River, seizing the opportunity of the
pitiful state of abandon in which all that area now lies, now descend to the
How could the population survive now that rubber production had been
Tejo River committing acts of the worst kind. The set fire to dwellings,
reduced drastically and prices had plummeted? On estates near the municipal seat
..J
ambush and kill rubber tappers and arouse fear in every mind. In great
of Cruzeiro do SuI, rubber output fell to less than half of what it had been, but
-'
numbers, armed with rifles, and said to be led by former rubber tappers, the
agriculture began to flourish, not only to supply tappers and patrons in the forest but-
Breu River Indians have thus become a dangerous element of unrest in the
also to attend to the urban market for manioc flour, sugar, coffee and other articles
,.
38
previously imported. Sugar mills and pastures became part of the Moa River
landscape, favoured by its proximity to the town of Cruzeiro do Su1. 4 Mter the crisis,
similar activitis developed along the Jurua, taking advantage of the fertile alluvial
soils. The remote Tejo River area now received manioc flour from the Moa and
other rubber estates and not, as before, from the Lower Amazon (Tastevin 1914;.
Castelo Branco 1922:719 ff.). The Nicolau & Co. estates, under Nobre's
copaiba oil, andiroba oil, vegetable ivory (jarina), timber, as well as otter, jaguar,
deer and snake skins (Andrade 1936: 116-20), and in fact Cruzeiro do SuI exports its
high-quality manioc flour to this day. Local tinsmi!s and blacksmiths began to
supply work implements such as knives and buckets used in rubber extraction;
clothes were confectioned locally; river vessels came to be built in the region,
inaugurating a solid tradition of master-artisans that survives to date on the Upper
moving on to the city of Cruzeiro do SuI or to the Jurua tributaries, from where
comes most rubber.
This process explains why the rubber estates remained in existence. They
now became estates in which rubber was just another article in a diversified
agrarian-extractive economy, based on the domestic forest economy of rubber
ranchers as well. With a low population density (about 1 inhabitant per km2), the
authorities, along with the 1920 census, show the increase of agricultural and
economy based o~ abundant game and fish reserves, on fertiie swidden agriculture
with long fallow periods, and on an arsenal of forest resources created several
patterns of permanent occupation of the forest niches, whose fabric resided in
groups of two-three houses (colocar;oes).6
Rubber tappers and riverside cultivators did not necessarily witness a decline
in
their standard of
living. If measured in terms of quantity and quality of food .
.
.
consumption, in terms of work routine and in terms of personal liberty, the quality
of life for rubber tappers probably improved. At the beginning of the 1920s, Father
Tastevin commented that the tappers' diet had improved substantially with the fall
40
of rubber prices. Besides manioc and sugar, crops included rice, maIze, beans,
peanuts, watermelons, squash and potatoes - planted on the river banks during thedry season for quick harvest with little labour input, as one may witness these same
items planted and harvested along the banks of the Jurua between June and
October. In 1920, Father Tastevin observed:
... for the moment, the fate of agriculture is linked to the price of rubber,
(I
meaning that the higher the price of rubber, the more the patron is able to
buy goods and obtain wealth through trade; and the less rubber brings to
him, greater are his own needs and the need to feed his workers (Tastevin
41
what they had been during the boom. An economy that had dedicated 180 working
days to the production of rubber and the consumption of manufactured goods
(clothes, soap, fuel), tools and foodstuffs (pirarucu, flour) was transformed to one in
which only 90 days a year were dedicated to rubber extraction and the consumption
of outside articles, as it came to produce foodstuffs (meat and flour) and a few
manufactured goods. The population density that could exploit rubber in a stablefashion coincided with that which the forest could support, using near-indigenous
subsistence strategies (swidden agriculture with the use of iron tools, hunting with
the use of firearms, and general collecting activities).
1920:145).7
One consequence also noted by Tastevin was the constitution of families,
now a definite advantage, since
-,
. . . older b9Ys could tap rubber trees, while the father would fish and the _
subsistence sector (Weinstein 1983). To extend this notion to the Upper Jurua, one
(Tastevin 1926).
is led to admit that a subsistence sector developed after the boom, and that it was
accommodated within the rubber estates. According to Weinstein, the main effect
probably affected more the patrons than the tappers. An employee on a Tejo River
of the crisis lay in the insolvency of financial concerns in Belem and Manaus. In fact,
rubber estate recalls that during the worst period for rubber prices papaya leaves
the Tejo river ~bb~r estates of the Nicolau & Co., received from the bankrupt Melo
were used to wash clothes, embauba wax and rubber were used as fuel and the bark
& Co., were able to survive tlirough the combined strategy of diversification in
from certain trees (envira) was used to roll cigarettes, while match sticks were split
extractive activities on the estates (pelts, timber and rubber) and of diversification
of activities at the Belem head office (industrial production for the regional
market). At the same time, the company gave its managers or local bosses
was this productivity that permitted the purchase of work tools and necessary
~rticles, such as salt and ammunition, even when rubber prices fell to one-fifth of
the crisis years to buy out the estates by 1940, on the eve of the transformations to
be described in the following chapter. The tappers themselves became a category
7. My father's mother, who lived in the Purus valley, remembered the days
when her husband, a rubber tapper, received the patron's permission to plant beans
and manioc, as that time they began to have fresh manioc flour, instead of the sour
flour that had been brought in from downriver.
more like forest peasants and less like specialized extractors, while the verticallyintegrated properties of the boom years turned into estates oriented towards
subsistence production and regional market activities. Surviving local patrons lost
political and economic power, while rubber tappers developed families and
diversified their economic activities. Wallerstein's comments are pertinent here:
Since the world market had a lowered demand, it was not rational for the
landowners to produce at the same rate, or for some of them to produce at
all. We then saw occur what has sometimes been called 'inversion'.
Cultivated areas were left untended. The workers were permitted, nay
encouraged, to take up a plot of land and feed themselves off it. Trade with
the rest of the world diminished. Handicrafts, which had previously died out,
were revived. The commercial estate seemed to be reverting to the status ofa self-sufficient manor once again. The encomienda was transformed into the
hacienda. The landlord himself moved from the city to the rural area, to
partake at least partially in the ' isolated subsistence economy (Wallerstein
1979:124).
Some local patrons became impoverished landowners whose descendants are
rubber tappers or itinerant traders. Others continued to derive some income from
the leasing of rubber trails, though moving away from the forest; still others formed
small cattle ranches within the rubber estates. These transformations may be
illustrated by the personal stories of Tejo River residents. In sum, during the critical
years between wars, surviving rubber estates, like those in upriver areas of Acre,
remained marginally linked to the world economy, becoming estates with deeds I
(though mostly without legal validity) over forest areas inhabited by rubber tappers,
cultivators, fishermen, hunters, as well as small traders and artisans.
Authors such as Furtado (1959) have assumed that survival conditions in the
forest were intrinsically inferior to those of the arid backlands of the northeast and
therefore tappers, lured to the forest by the promise of fortune, would remain in
extractive activities only as long as chronic debts forced them to stay. However,
there are no reports of rubber tappers who starved to death as a result of the rubber
crisis, in contrast with the backland Northeast"s droughts. The forest had supported
indigenous populations with population densities comparable to those of the rubber
extractors for centuries, and indigenous subsistence techniques were easily adapted by
tappers with a minimum of imports, basically limited to iron tools and weapons for
hunting. In other words, it seems likely that rubber extractors had their own interests in
to assert
~e
Vaca Diaz, and allow for an explanation of the persistence of the estate system during
the crisis in the rubber trade. Native populations and natural resources were not simply
plundered as in the extractive enclaves. Instead, stable local systems of forest
exploitation by permanent migrants emerged, including bOth patrons and rubber
tappers. With the onset of the market crisis, these systems reallocated the respective .
proportions of export activities and productive activjties oriented towards internal
consumption. This conferred stability to the economic system, which lasted throughout
the period spanning the years between wars, without any assistance from the public
sector. This stability was not an effect of the coercive power exercised by patrons, nor
did it have to do with the weight of debts on clients. It resulted from the capacity of a
migrant population of rubber tappers, originating from semi-arid peasant regions, to
adapt to the tropical forest economy. This fact also allowed for the survival of patrons
-,
nor from the patrons' abandon of the estates as in the eastern Acre (Bakx 1987): the
fundamental transformation had occurred within the system itself during the crisis,
when a new social type emerged in the Amazon, the forest peasant.
,
Introduction
This chapter deals with the extractive economy of the Amazon region from
the Second World War. Amazonian rubber estates that survived into the betweenwars period gradually withdrew from the structures of international trade and
developed as local enterprises with diversified productive activities, in the process
retaining a population of forest peasants under the control of the resident patrontraders. Subsistence-oriented estates substituted the rubber properties that had been
oriented by capitalist accumulation on a world scale. Acre became the scene of
involuted rubber estates, where rubber extraction ceased to be a dynamic activity
involving fresh investments and new labour inputs. Through a mechanism that
marked other export econOlnies in Brazil, especially in the example of sugar
plantations, rubber estates became dualist enterprises, notably resistant to marketstimuli (Furtado, 1959; Boeke, 1953). Dualism, in the sense here employed, means
that both the labour force and natural resources may be used in two sectors: one
which produces foodstuffs and consumer goods that circulate at the level of local
markets, and the other dynamic sector producing commodities for the world market.
In this sort of enterprise, there is little room for technical innovation; price peaks
serve mainly to stimulate the expansion of the export sector; low prices are absorbed
through the reduction of imports and retraction of export activities. In both the
8. These new peasants were not caboclos (parker 1985) or the product of the
disinte~ration of indigenous societies. In Acre, the word "caboc'lo" applies specifically
to ~~Ia~S; but Acre tappers substitu~. an indigenous population th~t had been
declmatea or forced to retreat after prelImmary contact, WIth the exceptIon of a few
small groups recruited as agricultural or extractive labour (the Poyanawa and Huni-kui
on the Moa; the Arara, Iaminahua and Cashinaua on the Upper Jurua tributaries and
along the Tejo).
JLL
.
46
The economic policy defended by the patrons was quite simple. They sought
measures to protect prices without incurring in transformations of the property and
distribution structures. According to Barbara Weinstein:
... most of the proposals and suggestions made by the regional elite were
47
characteristics which may appear to' have been vestiges of a hypothetical labour
coercion inherited from the boom years in fact developed as a result of the new
clientelistic relationship forged between the federal government and the patrons.
"
Thus there emerged a sort of "second serfdom' for Amazonian rubber tappers.
Around 1932, the price of rubber reached its nadir. Japan's role in World
living, ... was an open invitation to smuggling. And the campaign to defend
War II and its subsequent control over Southeast Asia altered this situation
rubber reached its lowest point ~hen the . Paraenses demanded a higher
dramatically. The Japanese invasion cut off the allied forces' access to the principal
cultivated rubber exporters, which were located in the Dutch, British and French
(Weinstein 1983:229).
This chapter shows how all these measures, without exception, came to be
't.
!.:J
. colonies. Rubber proved of great strategic significance, since its use was vital for
inilitary ground ' and air craft tyres as well as for other applications, and had been -
adopted in the post-war period, when Ainazonian rubber patrons became a regional
identified as one of the major bottlenecks in the war effort (Martinello 1988:77-78).
Brazil exported rubber to Nazi Germany until 1942, when the Vargas government
clout. The estates received price, capital and incentives, furthermore enjoying an
decided to join th,e allied forces, a decision which included an agreement to ship all
additional feature not present during the boom: sanctions of the federal government
backing up the patron-traders' power of coercion over rubber tappers. During four
decades (1943- ~985), a period which witnessed a rapid and significant advance in
Rubber Credit Bank. At that point, prices immediately doubled, rubber exportation
activities of the rubber estates pressed the limits of their resiliency: no technical
became a state monopoly and the Brazilian government took on the task of
innovations were introduced, the extractive ecology remained intact, and social
relations followed the pattern established during the inter-war period, only now the
forest, areas rich in native seringueira rubber trees. Rubber extraction practically
rubber became a state monopoly: the nascent Brazilian industrial sector found itself
forced to buy rubber from forest estates that were financed by the government.
Politically, the State began to sustain the local patronage system so that many
(Rubber Credit Bank), with American capital owning 40% of its stocks (Martinello
48
1988: 156), allowing rubber patrons to continue to exploit the forest areas they
claimed for six years, independent of valid property deeds (land use could be proved
by receipts of commodities supplied to tappers). Rubber prices came to be set by the
bank, whose norms established that 60% of the product's value was to go to the rubber tappers, 33% to the patron and the remaining 7% to the owner of the
property, thus distinguishing labourer, trader-businessman and land owner.
The State took on the task of transporting labourers through the "Comissao
Administrativa para
Patron-tapper relations were regulated by contracts which stipulated a sixday work week in rubber activities, which virtually undermined domestic economic
activities such as agriculture, fishing and hunting (Caderneta do Seringueiro, 1943
and 1945). According to the contracts, sale of rubber to anyone but the patron
became a criminal offence, whether or not the tapper had any outstanding debts.
Another clause prohibited indebted tappers from leaving the rubber estates, though
debts (along with tapp.ers' labour) c(;uld be transferred from one patron to another
(Caderneta de ' Seringueiro, 1943 and 1945). These conditions reflected the character of government support of the rubber estates, recreating the commercial
monopoly that the patrons had been losing since the end of the boom period.
It is likely that the Cruzeiro do SuI authorities attempted to restrain itinerant
merchants by summoning wartime federal legislation, while on the other hand,
tappers and independent traders resisted to this. Local authorities now sought
official support, arguing that "contraband" was proscribed officially by the laws and
decrees passed by~the federal government. Thus, in 1946, Colonel Mancio Lima of
Cruzeiro do SuI wrote a letter to the governor of Acre, a military man named
Guiomard Santos, plea.ding his support against what he called "an open insurrection"
of rubber tappers and itinerant inqependent traders, presumably against the new
measures which legalised the monopoly. Lima argued that local Cruzeiro do SuL
authorities already agreed to reduce taxes on "itinerant commerce" and to postpone
2. Decree 5225, February 1943; Notice from the War Ministry number 1262
May 1943. Charles Wagley's classic study of the caboclo peasants of the Amazo~
was a direct result of American participation in the "Battle for Rubber". Wagley
yvor~ed fro~ 1942 .to 1?45 as a member in a public pealth program intended "to
proVIde medIcal protection to the producers of strategIc raw materials __ the rubber gatherers in the Amazon Valley" (Wagley, 1964:vii-ix).
the register of "rubber brands" that were to identify the origins of each latex ball.
Both measures were designed to restrict commerce outside the trade posts, but
according to Mancio Lima, these concessions had little effect because "the real
problem lies in the prohibition against purchasing contraband rubber smuggled
from the estates." Mancio Lima blamed especially the itinerant traders, since they
"incited disobedience to the laws governing the
exploitation of rubber
50
estates."(Colonel Mancio Lima to the Governor of Acre,in Guiomard dos Santos 1946).
The final balance of the "Battle for Rubber" was quite modest, though. From
(.
"
rubber prices came to be established by the government. 3 Now that the w'ar was
over, the only market for the Amazonian rubber production was the nascent
51
16,000 tons in 1942, output rose to only 24,600 tons by 1947 (Pinto 1984). This
means that the 50,000 who were said to have migrated to Amazonia would have
(IBAMA 1988:46),since the total output of the Amazonian rubber estates remained
added 8,600 tons or 170 kgs. per tapper, while the average annual individual output
stagnated from then on, always short of the peak of 42,000 tons reache~ in 1912. 4
Although the "Battle for Rubber" proved a fiasco in economic terms, its
unable to cope with the increasing industrial demand, and also that it involved
principal result lay in the important victory achieved by local elite of patron-traders.
higher costs than those of the plantation areas from where the product could be
For the first time in the history of Brazil, the regional interests of this elite came to
imported.
While Brazilian natural rubber production levelled off at around 20,000 tons
associated with a trade monopoly - re-emerged from its stagnation, now backed by
per year during the post-war period, demand rose year after year. Already by 1958,
credit from the government; externally, rubber prices and market were guaranteed
Brazil managed to supply little more than 51 % of the 43,000 tons of natural rubber
by the same government. By 1946 the war had ended, but until 1950 Amazonian
patrons forged a new battle, aiming to preserve the status quo of government
consumed nearly 300,000 tons of rubber, seven times more than in 1958, including _
nearly 210,000 tons of synthetic rubber; 60,000 of natural rubber imported from
dispute. The Brazilian constitution of 1946 determined that three per cent of the
countries like Indian and Malaysia, arid only 30,000 tons produced in Brazil (Ibama
federal budget was to go to the Amazon; in 1947, the state monopoly over the
1988; 1985:3-4; 1985b; Pinto 1984). The government supported synthetic rubber
commercialisation of all types of rubber was renewed to 1950; rubber imports and
whereby industries could import rubber produced outside the country after
1947, in Braga c.1965; cf. Pinto 1984:95-112), and in 1950 the state monopoly was
extended indefirutely (Law Number 1.184,30 August, 1950; in Braga c.1965.)
Now under the organ called Banco de Credito da Amazonia (Amazonian
Credit Bank), whose president was appointed directly by the president of Brazil,
3. Between 1953 and 1979, rubber prices were readjusted according to the
inflation or above it. When corrected for the inflation, prices varied from 100% to
230% over the 1953. levels, except for the period 1966-67 (98% and 92%) and the
years 1973-1974 (96% and 86%). Pinto 1984:130, Table 14.
4. Mter the peak of 42/44 thousand tons in 1912 (according to different
sources), Brazilian exports went as low as 6 thousand tons in 1932; they rose
gradually up to 14 thousand tons in 1943 when the "Battle for Rubber" started; in
1944 the rubber output (from extraction alone) reached the 20/30 thousand tons
interval, where it would stagnate until the present. mGE 1987:309.
52
53
demonstrating that they had acquired Brazilian rubber, paying a higher price for
At the same time, though, this political advantage bolstered regional employment
both than the market price (Law 5.459, 21 June, 1968, Art.22, 1). SUDHEVEA
~md
unfounded.
income - .an argument frequently used by the patrons, and not totally-
a Produ~ao
had defeated Ford in his attempt to develop similar plantations earlier in the
century. However, a basic flaw in the policy lay in the fact that it was based on the
I
Amazonian patron-trader class. No patron had ever invested his own resources in
rubber tree cultivation, neither during the boom years nor in the period of decline
that followed, and much less during the post-war years. Rubber tappers, who now
operated their holdings through a combination of extraction, agriculture and
collecting, were largely ignored (except in the final phase of the programme, ina
. 6.
few projects implanting small rubber-processing units). The programme did not
invest in small-scale; domestically-run production, in contrast to the policies adopted
in Malaysia, where in 1973 over 60% of the total output came from small properties
(Norgaard 1980:791) and where state agencies played an important role in the
cultivation of rubber and palm trees at the family level (Robertson 1984:264; Hill
54
55
1982: 141ff; Bahrin, n.d.). This neglect of the rubber tappers may have been the
development of rubber plantations in southern Brazil, where investments stimulated 'by
government prices led producers to substitute orange and coffee plantations with
rubber. 7 Thus, around 1988, Amazonian production stood at around 22,000 tons~ the
same as fifteen years earlier, which demonstrates that the increase in production was
based on the greater use of planted rubber trees in the South, while Amazonia used the
same labouring population and on the same extractive techniques that had been used
during the war. In '1986, the political manipulation of funds distributed by PROBOR
was exposed by the press nation-wide, and following an official investigation, the
projects were suspended in the midst of a scandal. 8
The Failure of Rubber Tree Domestication
One might argue that the rubber patron-traders who employed PROBOR funds
to speculative ends or to invest in land acted rationally, not unlike the tappers who
1972:11). Thus, a lease contract for a rubber estate was sufficient to guarantee
special SUDHEVEA loans, which were proportional to the number of rubber trails
on the estate and, consequently, proportional to the number of rubber tappers who ,
time to the forest subsistence and commercial economy - the capital offered to the
inight be employed. The rubber tappers' production thus constituted the ultimate-
patrons had a similar cost because it could be invested profitably in several activities,
guarantee for loans contracted by the patron-traders. Not unlike the wartime
l~bour had a high opportunity cost - since they could dedicate their
7. Sao Paulo newspapers I agricultural supplements often informed ~eir readers '
of the advantages of rubber cultivation under the governm~nt-suPI?Orted pnces. There
was however the risk of a future loss of government protectIon (while rubber-trees took
a minimum of 7 years to start producing) Guia Rural Abril l(voll), April 1987: 101105' FoIha de Sao PauIo 8/9/87; Jomal do Brasil 10110/87; 0 Globo 24/0811987;
FoIha de Sao PauIo 19/0111988; 0 Estado de Sao PauIo 29/05191, Atualidades
Agropecmmas n.48, Nov. 1979; Tendencia, Abril de 1980.
8. "PIano da Borracha fracassa, conclui auditoria do Minisrerio da Industria".
Folha de Sao Paul,Q 30103/1986 A summary of findings from an official report:
"Insufficient stocks in a deteriorated state, production smaller than expected _ .
liberation of funds based on political criteria" decisions bolstered by unreahsttc
projections and an accumulated debt of 1.7 million cruzados" (1.7 million dollars at the
~~.
'
57
56
financial market to cattle-raising on the floodplains -- rather than in the risky, long-
.,
b
ff
ana and cacao cultivation. The
labour and on a combmatlOn of rub er, co ee, guar
plantation was less than a day's journey by boat from Cruzeiro do SuI. Lobao
projects in Cruzeiro do SuI, on the Upper Jurua. Legal title or a lease contract to a
sanitation facilities, electricity, running water, a chapel and a store, as well as stairs
.
.
h h
The bishop contributed with
and walkways connectmg the nverport to t e ouses.
request PROBOR funds. At that point, the patron received the benefits of long-
. and c
I thl'ng
o . By 1983 , Lobao had created nurseries for rubber trees,
me dlcme
term loans (with ten-year payment plans and interest rates lower than inflation).
~ofiee and gua~an~, and had cleared SQI;}e 100 hectares for pastures and another
Local offices provided technical support and determined what procedures were to
be employed (clones, chemical pest controls and processing plants). But one obvious
alternative was the employment of loans to develop a cattle ranch under the cover
of a rubber plantation.
On the occasion of my first visit in 1982, the labour force was made up of
d ho had a full time routine
.
. h'
twenty workers who were eqmpped With c amsaws an w
In one typical case which I personally followed, PROBOR funds were used to
clear a forest area and transform it into pasture, thus investing in cattle and labour.
wage. This staff remained much smaller than that which Lobao needed, in spite of
A small nursery was kept alongside the the river to show to PROBOR inspectors in-
case of surprise visits. PROBOR funds also were used to pay for the land title to
advertised on the
heirs of the original owners, following a suit pressed by a lawyer in Manaus seeking
that they could earn two or three times the daily wage offered by Lobao (in nominal
to revalidate an old deed from the pre-1904 period. The whole area, now with a
values) by extracting wild rubber from the forest. Thus, on the plantation men were
consolidated legal title and with an established cattle ranch, was then resold and the
paid 500 Cruzelros' per day (women were paid 300) while a kilogram of rubber was
worth 360 Cruzeiros - toe smallest amount produced by a tapper in the least
land acquisitions.
Although cases like the above were the rule, at least three businessmen in
A tapper who resided near the river could dedicate his or his families' time to
raising beans, tobacco, watermelon, manioc, maize and other agricultural products,
rubber trees seriously, and all three projects failed. underscoring the social and
which could either be consumed or sold to river traders. On the other hand, the
ecological barriers to the idea of transforming the labour process through the
"Corner of Love" prohibited its workers from cultivating manioc or any other food
crop. Debt did not exist in the "Corner of Love", but workers did not have access to
The first case was the Carlota estate, run by Lobao, member of one of the
credit either, which meant they had to work continuously to meet their subsistence
pioneer families of the region. Lobao set up a plantation based entirely on wage
needs. The only free time available to fish on the nearby lake was at night or on
59
58
Instead of working for a wage, they preferred to "live on the riverbank, planting for
subsistence on the beach and makin~ manioc garden plots on the mainland."
for the comforts qf regular labour. By 1987, PROBOR had been discontinued._
The main reason for building the residential nucleus near a lake, far from the
There was no more money to invest. The project had been abandoned.
riverbank, was precisely to break these attitudes. The land agency, INCRA, when
The "Corner of Love" project may be used as yet another example of the
completing the formal definition of Lob:lo's property deeds, took special care in
"woes of managers of tropical plantations having to deal with workers averse to the
locating the 100 hectares of each tapper-squatter well away from the river, in the
depths of Lob:lo's estate. The idea was that these families, once removed from the
competed with the domestic economy for the scarce labour of the region. In
riverbank, would become a permanent source of wage labour for the enterprise. But
contrast, the rubber extraction for a patron combined with the agricultural, hunting
the riverside people never moved into the bush, giving up their new land titles in
and fishing activities carried out within a single family unit of production. A similar
order to maintain access to the abundance of the river banks, preferring to pay
"rights" for the use of rubber trails than property taxes to the land agency and
1989).
However, the other two cases of attempts to implant rubber plantation seem
to confirm Warren Dean's hypothesis contending that Amazonian plantations were _
.
.....
the labour he was unable to extract from the forest economy. Conveniently, INCRA
thwarted not by social barriers but by ecological ones. Both of these projects
was in the process of recruiting migrants for Cruzeiro do SuI, from where they were
managed to solve the labour dilemma by abandoning the idea of permanent wage
taken by plane and settled on a dirt road near the limits of Lob:lo's estate. A first
homesteads complete with schools and medical assistance. When they were left at
the newly-opened dirt road at the beginning of rains, the men stayed on in order to
both cases, the labour force was made up of rubber tappers also employed in the
keep their claims, while their families were forced to beg in Cruzeiro do SuI for food
and shelter.
Lob:lo made a final attempt to solve the labour question by recruiting Campa
(Ashaninka) Indians from the Amonia River, offering medicine and clothing
source to rubber tappers. I visited one of these projects, run by the Messias family,
(actually donated by the Church) as well as salaries greater than those offered to
in 1983 and again in 1987. The plantations failed, due mainly to plant disease. The
Brazilians. In 1983, when I visited the "Re canto do Amor" for the second time, the
heaithiest trees 'we~e of feeble stature and low yield. But the entrepreneur did not
Campa were installed in a large barracks, but they did not stay on long. Lob:lo had
61
advice imposed by the government agency, which demanded homogeneity in all
projects (any short-cycle or permanent crop was prohibited and the cultivation of
rubber trees had to be on completely cleared grounds), imported clones (native
rubber trees were to be uprooted) and chemical control of pests. Homogeneous
stands fell easy victims to plagues, chemical agents themselves caused the death of
thousands of trees (applied according to technical advice), and native trees which
~urvived inspectio~ and were not uprooted proved much healthier and greater in
diameter than the imported clones. According to the local entrepreneur, if he had
enjoyed the freedom to experiment with local species and to combine several crops,
the project would have been a success. This speculative opinion is supported by the
success of small stands of native rubber trees grown alongside local vegetation,
which I had the chance to observe on the banks of the Jurua and Madeira Rivers. 9
A
.). .
pro~pects
available in the region: rubber estates, sawmills, ceramics works, urban commerce_
and cattle ranching. These three "houses", mixed by intermarriage and which I will
call M, Sand C, are also the principal actors in the recent history of the Tejo River,
which they leased, one after another from Sao Paulo purchasers during the 1980s -
SuI also had access to Banco do Brasil credit for commerce with rubber tappers.
subject of the chapter that follows. Thus, seven marriages in the M family within a
During the early years of the 1970s, this easy credit revived the extravagance of the
single generation of new patrons united urban merchants, federal bank employees,
golden age of the rubber boom in Cruzeiro do SuI, only different because the money
municipal administrators and military men within a family block. 10 If one considers
invested in cattle, money markets, and superfluous luxuries came from the Brazilian-
the M, S, and C families as an extended unit, one could travel for days along the
Jurua without ever leaving its territory, and in the same fashion their stores
euphoria somewhat, and many property deeds (though legally dubious) passed on to
occupied entire blocks in Cruzeiro do SuI. The group succeeded in capturing public
the bank. One of the consequences was the possibility of traders who became
funds and channeling them to a new generation of sons and daughters. In contrast
prosperous during this period to become "rubber estate owners". Many heirs of
with the old elite, which sent its sons off to school in Rio de Janeiro, the sons of
these new patrons were educated in the tasks of management on the banks of the
9. The trif to Madeira River in 1986 was part of my role as an advisor to the
N~tional C~u~ci of Rubber Tappers. In the Rib7ira Valley, Sao Paulo, which has a
ram cycle SImIlar to tpe Amazon,. early pl~ntatlOn projects (in the 60s) also were
affected by the leaf blIght, and I WItnessed m the 80s that abandoned nurseries and
rubber tree stands flourished after becoming ''bush''. These stands were rediscovered
only in the 80s and exploited by the Japanese families of the region.
10. Four M's (from the same sibling group) married C's (three belonging to a
sibling group). These alliances included Ratrons, urban merchants, and Banco do
Brasil employees. Another three M's (from the same sibling group) married
members of the S family (a bank employee, a municipal employee and a rubber
patron).
rivers. At the same time, this group effectively managed to block out competition,
particularly from the declining elite of older proprietors. For example, one M leased
an estate from a Nobre heir, managing the estate with money from the Banco do
Brasil. After six years of profitable activities, Nobre reclaimed the estate and installed his own son, who recently had returned from Rio de Janeiro, as manager,
but he was unable to obtain a loan from the Banco do Brasil and as a result the
enterprise soon failed.
Social Barriers to Modernization
native seringueira rubber trees on family scale and as part of mixed crop. A system
based on small holdings and less capital-intensive, however would depend on
granting to rubber tappers formal rights to the forest. In other words, peasant agroforest systems based on public investments would imply an agrarian reform on the
rubber estates, but this was unacceptable to patrons. If these assumptions are
correct, then there was a social and political barrier to the domestication of rubber
trees in Amazonia, not only an ecological one. This social and political barrier was
the power of the patrons over land and public funds.
Why did the modernization policy of the 1970s and 80~ fail on the rubber
estates? First of all, government policy in the 1970s basically was misguided for
having been addressed to the wrong group. A policy seeking to start a technical
consider seriously the possibility that households could domesticate the forest with
funds on the supply of rubber tappers under the extractive economy (having profits
density of around one tapped tree by hectare and the 400 trees per hectare in
an enterprise that would take ten years to come to fruition. On the other hand, those
patrons who intended to "modernize" their estates faced the lack of interest on the
part of rubber tappers who for their part had the option of maintaining their
subsistence-oriented forest economy. Since the property structure was not reformed,
rubber tappers continued to be denied the status of productive agents, and tappers,
disciplined wage labour (Kenelly 1989) but also because extensive homogeneous
who in fact managed extractive activities and agricultural production, were thus
plantations proved ecologically unfeasible (Dean 1987). The failure of these projects
in the period that followed also bares these same factors, only now undermining a
government policy 'directed towards the wrong social group employing the wrong
agro-forest technology.
65
64
Property Titles
Conclusions
During the 1970s, the rise in land prices engendered by the government
policy of colonising the Amazon laid bare the fact that patron-traders in Acre did
Bakx in an article published in 1991. Bakx asserts that the State's action in the early
not possess valid deeds to the estates they treated as their own properties. Until
eighties "eroded the power base of the traditional landowning class", opening a
1976, public lands could be acquired either through the legalisation of occupation
vacuum for the appearance of autonomous rubber tappers, who opposed "violent
(in the case of squatters with up to 100 hectares), or through "regularised property
struggle as they attempted to halt the advance of the ranching front which had been
titles" of areas up to 3,000 hectares (2,000 in border districts), or still through public
facilitated by that very [State] intervention" (Bakx 1991:50). Then, Bakx examines
proceedings (for agro-pastoral projects), but none of these conditions applied to the
the "State measures which have sought to defuse violent confrontation in the Acrean
patrons who "owned" areas ranging from 40,000 to 400,000 hectares in the Jurua
countryside and so stem the rural-urban exodus that it engendered", focusing the
Valley. This situation changed with a law passed in 1976, allowing for "regularised
titles" over large areas (Magalhaes 1977), at the same time that the government of
PROBOR projects respectively (Bakx 1991:49). Since our own analysis stressed that
"
,
I
Acre began to adopt an aggressive policy attracting investors from southern Brazil.
the State supported the patrons both on land and on financial issues, as exemplified
This was when the Tejo River region, covering 6,000 km2 (about 20% of the total
by the PROBOR projects and by INCRA action, Bakx's argument calls for some
municipal area of 31,312 km2), a small part of Nicolau & Co.'s holdings devolved to
commentary. 12
~
Quirino Nobre, was transferred to Consulmar enterprises, based in Sao Paulo. This
We deal first with the land issue. As stated above, until the early seventies,
deed was confirmed in 1982 by the land agency during a "land regularisation
the patrons had no valid title to land. In Eastern Acre, as Bakx points out, there
project", when it was announced initially that rubber tappers also would have right
were titles issued by the Province of Amazonas (until 1898), by the Independent
to their own lots, which later was ignored. As a result of such procedures, within one
State of Acre (1898-1904); by Bolivia and Peru (prior to 1903). Bakx's affirmation
decade the Acre state which legally was made up of vast areas of public lands, came
that "Land titles not descended from or based upon titles issued by the above were
deemed invalid" (Bakx 1991:55) implies that land titles descended from or based
upon titles issued by the above authorities were "deemed valid". What is true is that
none of these , (never validated by the Brazilian Republic, and covering only a
fraction of the area occupied by the estates, usually 10% in upper Jurua) were valid,
but were used to claim valid titles (Magalhaes 1977, Silva 1982). Thanks to special,
12. Bakx focuses on the "differential response of the residual rural population
to State intervention" (1991:49). In 1980, the "economically active population" of
91,588 included 46 1886 'rural workers' (51 %) and 23,203 rubber tappers (25%). Is _
this a "residual" figure?
67
ad hoc legal measures adopted during 1975-1976 and to Incra "regularisation"
programs of the early eighties, the patrons were able to sell their claims (not yet
validated legally) to southern investors at a period of soaring land prices. This is
described by Incra (and apparently accepted by Bakx) as "the regularisation of the
situation of those large property owners with legitimate title so as to facilitate their
access to credit and federal incentives" (my emphasis). As for the squatter's rights
(these did have a legal basis to claim up to one hundred hectares only), simple
~ithmetic
family, "regularising" 100 hectares per family and "regularising" the remainder for
the "traditional" landowner (on the basis of a new. ad hoc legislation designed for
Acre only) was indeed a scheme for wholesale appropriation of the public domain
by patrons (in the Jurua, the tapper's share remained a dead letter). Thus, as a
..
of land conflict and friction (Brasileia, Eastern Acre) should not suggest that therewas any policy intended to create a significant category of rubber-producing smallholders in the Amazon. During PROBOR's final year, SUDHEVEA collected 26.3
million dollars in taxes on imported natural rubber (Allegreti, May and Rueda
1990:3); from which 226 thousand dollars were earmarked to pay PROBOR I
projects (dating from 1972-1977), 11 million dollars went to PROBOR 11 projects
(dating from 1978-198,1), and 10 million dollars were dedicated to PROBOR III
projects. How great was the tappers' share? In Western Acre they received nothing
at all. In 1988, 50.4% of payments by PROBOR for existing contracts went to
landowners and planters, while 3% went to "community assistance" and to the
"settlement of colonists and protection of Indian communities" (Allegretti, May and
Rueda 1990:9). This hardly can be called a "blow to traditional landowners". 13
Rubber patron-traders succeeded in institutionalising their monopoly over
commodity distribution and labour beginning in 1943, also managing to appropriate
---
public funds on a large scale in the 1970s. Between 1975 and 1982, they used the
state machine to take control and then sell land titles, when the focus of the
Amazonian economy began to shift away from extractive activities to stock raising,
agriculture and lumber exploitation. The "traditional" system of the rubber estates
and the "debt slavery" associated with this system
continued in existence in
Western Acre, stimulated by a final flurry, both utilising federal funds in the socalled development plans between 1975 and 1985, and obtaining money from the
sale of land titles.
owners in Eastern Acre were able to convert land into money, leaving to newcomers
the task of actually evicting the squatters. In Western Acre, where land was
purchased only for speculation, the sellers did not bother to leave.
Now we may deal with the issue of financial support. The case of PROBOR
III investment in rubber-processing units for rubber tappers' cooperatives in an area
13. In Malaysia, the main rubber producer in the world (1.6 million tons by1980) the total planted area is 1.7 million hectares. From this, 1.1 million ha belong
to sm~llholders (less than 40 hectares per holder), producing 950 kg~ rubber ea~h
(Barlow 1978, apud Norgaard 1980:7.9~; COT-CAT 1981:31, s~ct.lOn 6.5; HIll
1982: 139-147ft). In Thailand (half a rrulhon tons of rubber, 1.7 1lllllion hectares),
95% of the output comes from 600,000 smallholdin& f~milies (COT-CAT 1981) ..The
small-holding sector in Malaysia was strongly subSIdIzed by governmental projects
(Bahrin and Perera n.d.; Robertson 1984:260-69,276-280).
Chapter 4
PEASANTS AND PATRONS ON THE TEJO RIVER ( 1980-1990)
Introduction
This chapter treats the relations between tappers and patrons in Tejo river
and in the sub-district of Thaumaturgo during the 80s. The focus is on problems of
hegemony in the forest extractive economy. The notion of hegemony involves a
process by which one group exercises domination through a combination of
"consensus" and "coercion" (Gramsci 1975: 1519, 1638 etc.). This Gramscian idea
leads to a concept of violence-biased regimes or "labour repressive" systems
(Barrington Moore Jr. 1966: 434,470). The cases described below furnish examples
of a "labour-repressive" system, which includes the ingredients of violence, contested
legitimacy, corruption and cooptation.
the patrons failed to dominate the tappers through direct repression alone - or
slavery. Purely repressive measures do not significantly increase labour productivity
in an extractive economy; nor do they attract workers to a labour-scarce area.
Instead, the combination of policies of terror and violence (bolstered here by judges,
police authorities and priests) and policies of material incentives, cooptation and
corruption made a lot of sense.
The Tejo River. 1980-1985
The rubber estates where field research was carried out have their main
headquarters near the mouth of the Tejo River, two hours by canoe from the settlement of Thaumaturgo. We may begin with a description of some events in the
Minas Gerais rubber estate near Thaumaturgo during my first visit, in September
tal
.......
--
70
tappers were exercising this very right, which allowed them to remain "free of debt"
with the patron-trader. If Magalhaes wished to acquire rubber under fair conditions,
through his relationship as supplier to Carneiro, he would have to compete with the
itinerant peddlers. Instead, Magalhaes chose to use government-subsidised violence
to impose an undisguised monopoly. It is important to stress that, for patron and
police authority alike, coercion found its justification not in tappers' debts to the
patron, but in the idea of a state monopoly over rubber, which was extended to
patrons who received state funds. 3
Even the principle that indebted tappers could not "sell rubber outside" was
disputed. When the patron failed to provide for basic needs tappers felt justified in
"selling his rubber outside" - that is, to itinerant merchants, local peddlers and even
neighbouring patrons. For instance, in 1982 one indebted tapper in the upper Tejo
bought some medicine from a river trader, paying for it in rubber. While the local
patron Furtado
(~t
the tapper, the accepted view among the local population was that the
circumstances justified the tapper's transgression: first, he had an urgent need of
medication for his son, and the patron did not have the necessary medicine in stock;
second, the patron refused to act as guarantor in the tapper's transaction with the
trader who sold the needed medication. Thus according to the rubber tappers, a welfare prerogative came before monopoly considerations.
Tappers also saw the violence itself as illegitimate. A head of house valued
moral values of personal dignity and independence. 4 For these heads of houses, the
levels by the tappers, and the repressive wave set off by Carneiro meant that many
1. Informal conversations (September 1982-November 1983, July 1986, JuneSeptember 1987, in January 1988, at various times 1989-91. See O'Dwyer, 1989.
2. Renda will be translated either as "trail rights" or as "rent" (for a moredetailed discussion, see Chap.5).
3. The tapper's view of rules is consistent with th~ pre-war period situation.
The patron's interpretation had a post-war tone, assurmng a state monopoly over
rubber trade.
4. In 1982, the case of the rubber tapper who killed a pat~on who .decided t,o
attempt to expel the tapper personally from his own house was stIll fresh III people s
memories.
humiliating invasion of their homes or canoes and the confiscation of their personal
reopened the trails, whicb were in the bush, since no one knew where the
possessions by the police was unacceptable. To be arrested and beaten by the police
Indeed, the patron's ownership was dubious. The Correa company took
The notion of rights payments or rent was part of the "common sense", but
Damiao and his brothers to court only in 1989 for lack of rent payment ("trail
its rules also were the source of conflicting interpretations. For the tappers, the
rights") betwee~ 1982 and 1989, but the real reason was that Darzinho and his
payment of rent was illegitimate when the tapper was responsible for the opening
and maintenance of the trail. One well-documented case occurred along the
Damiao's brothers, decreed his eviction of tr.e tapper from "rubber trails", but not
from homes and crops. Darzinho obstinately refused to leave the estate: the forest
In 1982, Correa & Irmaos company bought up the property titles, which later
were legalised by INCRA (the federal land agency). Though the titles covered an
where he and his brothers lived had abundant game, rubber, livestock and
agricultural produce. 6
area of 14,000 hectares, the company sought to gain control over an area of 114,000
hectares, within which lived Damiao, Darzinho and his other brothers. Correa
wished to collect rent from Damiao and his brothers, but Damiao did not agree. 5
"I haven't paid rights since I arrived, since I've been there seven years (from
1984 to 1991 on the
s~,me
powerful Melo & Co., which, bankrupt in 1918, sold out to Nicolau & Co.; in 1942,
tapping site), and never paid rights. Even when I
during the wartime,phase of rubber production, Nicolau went under and sold out to _
came down from upriver there was no payment here since three or four years
his local manager, Raimundo Quirino Nobre; and Nobre sold property titles to
before (rights had not been paid for this spot since 1981). I was the one who
Evilasio Maia in 1967, who passed them on to Consulmar Industrial Enterprises
between 1974 and 1976 (via its subsidiary company Santana Empreendimentos
5. The biography of Damiao illustrates several points' made in chapters I-Ill.
In 1913, Damiao's father arrived from Ceara on the Cruzeiro do Vale estate,
openin& a tapping area at age 13, and remainin~ their until his demise at 65. Mter
the CriSIS of 1913, the family began to plant mamoc, and according to Damiao "... the
original estate did not have this captivity that it now has.( ... )At first you could sell
rubber whenever you pleased, mainly to the local peddler; the estate owner was a
sort of small tnidee ... he [the patron] went as far as to split matches to help thepeople; from a little kilo of salt he would give a spoonful to one, a spoonful to
another ...". Mter 1943, prices improved - "the merchandise was favourable".
"Nobody took out [i.e. sold to outsiders] a single kilo of rubber", but on the other
hand rubber tappers often received "a small bonus" at the end of each year, taking
advantage of nsmg prices. In the late seventies and early eighties, "the fatron did
not allow the tapper to sell the product outside, since it was financed; i he sold a
kilo out, he was put out, and then they created submission to not sell rubber outside
... Then they started getting ambitious." (Damiao da Silva, 1991).
Agropastoris S.A.).
When Consulmar enterprises bought up the rubber estates on the Tejo
(including Restaura~ao), the former patron Geraldo stayed on as local patronlessee. With his demise in ' 1980 the change of ownership had a local impact. One
6. Correa & Irmaos vs. Manoel Gon~alves da Silva (Case Number 6.735/89,
F6rum de Cruzeiro do SuI). "Where I live is a good place to s~ttle. There's a lake,
good for fishing. Durin~ the summer (dry season) nobody worn~s much about food.
Summertime there is fish there is game, all kinds of game." WIth the onset of the
cooperative in ~98?, .the 'rubber ar.~a beca.me a "mine". ~ ...) They're getting on us _
because my busmess IS good" (Daffilaoda SIlva, recorded III 1991).
75
74
trader needed only to go to the head office in Cruzeiro do SuI to receive payment.
preferring to build a runway on the Peruvian border for suspect activities. The Tejo
At the same ti~e, Co~rea promised the tappers that he would supply the estate to
That the Tejo became a "river without a patron", to adopt the terms used by
the tappers, meant that there no longer existed a strong patron. The smaller patrons
February of 1983, the tappers began to realise that Correa was a wolf in sheep's
who lived at the trade-posts did not possess enough power or prestige either to
clothing. In the annual balance sheets for 1982-83, Correa charged the tappers 60%
secure bank loans or to mobilise repression - this power, as stated in the previous
interest over the debts bought from the itinerant merchants, in a year when ann~lal
chapter, was monopolized by a cartel of Cruzeiro do SuI families, and the minor
inflation stood around 30%. For their part, the itinerant traders noticed that their
patrons of the Restaura~ao area needed a strong patron above them in order to face -
"orders of payment" could not be discounted in cash without waiting for weeks, and
since they owed their own creditors in Cruzeiro do SuI on short term, the delay in
The Correa company came up from a humble background and it is said that
payments meant they would have to pay considerable interest. Thus Correa,
the "old Correa man" gathered his wealth by selling gramix6 or muscavado sugar.
although appearing to negotiate with tappers and itinerant traders alike, in effect
The old Correa succeeded in accumulating capital and lands in the 1970s, and was
severely punished both for dealing rubber "outside the trade post". When a few
the first to take advantage of PROBOR funds. Sebastiao Correa, one of the sons of
peddlers returned to the Tejo to resume trade with the tappers, Correa met them
the old Correa, entered the Tejo River in 1981 during the "patronless" period to sell
with law suits. O~ a brief visit to the post while awaiting the rains, I witnessed and
goods. Enthused by the high output of the rubber tappers, Correa obtained a lease
contract with Consulmar enterprises of Sao Paulo and in 1982 introduced himself as
Restaura~ao
estate.
Correa's first problem was the competition from itinerant river traders who
crime against the patrons' monopoly, the most serious offense to the customary rules
of the rubber estate. Patrons like Correa, who used tame methods, also were
dispersed the product. Through a Banco do Brasil loan, Correa bought a great -
economy based on trade arrangements with autonomous forest peasants was the
7. "When I was a boy, each of us had only two sets o~ clothes. My. father didn't
owe anything to anybody ... In my day there wasn't any radlo on the TeJo. There ~as_
bnly one engine. When the 'black mule' came about, by God, that was somethmg
amazing. Nowadays, who doesn't own a wat~h? Watches and record plaxers .
Nowadays everything is good."(Noe, a middle TeJo rubber tapper and a marretelfo).
"The TeJo River was too disorderly. Many people no longer worked. Th~y bought on
credit from the patron and then went to the hmterland and kept on buymg from the
river trader. By the end of the year, they had no~hing to pay the patron
with."(Timoteo, an upper Tejo rubber tapper and marretelro).
77
76
great range in individual productivity. Some tappers had an annual output of less
On the upper side of the range of productivity, the real issue was that these
than 400 kgs and remained usually indebted; the average output was around 600 kgs.
tappers could strike positive balances in their accounts with a cumulative effect. One
In the Riozinho igarape, almost half of the global output of around 42 tons came
method to deal with those cases was to push high-value merchandise to these
from 14 households (of a total of 66) producing over one ton of rubber.
tappers, in order to offset their credit.8 The largest producers received from the
On the lower side of productivity, the patron tried to evict the "lazy" tappers,
trade post Japanese watches (Orient and Seiko were the preferred brands), six-band
to be replaced by more productive ones. The basic problem, then, was not to
and to attract others. He could simply cut off supplies to tappers with small outputs.
It was said that Correa planned to expel all tappers who produced less than 300 kgs
individual productivity. The patron could identify a group of men at a distance and -
comment: "Look, those are the Cunhas. They're all one-ton tappers!" Another
when they are five or six (in a single settlement), like the Nonatos. Those who work
method was to offer larger stocks to be re-sold at a minor trade-post, and also the
alone extract little latex, they have other work, these are not tappers, they are a
job of collecting debts from the other tappers for a per cent payment. As a
waste of time." Widows and couples with many small children, who needed to
consequence, the tapper with a surplus was transformed into a petty patron indebted
produce food on their garden plots, were included in the "hit-the-road" category.
single woman head-of-house with several children living in the Riozinho Estate
1983, the abundant merchandise began to disappear, and even the manager had to
Restaura~ao
'
trade post in 1983, the manager told -
buy powdered milk from an itinerant merchant to feed his child. Correa now refused
her: "Soledade, your debt is too great. You owe 48,000 Cruzeiros." Soledade replied,
to make cash advances even to the most productive tappers who needed to go to
"that's not right." The manager: "I am right, Soledade." She said: "My debt is 32,000
town to treat illnesses in the family. As such, Correa began to show that he was not
Cruzeiros. You are stealing." The manager then said: "Correa's orders are to sell
only to those who have rubber." To which, Soledade retorted: "I have some rubber.
Early complaints already surfaced at the end of the first year (1982-83) when
l "1t up as SUIts
. me best.
The merchandise is yours but the rubber is mine . And I ' lgIve
the tappers realised that the customary practice of annual advances at fixed prices
But I will pay my bill only at the end of the year." The manager stood firm:
"Soledade, clients in debt cannot sell their rubber outside." From then on, she did
not return to the trade post, and began to buy from a Tarauaca river trader, through
a clandestine trail that cut through the back reaches of the estate - beyond the reach
of the estate's supervision.
was abolished. Under the new system the rubber turned over at the end of the year
commanded three different prices. Should the client refuse this system, the
alternative was to turn over rubber at current prices but to pay interest on the value
PROBOR projects were interrupted amidst a public scandal. Rubber prices began
"Now with Correa here, it's difficult to take out any savings. In the old
the same time when the Banco do Brasil raised interest rates on their loans to
days, merchandise was sold at a price that did not change all year, and rubber
rubber traders. At the same time, union organisation backed up the tappers'
was bought at the price it had at the end of the year. Now prices go up with
rubber prices. In the end, the system of three prices and the interest system
A
j -
This was the context for the transition from cooptation to violence in the
interior of the Tejo River area, as the Restaurac;ao estate's lease was transferred
from Correa & Irm'aos to Cameli & Filhos in 1985. Already in 1986, Cameli
legal actions against them. Most itinerant traders who stayed on were sons, sons-in-
iaw or relatives 'of the resident rubber tappers. How could one prohibit a son from-
visiting his father - even if he brought merchandise that may draw the interest of
rubber hidden in the forest. When the tappers did not come forward with amounts
neighbors willing to buy and pay in rubber? There were other reasons for protest:
of rubber great enough to satisfy Banha's demands, the police would take sewing
Gedeao, book-keeper of the central post, himself once a rubber tapper, was famous
machines, diesel and gasoline engines, and even milch cows. This is a sample
for his accounting errors, and I spent long hours in each household that I visited that
During the dry season of 1983, between August and October, goods became
to.pay up my account. When the other patron came, with this Manuel Banha,
scarce, but they were not replaced with the advent of the rains: now the patron
they went into the estate, taking the police with them, and went as far as my
awaited the rubber harvest to cover his own expenses, which would start only with
house. I wasn't there, my wife was. They already showed up with the police
the next rain period by December. Finally, in the third year, from june 1984 to 1985,
there, ordering me to present myself that same day, though it was already
Correa abandoned altogether his incentives policy. Debts contracted by tappers still
night, but I had to be there to settle accounts. Or else he'd take away the
outstanding would have to be settled by force. But instead of undertaking this task-
cow, the only thing I owned. I'm father of six children, it was the only thing I
personally, Cotrea transferred his lease to the estate to another important family of
owned, this cow, to give milk. I went there, and the police made me turn over
the cow. They went and got the cow, and still made me go back half hour by
water. There I ran away to Vila Thaumaturgo. I got there and they kept my
cow. They got it inside my house. That's why I became so disgusted. Because-
trade post and demanding a 30% reduction on all debts as well as the
cancellation of debt and rent payments of a tapper who had been arrested and
punished by the Thaumaturgo police during the previous year at the behest of
Armando Geraldo. Claudino also . counselled the tappers to refuse trail rent
. <,
while at least fifteen had the courage to accompany Ginu to the estate's
headquarters, where Ginu gave an inflamed speech demanding the immediate
evacuation of the police. The police retreated immediately, for they did not wish to
risk an open conflict with the tappers (each armed with a hunting gun) on their own
grounds, in exchange for a meager supplement to their regular salary. Following this
incident, even those tappers who had looked down on the union began to admit that
it served some purpose. In 1987, when I returned to the Tejo area after a three-year
absence, Chico Ginu's name was known even on the Lower Tejo, and the event had
The union began to demonstrate that the rubber tappers could act
collectively, provided adequate leadership and coordination. More important, one
been registered in song and in verse. Though many tappers never recovered their
possessions confiscated by the police, they had achieved an important moral victory
and, moreover, the union had recovered its own self-respect.
. The role of the union was the more significant because by 1986 it was largely
discredited in Tejo headwater estates as a result of Correa's shrewd tactics in the
previous years. In 1978, a rural labour union had been organised in Cruzeiro do SuI,
10. This account registered by me is recorded in Procuradoria Geral da
Republica 1989a.
11. In 1980, Raimundo Lino (from the rural ~nion of Tarauaca) ~dvised
suggested that Upper Tejo rubber tappers open a trail to the Tarauaca RIver. In
1983, this clandestine route still was in use.
12. Geraldo's widow refused to cash a money order sent b~ a manager, a~d
Claudino cashed it. When the manager collected the correspondmg rubber, lo_ao
Claudino assembled a dozen tappers and headed for the trade post C?f Restaura~ao. Claudino asked the group: "If this happened to you, and they came mto your home
to take your rubber, would you have the courage t~ go ~nto t~e boat to get you~
rubber back?" As the story goes, the answer came m umson: I would and I do!
Then Claudino said: "Then let's go!"
13. I witnessed some such arbitrations in 1982 and in 1983 in the Riozinho
Estate.
82
of Claudino's plans, if successful, would deal a serious blow to the estate monopoly.
15% of the expected return of 15 tons of rubber. 16 This step proved fatal for
Union members were to contribute ten kgs. of rubber apiece to buy a boat with a
Claudino's authority. In the tappers' view, Claudino had been bought by the patron.
He was "weak".
cost of 800 kgs. On the Riozinho and Manteiga alone, where there were 110 rubber
the tappers not to pay Claudino "if they didn't feel like it". Neither Claudino nor
Roberto collected a single kilo of rent rubber. Claudino had not only unpayable
merchandise purchased with cash, thus reducing the price of goods to at least one-
debts but also an unfulfilled contract with the patron. In the upshot, in January 19K\
half. 14
staff and had to cater for large meetings. 15 Claudino built a large, paxiuba wood
his debts and hi.s contract - also gaining the post of manager on a Correa estate on_
the Jurua Mirim, and a promise of a Banco do Brasilloan.
lodge, and planted a prodigious manioc plot that was to provide food as well as a
source of income. These were investments that most tappers would refuse to cover
through increased union dues. The patron Correa, instead of contesting Claudino's
other hand, though, they also show that independent actions at the local level were
limited by lack of material and symbolic resources. 1)1e tappers' movement against
rapidly, for he hoped to offset his expenses with the expected returns of the union
leadership had to be as strong as the patron, able to muster its own client following,
In 1981, Claudino had recommended that no one pay rent on rubber trails,
to extend credit, to substitute the patron and face him in the arena of legitimacy.
but in 1982 the union president Saraiva asserted from Cruzeiro do SuI that these
Under these conditions, any tapper movement without external support comparable
rents or "trail rights" were to be paid up. The patron then proposed that Claudino
to the support offered by the state to strong patrons was doomed to fail.
himself collect these rents - receiving a percentage of the rubber collected. Thus, in
Although defeated in one battle, the union had scored a point in a larger war. _
the final months of 1982, as the period of rent collection drew near, Claudino signed
More than simply an alternative local authority (and also an intermediary to the.
an agreement by which he was to collect the rubber trail rents for 1982, receiving
hospital, the land agency, the pension institute, and other institutions), it provided a
vehicle through which the dissatisfaction of the tappers could be formally
14. This followed the example set by the most productive rubber tappers such
as Joao Nascimento, Chico Farias, his brother Cha~as Farias, old Ginu man, Chico
Roberto, and Joao Nascimento. Old Ginu's son, ChlCO Ginu, became a union leader
along with Chico Roberto, under Claudino.
15. Chico Roberto, a tapper recruited to assist Claudino, received 10% of all
union dues plus the promise of 500 kgs. of rubber per year, to compensate his losses
while away from the rubber trees.
articulated, along two important notions. First, debt disputes between tappers and
16. The Restaura~ao rubber estate had around 224 households (an averag~
of two rubber trails each). Since each house was to pay 66 kgs. of rubber for trail
rights, the total expected as rent was 14.8 tons. The amount promIse.d, 15%, thus
represented around 2.2 tons. Of this total, 500 kgs. would be paId to ChlCO Roberto.
84
patrons no longer were to be consid~red criminal issues but rather labour questions.
Thus they were not to be settled by the police but rather by labour courts andunions. Second, the notion that tappers possessed rights over the land they worked,
which in their case implied rights over forest resources, meant that they could not be
expelled from the area they occupied. Thus, the issue of trail rents should be treated
as a land question and not as a criminal case. This was a prelude to the legitimacy
crisis that followed.
The Tejo River 1985-1990: Crisis of Hegemony
85
.insisted on two issues: the new patrons' neg 1 ectof rubber trails and their plans of
large-scale lumber extraction, with its destructive effects on the peasant economy of
the tappers. The new patrons had no interest in investing in the conservation of the
rubber trees. They did not employ mateiros (forest keepers) and they urged young
and recently arrived tappers to employ predatory techniques, which could foster
high productivity during a year or two, but which killed the trees after a short while.In Ginu's words, repeating
much as killing one's mother, whose milk provided their life support.
In 1987, the Cameli group planned to move from rubber to timber extraction
In January 1988, Chico Mendes and other National Rubber Tappers' Council
on the Tejo River. They owned sawmills in Cruzeiro do SuI and heavy equipment
already used in other areas near the Tejo (including the Campa, or Ashaninka,
SuI. Macedo, son of a tapper, had been a rubber tapper, tractor mechanic, boat
pilot, cultivator and, finally, Indian scout for FUNAI (a post that did not require
seeking to confirm the area's reputation for mahogany, and also selected areas for -
formal training), where he gained notoriety for his work in founding indigenous
headquarters and other functions, some tappers realised that these plans were to
cooperatives in Acre, particularly in the upper Jurua valley where he had been born
at the headwaters' of the Tarauaca river. Macedo began his Council activities by
organising a meeting in the interior of the Tejo area and in July 1988 he developed a
project to create an Extractive Reserve on the Tejo.17
began to establish relations with social movements in other parts of Acre and of
around 700 rubber tappers, as well as women and children, coming eighty-five
rubber estates of four municipal districts in the states of Acre and Amazonas. 18 The
to guarantee their permanence in the forest. In 1986, the Council began to organise
central themes discussed were rent payments and patrons' violence. At the end of
municipal chapters, and that year a meeting was held in Cruzeiro do SuI with forty
the conference, sixteen delegates were elected to represent the tapper population on
participants. Between July and August of 1987, Ginu and Roberto took advantage of
the presence of the anthropologist to gather further support to their union and to
spread the Council's news. Ginu, now the main union's delegate at Restaura~ao,
86
the Council. The union had worked for many years with agricultural labourers in
.h
areas surrounding Cruzeiro do SuI and intervened to med'late agreements Wit
patrons on the rubber estates - but by then, under the presidency of a cousin the
87
determined operating rules, fees to be collected, credit mechanisms and the nature
of the commodities "to be purchased; the terms of the contract with the bank also
were discussed, "int"egrated and applied to the association's bylaws, which were to
~io
murder, which had occurred earlier that month: international politics reached the
Upper Jurua and altered the local correlation of forces from without. Already in
January 1989 the first working meeting took place in Vila Thaumaturgo, involving
19
During the second half of the 1980s, state economic policy shifted from
rubber to other activities, notably stock raising and timber extraction. In 1988, the
state government of Acre was preoccupied mainly with obtaining funds from
multilateral banks to complete road projects connecting Acre to the rest of the
country and to the Pacific Ocean, while at the same time negotiating with
prospective lumber importers. Seeking to show an ecological concern capable of
placating the protests of environmentalist lobbies against World"Bank loans to pave
roads, the state Secretary of Planning in Acre State backed up the National Council
of Rubber Tappers' request to the National Development Bank (BNDES). At the
same time, it began to announce since January 1988 that the state's development'
policy no longer emphasised cattle raising and now was to be oriented by agro-forest
strategies, including the Extractive Reserves proposed by rubber tappers.
In January 1988, the state government inaugurated an "extractive reserve"
covering 30,000 hectares in Sao Luis Remanso, an area that had not made up part of
the plans of the tappers' organisation and where the federal government had
intended to implant an agrarian colonisation project. Thus, the suspension of the
World Bank loan to pave the BR 364 highway, which had resulted from Chi co
Mendes' expose made during a trip to the United States and reinforced by
environmentalist lobbies, had hopes of a reversal. This explains how the BNDES, a
federal bank for development projects, came to approve in the end of 1988 the
Tapper's Council project with the support of the Acre State's government.
"community managers" chosen throughout the Tejo area to manage financial funds,
construct new storage posts and operate the boats. At the same time, the meeting
19. The planning of activities (bylaws, maps, family rolls, budget, profit
margins, etc.) is reported in Almeida 1989.
88
When the Tejo-based extractive reserve started with the tapper's managed
89
on the fact that the Banco do BrasH had extended a loan to Messias (who sublet
cooperative in 1989, however, the situation changed and the State government
from Cameli, who in turn was Consulmar's lessee) to "finance the natural rubber
turned soon against the project. In the upper Jurua, the tappers' claim over an area
harvest", which was to be done through the "exploitation of the rubber estate." The
covering 500,000 hectares directly affected the major local economic and political
loan was to be used to purchase "utensils and tools" as well as "consumer goods for
bosses, the Cameli and Correa. Through Macedo's actions, the National Council of
the sustenance of the rubber tappers." As collateral for the loan, Messias placed "all
Rubber Tappers dealt a blow to the patrons' hegemony at the very heart of one of
the ' 'traditional''-ar~as. To be sure, in 1988 the Cameli family held a lease over the Tejo area, which they sublet .to the Messias family; while developing their plans (/
extracting high-value timber.
the natural rubber extracted from the estate", which meant the rubber that was to be
extracted the following year. The lawsuit
claim~d
the patrons from honouring their debts to the bank. The patrons' lawyers argued
that the presence of Macedo and Ginu on the estate was "violation of property",
A cooperative with sufficient financial clout to cover the entire Tejo Basin,
based on the idea that bank obligations created monopoly rights:
which included at least 400 families, represented a blow that the patrons' had never
It is necessary to emphasise that the first plaintiff will not be able to meet his -
felt before. By April 1989, around fifty tons of merchandise were stocked on three
obligations with the Banco do Brasil if the invasion should take place, which
boats in Cruzeiro do SuI, the first of them called "Chico Mendes". In order to
forestall the cooperative's operations, the Camelis who leased the estate from
Worse yet:
Consulmar, their subletters Messias and other lesser patrons joined forces and filed
It is true that the crisis brought about by the activists has generated a great
a lawsuit. The litigation sought specifically to prohibit Antonio Macedo,
deal of uneasiness and dissatisfaction, especially among banking/financial
representative of the tappers' Council, and Chico Ginu, union delegate on the Tejo,
institutions, which already are refusing to fund future harvests and,
from entering the -area with the boats loaded with merchandise. As a reaction,consequently, considering the fact that extraction is the mainstay of the local
Macedo drafted community managers Dolor, Damasio, Rubenir, Pedrinho, Gomes,
economy, one can easily imagine disorder taking over the region. 21
Leonardo and Damiao - all Tejo rubber tappers - to take their canoes and appear in
Patrons invoked not law, but tradition. Citing directly from the patrons'
Cruzeiro do SuI on the eve of the court hearing, set for April 25. While Tejo rubber
lawsuit:
tappers thus were mobilised in Cruzeiro do SuI, in an unprecedented show of force
In the municipal district of Cruzeiro do SuI as well as throughout the Juruabefore the court, the patrons counted on the presence of experienced lawyers from a
Valley the payment of rights by rubber tappers to estate owners has been
Rio Branco law firm, who also represented the Uniao Democnitica Ruralista
(UDR), the powerful national association of large land owners.
The lawsuits filed by patrons constitute rare written examples of how labour
relations on the rubber estates were perceived by patrons. "Proof' in this case rested
consecrated by tradition and through the practice and custom of over one
20. Case number 6.673/89, 14 April 1989, Comarca (Court District) of
Cruzeiro do SuI, p. 5.
21. Ibid.
91
90
hundred years, with the occupation of tapping sites and rubber tree trails,
from where latex is extracted ... The estate owners, proprietors and exploiters
trump for the federal government, but the Upper Jurua Extractive Reserve Project
conflicted with the Camelis' interests, which made up an essential part of the
ordinarily are those who outfit the rubber tappers with merchandise, they are -
political support upon which the Acre state government rested. World politics began
to affect the local scene, pitting Acre state government, now in retreat in the Jurua
extending and raising the nations sovereignty to the most remote corners, in
case, against the federal government - creating space for the activities of local
many cases where the devil himself refuses to go to seek his souls. And in
decree transfor~ed the entire area into the first extractive reserve to be defined by
and rubber tappers. are always honoured. The payment of trail rights is
federal law, and in January 1992 the area formally was expropriated. The Tejo River
universally accepted throughout the Jurua Valley, and that is a public and
well-known fact. 22
spread to several other estates, where five new tapper associations were formed and
While lawyers argued for the legitimacy of patron traditions, the rubber
where rents also ceased to be paid.23 Tappers in other zones near the Tejo river
tappers now found allies of their own at the national level, who sought to defend
also contested prices and the repression of free trade, and asserted that rent
payments for rubber trails should not be paid by old-time residents who had opened
their own trails. Now, all these issues were came to be formulated under a general
and a lawyer from a non-governmental organisation, the tappers faced the patrons'
argument: that tappers had rights to the forest they occupied and to free trade, as
.. . .
..
24
well as access to publIc InstltutlOns as CItIzens.
lawyers and managed to block the lawsuit against them, forcing the judge to abstain
from any decision. Community managers took a canoe convoy up the Tejo River,
Conclusion
carrying merchandise on such a scale as to humiliate the patrons. From this date on,
the Tejo River monopoly ceased to exist, and the tappers suspended trail rent
payments definitively.
At the same time, the federal government under President Jose Sarney,
through the powerful Secretary of Environmental Issues Fernando Cesar Mesquita,
seized the opportunity to demonstrate its own humanitarian and ecologicallyminded policy towards the Amazon. In the international context of the 1980s, the
22. Ibid.
---
-----
23. Ibid. In i991, Judge Heitor Macedo reopene~ old cases a?d ordered the
expulsion of tappers from other areas, such as the Cruzeuo do Vale flver.
24. The trail rights and monopoly issues generated a polemical debate in the
Rio Branco (Acre state capital) press. O. "Gazeta" and "0 Rio Branco", SeptemberOctober 1988. The poleJ?lc broke. out ~hen the C~meli fa~ily paid a reporter from
"0 Rio Branco" to publIsh an artIcle With accusatlOns ag~mst ~aced,?, myse~,f and
other persons, to which I wrote a page-length answer, publIshed m the Gazeta.
92
93
patrons' undeniable violence and its association with debt. I will argue that violence
education and public health. Thus, the legitimacy of the monopoly over commerce
the tappers utilises nature and commodities. These resources are monopolized by
In the first situation, the local struggle remained based on the "moral
the patron-traders supported by customary rules and police power, and this
of the forest. Later, the global struggle incorporated the goals of technical and
cultural advancement on the part of rubber tappers. A global economics logic would
predict the disappearance of wild rubber extraction in tropical forests. But between
the world system and real historical events lie national and local political structures.
the first situation, tappers accepted that patrons were owners and that they were
affect power relations, or, put in another way, local class struggle.
unacceptable the abusive commercial gains derived from trade monopoly and the
iIse .of violence to back up such abuses, as well as the illegal physical expulsion of
settled tappers which derived from land monopoly. In the second situation, the
exploited group, bolstered by the intellectual and material support of outside allies,
begins to formulate alternative principles of legitimacy and means of opposing
themselves to violence (Cf. Lenin, 1975 (1902), cap.II, a.).
In the first situation, rubber tappers are based on what Gramsci called
"common sense" - which justified a permanent, undeclared war on the patrons'
monopoly over commerce and land, assuming the forms of smuggling, hearsay,
individual resistance and strikes, or the ''weapons of the weak" in Scott's terms, while
the patrons employed violence and individual corruption as weapons of the
powerful( Scott
In the second situation, the patrons lose their control over both "good sense"as well as over the state apparatus of banks, lawyers, judges and churchmen. The
tappers begin to articulate their struggle against arbitrary violence and monopolies
by adopting the language of land rights, free trade, individual freedom, and access to
95
Chapter 5
TRADE POSTS
Introduction
This chapter deals with the trade post institution as a commercial system, not
as an institution for labour management. We are thus making a clear distinction
between, on the one hand, units of customary appropriation of resources and of
trade-monopoly, and on the other hand the units of work and of natural resources
management. The former units are the seringais or rubber estates, headed by
patrons; the latter are the
coloca~6es
leased from the patrons who often also act as traders who advance merchandise to
tappers.1 The control of access to estates or the monopoly over trade does not
guarantees also the-control over the work processes.
Weinstein advances this hypothesis with reference -to the boom years
(1983: 182ff). Bakx, to the contrary, asserts that there is a "production relation" and
not an "exchange relation" between the patron-trader (seringalista) and the rubber
tapper (Bakx,1986:72,74): in other words, that the seringalista
is not only a
shopkeeper but also a taskmaker. We should first mention that the term seringalista
(seringal entrepreneur) accepted by Bakx is a post-war neologism to replace the
term patrao in official jargon only. The detailed argument proposed by Bakx in
support of his thesis mentions that the tapper "must tap his estradas each day,,2
---
(emphasis mine), when both field research and census data show that the average
they did not manage the actual labour process (Chapter 3); and that the patrons'
working week has four tapping days in contemporary estates under patrons and
main concern was to ~aintain a monopoly through coercion and incentives over
older estates did not enjoy a higher productivity; that "cooperative activity... tended
autonomous forest peasants (Chapter 4). In this chapter the role of the trade-post
to occur under the direction of the seringalista, e.g., in the formation of teams to
institution will be described in more detail. The main feature which will emerge
open up new estradas or guide ox trains and so on", while in the Upper Jurua such
activities ended with the rubber boom and tappers could open up new trails on their
own even before; he also argues that the patron employed "armed guards, 'fiscais',
to police the estate" and that "the fiscais checked that the seringueiro tapped in the
prescribed manner ... [and] kept out the river traders", apparently mixing up the
description is that the barracao, represented by the patroes, operating with long-
patron" thesis that "the seringueiro was also obliged to stamp the pelas of rubber
on a itinerant retail trade, and by the permanent marreteiro petty trade. And
with the mark of the seringalista", which fact "had the effect of re-enforcing the
although this structure of trade posts, itinerant peddlers and forest shops channeled
notion that the pelas belonged to the seringalista and not to the seringueiro", when
the rubber produced by the Tejo river's 491 houses in an annual ecological and
under the "embarked rubber system" the existence of both the patron's and the
economic cycle, establishing links through debts and credit with even the most
individual tapper's mark in fact asserted the individual tapper's ownership until the
remote units, the real production process almost always occurred within the forest
rubber was transferred by sale to the trade post. Finally, "the fiscais were ordered to
destroy any such subsistence plots that were discovered on the seringal", which
presumably refers to the boom period (Bakx 1986:76).3 Thus, Bakx's argument that
tappers were labourers under a capitalist boss (Bakx 1986) and his later argument
that a forest peasantry of tappers only emerged in Western Acre during the
seventies (Bakx 1990:50) is far from being conclusive.
We have argued instead that the peasantization of Acre rubber tappers in the
upper Jurua valley can be traced to the end of the rubber boom (Chapters 1-2); that
without a priest, a school, a few shops and a television set. A 1991 census of the
patrons were unable to modernize rubber production because among other factors
Thaumaturgo district (not included Vila Thaumaturgo, nor the Arara-Jaminawa and
3. On these points, see chapters 1-4.
Camp a Indian territories) revealed a total of5,983 persons in 856 houses, over an
98
area of 5,062 square kilometres. 4 The Tejo River valley, the richest in population
and rubber output in the Thaumaturgo district, had 3,432 inhabitants in 491 houses
99
of beans, maize and tobacco, while rubber was the seringueiro's main currency, both
barranqueiro and seringueiro were cultivators. Thus, over 90% of the Thaumaturgo
spread over an area of about 2,644 square kilometres, which excluding the Araradistrict houses had their own manioc gardens, irrespective of being "rubber tapper's"
Jaminawa Indian territory at the Bage River headwaters not counted, leaves around
2,362 square kilometers (Map 3; Table 5.3). Field research was carried out mostly
The Tejo basin itself is clearly divided in "lower course" (0 baixo) and "upper
on Riozinho, an igarape at the Tejo headwaters with around 390 inhabitants living
course" (0 alto), and in "banks", the margem, and "hinterlands", the centros. On the
in 66 houses, on an area of approximately 230 square kilometers (Map 4; Table
5.3a).
Lower Tejo there were the rubber estates of Boca do Tejo and Iracema, holding
about 13% of the total Tejo popUlation. On the middle course of Tejo river both
hinterland estates (the Bage estate, on the Bage parana and the Chaleira igarape
with 21.8% of the Tejo population) and riverbank estates (the Horizonte, the
Fortaleza and the Maranguape With about 21.1% of the total Tejo popUlation)
which increase as one goes farther away from the banks of the Jurua, reflect the
existed. But the most populated zone was that of the Tejo headwaters, were the
basic feature of rubber extraction in the Thaumaturgo district. It is an activity
Tejo river branches off into several small courses - Dourado, Riozinho, Manteiga,
oriented towards the hinterland zones (centros), away from the riverbank areas
Camaleao, Boa Hora, Machadinho -, with their own trade posts, which together
(margem). The Tejo River is the main example on the Upper Jurua of this sort of
form the
accounting for -
hinterland specialized in rubber production, and Riozinho is the Tejo River's pearl
as far as rubber extraction is concerned.
The remaining Thaumaturgo district inhabitants live either on other Jurua
river tributaries (Amonia, Arara, Sao Joao do Breu, Acuria, Breu) where they are
Although the overal density of settlements is higher the farther one goes
seringueiros (rubber tappers) as on Tejo River, or on the Jurua river banks where
away from the margin, it is accompanied by spatial dispersion of the rubber tapping
sites as they approach the Tejo headwaters, when the waters split into a seemingly
endless series of igarapes, paranas, brooksan-d footpaths. Rain rarely falls between
rubber tappers, while the remaining were either river bank cultivators, or artisans,
the months of June and September. The water level of the Jurua diminishes laying
or worked on a dozen small cattle ranches along the Jurua banks. Note however that
bare huge sand banks, while the Tejo itself becomes a thin veil of water covering a
although the barranqueiro's main purchasing means derived from the seasonal crops
shallow, sandy bed, clogged with tree trunks which block even the lightest uba and
interrupted by rapids. By then, one drags canoes, axes are used to clear the way, and
even the flat-b~tto~ed uba must be hauled land sometimes. The smaller streams
such as the Riozinho allow only foot traffic. Between October and December,
Dourado, Manteiga, Ri6zinho, Camaleao, Boa Rora, Boa Vista and Machadinho.-
sporadic rains cause repiquetes, or flash floods, but only between January and March
These had each their own petty resident patrons (often marreteiros which made
the water level of the Tejo make access possible to the middle course of the river,
while during the heavy rains even the Upper Tejo temporarily accommodates boats
and the Riozinho can be negotiated. For the 490 households spread throughout at
rubber estate. 5
area of 230,000 hectares of the Tejo river, both the supply of merchandise and the
shipment of rubber are conditioned by nature's rhythm and by the fractal geography
the hiemrchical multiplicity of patrons. First one must distinguish between lease
patrons and patron-traders, each constituting its own structure, within of which one
may identify the three-tiered tree-like structure. As a result of these concurrent and
spreads out to successive secondary branches until it reaches its leaves. The main
trade post at the Tejo river's mouth (the seat of a major estate) feeds several of the
persons and roles within Tejo river. There was the major patron-trader with shops in
Tejo middle estates (either on the banks or on hinterland tributaries) and the Upper
Cruzeiro do SuI and who supplied merchandise to the main trade post at the Tejo -
Jurua's middle trade post; the latter branches off into several tributaries and
mouth; there were the mid-sized patron-traders (who sometimes lived on the Tejo
corresponding minor trade posts. Thus, a major estate splits off into middle estates,
River for two generations and were supplied by the major patron-trader, thus
and these finally are divided into minor estates, the smallest units with their own
commercial network and system of rent payments. Let us consider this fractured
structure within the Tejo River hinterland, where all three levels - major, medium
a single, main,
property structure based on the payment of rent (Diagrams 5.1 and 5.2). At the top
of this property structure was the Sao PauIo company which claimed the whole Tejo
strong Cruzeiro do SuI patron (during the 80s, the Correa & Sons Co. or the Cameli
River Estates' rent. Below this there was the Cruzeiro do SuI company (Correa & .
& Brothers Co.), with a main barracao (storehouse) at the River Tejo's mouth. The
Brothers in 1982-1985) who leased the Tejo River Estates, which could sub-let parts
medium estates under the Tejo Estates were the Horizonte, Fortaleza, Bage and
of the Tejo River Estates as exemplified by the Cameli & Sons' sub-letting of the _
Restaura~ao,
Restaura~ao Estate to the Messias in 1988. But there could be minor patrons on the
Restaura~ao
Tejo who were not patron-traders and who only collected rents, as in the case of the
At the top of this hierarchy, there were major patron-traders, wealthy men
In 1991, in the Tejo River Basin as a whole (2,632 square kilometres), there
with business and luxurious residences in the town of Cruzeiro do SuI; at the bottom
were around 1,414 trails distributed among 166 coloca~6es, but only 982 active trails
of the ladder there were minor patrons-traders who lived on the Tejo River and -
were rented by 491 houses (Table 5.3). On the Restaura~ao Estate (around 1,156
included
square kilometres), in 1981 there were around 657 trails from which 451 being
son was also a rubber tapper. Sales, a minor patron-owner without capital to run a
rented by 205 houses in 105 settlements (Table 5.2, 52a). Houses are situated in
trade in merchandise, also had sons working as tappers. A forest peasant could pay
groups of two or three on a central cleared area within each forest settlement,
rent to a lease patron who resided on the estate, while at the same time maintaining
house uses a section of over 5 square kilometers as if it were a slice of a pie (Table
In short, though the tree-like structure of the estates made each forest house
part of a vertical chain, consigning each person to at least one patron, a partial
Rent was the responsibility of individual house heads, whose names were
discontinuity between trade and ownership, together with the hierarchy of minor,
registered in the main trade post along with the number of trails rented. There was
medium and major levels, made the resulting structure more open to choice and
no record of "tappers", nor were the "trail rights owners" (titulares) mistaken for
tappers (facas or "knives"). The number actually rented ranged from a minimum of
one trail to four and even more trails. A house head (either man or woman) who
rubber trees owned by someone, and the supply of merchandise to the Tejo River
rented a trail was not necessarily a rubber tapper. Effective labour on the trails, and
the number of trails rented, was a domestic issue, to be resolved by the house head
renting the trails from the patron. In most cases, labour came from the domestic
group, though there were cases where employees were hired by the house head, and -
The main natural resource of a estate is a set of wild seringueira rubber trees
found along the rubber trails or estradas. A rubber trail also constitutes a measure
of labour, production, as a number of rubber trees which are tapped in a day's work.
A number of rubber trails, usually around a central clearing, makes up a
(a forest settlement). The
coloca~ao
coloca~ao
continuous territory comprising a group of trails. While each trail was rented by an
where one house took on other houses as its own clients in private arrangements.
In principle, trail tenants paid thirty to thirty-three kilograms of rubber (from
which the usual tare for excess water was not deduced) per trail composed of 120
trees. Rent could not be paid in cash or with any product other than rubber and did
not depend on the productivity or the exact number of trees. Since each house
ordinarily rented two trails, the average rent on the Riozinho Estate stood at 66
kilograms of rubber per house. This amount corresponded to about ten per cent of
effectively were occupied, while 31 % were not. These proportions were reproduced
on the smaller Riozinho Estate (Table 5.2a; cf. Table 5.3). It should be kept in mind
Reports from the boom years (1870-1912) mention that rent was paid as a
and loans. Thus, the low overall occupation of trails could only reflect the scarcity of
middle Jurua river; as it is reported for the Tejo river, it was a fixed amount of 66
labour, since there yvas no sensible unemployement in the Cruzeiro do SuI region at_
kgs per "pair of trails" (a "rent in kind" system) already in 1899 (CabraI1949), which
the time.
is confirmed by old tappers who say that rent exists in this form "since the beginning
In 1982, the new patron Correa took control over Restaura~ao. One of .bis
of time". Although rent averaged 10% of production6, the point about a "rent in
measures was to introduce the "half-trail" unit as accountable for rent payments.
With this innovation, Correa was reacting to the tappers' practice of increasing the
to the house head; furthermore, rent in kind was not affected by price fluctuations
number of trees in each trail, in reality incorporating well above the standard 120
and could not be manipulated by the accounting system. Such a system was
trees each for extraction in 'each ,trail, while paying rent for only one. In effect, this
meant that houses opening new trails were asserting the principle of not paying rent
trade-
for new trails. As a consequence of Correa's measure, between 1982 and 1983, the
number of rented rubber trails in the Riozinho Estate increased seven per cent, with
Restaura~ao
coloca~a.o
coloca~ao,
Restaura~ao
an increase both In the number of houses paying rent (from 59 to 63) and the
number of trails per house from 2.20 to 2.23 (Table 5.6).
Rent owed
On the other hand, by the end of the first year of the Correa era, many
records showed
tappers had not paid rent. Thus, towards the end of 1983, though by the estate's
with 205 houses renting 451 trails; the total number of trails was
account the 63 houses owed 3,905 kilograms of rubber (on the Riozinho Estate),
657, which meant that 206 trails remained idle. In other words, 69% of the trails
only 2,662 kilograms had been collected (Table 5.6). Moreover, the actual amount
coloca~6es,
coloca~ao.
Restaura~ao
of registered rent represented for each house a payment of 62 kgs. of rubber per
6. Production was estimated in around 300 tons for 491 households' for an
average of two trails per household, total rent would be 32 tons of rubber an
estimate co~irmed by tappers and employees. This meant US$54,000 in i982
(Rubber pnce:. Cr$400; exchange rate: Cr$220 to one dollar; price paid to the
rubber tapper: US$1.80).(IBAMA 1988.)
house (and not 73.6, for the average of 2.23 trails). This occurred due to the
only three (90 kgs.); another house exploited 3.5 trails (as recorded in the trade
individual negotiations between the patron and the houses which had as an effect
the reduction of rent rates. For example, one house rented four trails but paid for
post), but paid 100 kgs. instead of 105; another paid only 45 kgs. for two trails. The
book-keeper, who had worked as a rubber tapper on the estate during half of the
harvest, was to pay only 30 kgs. for two trails. Two houses simply refused to pay rent,
with special reasons of their own: they were female-headed houses, one run by a
widow and the other by a single woman, each with several children, and when a
fourth woman asked the patron to open an account and put the trails in her name,
given that her husband was chronically ill, she pleaded that he "pardon the rent". But
this rule was not always observed, since a fourth female-headed house head paid
rent with pride.
. In sum, reasons alleged for reducing or not paying rent included thus labour
jnvestments (trCl:il e.xtensions opened by tappers) and household handicaps (age, sex
and illness), but discounts could also be a result of bargain as part of the final
negotiation between patron and house head. The fact that rent allowed for
exceptions can be interpreted as suggesting that the institution itself was seen as
abusive. This may explain also why, contrary to what happened with rubber used to
"pay up accounts", rubber destined to pay rent was not subject to tare deductions:
"rubber to pay rent is tare-free", it was said, while the patron of Restaura~ao
claimed that he charged rent only because he had to pay his own landlord, although
the deeds to back up such an allegation were never presented to the tappers.
The patrons who claimed rights to trail rental fees never proved their
possession to the rubber tappers; on the contrary, it was the rent payment made by
tappers which, in a sense, proved ownership as based on "custom", while "custom"
itself prevailed only because it was sustained by municipal police authority since
outright non-payment could justify the eviction of tappers with the support of court
Credit and debts, advances and payments constituted the sources of wealthon the rubber estates. The financial and mercantile capital circuit in these estates
worked in the following fashion. Three large public banks - the Banco do Brasil,
Banco da Amazonia and the Banco do Estado do Acre directly financed the largest
patron-traders in Cruzeiro do SuI provided only that they demonstrate rubber estate
leases, giving to them loans to be paid within one year. Such annual loans were paid
with interest set lower than the 1982 inflation rate, with the future rubber "harvest"
serving as collateral (in the cases of sub-patrons who let a estate, and did not own
property themselves, rubber was the only guarantee offered). 8 The size of such bank
loans was determined according to a rule of thumb: each pair of rented rubber trails
(and thus each house) was worth 400 kilograms of rubber as collateral, the average
productivity per worker in Acre during lean years. Thus rents guaranteed financecapital.
On an estate like Restaura~ao, which had 225 "pairs" of trails rented to house
heads in 1982 (or 451 trails), a loan could r'each as much as the value of 90 tons of
harvested rubber, or around US$162,000 in money. This money was presumed to be
used in the purchase of merchandise, classified as "work instruments" and consumer
goods. Fabrics and hardware were 'acquired in Sao Paulo, Brazil's industrial
capital. 9 Cargoes were sent by ground transport overland to Belem, and from there
floated up the Amazon and Jurua Rivers. Major patrons such as the Cameli
possessed their own rivercraft which also transported rubber and bulky merchandise
decisions (Table 5.7). Rent was not so much a means for extracting a major surplus
from tappers, as a means for legitimation.
8. Furtado and Cameli vs. Macedo and Melo, lawsuit reproduced in Santoro _
harvest, was to pay only 30 kgs. for two trails. Two houses simply refused to pay rent,
with special reasons of their oWn: they were female-headed houses, one run by a
widow and the other by a single woman, each with several children, and when a
fourth woman asked the patron to open an account and put the trails in her name,
given that her husband was chronically ill, she pleaded that he "pardon the rent". But
this rule was not always observed, since a fourth female-headed house head paid
rent with pride.
Credit and debt~, advances and payments constituted the sources of wealthon the rubber estates. The financial and mercantile capital circuit in these estates
worked in the following fashion. Three large public banks - the Banco do Brasil,
Banco da Amazonia and the Banco do Estado do Acre directly financed the largest
patron-traders in Cruzeiro do SuI provided only that they demonstrate rubber estate
leases, giving to them loans to be paid within one year. Such annual loans were paid
. In sum, reasons alleged for reducing or not paying rent included thus labour
jnvestments
and illness), but discounts could also be a result of bargain as part of the final
negotiation between patron and house head. The fact that rent allowed for
exceptions can be interpreted as suggesting that the institution itself was seen as
abusive. This may explain also why, contrary to what happened with rubber used to
"pay up accounts", rubber destined to pay rent was not subject to tare deductions:
"rubber to pay rent is tare-free", it was said, while the patron of
Restaura~ao
claimed that he charged rent only because he had to pay his own landlord, although
the deeds to back up such an allegation were never presented to the tappers.
The patrons who claimed rights to trail rental fe~s never proved their
possession to the rubber tappers; on the contrary, it was the rent payment made by
tappers
which, in
a. sense, proved ownership as based on "custom", while "custom"
.
.
.
itself prevailed only because it was sustained by municipal police authority since
outright non-payment could justify the eviction of tappers with the support of court
with interest set lower than the 1982 inflation rate, with the future rubber "harvest"
serving as collateral (in the cases of sub-patrons who let a estate, and did not own
property themselves, rubber was the only guarantee offered).8 The size of such bank
loans was determined according to a rule of thumb: each pair of rented rubber trails
(and thus each house) was worth 400 kilograms of rubber as collateral, the average
productivity per worker in Acre during lean years. Thus rents guaranteed finance capital.
On an estate like Restaura~ao, which had 225 "pairs" of trails rented to house
heads in 1982 (or 451 trails), a loan could r'each as much as the value of 90 tons of
harvested rubber, or around US$162,000 in money. This money was presumed to be
used in the purchase of merchandise, classified as ''work jnstruments" and consumer
goods. Fabrics and hardware were acqtiired in Sao Paulo, Brazil's industrial
capital. 9 Cargoes were sent by ground transport overland to Belem, and from there
floated up the Amazon and Jurua Rivers. Major patrons such as the Cameli
possessed their own rivercraft which also transported rubber and bulky merchandise
decisions (Table 5.7). Rent was not so much a means for extracting a major surplus
from tappers, as a means for legitimation.
8. Furtado and Cameli vs. Macedo and Melo, lawsuit reproduced in Santoro _
. . '
"
,
"
'
f,
109
108
for Cruzeiro do SuI commerce, such as gas. Major patron-traders also managed to
stocks (in the principal, medium and minor posts), and in the tappers homes as
avoid paying sales or value-added taxes, by declaring their companies as part of the
paid in rubber at the end of the yearly 'harvest'. The corresponding chain of debts
costs over the course of 5,300 kilometres, by land and by river, from Sao Paulo to
linked rubber tappers to the minor deposits, the minor deposits to the mid-sized
tra~port
achieve profits. lO S"ince the major patrons also owned stores in Cruzeiro do SuI, part -
trade posts, the mid-sized trade posts to the main trade post, which in turn owed to
the bank (Diagram 5.2).
TIe average expected output of the Tejo Estates was 300 tons of rubber,
tappers, and to river traders who sought to replenish their stocks for resale on the
which would mean 612 kilograms produced on the average by 490 households.
Deducting around 30 tons of rent payments, the remainder should correspond to the
Between March and April, at the beginning of the tapping season, the major
~al~e of debts for ~erchandise advanced to the rubber tappers. As we have seen,
patron in Tejo river supplied the principal trade post - the Boca do Tejo trade post,
the bank loan could be worth 196 tons (490 pairs of trails worth 400 kgs. each). The
205 houses of the Restaura~ao Estate were expected to produce something on the
order of 120 tons of rubber, and on the Riozinho, a smaller estate, the
stocked at the Boca do Tejo post, while the rest was sent upriver on flatboats to the
post. On the
After collecting the rubber from the seringal houses "as payment for the
"merchandise advance, the main patron would sell rubber to refining plants in
Restaura~ao
smaller trade posts such as the Riozinho's, now transported in canoes or flat- -
Cruzeiro do SuI (there were two plants with about 25 employees apiece), and pay
bottomed, one-ton uba boats, as long as the water was deep enough. Up until this
back the banks, keeping a profit. The plants (usinas) then would supply tyre
point, the merchandise was transported in the patron's vessels and by his employees.
manufacturers in Sao Paulo (such as Goodyear, Pirelli and Michelin) which must
The rubber tappers themselves carried supplies on their backs from the minor trade
With the"rubber bought from the estates in Cruzeiro do SuI and surrounding
for at least two and up to six hours in the furthermost forest settlements. There were
municipalities along the Jurua valley, the processing plants exported as much as
3,000 tons a year (figure from 1982). Including the production of the other river
settlements. By the end of the whole process, the merchandise was split up in several
valleys in the state, Acre contributed with 13,000 tons of a total 20,000 tons of
Restaura~ao
111
110
wooden edifications including the trade post, the residences . of the administrative
The assistant manager helped weigh rubber and also attended customers in the
personnel (manager, assistant manager, and book-keeper), and the houses of other
store. The book-keeper maintained the individual accounts of rubber tappers and
residents. Only the trade post was more impressive, with piles of rubber on the
performed the general accounting of the store. Though the residents of the main
veranda and a variety of merchandise in the store: six Elgin sewing machines, five
settlemerit depe-nded in part on warehouse-food stocks, most of their basic food was -
three-band radios, three chainsaws, three gas stoves, five engines (two 10-HP Briggs
bought frofu an autonomous cultivator and from neighbours who sold meat and fish
& Stratton, one 3-HP Montgomery, one 8-HP Tiete and one 3-HP Branco), and
twenty-five wristwatches and clocks by October 1982. Besides these ''valuables'', one
The trade post, located on the top of a steep bank, was connected to a
rudimentary river port by a stairway. Among its rivercraft, the estate had a baleeira
biscuits, coffee, sweets, cooking oil, soap, powdered milk, cosmetics, lighters, combs,
medicine. Basic goods such as regular soap, salt, kerosene, dried beef, beans, sugar,
diesel engines, which could carry lighter loads, including two flat-bottomed canoes
~obacco
(ubas), used to go up the Tejo during the dry months between June and September,
when even the flat-bottomed baleeira would get stuck. The rivercraft were operated
by two permanellt employees, Dico and Zuzu, who live in the Tejo's mouth
The main store had a large, 150 kg. scale and a counter with a few smaller
headquarters and not on the Restaurac;ao Estate. Emergency trips also could be
scales and adding machines behind which the manager or his assistant occupied a
conducted by a -resident of the other bank, Ostemo, who used his own canoe and -
desk and jotted down the clients' purchases. In the back of the building there was a
charged for his services. Dieo, Zuzu and Osterno played an important role in "taking
warehouse with stocks, as well as the patron's or manager's office, where records of
the rubber out" from the tappers' homes, and in accompanying the canoes guiding
In 1983, there was a mateiro on Riozinho Estate, hired by the new patron.
The manager's residence, where the patron stayed during his visits, stood on
His job was to visit each rubber trail at least once every year, inspecting its state of
one side of the warehouse. The assistant manager's house was on the other side, and
preservation. Before being hired for the job, this man, a Riozinho resident, was
the book-keeper's house behind it. These buildings differed from most rubber
himself a tapper and also a part-time tinsmith. The mateiro (literally, ''woodsman'')
tappers' houses for being made of timber planks instead of palm tree bark, and
tapping flag on a tree when he found damage done. In such cases he made a mark
on the damaged tree or flag, an "x" for instarice, which the tapper would encounter
112
on his next journey to that trail. The prohibition would be valid for the whole season
teacher was the blacksmith's daughter-in-law and did not receive any remuneration
or for the whole year, and it could be extended to a whole trail. The mateiro could
for the classes she taught in a one-room school without walls, built by the men of her -
also suggest the expulsion of a tapper as punishment for serious damage inflicted to
house. The carpenter frequently was called upon to erect wood constructions or to
a tree.
build canoes. The two tinsmiths were rubber tappers who bought their raw material
While occupied permanently with the hundreds of trails he had to visit each
(oil cans) from tappers and sold pails or tin cups to the trade post or directly to the
.---/
year on a mid-sized to large estate (small estates rarely could afford hiring a
tappers.
woodsman), the mateiro could not keep tabs on the daily work routine of the
In the past, Restaura~ao's central settlement had a church, known for haviug
individual tappers, and his task was concerned with the technique employed, aiming
the largest bell in the region. There was no permanent priest, and the Restaura~ao
to avoid the dilapidation of the rubber trails, the rubber estate's natural. capital. The
received the visit of clergy only during the annual desobriga ceremonies of baptisms
rubber trails, not the land, the constituted natural resource holdings that made up
and marriages. By the early eighties, a priest still visited the area for desobriga once
the rubber estate: it made sense to preserve them. Depredation of the trails and
a year. The main role of the--church in old days, however, was to hold the annual
smuggling of goods constituted indeed the two majo~ crimes against the rubber
novenas: nine days of prayer and commerce, during a fixed period of the year,
commemorating th~ patron saint's day. Several novenas are celebrated to this day_
The Riozinho tappers, particularly the older ones, refused to accept the new
along the Jurua, which function also as commercial fairs, drawing together a great
mateiro's authority. They argued that he did not have enough expertise on the job,
number of river tr~ders, rubber tappers and Indians. In the past there also used to
and some alleged that they could teach him a lesson on tapping techniques. The
main patron did not insist with the mateiro nor replaced him. The fact that the new
hung, to accommodate men and women who had come to the post for religious
patrons leased the estates for only three years may explain why they did not show
feasts and commerce. The central settlement also had a levelled area, which served
Besides its commercial function, the central area of an estate also gathered
When tappers visited the central post for sporadic purchases or in case of
together other -seivices. There lived at the Restaura~ao head office area a -
emergency, this would involve often spending the night, when tappers lived far away.
blacksmith and a schoolteacher, and further away there were a carpenter (two lived
Patrons or managers were then expected to extend hospitality, offering lodging and
in the forest hinterland). These people were not trade post employees, but settlers
meals to tappers during their visits to the main trade post. But tbis rule ordinarily
whose residency rights derived from the fact that they rented rubber trails. The
was not observed, and only the friends of the patron or the manager received now_
such treatment. For the other tappers, there was the alternative of seeking shelter
11. The Residents' Association planned to reinstate the use of mateiros after
the patrons left. Cf. "PIano de Uso da Reserva Extrativista do Alto Jurua
(UtilIzation Plan for the Upper Jurua Extractive Reserve)" unpublished document
'
December 1991. See also Almeida 1991.
with the residents of nearby house (though there were plans to erect a new hostel),
and buying biscuits, sweets, condensed milk and canned beef. Cacha~a, or cane
114
115
brandy, was not sold because the patron feared that it would result in violent
incidents involving drunken rubber tappers, but the tappers would substitute it with
pure alcohol mixed with artificial fruit juice bought at the store, or would purchase it
from the river traders.
Besides the major trade post, other smaller storehouses or simply "goods
deposits" existed on the minor estates along the small streams that formed the-
to autonomous niarreteiros again. Thus, the minor posts were at the fringe of the
formal trade-post structure, and continued to operate without it, being then supplied
by itinerant traders or buying at other trade-posts downstream.
Recruitment. Rubber Removal and Weighing
There were two ways of becoming part of a rubber estate such as
headwaters of the Tejo. The managers of such deposits supplied merchandise over a
Restaura~ao.
smaller area, collected rent and helped the main trade-post personnel. They might
colocaltao. This consisted in allocating a set of idle trails on one of the settlements,
be recruited for the position by the current patron, as was the case of Timoteo on
for which the person would now pay a rent. At the same time, the new trail lessee
the Riozinho. Until the previous year, Timoteo was one of the most productive
would have an . ac~ount open at the trade-post. The patron could also recruit a_
tapper el~ewhere, as Correa did in Cruzeir?"do SuI ~nd in "ila Thaumaturgo. The
Restaura~ao
by his
One was to ask the patron or manager at the trade post for access to a
other was to obtain trails by transfer from another rubber tapper, or to be recruited
by another tapper to work on the trails rented by him. Tappers who received rubber
Timoteo's two grown-up sons tapped rubber trees and himself and
his wife worked on the manioc garden. He continued to be a house head paying
Th
.
ese marreteuo-managers received
open an account at the trade post. But this was not always necessary when a house
head who exploited a trail was the client of another tapper, who possessed rights to
the trail and was responsible for accounts with the trade post.
merchandise from the central trade post and would resell it to tappers, using the
Between 1982 and 1983, in the Riozinho, I recorded six cases of rubber
rubber to pay for the products and drawing a profit from price differences, but also
tappers who were recruited directly by the trade post for the Riozinho, increasing
would assist the trade post employees in the transport of goods and in the reception
the
rent-payers to a. total of 66 households (three other houses did not pay rent).
.
One of these quit his former patron Carneiro; another quit a seringal on the
Tarauaca river on the over the backwaters of the Restaura~ao, bringing with him his
Restaura~ao
own client. Both were cases of tappers unhappy with their former patrons and with a
and the implantation of a cooperative did not affect the three main
balance. Besides these six new tapping houses, other fourteen new house were
formed, of which ten as the result of newly married young men who came to occupy
part of their father'S, father-in-Iaw's, or brother's colocaao, or an idle trail within
the estate where they already lived. Other four new houses moved in from
~eighbouring i~ar~pes
Restaura~ao),
repiquetes _ flash floods that offered rare opportunities to float the rubber bundles,
which weighed more than 50 kilograms apiece, downstream. Patrons expected the
answering the call of relatives or acquaintances to occupy idle rubber trails. These
tappers to help to transport the accumulated rubber of their settlements to the trade
twenty new houses did not increase significantly the total number on the Riozinho
post, I witnessed cases in which five tons of rubber remained stocked in hinterland
~me
areas of difficult access from one year to the next, because the crucial rains were
missed. Once removed from the interior of the estate and placed at the trade post,_
migration or death.
A patron or his manager in principle had the authority to evict a house head,
the rubber bundles were tied together in long floats weighing scores of tons, which
in turn were piloted down-river by canoes, to the mouth of the Tejo and on to
smuggling rubber out while in debt to the trade-post, or failing to pay the rent, or for
doing damage to the rubber trails. If a house head refused the notice to leave, the
Cruzeiro do SuI.
"Weighing the rubber" constituted another important activity exercised by the
patron could resort to the Cruzeiro do SuI police. The patron would also cut off
trade post. The operation ostensively consisted in determining the weight of rubber
credit to the house in question. Attempts to expel tappers by merely cutting off their
stocks in the hinterland settlements, along the year. But in reality the operation also
credit at the trade post did not succeed in the cases I observed on Riozinho , since
meant the nominal transfer of rights on rubber stocks to the trade post: this was
rubber tappers always could make clandestine supply purchases from itinerant
I did not witness any successful eviction, which may have been a consequence
presence. Neither ~ves nor children could substitute the head during the weighing. _
area, it seems likely that the union played an important role in reducing evictions in
And there were house heads who refused to liquidate. These cases involved highly-
productive tappers who' disagreed with the price and interest rules of the liquidation
improvements resulting from their own labour, which implied the payment of
process, but also included tappers with low productivity who had personal disputes
with the trade post. Since the weighing procedure involved the political relations
At the end of January and the beginning of February, the trade post
between the trade post and house heads, sometimes the manager himself led
weighing sessions, negotiating directly with the tappers. This occurred on the
from the densely-forested hinterland took place during heavy rains, or using the
Manteiga River, an area where the rural union's strength was felt in 1982-83, and in
119
home showed one result, while the manager's scale showed another), over the
number on the scale (the result could be altered by a crooked pointer or by faded
to the households and not (as they argued in legal cases attempting to preserve their
monopoly) to the bank, or, by extension, to the patrons who were financed by the
out "green", gradually becoming "dry" with time), but the estimate usually was an
arbitrary exercise, varying from five to fifteen percent of gross weight. River traders
bank.
The trade post did not have personnel to construct or maintain the
footpaths, bridges and stairways that connected the main settlement to the tappers'
homes, nor did it develop any regular or even irregular transport system (using draft
animals or boats) linking the trade post ~o the colocac;oes, which took two to six
hours to reach on foot (see map 4).
were reputed to cheat tappers both on the weigh-scale and on the tare discount,
Supply of Goods and Accounts
This is why I tell you that lam surprised with my account. I didn;t buy
four visits to the Riozinho estate, which accounted for the product of the first cut on
the Jurua (from April to July), or the dry season harvest. A total of 68 households
only sugar, coffee and oil. What surprises me is how the others manage to
were visited for weighing, though not on the same occasions. The total production
pay. Because I buy none of that. It's been three years since I bought a knife.
for the house ... Luxury, no sir: coffee is food, oil is food, sugar is food. One
can live without these articles, but in ,a disgraceful mariner. Look: we can live
innovation to which we will come back), and an average household output of 294.5
kgs. per half season. This production of 64-kilogram bundles was "to settle accounts",
Merchandise introduced from Cruzeiro do SuI was stored at the mouth of the
and did not include rubber sold in smaller amounts of ten to thirty kilograms, to the
Tejo, then transported to the Restaura~ao warehouse and finally distributed to the
river peddlers
and marreteiros.
Individual household
productivity varied
Ze
considerably, ranging from under 100 kilograms (two female-headed households) to-
over 700 kilograms for the half-season (Tables 5.8 and 5.9).
kept at the Restaura~ao trade post, composed of stocks from the previous year
Besides the apparent function of transferring rights over rubber from houses
(including those in the smaller deposits) and the merchandise received during the
to trade-post, another purpose behind the weighing of rubber was to evaluate each
current year from the main trade post at the mouth of the Tejo. In order to
appreciate the _si&nificance of these stocks, which were to supply over 220.
be supplied over the course of the year. But an inevitable conclusion is that through
households (almost 70 on the Riozinho Estate), goods may be divided into several
the weighing procedure, the patrons implicitly recognised that the rubber belonged
specific categories.
121
120
Goods that were used throughout the year (not including long-term
durables), which had to be acquired at the beginning of the year, were called
. ..
.. hI
db ause some households did not
not avalla e an ec
IS
"estivas" (gross goods). The "estivas" included articles for domestic consumption
(salt, soap, sugar, cooking oil, powdered milk for children), hunting needs
(ammunition), and fuel for engines used in the flour-making process. In contrast
~th the "estiva~", t~e "luxos" (luxuries) and "vicios" ~("Vices" or "habits") were articles
COMMENTS
NAME
Vicios (vices)
?
Miudezas (trifles)
Sewing
Cig.smoking
Body care
Writing
Production goods, durable
Utensilios
Agriculture (Funds)
Tapping .
Fishing
Building
Household equipment, durable
Kitchen
(Funds)
Other
Tecidos
Cloth
Garments
Hammocks
Moveis de Valor
Motors
Machines
Investment Goods
(Long term funds, saleable)
lasting articles with high unitary values, including gasoline and diesel powered
engines, chainsaws, mattresses, suitcases, gas stoves, sewing machines, radios,
watches and clocks. The term "m6veis de valor" suggests that these goods
constituted investments, which evidently is the case with gas stoves and frequently
122
years, which may be repaired or substituted over time ("utensflios"), and (c) those
durable items that may be kept as property holdings ("m6veis de valor"). Another
important additional classification distinguished between basic items in any of the
123
a tapper calculating -the depreciation cost of his equipment (either that used inrubber tapping, or to make flour) in his assessment of the total value of the products
of his labour (rubber or flour).
above categories (such as salt, soap, petrol and ammunition, which always are
needed in day-to-day consumption; or such as the flour-mill engine, a permanent,
long-term necessity) and non-basic items (whether "luxos" or ''vicios'' for ready
consumption, or "m6veis de valor" kept over long periods).
Restaura~ao
Restaura~ao
BASIC Itens
NON-DURABLE
Food
Soap (coarse)
Ammunition
Fuel
Basic Miudezas
Basic Cloth
DURABLE
NON-BASIC
Luxury food, vices
Soap ("sabonete")
between January and August 1983 (not including the goods stocked over from 1982)
had a total value of 29,000 kilograms of rubber, using the final prices paid by traders
at the smaller posts to the ;maintrade-post. At least -another 15% must be tagged on
to
? Luxury miudezas
Luxury cloths
Utensils
Valuables
Kitchen sets
Valuables
As equipment
stock or display
Restaura~ao
~ccount
for the small trader's commission, resulting in the total value of 33,000 _
kilograms of rubber at tappers' prices - a little over one-fourth of the total expected
. R estaura~ao.
- 13
output of all the minor estates composmg
Once transferred to the tappers' houses, the goods were converted into a
growing debt to - be paid in rubber (Table 5.9). In a parallel process, houses
accumulated stocks of rubber until it was transferred all at once to the trade post.
The tapper did not know at any given moment the precise size of his accumulated
In principle,-the "goods of value" may be used not only as accumulated wealthor equipment stocks but also as capital: this was the case when a motor or other
instrument was lent in exchange for some sort of gain. Indeed, I observed on
different occasions payment for the use of a chainsaw or a flour-processor (equipped
with a 3 HP engine, a metal plate and tools to grate and press manioc), both of them
expensive equipment. But these (rare) payments were often part of a transaction
that included the purchase of manioc (together with the right to use the manioc
debt; an d, through the weighing procedure, the patron had only an approximate
estimate of the final stock of rubber.
River Traders and Forest Petty-Traders
Traders who carried their stocks on boats and canoes and sold merchandise
at the tapper's door for payment in rubber, the regat6es ("river traders" or "river peddlers"), are as old as the trade-post itself. These river peddlers represented, as it
were, an 'informal market' in the extraction zones, side by side to the monopolistic
owner's equipment) and the hiring of the chainsaw owner's labour. I never witnessed
13. The following is based on trade-post bills and stock inventories, 1982.
125
124
trade-post system. The regatoes were autonomous traders, without any connection
either to the (nominal) owner of the rubber estate or the patron-trader; they
possessed only small boats of one to five tollS, and purchased goods from urban
merchants on relatively short terms (three months) and high interest (30%), in
quantities up to the equivalent of five tons of rubber. These traders sold rubber to/
the same processing plants in Cruzeiro do SuI that bought rubber from the patrons.
While the trade posts faced the problems of storing bulky goods for the
totality of houses and for year-long periods, and transporting them throughout the
forest, restricted by the rain cycles and having to project the annual output and
payment capabilities of their clients, river traders operated on a small scale and with
a restrict temporal standard. The river traders' goods were light in weight, "nonbasic" in character, and turned over quickly. On the voyage upriver, goods were left
directly at the tappers' homes, and on the return trip, the itinerant trader collected
payment in rubber. A river trader would wait patiently in the vicinity or even at the
tapper's river outlet until he received payment. The river traders boat also was his
home, with a small cover and a fireplace in which to prepare food.
Though each journey yielded only a small amount of rubber, the river traders
could return often with with fresh merchandise, turning over their petty trade capital
several times during the year. While affording the tappers the opportunity to buy
goods and sell rubber in direct payments, they cut into the patrons' monopoly.
Though they did not supply households with the "estiva" portion of the monopoly,
they substituted the trade post by furnishing some non-durables (except food) to
houses whose supply-line had been severed because of low productivity or relatively
high debts (which means that supply did not correspond to production during the
powdered milk for children, medicine for the sick, alcoholic beverages or "luxury"
items for adults, such as candy, shoes or perfume.
As a canoe approaches on the bend of the river silently (to save fuel but also
because the river is shallow, making it necessary for two men to move the boat with
poles), the
coloca~ao
under a plastic cover, the boat discloses wooden crates with a great many small
articles. A yOl:ng tapper buys shoes, another purchases a can of powdered milk, and
an old man acquires some sweets. These purchases will be paid with a fewkgs each.
In 1983, after siX months under a new and powerful patron, there were in the RestauraC$ao six tons of rubber (about five percent of its total estimated produdion)
committed to eight river traders who operated in the hinterland, including an
itinerant photographer and a dentist. This was the rubber still outstanding, that is
which had not already been transferred directly to the river traders, who were
waiting to receive it. This also was the debt bought by the patron, under the
condition that the traders agree to leave the area, as mentioned in a previous
chapter). Taking into consideration only these cases, which were recorded in the
trade post, an average river trader had a credit of 774 kilograms of rubber to receive
from 24 different clients (on the average, 10 from Riozinho and 14 from the other
sub-estates of RestauraC$ao); the average debt per client stood at 40 kgs. of rubber
(Table 5.10). ,The real volume negotiated by river traders over the whole year musthave been much larger, and the traders did not disappear after the agreement with
the patron. 14 It was not uncommon to find as much as thirty percent of the total
house budget committed to the river traders and local peddlers. The rubber tapper
Ferreira had the largest debt on the Riozinho Estate, in a case of high volume
course of the year). If a river trader was in reach, a tapper who was not being
supplied by the trade post could easily produce Within a couple of days ten sixteen to
twenty kilos of rubber (worthy approximately one U$ dollar a kilo) to pay for
14. In the following year, the patron took the issue to court.
126
purchases from the flver traders: sixty per cent of Ferreira's 490-kg. rubber
production went to the itinerant traders (Table 5.11; Table 6.3, Account 11).
The informal forest market included another category. The "marreteiros" are
small traders with fixed residence in the estate hinterland ("forest traders") (Table
to haggle, and mayor may not be related to the term "regatao". "Marreteiro" comes
from the verb "marretar", taken unambiguously throughout the Amazon only to_
tappers' in a wider sense (family members, employees or clients work trails rented
designate the act of buying and selling for profit, using bargaining or any other
by the head). As an example of the continuum going from tappers to petty traders,
Ferreira's neighbour, also his stepson, whose household also produced considerable
itinerant "marreteiros".
merchandise to other houses for profit. In another house on the same settlement
main trade post, or a "minor patron" as I have called them sometimes, occurred
(coloca~ao),
from time to time - this constituted one of the ways in which the trade post absorbed
working as a tapper but also acting as a marreteiro (he was married to a regatao's
petty forest commerce into its own structure, transforming the "marreteiro" into a
and former minor patron's daughter), selling the goods he brought in his luggage. At
trade post debtor while at the same time holding him responsible for the rubber
one point, he enjoyed credits of 106.75 kilograms of rubber, spread among seven
clients, including his own stepfather Ferreira. Thus, in this settlement of three
houses, one belonged to an indebted house head, Ferreira (who owed heavily to
peddlers and the trade post alike), while the other two were headed by stepsons, one
Restauracsao Estate held on for ten years under three different patrons, and in times
houses as a whole included agricultural activity, tapping and hunting just as the
managers". This continuity, rooted in their participation in the forest economy and
others. These marreteiros sold at higher prices, which was inevitable since they had
in local social networks, confirm that the "marreteiros" were not simply an appendix
to make purchases from river traders, from the trade post or more rarely directly
to the estates' monopoly structure. Instead, they should be seen as part of the forest
from urban merchants. The patron tolerated more widely the marreteiros than the
peasant economy, of which tapping, cultivation, hunt and trade were part-activities.
river traders, because unlike the itinerant merchants, the rubber they received as
payment was stored along with the other rubber accumulated in the house, and sent
along to trade post at the end of the year.
129
128
In fact, in 1987, the year when the trade post was closed up after a rebellion
fifty to sixty kilogr.ams apiece; the latter, in the past, suspect small bundles, the_
had expelled the Camelis' collection team in 1986, there was an immediate
proliferation of both river traders and "marreteiros" - four of these within Riozinho
ten to twelve kilograms, were suspiciously similar to "principios". Since patrons were
alone, all of them also heads of forest houses. It would be difficult to estimate the
concerned first with preserving the monopoly, they were suspicious of the plank
volume of rubber circulating within this horizontal forest trade, precisely because
system. Thus, the patrons' role in the productive sphere consisted, in this case, of
of the patrons. When the patrons introduced little merchandise, the number of
The data presented in this chapter show that the trade post that I studied in
itinerant traders increased to meet tappers' demands, as did the number of tapper-
the 1980s did not control the labour process and only partially realised its function
i'marreteiros" who ' acquired goods from the river traders to resell to .their -
neighbours.
Conclusion
A consequence of the above argument is that trade-post patrons were not
concerned with the efficiency of the productive process itself. On the contrary, as the
following example illustrates, the "shopkeeper" character of the patron-trader
entered in direct conflict with productive efficiency. During the early 1980s, one
family of tappers on the Riozinho Estate began to construct their own rubber
presses to replace the smoking house. Between 1983 and 1987, this technique spread
to most other houses in the district. The method proved labour-saving and in
principle allowed for daily productivity to double, without additional capital costs.
Meanwhile, the patron of Restaura~ao sought to discourage the use of this method
introduced independently by the rubber tappers, asserting that "he did not know
how to work with rubber planks." The main reason for the patron's opposition to this
technical innovation lay in the fact that the final product were light rubber planks,
easier to smuggle and with less water content (the tappers claimed that tare should
be smaller or not deducted at all). Planks diluted the distinction between "rubber
used to settle accounts" and "rubber used to pay river traders", the first as bundles of
16. Cf. Pacheco de Oliveira 1971; Santos 1974:19; Weinstein 1984.
131
Chapter 6
DEBTS, RUBBER AND MERCHANDISE
Introduction
The debt-peonage system was deceptively transparent, and the casually
authoritative way that it was and is still today referred to by outsiders, whether
'mere' travelers or sociologically astute historians and anthropologists, helped
mystify even more the network of histories, moral obligations, and coercions
ensuring that, just as indebtedness ensured peonage, so peonage ensured
advances of credit (Taussig 1987:63).
The case of the Tejo and Riozinho estates suggests that patron's power did not
follow a syllogistic scheme going from debt to enslavement to control over labour.1
Both rent payments and rubber collected as debts involved conflict. Taussig observes
that debt-peonage may be narrated as a story "whose plot moved, in Gramscian terms,
from dominio to egemonia, from the opening blast of brute force (the correria or slave
hunt) to the succeeding phase of debt-peonage and the subculture of mutually respected
obligations it assumed" (1987:64). This suggestion of a combination of physical and
economic coercion with selective and differentiated incentive contracts will be be taken
up in this chapter as an alternative to the theory of debts as a facade for disguised
forced-labour camps.
The "subculture of mutually-respected obligations" (in fact not so much
respected as having a claim to being respected) is summarised by Taussig in the
postulate that "just as indebtedness ensured peonage, so peonage ensured advances of
credit". Weinstein has suggested that debt relations - beyond obviating the use of
1. For such syllogistic reasoning, see Furtado 1962; Cunha 1967:24 ff; PauIa
1979; Teixeira 1980:262, 222, 245; Silva 1982: 18; Pinto 1984; Bakx 1987:77-8;
CoelTio 1982:78; Coates 1987:94; Bunker 1985:67; Martinello 1988:45; and Geffray
1991. Teixeira and Bakxare anthropologists; only Teixeira provides actual data on
debts and work. But his own data show that tappers were indebted only "on the
average" .
money and allowing for the imposition of "forced commerce", an expression I borrow
from Bhaduri (1983) - constituted a demand of the tappers themselves, who expected
patron-traders to provide long-term advances in kind (Weinstein 1983:23-25). It is not
difficult to illustrate this with examples. A Jurmi riverbank cultivator who neither paid
rent nor was a trade post client complained about an itinerant trader who wished to
settle his credit balance in cash, since this sort of "account liquidation" would represent
the rupture of a tacit contract implying the continued supply of goods, even when he
did not dispose of a positive balance. In this case, the preservation of debts served to
guarantee the continuity of mutual obligations, and in particular, to establish a
continued flow of commodity advances. In this respect, the aviamento supply relation is
not simply the result of coercion or of a paternalistic/clientelistic culture. The
aviamento supply relation represents a long-term exchange agreement which is
pervasive in the rubber, fishing and agricultural economies throughout the Amazon,
of violence in maintaining debt relations and, at the same time on the role of ,an
Rubber tappers began their purchases between March and April, at the
ideology based on moral kinship ties (Geffray 1991) in reinforcing these same
relations? This may be due to the inability to distinguish between the aviamento relation
Restaura~ao
.f rom the mouth of the Tejo (the major trade post) in June, September and
We shall argue that, even though the rubber tappers may consider the advance
November. A house-head living on the Riozinho Estate had the option of going to
credit system (and the corresponding debts) as legitimate, this does not mean they
accept as equa,lly legitimate either the monopoly clauses demanded by the patrons
an additional distance of three to six hours on foot. Once at the trade post, after-
Of
Of
some informal conversation and a rest, he would buy goods on credit, without
having to deposit any money or rubber at this point, and usually not taking into
account the price of the products acquired. The articles purchased were recorded on
a "purchase receipt" or notas de compra. In principle, a copy of this "receipt" would
,
be given to the house head, the only person in his house allowed to buy goods at the trade post. "Purchase receipts" were then transcribed in a notebook for each house
head. In January, at the end of the tapping season when rubber was turned over to
the trade post, these accounts were tabulated in the form of balances. Though
double-accounting was used, accounts were complicated for several reasons,
including the existence of multiple categories of debts and credits, the presence of
two accountable currencies (money and rubber), the variation of prices over the
course of the year and the levying of interest and "commission" charges. Only rarely
did a Riozinho house head know how to read, though most were quite capable of
performing mathematical operations in their heads (Table 6.1).
The debit column of accounts included the closing debt of the previous year
(in kilograms of nibber, converted into Cruzeiros at the current price quotation); the cash value of the dated purchase receipts (including a "commission" charge if
rubber was to be paid at, the final price in the season); and debts for cash advances
(with 60% interest). Debts that one tapper owed to another also were accounted for
here (in this case, they were tabulated also as the other's credit). The sum of the
debt column or current gross debt was expressed in money. '
The credit column included the "liquidated" rubber turned over to the trade
post and converted into money (at distinct prices corresponding to the periods of
purchases, if the merchandise was debited without "commission" or interest; or at
the final price if goods purchased were charged interest), plus credits in Cruzeiros
for transactions between rubber tappers and for products or services sold to the
patron besides rubber. These itens made up the gross credits.
The final money balance was calculated in the equation of gross credits
minus gross debts. If this balance was negative (a net debt), it was converted into
rubber at current prices and forwarded to the following year's account (if rent had
not yet been paid, it would be added on the next year's account). If the balance was
positive (net credit), it would remain as a credit in money on the next year's account
- though, at the rubber tapper's criterion, it might be converted either into goods or
cash for""rubber surplus", in this order of freqt;ency. A further type of transaction
~as sometimes ~egistered separate from the accounting procedure described above:
the encomendas (advance orders), which involved particular goods, usually of high
value, ordered specifically for an individual rubber tapper, in exchange for a
previously-negotiated amount of rubber (a shot-gun was atypical case). These
encomenda transactions were liquidated with payments in rubber and were not
treated as the usual accounting debt, even 'though there was a period of months
between the order, delivery and final payment of the goods in rubber. They were
seen as "cash" and Were not affected by uncertainty on the terms of exchange; they
did not affect the previous balance, just as was the case when the trade post turned
over cash directly in exchange for rubber "in cash", i.e. surplus rubber delivered on
the spot or not, in a previously agreed amount. ,
Transactions between rubber tappers havin~ both accounts at a single
patron-trader could be cleared through the trade-post accounting, generally in the
case of somewhat elevated sums, such as those involved in the sale of garden plots
or livestock. They were recorded informally on the spot, without requiring the
presence of a trade post representative, on "receipts" (simple pieces of paper written
by a literate person) passed from one tapper to another, who authorised the trade
post accountant to transfer the amount from the buyer's account to the seller's. For
instance, when in 1983 a house head at the Jarana settlement decided to move to
another settlement, he sold his manioc gardens, coffee and coconut palms to
136
Manipulation of Accounts
The stages leading up to the final balance allowed for an ample margin of
account manipulation by the tra de pos t ,
. h'
throughout the year _ the many purchase receipts and the weIg mg
frequently included mistakes, often outrageous ones, such as debiting the same _
not distinguish the sources of the rubber delivered by a house head, the debit due to
purchase to two different accounts from the same coloca~ao (settlement) - and often
purchases was subdivided into direct purchases by the house head and by
.
. h 't I b rinth of different prices,
occurring in the accounting process Itself, Wit 1 say
.
I
h' h in most cases rendered the final
interest rates, currencies and payment pans, w lC
.
either. Ideally, a trade post should possess its own money supply, but money was
somewhat like certain kinds of merchandise, such as medicine, which had only
occasional use and in emergency situations (cf. HUMPHREY 1985:52). Cash
advances received different treatment than the advance supply of goods, because
they drew interest (called juro) and because cash advances only could be justified to
bber
ru
obtained through petty forest trade, the house-head's credit could also included the
result practically impossible to estimate beforehand with accuracy, and only in rough
terms within a broad margin of error.
However manipulated, the double accounting system recognised in principle
'vely commercial relation with the trade post - a
that the tappers had an exclusl
a monopoly, could not be
relation governed by prices, which although fixed by
.
ted On the other hand, the context for this
altered once a transactIOn was consumma .
.
d
brutal difference in the degree of information and
accountmg proce ure was a
139
138
both commerce and prices were mere theatrics, masking the expropriation of rubber
by the trade post in arbitrarily-assessed quantities, in exchange for goods whose
quantities also were assessed arbitrarily by the same trade post?
In considering trade post commercial strategies, we shall see that this clearly
two or three times during the year, which apparently made many rubber tappers
was not the case. Though manipulated, the gross debt was in fact tied up to the
think that this system worked in their favour. This hypothesis may serve to explain
why, in 1982 and 1983, most of the rubber tappers on Riozinho Estate refused to
accept that their rubber be settled at "several prices" at the end of the year - which
commodity prices .spread throughout the rubber estate, quoted in rubber. Rubber
in effect meant that the monetary value of their purchases were expressed in rubber
tappers can easily .assess the inflation ' along the last years quoting the successive
at different rates (of rubber) throughout the year. The tappers wished to derive
rubber-values of powder milk or of cooking oil; thus, "in that year (or at the time of
benefits from the "higher price" of rubber at the end of the year when their yearly
patron X, a can of powder milk cost one kg rubber; now it takes four kgs rubber to
buy the same" is a typical and accurate comment about the fall in real value of
over their goods purchases, whose prices they insisted maintaining constant during
rubber from 1985 to 1992. Hence, the tappers' insistence in maintaining steady
the year, for the right of having the "final price of rubber" paid for their product.
prices for goods all year long makes sense, since under the assumption of constant
With the escalation of Brazilian inflation in the 1980s, these rules became
Patrons could have argued that prices kept constant over the year had to be
much higher than prices on the urban market, since they both anticipated future
"surplus rubber" or borracha de saldo could be converted into cash: this could be
inflation and paid profits. Tappers reasoned that, given constant prices all season
used not only to buy goods "out", but also at prices much lower than the standard
long, they stood to gain with the rise in rubber prices if payments for rubber were set
trade-post price. In order to prevent tappers from achieving positive balances, some
at the final price quote, in a system where official price hikes were first announced -
patrons (especially those possessing bank loans) pushed greater amounts of goods
informally and later by radio. In the 1970s, the Brazilian government established
upon their clients. 4 Heavy debts or high level of purchases came thus to be
several radio stations in remote border regions as part of a national security policy.
associated with workers who were both hard workers and "courageous", or
The Cruzeiro do SuI radio station transmitted personal messages from town to
spendthrift tappers - "thriftiness" was not a general rule (cf. Gudeman 1990: 162ft).
tappers in the forest, as well as messages from patrons. But until 1982 messages with
We must thus bea~ in mind that the notion of "heavy debt" was not absolute but
4. We assume here a patron who had access to generous bank loans. But
remember the alternative strategy of seeking to thwart the exodus of "surplus
rubber" by summoning the police.
rather relative to the productive capacity of a household, since the advance (and
corresponding debt) of a 'valuable good", say a gasoline engine, worth 400 kgs. of
rubber would be an exceedingly high risk for a trade-post if it benefited a house
whose annual output was 400 kgs. (it would simply leave the debt outstanding at the
~nd of the yea~), while it might represent an investment in the case of a house
capable of producing 1,200 kgs. (who would convert it into an increase in the rubber
productive tappers on Riozinho who, during the 1980s, in making their annual
purchases with balances left over from the previous years and during a period of
rising rubber prices, were able to remain "in the black" and to accumulate wealth in
the form of "valuables" and cattle.7 For these tappers, the next step was the patrons'
attempt to recruit them as SUb-patrons, converting them from creditors into debtors
of capital advanced in the form of merchandise.8
product turned over to the trade-post).5 Thus, "heavy debt" was not a synonymous
for unpayable debts. In considering this accounting strategy in detail, one finds that
the accumulation of high gross debts over the year were stimulated, but they could
represent at the end of the year a low debt in net
ensuring a cash balance ("free of debt") at the end of the year, provided that the
!liles were observed. In this case, the patron would have to adopt another strategy to
keep the tapper's rubber within the trade post. One such strategy was that of
reducing his own profit margin per unit good (although not total profit), by selling
goods at "city prices" in exchange for the "surplus rubber". Thus the trade-post could
use both a monopoly price and a "free-market" price.
direct
from the .forest
with their own work"); (5) an inescapable monopoly (clients_
.
.
:
could not leave the game because flight would be impossible). But it should be
cle~r
both from the historical record and from the ethnographic description that these
.
? A house. of means might also use livestock as collateral for loans,
Increasmg the secunty of patrons' investments.
6. Here, for simplicity, we identify highly-productive house heads as "hard
~o~k~rs". Nonet~eless,. a la.rge rubber output might be produced with little
IndlVIduallabo~r mp?t In a bIg house; and a low output of rubber might be related
to hard work. m .a~ncu1ture an~ .hunting. Thus, "hard working houses" is not an
assessment of mdiVIdual productiVIty.
142
conditions were not necessarily valid, and thus there
IS
no reason to expect
.Juvenal from the trade post, and luveila1's purchases were made through Chagas'
account.
If they did in fact occur, one might expect certain consequences, to wit: (6)
In February 1983, Chagas obtained a copy of his final balance from the
tappers would be homogeneous (differences would be erased by the trade post,
Restaura~ao
trade post. Chagas had a ton of rubber (1,061 kgs.) to his credit, which
which would appropriate productive margins for itself); (7) there would be no
was weighed at three different occasions, "corresponding to his purchases and his _
accumulation on the part of rubber tappers; (8) bargaining between tappers and
patrons would have predetermined results and would not allow for tapper gains. In
post. 9 Once rent was deducted (90 kgs. for three rubber trails), 971 kilograms of
the remainder of this chapter, it is suggested that (6), (7) and (8) do not occur, based
"tared" rubber remained. By his own accounts, Chagas should have had a debt-free
on real examples and on statistical data on debts. In other words, social
balance. My own estimates concluded that around 611 kilograms would suffice to
differentiation among tappers, local accumulation and even the exploitation of one
pay his gross debt; the 360 ks. of "surplus rubber" already had been exchanged for
tapper by another all can exist precisely because there was no absolute monopoly (a
cash in January, when Chagas used it to purchase articles at the Vila Thaumaturgo
combination of perfect, total and inescapable monopoly as in 3, 4 and 5) on the
novena. However, contrary to this expectation, Chagas final negative balance rose to
rubber estates. -Thus, a conflict in the sphere of distribution exists between the nearly 100 kgs., which could norhe explained even through an "error" (although the
accumulation of mercantile monopoly profits by patrons on the one hand, and the
accumulation of productivity gains by rubber tappers, on the other.
trade post accounting showed only 581 kgs. of rubber on Chagas' credit, an "error" of
30 kgs. according to our own calculations). The debit apparently had been
calculated according to the accounting rules described above. The 100 kgs. of net
households come from the same coloca~ao or forest settlement where I lived and,
Chagas first thought that the patron had charged interest on the money paid
consequently, where I was able to observe in detail the movement of debts between
out in January in exchange for his "surplus rubber", which would have been contrary
September 1982 and November 1983. The "Tacaratu" settlement was occupied by
to the accepted rules. He arrived at this conclusion when he noticed that part of his
four houses. Three were located in the principal clearing and had independent
account included a a 60% "interest" charge. According to the current rules, 60%
accounts with the trade post: one house where old Ferreira lived with his wife
interest was levied on cash forwarded - but with the exception of payments for
Nazare; Nazare's married son Chagas' house; and the residence of Assis, another
married son of Nazare. Juvenal, Ferreira's brother, lived in a separate clearing, in
the same settlement. Juvenal was Chagas' "client". Chagas rented the trail tapped by
9 The cases under study exemplify how my own efforts to understand the
trade p~st accounting based on both current receipts and final account statements,
assisting the tappers of many other houses. During 1982-83 I never. had access to the
trade post general balance sheets. I obtained a copy of these only III 1987, thanks to
an ex-employee.
145
144
"surplus rubber". Chagas' reaction was filled with indignation: "I'm going to tell him
batteries --, along With small expenses with food. Chagas thus self-financed the
consumption of-durable and non-durable goods for the following year. He was able-
to estimate his surplus because he had opted to pay the so-called "commission on
precisely the amount of "surplus rubber" available for cashing in and the trade post
goods" (a mark-up of 30% on merchandise prices) in order to turn over his rubber at
had
this
the year's final price quotation. Adding up his purchases at the trade post (including
"surplus rubber money" had to be available before January 20, first day of the Yila
the 30% "commission''' a term which distinguished this charge from "interest"), he
Thaumaturgo novena, when minor and major patrons, river traders and local
now arrived at the total of nearly 408 kgs, which together with the rent and to his
peddlers, urban merchants and neighbours get together for a week at a riverside
surplus rubber added up to 858 kgs or 80% of his delivered rubber (Table 6.3). The
bazaar. By that date, he had not transported all his rubber to the trade post yet, nor
272 kgs of his client's purchases at itinerant traders (plus 30 kgs not accounted for)
was his final balance prepared, but the trade-post acc~pted to grant him money for
added up to 29% of the delivered rubber, leaving an excess of 9% over the delivered .
his "surplus rubber" which he threathened to take over to Yila Thaumaturgo anyway.
rubber (Table 6.3). Now the interests charged by the trade-post on his client's
from the trade post, for the 360 kgs. of rubber handled as "surplus rubber". First,
excess of his delivered rubber. The problem with luvenal's account was that most of-
Chagas bought basic provisions ("estiva" goods) for 1983 at the novena, at market
his purchases had been made from river traders, and were increased by 60%
prices, from his own patron (there competing with the other merchants), spending
"interest" charged by the patron, who had by the end of 1982 bought up the river
only thirty percent of his cash (50,000 Cruzeiros): 36 litres of kerosene, 60 kgs. of
traders' debts in exchange for their promise of leaving the area (see Chapter 4). This
salt, two sacks of sugar (60kgs), a box of soap bars, eight kgs. of coffee, three kgs. of
interest added up to 102 kgs. of rubber, about the final (and unexpected) debt that
shot for hunting and a kilo of gunpowder. It also should be noted that these goods
Chagas found on his account Apparently,Chagas was aware of the volume of his
were not readily available at the fair, since the patron displayed stocks composed
client's debts, but he did not suspect that the patron would levy "money interest" on
~ainly of "luxury" ~oods; Chagas was to receive the basic goods he purchased at the
these debts. The patron sought to justify his actions by claiming that in order for the
fair directly from the Restaura~ao trade post, once he had returned to the estate.
river traders to leave the area, they had to be paid "in cash" (when a 60% interest
The only difference lay in the fact that the transaction, settled outside the estate and
was charged). Mter going over the details of his account, Chagas redirected his
independent of the debt mechanism, carried much lower prices. Chagas spent the
indignation towards his own client, accusing him of falling deep in debt to the river
classification outlined in the previous chapter), bought from different river traders __
the hope of landing a wife. As a matter of fact, this was not true, since Juvenal's
a record player and two records, clothing, scents, a complete set of new pots and
major purchase was a new shotgun, really an investment and not a "luxury", but one
pans for the kitchen, a metal cabinet in which to store flour, two metal tubs,
which he probably would not be allowed to obtain from the patron, since he was a
accepted
this
calculation
in
January.
For
Chagas,
146
tapper of low productivity (also, Chagas charged 25% higher than the trade post
At the end of the year, Ferreira owed 274 kgs. of rubber, while he had
prices). Chagas then sought out the patron Correa and complained that he had not
produced 499 kgs.Th.iswasa- heavy debt -: that is, 55% over his total output.
authorised luvenal to make such purchases from the river traders. The patron
Taking out 33 kgs. forre~t (though Ferreira tapped two trails depicted in Map 5, he
replied that, in any case, since Chagas was luvenal's patron, he was responsible for
refused to pay rent on the one he personally had opened), 466 kgs. remained. But
Ferreira had paid 278 kgs. to river traders and local peddlers, leaving only 188 kgs.
Though neither Chagas nor his wife Maria could read, there can be no doubt
of "account rubber" (the trade post recorded only 177). The rubber turned over to
as to the fact that Chagas had calculated his account so as to . remain with a zero
the trade post, at 548 Cruzeiros each kilo, would yield 97,000 Cruzeiros in money
balance at the trade post, and to maintain a stock of goods - while the patron had
terms, but his account owed 247,542 Cruzeiros. His debt corresponded to 274 kgs. of
insisted that he purchase much more, trying to seduce him with a diesel engine and
rubber, almost exactly the amount he had spent in purchases with the river traders
could resell at a profit to other rubber tappers. lO When Chagas refused to fall into
Unlike Juvenal, however, Ferreira did not pay. high interests, since he usually
what he considered a "trap", the patron called him a coward: "By the Virgin Mary:
paid for his purchases directly in rubber, and generally to socially close river traders:
are you afraid of bills?" Chagas played safe, but the patron was clever, as Chagas
admitted: "Nobody fools him. He never was fooled and never was marretado (Le.
Unlike Chagas, and like many tappers, old Ferreira (Chagas' neighbour and
Tim6teo, a river trader whose father was patron of the minor trade post of Riozinbo
stepfather) easily was satisfied with the vague notices of his account offered by the
(Table 5.9A). Had Ferreira not bought anything from the river traders, he would
trade post manager, such as: "Your bills with the river traders have already been
have to direct all his production to square away his account at the trade post. Buying
paid off ..." Or, to cut back on trade post supply advances, "Ferreira, your account is
getting a little heavy ..." When Ferreira asked to see his account, the book-keeper's
Taking a quick glance at the third house of the settlement, headed by Assis,
wife sought to appease him, saying: "Padrinho, you're not in too deep with the
one also finds aheavy debt as a proportion of the tot,al product. Assis produced 700
kgs. of rubber during the year, about half of what a young and strong tapper like
10. Though illiterate, Cha~as could propose and solve problems such as the
following. What is better: pay a bIll of 5,000 Cruzeiros to a river trader in rubber, at
the current rate of 800 Cruzeiros, or transfer it as a "trade post order", waiting for
rubber prices to reach 1,000 at the trade post, and then paymg 60% interest on the
account?
11. Note that a tapper is the godfather (padrinho) to the trade-post's clerk's
wife. The clerk Gedeao first entered Restaura~ao in 1982 as a rubber tapper,
himself was capable of produ~ing, though no one could accuse Assis of not being a
hard worker. Alont!, ,vithout grown sons or hired hands, Assis tended one of the
largest manioc plots in the region and was proud of it. Of 700 kilograms of rubber
marrying a local woin~. He l~ter was r.ecruit~d by the patron as an accountant,
.
since he proved capable III readmg and ,anthmetIc.
produced, Assis bought goods worth 300 kgs. (a propOrtion similar to Ferreira) from
dozen bead of cattle ' (descended from a cow acquired by his first wife), thirty head
river traders, and after paying rent, the remaining amount of rubber proved insufficient
of swine, chickens and ,ducks. Chagas did not need to buy flour or meat (which he
to settle his account. Like Ferreira, his debt to the trade post added up to about the
would not find for sale anyway), and he had slaughtered an ox each year during the
. same amount he had spent on purchases from river traders, also with direct payments
last two years, whose meat he sold to tapper neighbours in exchange for rubber.
Note however that besides having sons able to work (although only part-
time), Chagas's rubber included that delivered to him by a client who worked a third
range of rubber productivity (500 to 1,000 kgs), the patron's account manipulation, and
~bber trail, th~s i~creasing his total volume of rubber production; and two also
the diversity of strategic responses on the part of the house heads. They also show how
"debts" only have meanings as part of a different productive capacities and different
benefited from the settlement's working team including his stepfather Ferreira, his
purchasing strategies. Ferreira' s debt was high in relation to his productive capacity
brother Assis, and his client Juvenal, who lived in the forest a short distance away.
(55% over his current output), while Chagas' debt was low (9% over his current
By contrast, Ferreira, Chagas' stepfather, godfather and neighbour, was too old to
output), and thus Ferreira owed half of the following year's expected production, while
work efficiently as a rubber tapper and, following a stroke, had lost much of his
Chagas owed less than one tenth. At the same time, Chagas had by then purchased his
former hunting ability; his stepsons all were heads of their own houses and did not
basic needs for the next year (and had in fact accumulated over the year), while
have time to contribute much labour to their stepfather's unit; his wife also was old,
h~vier
12. On the Born Jardim settlement, the record Nascimento's house had a net
debt a.t the end of 1983. B~t on that.year, Nascimento (1,800 kgs) bought a house in
Cru~lfO do SuI, along WIth a refngerator, a fan, a Jug and some chairs and in
preVIOUS y~s he had purchased a ~asoline engine, two Singer sewing machin~s, a gas
stove, flashlights, shotguns, a radlO and two Japanese watches. He Along with the
cattle he kept near the house, these were bonuses paid for by the labour of four adult
sons, as well as by Nascimento's own solid management, as he was one of the first to
buy his basic needs at urban prices wit.1 his "surplus rubber" (borracha de saldo). In
of another, thus avoiding dealing directly with the trade post. This was the case of
Ivan, a tapper who loved the forest and its game, and who was less interested in
larger agricultural or rubber production. He had planned to payoff his bills and
leave the estate by 1983, but he stayed on another year because he did not have the
"positive balance" he had expected.
150
In the foregoing analysis, we have assumed that the basic strategy adopted by
the trade post sought to maximise the total volume of rubber. In order to do this, the
trade post sought to maximise gross debts, subject to the productive capacity of each
house. This was achieved by increasing the volume of commodity advances (at a
given price that already included profits and costs), and then imposing an additional
mark-up over the capacity of the household to pay. Had this strategy been executed '
perfectly, two consequences would have resulted. First, the gross credit of the tapper
(result of rubber turned over and other transactions) would be canceled by the gross
debt (merchandise advanced), and thus all the tapper's rubber would be committed
to the trade post. This effect would increase the proportion of the total production
bought up by the 'trade post at monopoly prices (as "debt rubber"). This was a
monopoly effect. Second, rubber tappers would be encouraged to produce more
rubber as they received more merchandise. This second effect influenced the total
volume of production: it was an incentive effect. 13 Such a basic strategy, however,
could only work "on the average".
151
Restaura~ao
trade
post. 14
It should be noted beforehand that the "rubber harvest" refers to the period
covering roughly April to January of the following year (hence, for example, the
"1983-84 harvest"). A "harvest", or the period of rubber extraction and production,
varies from house to house. Another useful preliminary observation is that final
balances are established beginriingin"February with fixed dates, which also vary
depending on each house's 'harvest end. Between the end of a harvest (that is, when
the rubber is "liquidated" and the accounts tallied) and the beginning of the next
(when, for accounting purposes, the net difference left over from the past year is
recorded as part of the "debt" or "credit" for the year to come), there is a lag of a few
months during which a house accumulates a growing debt, which may spread from
January to May. In other words, at the beginning of the harvest, a tapper's account
would include not only the net difference left over from the previous year, but also a
series of expenses from the inter-harvest period.
In the 1983-84 harvest on the Riozinho Estate data from 31 houses give a
volume of 150 kgs as the average net debt (Table 6.4). In order to estimate the total
Accounts included annual flows (gross debts and credits that grew at
was
different paces throughout the year), while the final balance was composed of a
cases (Table 6.4), which suggests an yearly product of around 18,338 kgs for 28
defined net quantity determined only at the end of the rubber harvest.
houses or 655 kgs/house/year. We thus estimate the average net debt as 23% over
Even the trade post faced the unpredictable nature of this process. Hence,
the yearly output. When we project these figures for the whole Riozinbo (assuming
the trade post adopted a global policy dealing with averages and totals, around
which houses were expected to oscillate between greater or lesser margins. By
quantifying the variables mentioned, these hypotheses can be checked with data
13. It should be noted that these effects of differential supplying would be
superfluous in systems with direct repression of the labour process or with an
abundant labour supply.
14. The data cover the 1982-83 harvest (balances of. Ja!1uary and February
1983, obtained at tappers' houses); the first half of 1983 (welghmgs of August 1983,
covering half of all houses); the 1983-84 harvest (balance~ from trade post books,
covering 70% of all houses). In these different sources, an~ In the same sources~ pa!t
of the data originally appears in money va.lues (~ruzerros, converted by me In
Dollars and Pounds Sterling) and part m kind (kilo~ra!llS .of rubber). There are
inconsistencies, ' but the data should be taken as mdlcatlOns. on the order ofma~nitude of debts (both global and average), as well as on the dIfferences between
indIvidual producers.
I,
152
66 productive houses), we get a total output of 43,230 kgs (43 tons roughly), and a
total debt of 9,900 kgs (roughly 10 tons). These figures are within the order of
magnitude which would be expected on the basis of interviews.
Clearly, the distribution
sample had net debts, and the average among those with net debts was 273 kgs
(42% over the average output of 655 kgs/year), while the average among those 35%
houses which manage to strike net credits was only 73 kgs rubber (11% on the
average output). The data from this harvest show, however, a very high variation between individuals houses. Individual balances for 1982/83 ranged from net credits
worth 200 kgs rubber to net debts worth 898 kgs of rubber.
According to trade post balance sheets (obtained only in 1987) for 48 houses,
the average gross debt stood at the end of the 1982/83 harvest at 675,000 Cruzeiros
(US$675 or 452 at going exchange rates), while the average gross credit was
584,000 Cruzeiros (US$584 or 391); the average balance was thus a debt of 91,000
cruzeiros. At the rate of 500 cruzeiros to a kg rubber, this figure corresponds to 182
kgs rubber (compare with the figure of 150 kgs obtained on the basis of individual
accounts and interviews). Based on the trade post data (Table 6.5), we see also that
the gross credit (average over 48 houses) included 525,000 Cruzeiros in rubber
turned over (483 kgs. of rubber, without including rent) and 59,000 for credits of.
and the month of July 1983, resulting in the transfer of a debit of an average Of 399kgs. of rubber in July into the account. 15
In the trade-post books, the gross debt registered in July 1983 (incorporating
the net debt of the 1982-83 season and the inter-harvest debts) appeared as a single
item in the final gross debt of the 1983-84 harvest, which averaged 675 kgs. of
rubber. In other words, debts left over from 1982-83, along with additional
purchases made
befo~e
Once the tappers had turned over the rubber from the 1982-83 harvest, the average
final net debt remained at about 68 kgs.
1984) observations begin by explaining how debts were transferred from one year to
the next as a consequence of the incongruity between periods of production and
periods of advance supply:
The 1982-83 negative balances were increased by the purchases made between the
liquidation of the previous harvest (sometime between January and March of 1983)
15. Note that for July 1983, .I converted .debts into money expre~sed in
quantities of rubber, using current pnces. In practice, t~e trade post used dIfferent
prices depending on individual arrangeII?-ents made With tappers. Therefore, the
figures in rubber must be seen as mere estimates.
154
Almost all the clients owe at the end of the tapping season and that debt
remains to be paid with the production from the following season, and then
comes another debt and so on. Like, for example, these debts left over from
83 will be paid off only in 85 with the production from 84. That's because one
year's accounts only are settled the other year in April or May.
In 1983, explaining the role of permanent indebtedness in the accounting
aggregate gross debt in money terms) as well as the total volume of rubber turned
over by the tappers of within each category (gross credit in money terms). We find
the trade post acting upon these macroscopic variables through the fixing of prices
for rubber and for merchandise (within the different classes of goods), increasing or
reducing the total volume of merchandise available at the post according to each
kind of good. In acting upon these variables, the trade post affected the total
maintain profits and the average debt; it did not require a mechanical control of
invididual cases.
Excessively high profit taking (for example, much higher prices for "luxuries")could lead to a reduction in the amount of rubber turned over for these goods,
hence reducing global output; excessive profits in the "basic goods" sector could lead
to an increase in net debts without a corresponding increase in production - up to
the point where other control variables would step into action, such as the use of
physical violence to stifle river traders' activities or the seizure of tappers'
possessions.
In other words, the same macroscopic state (for example, forty tons of rubber
distributed in 480 kgs. per account for advance purchases and 60 kgs. for rent) could
result from a great number of different microscopic conditions. On a hypothetical
rubber estate with two households, a macroscopic state characterised by an average
production of 600 kgs. of rubber could occur when house A produces 400 kgs. while-
merchandise.
the output of house B is 800; or when house A produces 800 kgs. and house B 400;
Average accounts included thus costs, profits at different points of the
mercantile hyerarchy, individual bargains and some tapper's refusal to settle their
entire bill, as well as an average debt.
On the Riozinho Estate, the average debt of about sixty-six houses
represented a macroscopic variable, as did the total volume of goods advanced (the
or when each house produces 600 kgs. and so on. Thus, the persistence of debts on
the average is consistent with the oscillation of individual houses between debts and
credits. Some macroscopic states allow for a great many micro-states to exist
(metaphorically, these states have high entropy), while other states do not permit
micro-state variations (low entropy). Hence, considering the hypothetical example
Just as the political relations between patrons and rubber tappers have been
presented as a combination between violence and non-consensus legitimation in the
Chapter 5, in this chapter we have argued that the advance commodity trade is an
arena of dispute. However"it may appear as though I agree with a hypothesis
concerning the . "morality of patron-client relations", because I have argued that.
tappers accepted the principle of legitimacy with respect to the advance supply and
debt system. The term "debt fetishism" may be taken as a label for the idea that the
commodities in circulation become identified with the persons who receive them as
advances and who by extension would become the property of creditors (Taussig
1987:70). However, we have reiterated that "debt fetishism" is denied by the rubber
tappers who see themselves as autonomous agents. Therefore, we need to look more
on the notions of legitimacy and morality of exchange.
The terms of trade could be considered unfair by the tappers, even though
the existence of advances was viewed as legitimate. In conversations held at a
I
distance from the trade post, it was common for tappers to entertain the notion that
the patrons cheated them - irrefutable evidence of this comes from the fact that.
many patrons turned rich simply buying and selling rubber, the only source of wealth
that could be converted into the merchandise that came from the distant world
"below", or downriver (cf. Allegretti 1979). But this should not affect the legitimacy
with regard to debts. A debt constituted the counterpart of an advance or credit.
Unavailable to tappers through public banks, credit was monopolised by the
available, just as the lease-patrons afforded the only access possible to forest
resources. Rubber tappers were
concentrated both forest trails and credit in the hands of a few. But they were also
which 'assumed ",the' formof bi-Iateral obligations: the patron was to continue
supplying tappers (eveQ in the in the case of illness or other disability) and to extend
debt payments over several years; the tapper remained committed to "paying off' his
debt sometime i.n the future.
demonstrate that patrons, river traders and local peddlers (actors in the forest trade
scene) may be distinguished by the nature of their stocks as well as by the different
terms of trade each adopted.
If indeed the content and terms of trade were different between these three
types, whether or not a moral difference existed remains to be seen. We argue first
that all three categories share a common core of ma~retagem, the act derived from
In the global forest economy, patrons, river traders and local peddlers were
complementary figures, occupying different commercial niches. Resident peddlers
("marreteiros") offered goods in the neighbourhood to tappers whose credit at the
trade post had been restricted, or when there were no river traders nearby, serving
the verb marretar which means to purchase with the intention of reselling for profit,
or "to cheat".16 This term applied especially to the small transactions of local
peddlers, involving bargaining and immediate profit turnover, but it also was used
by tappers when criticising patrons, who were supposed to be more than a simple
somewhat as forest "drugstores" accessible at any moment provided one had "cash
marreteiro - insofar, that is, as he fulfilled his obligation of extending long-term
rubber" to pay with; river traders sold low-weight "luxury" and "vice" goods in retail
and on short credit terms, lasting weeks or at the most months; patrons operated
with basic ("estiva") goods, such as salt, soap, fuel and ammunition, on a large scale
and with long credit terms of a year, but which could be transferred to subsequent
credit.
As we mentioned above; the marreteiro lived in the neighbourhood. Thus, he
years. In social terms, patrons, river traders and local peddlers could be found on a
but, as neighbours and relatives of their clients, they failed to offer services normally
scale ranging from distant to close: strangers living in town, acquaintances, or
neighbours and kin. On a temporal scale, they ranged from long-term to short-term
credit: annual debts that could be extended to several years; short-month terms, or
immediate payment. In material terms, they cover a scale ranging from large stocks
of basic articles to small stocks of superfluous items: from the "estiva" of work
implements and current consumer goods, to luxury articles, or liquor and odds-and-
16. The dictionary meaning is "to cheat". ~ccording ~o Miyazaki and On?, the
"term marreteiro is used to designate one who mterferes m ~n already-establ~shed
transaction (advances-debts), selling goods cheaply or bUyIng. th~m(M~xpen~vel~
breaking the existing harmony. It possesses a derogatory meanmg.
lyaza anOno: 366 and Santos 1989:53).
as marreteiro he would sell remedies for profit, he now did not seem ashamed of
asking for medicine in his role as "relative" (and thus would have to settle the
amount acquired in kind, following the tappers' ethic). Such ambiguity would not
arise had it involved a patron, precisely because he was not a relative and,
furthermore, would be expected to advance medicine as his part of the quasicontractual relations he had with the tappers (who would pay usurious prices).
For many rubber tappers, cash loans constituted the only situation in which
interest was acceptable. When one tapper demanded that another pay "interest"over the time his rubber was held over at the second tapper's house, union delegate
Chico Ginu determined, in a public meeting, that "rubber doesn't pay interest, only
money does." Once fixed in rubber values, debts extended from year to year could
not be altered. The confiscation of goods to payoff debts was illegitimate _ though
payment for them could be extended, tappers assumed that payment should not be
compounded by cumulative interest nor should .
cacha~a
brandy
on our way home through the forest after making a purchase at the Riozinbo trade
post, of the amount of labour that would be necessary to square away that particular
bill at the prices charged, arguing that he at any rate would not pay the total amount
charged. In many
co~versations,
becoming simple marreteiros: that is, they only undertook marretagem, or the
middleman activities of buying to sell, and selling to buy. One tendency that
developed beginning in 1982, when the new patron implanted a system using three
different rubber prices, "the three price system", confirmed this feeling. This _
.
tendency was to intensify after 1985, when the next patron began to reduce the time
between purchases and payments and to collect debts on short notice, calling on the
police to seize tappers' possessions in cases of debtors who fell behind on their
payments - which provided the setting for police violence on the Riozinbo and
Restaura~ao estates and for the "strike" led by Chico Ginu, both in 1986. 19 The
same tappers who accepted, within certain limitations, the tacit contract involving
reciprocal obligations with patrons, also were capable of open rebellion when these
obligations clearly were violated. Precisely because there was a consensus around
the fact that patrons also grew rich from the labour of the rubber tappers, patrons
tapper of high ' productivity, or refused to supply goods over a long period, or
most revered "saint" of the Upper Jurua, developed a reputation for refusing to
touch money under any circumstances.
19. See Chapter 4. The "Alagoas Estate Rebellion", another document~d
example of rubber tapper revolt against the trade post, also ~as set off by c~anges III
rubber and commodIty price policies, which also accomparued a change III patron
(Allegretti 1979).
162
There then would be no reason for rubber tappers to pay for their own long-term
words, goods and services 'were exchanged using labour-time as a measure - child-
commitments.
days were converted into adult-days, rubber-days converted into flour-days, and so
The
tappers~
on. Finally, in the commercial sphere, the agents essentially were unequal parts
associated with the concept of profit) naturally belongs to the age-old tradition
(patrons, river traders and the "social outsiders" who were the marreteiros), and the
according to which the accumulation of wealth was considered an objective that was
goods were merchandise introduced by social outsiders and produced by the tappers
incompatible with the continued reproduction of the good, a tradition which goes
(rubber); the underlyiI?-g ideology was that of marretagem - the notion that someone
back at least to Aristotle (Aristotle, Politics, 1,8 and Nicomachean Ethics, V,3; Marx
n.d. [1987]:152; Parry & Bloch 1989:4; Gudeman & Rivera 1990). One explanation
for the form this view assumed among rubber tappers is that different spheres of
cooperative action and consumption, which were the extended houses or macro-
circulation coexisted on the rubber estate, which may be grouped as trade (trade
houses of the hinterland settlements. The labour-value code was employed to gauge
post, river traders, local peddlers), barter (between houses independent from one
another) and the exchange of gifts (between houses within a single interconnected
settlement). These different spheres of circulation were not coterminous, and had
and patrons were not expressed in either of these codes. How could one establish
different uses.
Within the sphere of gift exchanges, the agents were co-habit ants of a
watch from the free-trade zone of Manaus, against rubber? How can one speak of
hinterland settlement who also maintained a game meat relationship, in which fixed
portions of any slain game animal (though sometimes including fish or other articles
where it was inconceivable that a rubber tapper be received and given food, as was
collected) were transferred to neighbours: a side, a rear quarter, a front quarter, etc.
the custom of the forest?20 In these cases, the appropriate code was that of_
The reciprocal notion of bestowing neighbours with game meat was associated with
the mystical danger of the panema, or bad luck that might strike a hunter as a
house: in sum, this regime was marked by the dangers of excessive proximity
Exactly how great is the difference between "another object" M' and the
(Galvao 1951 and 1976; see also Chapter 7, 8). In the barter sphere, agents were
previous one M? The marretagem ideal does not determine how much. The trader's
house heads from neighbouring settlements, and the items exchanged were either
own cleverness and ability establish this difference within bounds given by the
the products of labour (flour, rubber, baskets, manioc) or labour itself (days offered
20. Even. tappers. ~ho. we~e blood relatives of the patron did not have any
intimacy or "SOCial kinshIp With hIm.
by children or adults to execute specific tasks); the rule of exchange was based on
164
market (prices heard on the radio) or barter practices. There may be a minimum .
[1966]:471). It is obvious, for example, that a 1or d wh0 t akes the wives of 'his
and maximum profit for these traders, but the exact point depends on the agents'
subjects, who steals their food and who does not preserve peace is an exploiter. In
personal qualities and is not determined a priori. Distribution in this competitive
attempting to improve upon Moore Jr.'s formulation, James Scott fleshes out the
arena is determined post facto by the relative force of the agents involved, by the
idea that there is a "subsistence ethic", which is obvious to peasants: "a right to
capacity to bargain and by resistance. Therefore, in essence the marretagem trade
subsistence ... forms the standard against which claims to the surplus by landlords.
theory differs from a deterministic theory of exploitation - according to which the
surplus value rate is fixed by a minimum of subsistence consumption. 21
and the state are evaluated" (Scott 1976:7). There is much to say in favour of these
ideas in the case of peasants, who have weathered famines and natural catastrophes,
and who may be treated as "averse to risk", but this focus distorts the situation of
rubber tappers that I have known. Tappers did not seem to adopt an "image of
limited good" when wealth derived from their own labour - which is consistent with
their folk "labour-value theory" in rubber and manioc flour regimes. Though tappers
made pacts with the "rubber tree mother" (and not with the devil) to increase
productivity, and could use analogous pacts with "mothers of game", thus increasing
hunting "happiness" or luck, these pacts were not used to criticise hard-working
t~,ppers
perceive "thriftiness" (cf. Gudeman & Rivera 1990): on the contrary, a category of.
tappers seemed to have a Dionysian urge to acquire goods and novelties, not unlike
that described among certain indigenous groups (Hugh-Jones 1989), which urge was
compatible with the notion that incurring on high debts was a proof of "courage" in a
high-productivity tapper.
To be deprived of powdered milk (imported from Holland), of a shirt
(introduced from Sao Paulo), of a Seiko watch (from Japan) or of a gasoline engine
(imported from the V.S.), and to be deprived of health services ,and education
(attributed to patrons' greed and not to natural factors) are more significant as
21. That this theory comes from Marx is debatable. Marx said once that "The
matter resolves itself into a question of the respective powers of the combatants"
(1950(1898):401-402). The pomt is taken e.g. by William Baumol (1979:ix) but see
'
Morishima & Catephores on the same (1978:106).
While in this economy the relative abundance of manioc depends on the personal effort
exp1oitation hypothesis: the rules for leaving the game must be specified. If, for
of the producer and the availability of game meat depends in part on mystic factors, in
example, the rules det~rmine that the tappers "take the forest with them" in leaving_
the rubber tappers opinion shortages of imported goods (whether consumer goods or
I
the rent-patrons' game (in the case of agrarian reform), and all the tappers
articles used in production) clearly were linked to the traders' greed. Given this
obviously become better off (because they cease paying rent) and the only ones
perspective, an assessment of changes in the level of 'exploitation carries much more
worse off are the patrons, then this suggests that the prior situation was one of
weight than a fixed idea of !'subsistence". While not involving a drop below the
exploitation (of a feudal sort), which only was sustained by the monopoly control
"subsistence level", since only part of the tappers effectively could acquire "luxuries"
over the forest by the patrons, as a consequence of political dominance. Hence, this
that might be substituted by cheap local products, the loss of access to canned butter for
exploitation was based on the monopolistic appropriation of the forest. However, if
women following childbirth, or to Dutch powdered milk for small children, or to
the rent patrons had provided services that the tappers could not execute without
ground coffee from Sao Paulo for workers represented real deprivation. These changes
them (for example, the conservation of trails), then the tappers situation after the
reduced the economy's entropy, that is, they limited the degree of dispersion in the
individual houses' patterns.
expulsion presumably would have worsened, and the prior situation could not be
considered one of exploitation in Roemer's terms.
r~lationship of dominance
exploitation of a coalition of rubber tappers on an estate is the possibility that they may
abandon the game (say, by expelling the patron), improving their own lot (and not
making it worse for any of the coalition's components), while the only ones worse off
would be those outside the coalition (patrons and managers, in this case). There is a
If the tappers' coalition "takes the trade post" when withdrawing from the
game, which means access to financial credits, and all the tappers become better off
,
(because they no longer have to pay abusive interest and monopoly prices) and the
only ones worse off are the patron-traders, this means the prior situation was one of
exploitation (moved by usury capital). This exploitation was based on the
monopolistic appropriation of access to credit, guaranteed by the patron-traders'
political dominance. It should be noted that the existence or absence of exploitation
depends on the capacity of exploited groups to organise themselves: in the rubber
estate example, even eliminating monopoly barriers, the rubber tappers still must
overcome problems of communication, organisation and leadership, which
accompany the difficulties involved in distributing goods throughout the forest
(transport, storage and distribution). Analogous to the hypothetical situation
described in the previous note, if patron-traders had offered services of
communication and articulation that could not be substituted, the tappers would be
168
169
terms. It is worth mentioning here that the cooperative constitutes a solution for
Chapter 7
taking over the economies of scale formerly controlled by the trade post.
Introduction
As employed here, the concept of house refers to a unit formed by a group of
to maintain a stable commercial monopoly over the rubber tappers, based on the '
maintenance of average debts that also were stable over time. At the same time, we
have endeavoured to show that the existence of average debts and high profit
margins was compatible with the existence of individual variations between rubber
inform people's actions over nature with the intention of reproducing a good life.
tappers, emphasising the economic mobility of some tappers and the relative
Thus defined, the house is a discrete unit of technical action that comprises a group
of individual producers (or houses) is consistent with a model of several houses with
It is also a unit of social action, which establishes relations with the external
Chayanov's). On the other hand, we have suggested that the possibility of variation
world at given points in time .and is reproduced over time. As a unit of economic
houses show that the profit margins of the trade post were not the maximum
consumption, each house is represented by its head, who also exercises the function
possible. There was a conflict between the patrons' accumulation, on the one hand , .
of deciding over the use of material objects, of labour, and of natural resources, as
and the freedom of houses to establish their own economic strategies, on the other.
well as over which technical procedures are to be adopted. The head also answers
for the house hi social interaction, commerce, alliances and conflicts. It is he who'
keeps an account at the trade post and who pays rent for the use of rubber trails; in
1. Here, the "house" (casa) corresponds to the Greek oikos and the "house
head" to an oeconomicus, or "estate manager" (Xenophon 1990: 271; Aristotle 1977,
1977a; Descola 1987; Almeida 1988; Gudeman and Rivera 1990). It should not be
confused with the sljghtly more specific uses whereby a "house"or "maison" refers to
a kinship group defined by "claims on a particular estate" and perpetuated through
cognates or adopted heirs (Uvi-Strauss 1979; Goody 1990:469), or with the notion
of a household or a domestic group (Netting, Wilk and Arnould 1984). The use of
"house" as a translation for casa in the sense defined above should also preserve the
association of cas a in current Portuguese with a physical building (extended here to
a territory and its resources), a household ("family") and a home (Portuguese lar, a
domestic fire or, more generally, a set of technical activities involving a
household)(See Pina CabraI1991:128ff).
.
170
principle, he also decides when to torch forest clearings for cultivation, how many
, 'the individual-fantily'." Furthe'rmore, at the local level, the basic ecological character
trails are to be tapped and when to sell an ox. The house group is composed of
,
. 'III th e fac t th at "e ach environment affords a
of the extractive economy
resIdes
relatives, children, godchildren and domestic hired help. These people have their
resource for trade purposes which could best be exploited by individual families
own interests, which may come into conflict with the house head's orientation; for
controlling these products within delimited territories." The "gradual shift from a
example, individuals may have personal rights over part of the house's property
(such as head of cattle or of other livestock, personal objects, etc.); and there are
culmination point may be said to have been reached when the amount of activity
cases when the house head is a woman or even another relative who exercises a
considerable degree of power over house decisions. The incorporation of members
devoted to pruduction, for trade grows to such an extent that it interferes with the
and makes their
aboriginal subsistence cycle and associated social organization
into the house group (household) occurs mainly through marriage, adoption, fictive
Steward and Murphy's hypothesis, which was formulated around the case of
the Mundurucu Indians who had become rubber collectors, might explain why
even when spatially discontinuous and accommodating persons who do not eat
rubber tappers did not form villages, rather becoming spatially distributed in
isolated family units through6utthe forest. This would have to do with the
under the same roof. On the other hand, houses have an ethical orientation, in that
maximisation of extraction returns; Such dispersion means more than simply low
the house is oriented towards its members' welfare, and decisions should ideally be
demographic den; ity, which can be compatible with the existence of an elaborate
weighed with the criteria of justice, equality and sensibility, to which in principle
social organization '(say, ten houses grouped together in a central clearing, exploring-
subordinated. 2
to the level of the "family", expressed physically in the isolation of individual units
units involving family relations, and consumption units with a common budget -
Julian Steward and Robert Murphy have written that extractive economies,
(let us say, ten isolated houses exploring three square kilometres each). It would be
exemplified by fur collectors and rubber tappers, are characterized by "an almost
complete economic dependence upon trade goods which were exchanged for certain
would lead to an increase in entropy with regard to spatial organization and social
life. Families would represent the highest form of social integration at the local
whieh "led to reduCtion of the local level of integration from the band or village to-
level. Trade would integrate these sociologically minimal units at a higher level,
through visits punctuated in time and involving clients distant in space. The world
market therefore would provide the only real social articulation of these houses
was gIven, In
the sense that we described a vertical, fractal-like commercial hierarchy whereby
forest houses are linked to national urban markets through the trade posts. But
domestic group or household which is part of the definition might not compose a
.
brothers, fnends,
a man an d h"IS h"ne d help) " Instead , our definition focuses on the
nature between patron-traders and houses, partially legitimized by rules which were
fact form a group of relatives, and these relations provide the principal language
men and women who are of working age, of growing boys and girls who work part-_
houses are not however isolated atoms. They are found in groups using the
,
time
time' and of children and aged
persons of both sexes. 0 v e
r , therefore , a house
coloca~6es,
compositio~
grows and mo d"f"
lies 1"tS
, insofar as new children ,are born, old peop e
h
expire, while other--.. members come and go. In sort,
whl
1 e h 0 uses may be described
macro-house networks, which are related through common social activities and
share a moral identity, constituting thereby systems which are not isomorphous
,.
., .
as stocks of people at a point of time, they also may be COnsld ere d as flows .4
Focusing on the population of Riozinho Estate, part of Restaura~ao Estate,
within the larger and smaller rubber estates. Rather, they constitute horizontal,
" .'
this chapter describes both the internal orgarnzatIOn
0f
flexible social structures, which overlap across the estates' boundaries, and although-
articulation between them. At the end of 1982, a popUlation survey revealed a total
subject to fragmentation and change, are real enough (Cf. Tanner 1979; Morris
of 370 persons distributed among 68 houses on Riozinho (Table 7.1). This total
included 109 men and 95 women (this classification covered residents aged ten and
up), along with 90 boys and 88 girls. Eight men appeared as resident hirelings. One
174
The number of "men" defined as above corresponds approximately to the use
of the term "faca" (knife), used to denote the real or potential number of tappers in
each house. Thus, when the Riozinho patron composed an informal list of "facas", he
counted 100 "facas" for 63 houses, or an average of 1.6 cutters per house, which is
the ratio found ~n t~e demographic survey data (Table 7.1, first column 1).
Though in the under-ten category the number of boys and girls was roughly
equal, shortly later the number of females would fall off in relation to males (7.1).
The greater proportion of men among the adult population may be explained by the
fact that Riozinho attracted male rubber tappers. S In effect, besides the eight house
hirelings in the popUlation, there were houses composed only of men. For example,
one house was headed by a man who lived near the Tejo River and rented a series
and with his surplus production moved to Cruzeiro do SuI. There he kept a manioc
plot outside of town, sold bananas to the urban population, and sent his young son
to work seasonally as a tapper at Riozinho. Therefore, in such cases the ostensibly
specialized male labour in effect contributed to houses located outside the estate. In
1982 and 1983, the patron's recruitment strategy was responsible for the attraction
of male work groups, that is three groups without relatives on the estate, since they
had permanent
reside~ces
employed by the trade post. By 1984, part of these groups had constituted their own
houses within the estate, developing domestic economies. 6 After all, the status of
speCialized worker -was not at all a stable one. All things considered, then, we may _
assert that the popUlation of Riozinho belonged either to local houses (structured
In other three cases, groups of brothers (or of a father and sons) worked on
the rubber estate while having another residence in town. In another case, a house
head residing at the mouth of the Riozinho kept an unmarried son working at a
married son's house: here, one brother was a house head himself, while the other
worked to help support their father's house. Chagas Farias, left the estate in 1987
5. Anot.her re?son is the fact that women marry at younger ages than men:
the age of 12.ls ~onslder~d acc~p~able (abducti<:n of 13 to 14 year old women was
popular on RlOzlnhO), whIle 16 IS Ideal for marnage. In contrast men married after
reaching 18. Thus, new houses were formed with the early exodus of women from
their parents' houses, often marrying men from outside the estate. Therefore on the
'
average, families with children over age 10 had less females than males.
est~te,
after
concludi~g that the ~ost of food (which they did not pro~uce) absorbed theIr surplus
rubber production. A third, with two sons, developed mto two other local houses
after the two were married.
-
176
parent's house before marriage. These "employees" all were 17 or 18 years old in
Riozinho. 7
177
groups include individuals who are related as parents and children, brothers and
harmony always was preserved. Intra-domestic conflicts were especially common in'
houses recomposed through second marriages. One reason is that children might
new home could be located in the husband's father's neighbourhood, or near the
dispute the property accumulated by the first house - as in the case of a dispute over
the offspring of a cow that had been raised by a mother who died.
house of his wife's father, or near the homes of his or her brothers or sisters. A
recently-formed house might fill up rapidly with the head's adult unmarried
women whose husbands were disabled .. Three female-headed houses were recorded
?rothers, who on that occasion leave their father's house. Later, these brothers will
in the 1983 population survey (the 1982 census provided the base for numbers cited
form their own houses, while the first house itself expands with the arrival of sons
, above, though detailed information was available for only 60 of a total 68 houses).
and daughters. When these children arrive at working age, the house reaches its
Two houses were run by the widows Isaura and her daughter Nazare, living on the
maximum potential for producing rubber, foodstuffs and game. Older houses lose
same
their adult sons and in many cases lose one of the head couple as well. These houses
two houses, Isaura's sons and married daughters also lived in the neighbourhood.
frequently are recomposed through the incorporation of new couples and relatives
They did not pay rent (since they considered charging rent to a woman abusive) but-
they had their own manioc .crops, and Nazare had a rubber trail, which was tapped
Only in rare cases, old women, incapable of working and without a husband, become
by her sons. Soledade, a single woman with several male children, headed the third
dep~ndents at their relative's house (though I did not see any cases of older men
house, where the boys all produced rubber in small amounts (in her smoke-house,
living with a son or son-in-law). Apparently, older men often can find new wives,
and these houses, in turn, become technically viable when they manage to recruit
Soledade had several quantities of rubber in sizes corresponding to the ages of her
~oys and girls as f~ster children. In sum, a house may experience a cycle beginning
coloca~ao,
sons).
The importance of having a house full of different people resides in the fact
that the house's productive structure requires not only adult male rubber tappers,
but also cultivators (adults and children of both sexes who clear gardens), hunters
(adult males and boys), crop-tenders and housekeepers (women and girls). In terms
of the people who make up its composition, a house is a technical unit whose
specialist in raising ducks - one of many features of the domestic economy, with its
own peculiar technical problems (ducks spend most of the time in streams, facing
technical cooperation, as well as figuring as a "moral" unit under the leadership of-
the danger of flash floods; they lay eggs at the edge of the forest, etc.). One idea
an older house. Macro-houses are social groups of a size appropriate for the
behind the house is that a group composed of men, women, boys and girls must
remain economically feasible over a period when people are born, die, and change
status, as boys become men and girls turn into women. This, incidentally, provides
macro-houses form technical units for hunting, for making flour from manioc, for
an explanation for the high proportion of idle trails. Since each house is a process
clearing forest areas in preparing agricultural plots, and for torching these same
areas, and they are physically neighbour houses. The individual house heads of a
macro-house (for example, a father and his two sons, each with their independent
houses) may act as a team, each commanding the production of his own unit. A
coloca~ao
territory is marked by
a well defined number of rubber trails radiating from a central clearing. There are
no physical boundaries separating colocas;6es, but the areas crossed by rubber trails
macro-house does not need to have a head, though it might have leadership.
Observe that the set of houses that uses a settlement does not coincide necessarily
with what I call a macro-house. Houses claim rights over parts of a settlement (on the basis of renting trails in that settlement) and manage them, but houses could
either enter a settlement being ' recruited by the trade-post, or through direct
starting from a single clearing are considered part of a single settlement. In 1982,
there were 26 of these settlements in Riozinho, within an area of approximately
23,000 hectares (Table 7.2, 7.2a). Therefore, each settlement held an average of
macro-house, of one macro-house and one or two additional houses, and of
around 850 hectares of area (8.85 sq km), and an average of 2.79 houses, so each
settlements with one or two isolated houses (Table 7.5). Note that the average
house accounted for an average of 338 hectares (3.38 sq km). The general
number of houses in those settlements with a single macro-house was 3.13, rising to
gopulation density stood therefore at 1.7 persons per square kilometre in the
3.71 in those settlements with one macro-house plus isolated houses, and to 4 in the
Riozinbo (Table 7a).
In the Riozinho, leaving aside one empty settlement, the number of houses in
settlements ranged from one to five (Table 7.3). These cases usually led to a
situation of potential conflict. ' The potential conflict involved in the use of a
180
case of two macro-houses. The settlements with one or two "isolated" houses (with
forest peasant's house or macro-house in the upper Jurua. Forest houses and macro- _
an 1.3 houses per settlement) might have been abandoned recently, altough not
houses were not a vehicle for land property.9 This is one of the reasons for the
emphasis given here to their role as units in the labour processes, while they
Note that the case of two separate macro-houses sharing a single settlement
house (macro-house) resides thus in the set of persons which as a whole make up a
viable labour unit, .endowed with the necessary knowledge, sharing a movable
seen later, this corporateness includes also a mystical component more apparent in
was rare (
coloca~ao,
gaining
that of defining leadership. This argument may draw support in the following_
settlement were about to establish an alliance through the marriage of one family's
between wives. Other reasons for moving could be the temporary exhaustion of local
son to the daughter of another. The couple was expected to remain on the
resources, was when a group of houses decided to look for a better environment
coloca~ao, but if this had been carried through, it would be the only case of a new
more rich in game, rubber patches or agricultural areas, often within a short
house pledging loyalty to two older houses at a single settlement: that of the groom's
distance of the first one -- a decision which again might be influenced by the existing
parents and that of his bride's. In the upshot, the marriage did not take place:
instead, each of the engaged persons married someone from outside the settlement
From year to year the distribution of houses was reshuffled (Table 7.9), and
(the young woman wedded virilocally on another coloca~ao, while the man married
coloca~ao
houses became distant from one another in spatial terms, the settlement split in two.
could be vacant, while others could crowded with as many as five houses
as an old macro-house. But these were transitory situations, since settlements went
Obviously, the problem with the proposed marriage lay in that the new house would _
through phases of abandon, reoccupation, growth, and abandon once more. This
was possible both because of a low overall density, and because under the patronproperty regime no permanent titles tied up a settlement territory to a given house
or macro-house. There was no land patrimony permanently associated with the
be subordinated to two separate authorities (of the father and the father-rn-law)
marriage. Under this last strategy, neo-Iocal marriages carry with them the original
within a single work space. This would be inconsistent with the political aspect of a
fraternal cohorts of the .couple (or at least part of them): therefore, these marriages
allow new houses to emerge free of paternal authority (or that of fathers-in-law),
while at the same time preserve technical and emotional cooperation sets of the
same age cohort. In this type of allied macro-house, ideological emphasis is placed
frequent and also were ''big'', followed by macro-houses of heads related as brothers
become subject to the authority and frequent violence of husbands when they marry
neo-Iocally, but carry the protection of their consanguineous relatives when their
differences, set the basic pattern of stable authority for the technically efficient
own brothers marry their husbands' sisters and become their neighbours. Having a.
by some brothers and fathers of a young married woman, often a marriage is broken
On the other hand, macro-houses with a head and two unrelated sons-in-law were
unstable. lO This suggests that one of the foundations of macro-house cohesion and
stability lay in a sort of principle of solidarity among fraternal sets. While the
permanence after
~arriage
(and corresponded to some of the most productive houses of the estate), some
married sons demonstrated a clear aversion to remaining under their fathers'
authority. The alternative was the neo-Iocal marriage pattern of the fraternal group.
Settlements with this sort of structure frequently were composed by a pair of houses,
where two married brothers or sisters live together in the same settlement. In some
cases, these fraternal macro-houses were associated with the exchange of brides,
which means that the settlement includes two fraternal groups allied through
10. In one case, the situation degenerated with a conflict between the
mother-in-law and one of the sons-in-law. In another case, two unrelated brothersin-law moved into the same coloca~ao, and an open conflict broke out between
them. Both conflicts had to do with the sharing of game meat, in one case associated
with witchcraft accusations.
185
184
production, peaks occurred twice a year, when trails were prepared for the latex
niece, godson or goddaughter of the members of the next house. When they belong
"harvest"; in agriculture, peaks occurred with the knocking down of trees, the
burning of these areas and the coivara, or final clearing of plots {or cultivation; in
agricultural tasks of different houses, except for the fact that the final product
hunting, the capture of deer could benefit greatly from collective action. Groups
belongs exclusively to one of the units. A smoke-house for treating rubber also may
be used by more than one house. Though each house has its own hunting and fishing
caring for children and domestic animals in common areas, and in raising a new
home. In general, one house hosted the labour of others (except on hunting
Once the rubber or flour is produced, and game hunted, the consumption
expeditions): in this sense, the labour process belonged to a particular house, which
process becomes the private matter of each house. Rubber belongs strictly to
at any given moment could benefit from the cooperative labour of the oth~rs. The
individual houses, though within each unit there may exist different rights over the
tacit understanding was that either the labour was reciprocal (to be returned at a
value of the product - including those of-the hired help, the young manor woman
later date) or was donated by a young house head who assisted his elderly father's
who helped out, or the young rubber tappers, who expect some remuneration for
house. The cooperation within a macro-house was thus quite different from that
their part in contributing to the domestic economy. Just as rubber is the property of
between unconnected houses. In the former case, labour assistance was not .
individual houses (even when stored in common areas or left in the smoke-house),
measured in order to be paid off later in precise quantities which distinguished half-
flour is stored separately in each house (in the kitchen or bedroom). Each house's
flour stock, which is refreshed every couple of weeks, is the very symbol of its
In addition to forming extended work crews for peak moments of the labour
process, macro-houses also shared work equipment in two important cases: the
do not contribute to a single unit that was to own the product: on the contrary, the
which included expensive equipment such as a small gasoline engine (3.5 HP) and a
divides the product according among the individual houses according to precise
There is never more than one flour-house in a settlement, and though the
rules. A fixed portion of the slain animal is sent (not shared on a same kitchen-
charge. The main manioc plots surround the flour-house, in a mosaic of areas
transitive verb meaning to share game meat with someone, so that "A neighbours
area, covering the same daily routine. The wife or child of one house can bring food
with B" with a portion specified as a quarter or a side plus half the spine. It should
or water to a work crew from another. This comes naturally, considering that the son
be stressed that such "neighbouring" relationship does not apply to domestic animals
belonging to the
in~Hvidual
hunting sphere, wbich ' is ' one of the most essential components of the rubber
tappers' happiness. Just as trust between neighbouring houses must be complete, in
a certain sense the -coloca~ao's core, in the form of a macro-house, is a moral unit -insofar as each house is responsible for the integrity of hunters from other houses
who have provided them with meat from the forest. And more: the relations
between male hunters critically depend on the behaviour of women in the whole
macro-house, the only ones to handle and prepare game meat from the forest, once
it is introduced into her kitchen. This point helps clear up the distinction between
simply occupying the same settlement and composing a macro-house. In settlements
with two macro-houses tension and even open conflict are common, and one
particularly clear expression of conflict is precisely the absence of game meat
neighbouring. 11
rubber tapper may increase through pacts with the "mae-de-seringueira", or rubbertree mother, but it is again essentially due to the house's own efforts. In each of
.0:"
these cases, success is a matter restricted to the house's work crew. But success in
hunting depends not only on personal ability and on one's own magic. In this case,
might call its members a macro-domestic group). "Family" (faITIl1ia) was the term
used to designate the kindred known by a given person -- that is, the people within
behaviour of the other houses. A hunter may be ruined if he or his dog becomes
the sphere of memory who were identified by expressions involving filial descent,
panema (a Tupi term for one who is unfortunate in hunting), a condition inflicted by
marriage (in certain contexts) and adoption. The micro (macro) domestic group is
others who consume the meat of forest animals (Galvao 1951,1976). What is crucial
about neighbourhood relations is that a house receiving game meat from another
easily might empanemar (make panema) the hunter who killed the animal. A hunter
becomes panema when a pregnant woman, or a woman during her menstrual
period, comes in contact with, or pass over, the meat or bones of the game. Thus
special care must be taken in disposing of bones, lest a woman in these conditions or
even bitches in heat contaminate them inadvertently. Thus, neighbours who share
game meat must be trustworthy since they represent a potential threat to the
188
composed of the part of the family that forms the house (macro-house), as well as
and other cooperative relations and who formalize this situation when house heads
Beyond this language of real kinship (based on filial, marriage and adoptive
Compadres, ritual co-parents, were chosen, and thus "fictive kinship" could also be
frequently became superimposed over real kinship. Thus, they would be somewhat
redundant if the only point in it were to contract a "kin" relationship. Indeed, the
infant baptism. For example, couple A and B have a child, C. They invite D to be -
practice of fictive kinship between dose relatives, such as fathers and sons or
brothers and cousins, suggests that this institution was more solid, in a sense, than
established, this ritually contracted kinship relation lasts forever, just like real
kinship. In addressing, it takes precedence over real kinship: a child does not call his
kinship, while ideally uniting people, at the same time creates new, contracted -
godfather by his name, nor by any real kinship term (for example, uncle,
formal links and barriers between houses. As emphasized above, the members
grandfather), but addresses (and refers to) him as padrinho and always asks for his
blessing when first meeting him in a day, a treatment which persists in adulthood.
members who had belonged to a single house before setting out to create their own
Co-parents now begin to address each other as compadre or comadre, leaving aside
proper names (even in the case of brother and sister). Fictive kinship ties may be set
godfather of his own grandson, father ana son can overcome the conditions of
on different occasions and rituals. Beside the main one at baptism (godfather of a
domestic authority and assymetry that had marked their relationship as part of a
~hild),
single house. The newborn child marks a phase in the progressive growth of the new
Sao Joao (Saint John) feasts (godfather of the bonfire), and at birth (godmother of
house, and now the symmetrical compadre relationship replaces in a sense the
One acquires relatives by birth, but one chooses compadres. Social groups
here designated as macro-houses may be linked by contractual kinship relations
unconnected houses may be forged by compadre relationships, at the same time the
institution separates houses that in a certain sense are too close.
among their members, and usually do. We might be inclined to conclude that the
institution of contractual kinship thus allows for macro-houses to develop a formal
constitution, even when its members are not related in the language of "rear'
kinship. This in fact does occur, in the case of houses in a neighbouring relationship
12. On Riozinho Estate (1982-3), the patron did not serve as godfather to any
child (and hence was not the compadre of any rubber tapper). In ~o cases, tappe~s
were genealogical second-degree cousins of the patron Correa, though thIS
or
genealogical tie was ignored conspicuously. According one of t.he tappers'Jei
man addressing a rich man as "cousin" (pri~o) ~ght appear as If he was as ng or
a favour on the presumption of a close relatIOnshIp.
Pt
191
In another instance, man and his sister-in-law (brother's wife) who become
compadre and comadre will have a an often jocose relationship, which carries
however greater overtones of sexual taboo than any other relatlonsh
h
lp, except t at
between mother/son and brother/sister. The role of contractual kin h t
s Ip 0 ensure ..
that a formal distance is kept between "too close" indd I .
.
IVI ua s IS partIcularly effective
m cases involving opposing sexes as brother-in-law and sisters-in-law who are
neighbours. While there is no rigid taboo Proscribing extra-marital relations
between a man and his sister-in-law th t b
b
, e a 00 etween a compadre and his
comadre is extremely serious. 13
participation in parties and bigger work crews). In a certain respect, they also
comprise units of collective and political action, coming to the fore particularly in
situations of conflict and vengeance against similar groups, while also in establishing
common alliances and strategies, for example in the development of a cooperative.
Often, these groups assume the appearance of a kin group drawing their roots from
close ancestors, such as a mother or a grandparent, whose name they adopt. 14 Also
common are marriage patterns between cousins, which repeated in subsequent
generations, tend to keep such kin groups relatively isolated (Table 7.7, 7.7a).
In order to assess the extension of these relations between macro-houses, we
adopt a narrow definition for the "macro-house network" as a set within which it is
of interaction within which people found marriage partners, held parties, and joined
together to face conflicts. If, on a map, we were to connect all the houses that had
ones (wife's brother, wife's father), always based on the house heads in tWo macro-
houses of sons and sons-hI-law: the Farias (linking three settlements, and whose
nucleus was the settlement occupied by Nazare Farias), the Castelo group (five
settl~ments (although often contiguous), do not have any technical function, that is
settlements, built around old woman Castelo's settlement), the Santana group (five
to say, do not act together at the labour processes. One might say that they are the
settlements, built around old woman Santana's settlement, and five more around
.
largest groups whose social activities are somewhat self-contaIn d ( .. t
e VISI s, courtmg,
. . 13. "A pior cama do inferno e a c
d
d
worst bed in the hell is the bed of a comp:~: Jfh~f!1pa re dCo1,;l) a comadre" ('the_
me.
IS coma re ,as some put It to
14. Hence there are the "Is auras" (first name of the grandmother), the
"Ginus" (the father's and grandfather's nickname), and similar technonyms in other
cases.
her sister's), the Isaura group (five settlements, old woman Isaura). Such blocks, made up by close settlements, involved at most three generations of house heads.
river, all of the houses connected to her's also moved as a block. Union leadership
mother and her sons and daughters. Often these houses were headed by widows
in Riozinho from 1980 to 1983 (Claudino, Roberto, Nazaro and Ginu) provide
(Isaura) or included a remarried woman with a childless husband (Farias and one of
another example. When Claudino, the only leader not originally belonging to the-
the Santanas); or, in the Castelo case, the houses had female heads. These
Riozinho, left the union, leadership became consolidated around Ginu, who
although the youngest of a group, belonged :to the broadest network around,
covering several sub-estates, being a grand-son 'of old Rita's. When a cooperative
groups and occupying contiguous territories. Thus, following forest paths, one could
was established in 1989, one-third of the thirteen local posts set up in the Tejo
cross the sub-estates of Dourado, Manteiga, Riozinho, Camaleao and Boa Hora
always, or nearly always, meeting houses linked to one another. In going back one
old Rita or her sister, along with their sons, sons-in-law and stepsons in the
more generation (now refering decesead ancestors), the division between some o(
the blocks describe above vanishes. The most notable case was that of the houses
linked to (deceased) old Rita and her sister, both Jaminawa Indians, whose
and distinguished, as in the case of Isaura's group (although the oldest to occupy the
daughters married rubber tappers and whose grandchildren also were married. The
same forest area continuously) and of "old Rita's people"; or the Farias, who stayed
two Santana groups (two within Riozinho) once were part of Rita's descendants'
on for along time on the neighbouring Tarauaca River and in which they counted _
patrons among their relatives; or the Castelos, or the Cunhas. As compared with
these blocks, temporary and isolated houses of workers from Cruzeiro do SuI proved
unstable. A typical group of young and single rubber tappers (the Fonteneles),
"family" atoms whose only unity lies in the existence of a common trade post above
recently-arrived from the Lower Jurua, managed to establish a solid local base when
them, forest houses join together first as macro-houses with a strong technical and
m<?ral unity, which is reinforced by real and fictive kin relations, and with a common
leader. In turn,
m~cro-houses
conflicts and in party fights, and married one another. The Santanas were allied to
the Cunhas and Nascimentos; the Cunhas had a strong alliance with the Castelos,
cemented by seven marriages in all; the Castelos exchanged sisters with the
Cassianos (also linked to the Cunhas by marriage); the Farias had one marriage
with a Castelo and one with a Nascimento. The marriages tying together the second
generation of Cunhas, Santanas, and Cassianos tended to be repeated in the third
.
. 0 r at least having relations
aparently, local residents had no taboo agaInst
marrymg
' h'IpS. 15
with their female cousins, and even showed a preference for such
reIatlons
As
hous~ n~tworks spread, they cannot impose themselves over rivalries and
. . . harmony m
. ki ns h'lp, and the unity of a network
conflicts. There was no mtnnslC
'
group could easily ero d e over the course of three generations. 16 The F
ariaS
0f
15. I heard a tapper sing the following verse which illustrates the attitudes
mentioned:
.
(also' married to a Castelo), jokes and suggestions came up about the chances of a
Cunha boy courting a Castelo girl, until the Cunha father mused: "No, you are
carnal cousins." At first sight, these relations may appear quite obvious, and
Riozinho provide an example of this, as the group was about to break apart. Old
residents. Just as these houses were not isolated from one another in their forest
woman Farias lived with two married sons and daughters-in-law (a Castelo and a
clearings, they were not isolated as a whole fro"m the external world.
(coloca~ao
From the starting point of a tapper house, one can follow downstream the
settlement. A bloody rift broke out between the Castelos and the Fonteneles. Since
property structure (over areas where rent is collected) or the commercial monopoly
" grams 5.1 5.2). But
structure (over routes where goods are advance d by patrons ) (D la
'
"
the Born Jardim Farias had three Fontenele sons-in-law, and the Tacaratu Farias
one might also walk across hilltops and streams, reaching the houses of neighbours
had a Castelo daughter-in-law, the result was a near-rupture between the two Farias
blocks, divided between their ties to the Castelos, on the one hand, and the
within forest macro-houses, and farther ahead come into contact with faraway
relatives (Map 5). These internal circuits, cutting accross vertical structures of
commerce, constitute local webs mapping the long-term interests of forest houses in
Conclusion
People inhabiting and forming families in the forest over generations were
able to constitute relatively long-lasting social connections, in spite of the distances
imposed by the tropical forest extractive economy. Although without landed
property and without settling villages, rubber tapper house heads nonetheless joined
together in macro-houses, which behaved as technical units, as well as in extended
macro-houses networks and blocks, which can demonstrate solidarity in common
~ctions.
This analysis, supported by the quantitative data from Riozinho Estate, may
be extended to wider areas of the forest. Indeed, along the forest paths, rubber
tappers kept relations with tappers on neighbouring estates within and beyond
Restaura~ao
regularly with the Upper Tarauaca and Jordao basins, and permanently with urban
wife's former husband's descendant, although also his own son's son). He
commented to his wife: "Look, Joao Nascimento beat up Chico Teixeira."
17. Parties - where tappers enjoy drinking and displaying valentia and other
macho qualities towards women - were reputed to be dangerous occasions, except
when attended exclusively by a close-knit macro-house group. TheX often ended up
in generalised fighting and threatening. On such occasions I identified many of the
cleavages between macro-house blocks here mentioned.
199
Chapter 8
1. "For it is possible for all the kinds of causes to apply to the same object;
e.g., in the case of a house the source of motion is the art (techne) and the architect;
the final cause (eneka) is the function; the matter (ule) is earth and stone, and the
form (eidos) is the definition (logos)" (Aristotle, Metaphysics, rn, ii, 6; also VII
.
passim).
200
are stored, and bananas ripened. Also in the kitchen is the iirau, a wooden platform
where women prepare game meat and wash dishes. Among the basic equipment of a
kitchen are pots, pans, plates, spoons, metal cups and a clay ewer with fresh water.
Adult women or young girls carry wood from the back yard into the kitchen, andfetch water from the igarape. At night, under the feeble lamparina light, the kitchen
floor is occupied by the complete domestic group of men, women, children, nephews
and nieces, stepchildren and hired hands sitting in a circle to eat the pirao of mixed
manioc meal and broth.
A house possesses various small house lamps (lamparinas), and larger forest
lamps (porongas), both using kerosene. Some houses are distinguished by a sewing
machine, a record player, a wall radio or a wall clock in the main room, and framed
family portraits taken by itinerant photographers; suitcases, perfumes and a mirror
in the bedroom; in the kitchen of some houses one sees a large set of shining pots
and pans, and rows of cooking oil here used to enhance the manioc meal or the
cooked broth.
A house structure may last ten years, with one or two replacements in the-
In the front room single men set up their hammocks and male visitors lodge ..
When they set out to work, the single men grab their well-preserved rubber tapping
blades from the straw roof, as well as the poronga lamp, a forest knife and a
smoking-bag with tobacco, lighter and cigarette paper; they may take to the forest
the shot-gun which also hangs from the wall. The front room is where men fill metal
or cardboard cartridges with shot, powder, beeswax and bark before starting out for
the hunt. The front room also may contain fishing nets and harpoon heads (the garden equipment, also used by women and children, is to be found at the back end
of the house). Visitors stay in the main room unless they are invited to eat, when
they pass to the kitchen. The front room is a place for conversation, card playing and
straw of the ceiling, as well as in the walls. As it gets older, it tends to become
infested with cockroaches, spiders and ants, the yard becomes impoverished as the
topsoil is washed away, and the remaining vegetation disappears, making an old
house terreiro, the patio, inappropriate for feeding chickens and other fowl or
cria~ao.
The kitchen usually has its own foundations, separate from the rest of the
dancing to the sound of record players at week-end parties. The main room walls
forest, and give their knives to the house head who will put them away to becan be decorated with illustrations taken from magazines, advertisements or school
books.
returned later.
A yard surrounds the house. It is a cleared area, sometimes swept with
- While the kitchen and main room have windows and unlocked doors leading
to the terreiro (and sometimes have no walls at all), the bedroom has no windows
and has a single opening to the corridor, closed by a curtain. This is where a married
couple sleeps on a hammock or in a bed under a mosquito-net, along with a variable
number of children and single young women on overlapping hammocks, suitcases
and personal possessions.
brooms. It has fruit trees, chickens and pigs, the smoking-hut and a suspended
garden where women cultivate tomatoes, onions and other plants used for
seasoning. Originally created as a forest clearing, a new yard always possesses a
stock of tree trunks to be used as firewood in the kitchen, and at least one calabash
tree which provides all-purpose gourds, the cuias (one also may find pepper shrubs,
lemon trees for medicinal purposes, annatto trees, avocado trees whose leaves serve
as
medicine, peppermint and other medicinal herbs, as well as flowers). At the '
edges of the yard, there is sometimes a field with isolated palms and other trees,
where cattle go to pasture, which is the final border between the house and the
forest. To have one's house separated from the forest by a beautiful campo is the
. the
' VISIt.
. . Rubb er tappers 'rties
pa
, with music from batterysold to the houses paymg
operated record players, last well into the night, and may extend themselves from
one day to the next without interruption.
... .
Seen from above, the yard and its surrounding campo appear as an isolated.
clearing in a dense forest. Seen from below, the clearing is connected to the other
components of the settlement by a series of paths and trails. Several paths lead to
other settlements, and at least one of them leads to the
ro~ado
~~ri~g the two days set aside for making flour every two weeks or so,
I
the entire domestic group gathers in the the flour house, accompanied by their dogs.
Every hand in the house (macro-house) 'is needed to make in two days the farinha
necessary to last for two weeks, at least one paneiro or around 30 kgs.
The well-trodden path leading to the garden plot is known in every detail by
the trees, vines and palms along the way. Other paths may lead to isolated garden
plots, especially rice patches that do not need to be close to the flour house. Garden
plots do not have fences around them, though at times the path is blocked off by a
gate so as to avoid the invasion of cattle. Paths leading to the rubber trails also issue
from the main clearing, and off in every direction, as a whole covering a roughly
circular area around the central clearing.
!
I
204
As a rule, the area marked by the trails also constitutes the each house's
hunting and collecting spaces, which is respected by the other houses of the same
which after fifty years only a trained eye can distinguish from the virgin forest; the
settlement and especially by the houses of other settlements, although there may be
forest at its clim~, the mata bruta. All of these systems and zones represent
conflicts between houses in specific cases involving trails which were unused, or
together a wealth of water, soil, plant and animal resources, organised within the
when a pursued animal is killed in another house's clearing. Sometimes it is not easy
collective memory as one organises goods on the shelves of a store. The men,
women and children of a house who go through these resource zones including
a newly formed section of one's own trail; or there is a problem when one hunter
forest, gardens, field, back yard and waters can name "1lbber tree by rubber tree, a
man. The general principles involved are that rights were obtained either by renting
ripe assai fruits, a curve in the river whose good clay will serve to smooth one's hair,
trails or by actually using garden plots, hunting and gathering areas. If only a small
a rubber tree that ripened untended in the forest and is ready to tap, a paca trail, an
number of trails are idle at a settlement which is already occupied by two or three
area with soils and plants that promise a strong manioc crop, a river curve teeming
with fish. They spontaneously comment on such topics when walking along the
forest paths or trails, just as they never tire of talking about rubber, hunting and
hon~y,
and in the "heart of the trails", planting and processing of manioc at the col6nias,
Agriculture: Varzea and Terra Firme
raising cattle, chickens, pigs, ducks and turkeys in the yards (allowing them to roam
and forage), processing rubber in the smoke house and making food at home.
The Forest
On the Riozinho Estate, the area of forest surrounding the central clearing of_
floodplain soils (Meggers 1977 [1971]); Roosevelt 1980; see however Moran 1979;
a hinterland settlement with two or three houses covers a continuous territory of 900
1984:378), but Tejo River has predominantly eutrophic soils, low in acids (IBGE
1990).
the degree of human modification effected upon them. The areas recently affected
soils. Floodplain cultivation dispense with the use of swidden agriClllture. Flooded _
by the human action of houses (terreiro and campo often abandoned and replaced
areas on the banks, beaches or forest are used in complete annual cycles, beginning
ro~ados
with the end of one rainy season (May) and ending with the beginning of the next
206
(December). Though with very steep slopes, the clayey, steep river bank frequently
is cultivated with maize - sometimes requiring a preliminary clearing with a
machete, around June or July, and the crop is harvested in October or November,
when the first rains come. Beaches are the convex and sandy side of the river bank,
formed by the deposit of sediments washed from the opposite bank. On different
beach levels, which possess distinct soil characteristics, cultivators plant beans and
inaize; watermelon'; potatoes. On the floodplain soil above the bank, also rich in sediments, they plant tobacco and manioc (in this case, it is a risky crop, since
protracted flooding can make the roots rot). In September and October, tobacco
leaves are prepared in riverside houses well into the night, while beans are shelled
and dried. On the Jurua itself, often cattle are raised on the sloping banks (feeding
on the vegetation grown after the waters recede) while crops are cultivated on the
beaches; other times, fences separate cattle from cultivated zones. Both the beaches
and the lakes are treated as a group of resources to be used by a group of houses,
which are not necessarily located nearby. Typically, a group of houses linked
through marriage, kinship and friendship (a macro-house) will divide both beach
and bank in strips, while maintaining common pasture areas and using the same
lake for fishing. -To 'have access to both the floodplain and the "mainland" represents the best of both worlds. This situation is characteristic of the Upper Jurua, where
the floodplain occupies a narrow strip of alluvial ground. Thus, a house may use a
mainland area located twenty minutes distance in the forest. With the horizontal
combination of mainland and floodplain niches, a single house may enjoy longlasti!1g manioc and maize plots (associated to a flour-house) on the mainland, and
temporary crops of beans and maize on the beach and cattle on the opposite bank.
A house of this sort then can combine seasonal crops (tobacco, beans), multi-annual
crops (manioc), rubber trails (perhaps rented from different patrons, and located
both on the floodplain and mainland) and oxbow lakes (maybe an hour's distance by
river). Ordinarily these areas all are rich in game, since houses are concentrated on
the margins and only
r~rely
Such an area with highly diversified niches sometimes is exploited by three or more
houses together.
Agricultural procedures on the hilly mainland of headwaters, essentially ,based on swidden methods, deal' with erosion by the judicious use of aceiro palms
and banana on steep slopes, of interspersed crops (fast-growing maize and manioc)
and of the protective action of remaining trees after the burning. Gardens may be
managed for periods of up to ten years (cf. Tables 8.8a and 8.8b), and final
abandonment may not be due to declining fertility.
Mainland swidden agricultural plots .require greater efforts than varzea_
agriculture. The procedure calls for the striking down of large trunks with axes, after
the undergrowth has been cleared :with a machete; it also calls for the coivara
(which is the clearing of remaining trunks from an area that already has been
torched); it calls for careful sowing and
harder as times passes, and wild animals corisume a significant share of the roots.
Although the floodplain shows greater diversity in niches, mainland territory
such as the Riozinho Estate have greater population densities (1.7 inhabitants per
square kilometre) than the region as a whole (1.2 inhabitants per square kilometre)
and than the floodplain in particular - which is immediately recognizable in the air
photography of the region. The main reason for this pattern is that the most
productive rubl?er areas are located at the mainland headwaters. It should be
observed that Tejo River headwaters are not new frontiers, and have been inhabited
at least since 1899 (Cabral 1949), when it had the same high rubber productivity of
10 kgs rubber/day as today (cf. also Tastevin 1925:422).
is followed immediately by their processing, involving all the house members in two
The continued exploitation of an area may last from three to ten years. The .
days of work, mainly to make flour (and sometimes along with baby food, cakes,
clearing of a new plot (first year) calls for an area of virgin forest (which might be a
juices). In short, the stock of ripe roots in the garden is transformed into a stock of
capoeira, or previously cleared area, provided that it has been regenerated for over
processed flour within the houses. If the plot is replanted as it is being harvested, it -
fifty years). The first stage, in July, involves the clearing of undergrowth by machete;
affords a second harvest in its third year of activity. During this second manioc
the second and most intensive stage, in August, is when the trees are stricken down
harvest maize will not be harvested, since it is not planted during the second year
with axes; thus exposed to the sun, the area remains abandoned during a variable
(ril~ybe becaus~ th~ y~ung plants do not receive sunlight because of the size of the
period covering the driest months of the year (August and September). House heads
manioc plants). It becomes more difficult to clear invading weeds; the size of the
determine the ideal moment for burning the vegetation, upon which the fertility of
roots diminishes. Manioc is not replanted a third time on a row (cf. Tables 8.3 to
the plot depends critically, gambling between the additional gain of greater
exposure to the sun and the risk of early showers. The vegetation that is partially .
8.3d).
At this point, the plot is left fallow during bne or two years (thus completing
burned is then removed from the area to be cultivated (a procedure called the
five years since the initial clearing). After the fallow period, the area will be covered
coivara). At this point (September-October), maize seeds and cuts from manioc
stalks are planted. By December, the maize becomes ready to be harvested and part
machete and a hoe), and the plot replanted with manioc, or with tobacco, sugar cane
is eaten as canjica (a sweetened maize dish); the rest is left to dry, serving a chicken
~r some other crop .. This operation is preferred by some older cultivators because it _
feed over the following year. Ideally - depending on the variety of manioc used - new
is not necessary to open a new clearing with an ax, though on the other hand
manioc ripens over a year, and is weeded twice during this period. Hence, the global
period between August of one year and August of the next is divided into an
This new cyCle lasts another two years (now seven years have elapsed since'
the initial clearing was made). The cycle may go on with perennial crops like sugar
cane, which remain on abandoned plots or during transitional phases when the land
mainland manioc plots is that the ripened manioc may remain an additional year
under the ground. The second year of a mainland plot cycle is dedicated to the
their first year, others in their second, still others in temporary fallow; some with
gradual harvest of the ripe stock, along with the clearing of second-year weeds,
manioc, others with tobacco. The plots are surrounded by ubiquitous banana,
which are more aggressive. During the second year, the cultivator may choose to
pineapple and papaya trees. According to rubber tappers, papaya trees grow
replant manioc as he harvest the first batch - a simple operation requiring only the
reintroduction of stem pieces in the open holes. The process of pulling up the roots
211
210
removal of the mass from the press and the preparation of the wood stove. Then
(Table 8.8a).2
comes the long and- delicate operation of baking the mass in order to obtain 30 to 45
ro~a.
Each cova
(pit) correspond to a maniva bush. One thousand covas, occupying around 0.1
Women and children prepare food, bring water from home and prepare
other products, among which the mass a fina (fine meal) made of starch flour mixed
down, burning, clearing the burned area and planting), and can yield between 15
with the liquid extracted from the manioc press, which is an essential foodstuff for
and 20 paneiros or as much as 600 kgs. of manioc flour (each paneiro weighs around
.infants. Another delidous though less frequent product are the manioc cakes (beiju
30 kilos). A typical plot measures 0.3 hectares and requires around 42 man-days of
and tapioca). The peel of the manioc root is taken to the house to feed the pigs.
labour per year (Tables 8.3 to 8.4). These figures do not include weeding visits, often
Rubber tappers take great care in preparing manioc flour, always concerned with its
done by women and children throughout the year, in casual visits that are not
quality. Men always take over at the critical phase of toasting the manioc .paste,
counted by the house head. In contrast, house heads count two or three annual
stirring it continuously over a hot stove (on an iron griddle 1 to 1.5 metres long,
weedings (for gardens planted in mata bruta area) during the initial year of the plot,
sometimes circular) with a wooden oar, until the past arrives at the consistency of a
each consuming around three man-days. This means an annual cost of six to nine
crispy, coarse meal so characteristic of the Upper Jurua. Women command the -
man-days for 1,000 covas (18 to 27 days for a typical roc;ado) for systematic clearing,
preparation of massa fina, which also is to~sted, though with the final consistency of
wheat flour.
becomes absorbed for two whole days. The activity sequence is as follows. On the
enough to supply an average house (including dogs among the consumers), requiring
first day, roots are pulled up and transported to the flour-house by adult males, are
an annual planting Of 3,000 covas. The average plot, of 3,000 covas, requires at least
peeled (by men, women and children), and are washed in water brought from the
24 flour-making sessions (two a month), each involving the labour of the entire
stream. Peeled and washed manioc is then grated (by an adult male, using a wheel
house for two days, or almost 50 days per year each house. The variation in the
with alurninum teeth powered by a 3.5 HP engine) and the paste is poured into a
amount of flour produced depends on the extent of work in the flour-house, and
wooden vessel. At the end of the afternoon, a man prepares a press, covers the wet
does not necessarily reflect more or less days dedicated to flour making (cf. Tables
mass with leaves and places it with a metal cable between pieces of wood, exerting
8.3).
pressure so as to extract the liquid from the fiber. The second day begins with the
If a house dedicates 150 days a year to rubber (about 120 days tapping, four-
days a week over eight months, plus 30 days to prepare two standard-sized trails)
2. Around the flour-house, an area frequently occupied by women, there is a
dense amount of domestic plants for medicinal and other purposes, such as the oaca
bushes used to prepare a paste used for stream fishing.
and develops a annual garden plot of 3,000 covas, it will be occupying 267 man-days
(150 + 117) to operate the rubber and maniac sectors of the settlement. Since the
213
total number of man-days per year is limited to 312 (52 weeks x 6 days),
Many animals are slain' simply because they are considered predators. This occurs
simultaneous rubber and manioc production take up 85% of available labour time,
with jaguars, who visit clearings in search of pigs to eat, but tappers also kill the
leaving approximately one day per week for hunting activities (not counting
tailed pacas, though they do not eat them, because this animal preys on the manioc
Sundays). The real income of the house would consist in 1,800 kgs. of flour (3,000
plot. Capibaras, though ordinarily not eaten by tappers, have been practically
covas x 600 kgs.) and 960 kgs. of rubber (120 tapping days x 8 kgs.). One must keep
exterminated along the Tejo River, maybe simply because they are easy to locate
in mind that these -are man-days, assuming that there is only a single adult male from canoes (on the other hand, their fat possesses medicinal properties). Animals
such as sloths and anteaters are "neutral", since they do not serve as food and they
Hunting
are not viewed as man's competitors ':; -though one type of anteater (the mambira) is
In the standard division of time during the week, Wednesdays and Saturdays
are the preferred days for hunting. If possible, tappers avoid hunting on Sundays; on
some places, hunting on Thursdays is strictly avoided. Rain season the ideal season
Among birds and reptiles there is also a pattern of avoidance, with an even
for hunting (the hunter's presence is less easily detected because the soil is wet-,
greater degree of exclusion and restrictions. The nambu forest partridges are
Riozinho rubber tappers show exceptions to the rule that hunters maximize
among birds analogous to deer as safe food), and seem to be as abundant as the
returns by seeking out species that have a greater presence in the total anim al
paca. By contrast, large land birds such as curassows, guans, chachalacas and
biomass (Robinson & Redford 1991) According to this rule, animals thal are most
trumpeters are rare in Riozinho, and I only saw them kept as domestic animals,
frequently
capt~red should
probably captured while small (the trumpeter is a guardian the domestic fowl).
gre~t
meat relative to the total amount available in the forest; they are both more visible
and give the largest return per effort. This rule is broken in almost every category of
serious shortages the sm,all, white egrettes, or the japinin birds, were never bothered
size. For example, both the tailed paca (paca-de-rabo) and the capivara have the
at the Tacaratu settlement. Among reptiles, river turtles have become very scarce
size of the paca (Agouti paca), but they are not eaten by rubber tappers. In a smaller
(though they
category which includes the larger monkeys, neither the anteaters nor the sloths are
4av~
h'l e th e sqUlrre
. 1s
are highly valued as food, tappers find repUlsive the forest rabbits, marsupials of all
kinds, ground and tree rodents such as the bamboo rats, along with weasels, wolves
employing diverse techniques and affecting different animals in quite unequal ways.
and bats. The status of small monkeys is less rigid, and they are often kept as pets.
This means that concepts like "depletion", analogous to "impoverished soils", are
W 1
215
techniques are involved (Vickers 1988: 1521). The rubber tappers themselves
distinguish demographic factors (high occupation density drives away game) from
technical factors (the use of pauIista dogs drives away game), and from a _
consumption perspective they are quite selective when it comes to game _
designating a restricted sub-group within each large animal category identified (see
years, and 'is 'l ike a grave ''illness affecting not only the hunter but the whole house
group.
Some very successful hunters attribute their success to the use of hunting
amulets, the cabojos. These amulets may be extracted from animals (such hair balls)
or can be prepared with the nests of rare birds (such as the uirapuru). These hunters
maintain their amulets secret. Forest animals may be protected by "fathers of the
Hunting. Panema and Reima
forest", who can punish hunters who hunt the same species repeatedly, or hunt on-
surrounded by dangers of a mystical origin. Hunters facing persistent bad luck will
tend to blame another person for inflicting panema on them. The ultimate cause of
because they can control the mystical forces that govern hunting. 4
panema lies in a woman who has consumed game killed by the hunter. A woman
Reima restrictions divide game meats into one category that is safe for all to
might afflict a hunter with panema either on purpose or not. A hunter becomes
inflicted when the meat or bones of game he gave (and did not sell) to another -
game meat is not enough to sustain a woman during the resguardo (post-partum)
house come into contact with a woman during her menstrual period or after
period, which might last a whole month, and during which she must avoid any sort of
childbirth; in fact a woman in this state has only to walk over an area where these
work, must remain secluded and must not eat reimosa food. A woman with sons who
bones are buried to affect the hunter. A bitch in heat can cause panema, which also
are hunters might allege that she cannot eat reimosa meats, which means that the
afflict hunting dogs in the same way, with disastrous results for a household. Even
hunters must supply the house with deer meat (the only large game that is not-
hunting trails, along which traps are placed, can become affected by panema.3
restricted), and of
non-~eimosa
During the post-partum period, women ideally should survive on a diet without
hunting larger game altogether, restricting his actions to embiaras, that is until he
forest foods - typically chicken broths flavoured with butter.5 Great ("lucky") hunters
beco_mes cured by herb baths and appropriate chants. The panema state can last for
, , ,3. I. attempted to photograph the setting up of a trap on a paca trail Thetapper s WIfe alleged that I co~ld empanemar or enrascar (get in trouble) the' trail.
The n.ext day we found a paca III the trap. From then on, I gained the reputation of
breaking panema states. It worked many times.
4. The Floresta settlement, on Riozinho Estate, had two feuding macrohouses in two different clearings (invasion of crops by pigs started it). The son-inlaw of the head in one macro-house was a caboclo who "closed off' access to the
clearing to the residents of the rival clearing (two brothers-in-law). The caboclo had
also the power to have all the women he desired, and was in conflict with the patron,
who thought that he produced little rubber.
5. During a meeting of cooperative managers in 1989, the only point raised
by women was to request the inclusion of canned butter among "basic commodities".
undeniably attract women. To be able to bring home food from the forest is a
was found to be around 2.33 kgs ' per day per house (mammals only), including by
order of weight pac.as, deer,.tapirs, boars, agoutis, monkeys, armadillos and squirrels _
(birds and reptiles also were not computed here).8 The contrast in daily house
dangerous to sick men and women, and to woman after giving birth. Both complexes
consumption confirms the current opinion that Riozinho has less game than the rest
affect the hunter's behaviour and consumption patterns. When combined with the
restriction of certain weekdays (Thursdays in some locations, Sundays and holidays
of the region.
Predatory action on tapirs, wild boars and, to a lesser extent, deer (on the
in others) and with limitations to the repeated preying on the same species, the
notions of panema and reima in a certain way limit the total pressure upon forest
animals.
confirmed by fact that tapirs are inexistent, and boars and large monkeys are very
rare on the Riozinho according to the rubber tappers (cf. Bodmer et al 1988). On
Hunting Productivity
The region easily could accommodate larger garden plots (within a total area
the other hand, the comparison with published data reveals that the paca (Agouti
1li!-ca) also was being hunted at greater levels than theoretically sustainable, though
of around 400 hectares of forest per house, and annual plots of 0.3 hectares); it
its scarcity was not noticed anywhere, even in Riozinho. One possible explanation is could allow for greater rubber extraction (one third of the existing trails remain
that pacas feed on the rubber tappers' garden plots, and have a short reproductive
idle). In Riozinho, density reaches 1.7 persons per square kilometer. No one is
concerned about the shortage of land for cultivation; as for rubber trails,
cycle. According to the tappers, as much as 25 % of the manioc roots are eaten by
pacas and other rodents, while ambushes set up at the plots rarely failed to produce
approximately 30% of the trails remained idle; but the shortage of game remained a
game. N ambu partridges, and land turtles are other examples. According to rubber
genuine source of dissatisfaction.
tappers, deer have become more difficult to find since the introduction of "paulista"
On the headwater Riozinho, the average daily amount of game meat per
dogs in the early 1980s. The tappers' impression is that Riozinho is a place rich in
house could be 1.48 kgs.1 On the Tejo area including the Jurua varzea, the average
rubber but poor in game (ruim de rancho or "bad for food"), where one faces
6. I witnessed a conversation in which a woman accepted to live with a tapper
~he ~ad never seen before, on the information that there was plenty of monkey meat
III hIS house .
. ", 7. O.n Sep~ember 1983, during eleven days, three houses (a macro-house) on
RlOzlnhO killed eIght mammals (47.4 kilograms of meat): one deer (1752 kgs 370/
of the total. volume of meat), two pacas (16.~4, 34.7%), two squirrels (0.692 1.5%)
~4~~rkadII~f (7.02, 14.9%) and two agoutls(5,68, 12%). These figures av~rage t~
.
gs.T game per house per day; 0.269 kgs/person/day' and 0479
kgs./day/km . Game meat was available on nine of the efeven days' f~r three days
the houses also consumed on duck, one chicken and several egg~' bananas and
papayas m,ade up meals on four days, and fish was eaten twice. M~nioc flour was
always avaIlable.
"hunger".
. . 8. Mammals Hunted, September 1991, 49 houses: Of 108 mammals and 799.132 kilograms of meat, there were 10 deer (175.17 kgs., 21.9% of the t!>tal
volume of meat) 20 pacas (227.2, 28.4%), 17 squirrels (5.882, 0.7%), 4 armadillos
(14.16, 1.8%), 26 cotias (73.84, 9.2%), 21 monkeys (63.0, 7.9%), 8 pigs (90.88,
11.4%) and 1 tapir (149, 18.6%). These figures average O 2.33 legs. of meat per
house per day; 0.33 kgs/person/day; and 0.77 kgs.jdayjkm .
219
218
gathered from' t?e ~orest, sucb as honey, palm tree wines, or palm hearts, or forest_
fruits.
Note that the presence of manufactured goods (purchased with rubber)
constitutes an essential component in the quality of life of a rubber tapper's house.
A house should never be deprived of manioc flour. The point in making the
Houses are not complete or even viable without hunting equipment (guns,
manioc flour is not to extract poison from the manioc. The varieties in use are not
ammunition), fishing equipment (nylon for nets and lead), agricultural tools (axes,
poisonous (at most some are bitter to taste). Manioc and bananas can be eaten
hoes, machetes), and extraction tools (blades, tapping knives, tins, pails, sacks);
cooked or roasted, but during a main meal these items do not replace flour. Indeed,
gasoline, metal plates for the flour-houses. Houses also need fabrics and thread, and
instead of making
flour~
soap and salt to be used in the domestic processes. These imported goods, acquired
tapper's home, the absence of flour is a sign of poverty and a cause of shame.
as means of production and paid for by rubber exports, indirectly enter as part of the
Manioc flour is a universal food, with no reima or restrictions due to sex or age.
production process in every sector. This means that tt is not possible to separate two
Children under one are fed on a diet of massa fina or a special manioc flour rich in
discrete technical spheres within a house: one of subsistence production and the
~ther of comm~dity production. This is the reason why I have suggested that basic
goods are those used as permanent flows, either as a flow of things (such as manioc-
part of every meal, in the form of a broth which together with the manioc flour
flour, game, salt, soap) or as a flow of services (such as the manioc-house, shot-guns,
makes up a pirao. Everyone appreciates the standard farinha and caldo (game
or tapping knives), in contrast with non-basic goods such as occasional food (fruits,
broth) staple, followed in some houses by coffee and the strong Jurua tobacco,
rolled into cigarettes or chewed. The use of ducks and chickens, eggs and fruits,
The line separating basic from non~basics cannot be easily drawn, since some itens
wine made from patoa or beans does not indicate only food variation, but rather
may classified differently by different houses. What looks like non-basic cooking oil
shows the absolute scarcity of game meat, which ideally should always accompany -
or coffee in one house is basic food (for male workers only) for another house in the
manioc. But game meat, contrary to manioc, is not a universal food. Women on
resguardo or protection after childbirth must avoid reima meat (which excludes
most. of the forest food), and should have domestic food or commodity food
new records, and radios all are highly valued by many tappers, who also enjoy
(chicken and butter), similar in this respect to children under one (powder milk ans
decorating the ~alls of their homes with pages cut out from magazines and books.
sugar). These restrictions also apply to sick persons who should also avoid fruits
Men, women and children great pride in their imported possessions, and
such as pineapple, orange, lemmon and papaya. Are luxury food, in a sense, both
manufactured articles may draw the same interest as forest animals and plants.
commodity food such as chocolate, guava paste, crackers or sweets, and food
Imported objects are used for personal expression and in communication, and the
fact that game meat cannot be exchanged for commodities - ' , in general exchange
. econOITIles
. lk
that of the rubber tappers are essentially linked to
Domestic
1 e
regimes distinguish that which is given and comes from the forest, that which is
the market both on the , production level (which makes it impossible to discern
exchanged in kind and comes from human labour, and that which is exchanged for
debt and carries overtones of exploitation - does not prevent watches, knives and
consumption), and also on the technical level (which makes it impossible to see a
1986, 1989). Such barter exchanges among friends put into circulation novelties
such as records, new dog breeds, or new guns, in circuits which sometimes lead to
the exchange of an old gun for a new one. Game meat, manioc flour, imported guns
or dogs occupy, in their own ways, a central position, whether affectionate or
cognitive, social or personal, among the rubber tappers' interests and exchanges.
This means that the global quality of life simultaneously is a function of different
categories of goods - deer meat, patoa wine, but also chocolate and powdered milk -, and that the consumption sphere is not closed.
Similarly, the forest peasant's technical knowledge should not be seen as
~losed.
have a keen intrest in mechanical processes. To take apart and rebuild a combustion
engine is a matter of pride to many rubber tappers. Watches and motors were
especially admired. Watches are tested for macho qualities. A friend of mine asked
during the noon break at the trail: How can all clocks be synchronised (among
themselves and with the sun)? The tapper mused that the first clock maker must
have "lost a lot of clocks" in trying to synchronise the hand with solar movements.
Rubber tappers also admired the builders of combustion engines. A toy car given to
a child immediately was taken apart by adults who were curious about the spring
mechanism. Interest in mechanical objects, from diesel motors to "super-sonho"
(supersonic corrupted to "super-dream") airplanes that crosses without making noise
~
was great.
Conclusion
The obje~tiv.e of this chapter was to outline the form of tappers' houses in the_
forest on the Upper Jurua, as an introduction to a more detailed analysis of rubbermaking in the next chapter. Rubber tappers are extractors, cultivators,
hunters/fishermen and artisans. They use fo_rest areas whose natural resources are
carefully mapped; they use technical procedures which depend both on the forest
niches and on imported equipment; they desire game and manioc flour as food, and
also imported commodities and are keen on new technical ideas.
21L
223
Chapter 9
HOUSES AND THE MAKING OF RUBBER
Structure of the Labour Process
We now introduce a definition of labour processes, which corresponds to the
definition of the processes of house maintenance during the short-term period of
technical activities. 1 This definition follows Marx's idea of labour, which is "above
all nothing more than a process between man and nature ( ... at the end of which ...)
emerges a result that already had been determined from the outset and, therefore,
ideally, in the worker's representation" (Marx 1969:149). Still according to Marx, the components of a labour process as "an activity oriented towards an end" are the
labourer, the object of labour (especially nature) and the means of labour (Marx
1969:149). This definition may be traced to Aristotle, who used it to describe the
existence of things, - substances - as in the articulation between matter and form,
through the action of a technical agent seeking an end. 2 In applying these ideas to
the study of the rubber tappers' economy, the labour process may be described by
specifying the technical agent (the working team of a house or macro-house), the
matter (the forest territory); the -tools (knives and so on), all this antecipated as an
image together with an .end (rubber balls and their commodity equivalent). In
quantifying these components, one may use the hours of a labourer, the number of
trees on a rubber trail, the cost (in rubber) of tools, the number of rubber bundles to be produced, and a final stock of commodities to be acquired.
224
A rubber tapper's house is thus a unit of the labour process, that is, it is atechnical unit. It also is a juridical unit, in the sense that the components of the
"leg"). Some girls would cut and save sinall quantities of rubber in order to buy shoes
or a new dress.
labour process are under the house head control (rule over the house people; rights
cutting and gathering the rubber of an entire trail by themselves, and able to smoke
decisions over the volume of production, over the use of labour-time, over the
(5) Hired persons, who took care of particular trails or worked in teams of
exploitation of trails and over the techniques employed. 3 I shall use this framework
two with the house head. They were remunerated either in goods (a shotgun, for
example), or in fixed amounts of rubber or money (to be handed over at the end of
the harvest), or received a percentage of the final product (in this case we properly
speaking a meeiro or sharecropper).
Labourers
In rubber production, the work group was composed of the facas (knives, or
available labour) of each house (house heads, dependent tappers "working for the
head", and auxiliary workers including women). Work crews varied from house to
house and included the following categories.
Dependent tappers who were sons of the house head worked "for their
father" in his house until marriage. 'During this stage, -. the young rubber tapper has
the capacity of an adult, but not the responsibilities of a house head with the maniac
garden. Tappers who are dependents of the head usually are distributed between
ages 16 and 24. A young tapper might leave to work either for a married brother ~
where labour relations and distribution might be more equal, or the work day less
intensive - or as a hired hand at another house, usually on another rubber estate, or
by establishing his own house by marriage. On Riozinho Estate, the five hirelings
aged 17 or 18, which suggests that the situation of hired help was transitory. Hired
help lived in the house and took advantage of the same services as the other
residents, such as food preparation and laundry, services provided by the women of
and remoVing forest resources. This is the case of wild honey extraction, when tree
tru~s are stricken ' do~ and the whole honeycomb is withdrawn; or of fishing and
For 1980, the population census for the State of Acre recorded 23,830
tappers for the whole state, but a census of agro-pastoral activities declared 38,000
workers in natural extraction for the same state. The difference may be traced to the
hunting. According to this definition, the collection of animal pelts through the use
of traps also is a predatory activity. A plant or animal population submitted to
human predatory activities mayor may not be in equilibrium (and equilibrium may
inclusion (or exclusion) of female labour and of labour under age 16. The effect of
or may not be a stable one) depending on whether or not the number of individuals
this counting procedure is illustrated in the table above. One could argue that the
removed from tile population is offset by natural reproduction.
number of 28 auxiliaries in 57 houses (probably underestimating the real
to the death of a popUlation: one might say then that unbalanced extraction
The image of the tapper's work as a stroll through the forest where he
iniply that a popUlation or group of populations remains roughly constant over time,
including that of predator/extractor, on the one hand, and that of prey and
Predatory activities are those which involve the permanent and irreversible
extractive resources.
From the foregoing, it should be clear that neither predation nor extraction
forest's stock of life into a flow. Predatory techniques include searching for, locating
may be taken as synonymous with either the absence of human investment over
nature, or with disequilibrium. As soon as they are part of labour processes and
.
5. Chagas's house (38. years) included two young tappers, aged twelve and
thIrteen. One of the two traIls tapped had 150 trees (since the standard trail had
120, this one was equivalent to 1.25 trails). The trail was divided into a main course,
tapped by Chagas (100 trees, or 0.83 of a standard trail), and a section tapped by his
two sons (50 trees, or 0.42 of a standard trail for the two). Cacundo (46 years) had
thr~e young "blades" (aged 16, 15, a~d 14). Usually three of them tapped the same
traIl of 140 trees (1.17 standard traIls) - In other words, at Cacundo's house with
more workers than Chagas', each worker tapped the equivalent of 0.4 trails pe~ day.
Similar data have been obtained from the two surveys in Riozinho (1982/83).
- --
.~
229
228
for other sorts of trails 'in the forest, which we call "paths". The latter are narrow,
grooved and directed roughly along a straight line; the rubber trails are wide, hardly
trodden, and meandering along a closed loop, though there are variations with
principal bends (volt as) and secondary bends (oitos), cul-de-sacs (mangas) and short
cuts (vara~6es) (Map 5).
The preparation of a rubber trail begins with three tasks:
raspar (scraping) -and. empausar (specially prepar;ng the trees).
ro~ar (cutting),
Ro~ar
essentially
ladders on trees to be milked, and to install or repair trunks over streams to serve as
. 7. Cf. Emperaire 3;nd Delavaux 1992:2. Trails are of course used not only for
tappmg, but also for huntmg and gathering.
getting it accustomed to the cut. It was important to accustom a tree during the-
bridges. Before proceeding to routing cutting, some tappers would spend several
days in a preliminary milking of the tree. This was called "taming" the tree, or
230
correct lunar phase and with the necessary regularity, in order to increase its
productivity gradually until reaching its maximum. By this time, the trail is ready.
The maintenance of a trail is a more difficult task than it appears to be at
Work Instruments
Rubber production required labour and trails, but it also demanded
equipment.
production belonged to individual houses: the tins, the smoke house, the press and
the tropical forest in Acre, makes the task of preserving a trail quite arduous. The
the rest of the equipment. To remove a rubber tapper's tins from a trail was the
taboca stands have an unpredictable life cycle, and they might fall at any point,
symbolic gesture meaning his expulsion - though these implements would be taken _
by their owner upon his leave. They never belonged to the patron.
fall. The forest surrounding and above trails invades them during the months they
The purchase of this equipment was not the basis for the dependence of
are not used for rubber gathering (August-September, and February to May),
tappers on patrons either, since most implements were produced simply and locally.
cluttering them with branches, vines, trunks and taboca bamboo. 8 Tappers not only
Only the raw matenals used in these implements; ' such as tin and iron,. were
maintained but also added new turns to old trails, sometimes making entirely new
necessarily imported from outside the regional economy. In other words, the
trails.
his contract, a fact made evident with cases of tappers subletting trails. Renting '
equipment.
The most expensive equipment used in rubber extraction consisted of tin
cups (to gather the l~tex that oozed from incisions), because of the number involved.
trail, planned and executed by a house head with the labour and equipment he
Individual tins could be made from scrap 'metal (vegetable oil cans were a main
commanded, created informal use rights over the new trails 9, but the tapper had no
source), and soldering was unnecessary since the coagulated latex itself sealed the
rights to compensation for the labour invested in the opening and conservation of a
~etal. Nonetheiess~ in"order to fit 240 trees,a house needed 500 tins or more. This
trail should they choose to leave the estate (in contrast with the case of agricultural
was not impossible for a young house tapper to ,invest. Part of the tins could be
improvements ).
purchase second-hand from other houses, and a young tapper might receive tins
from his father as a contribution to the equipping of the new house.
233
higher costs. The best blades were fashioned by master blacksmiths along the Jurua,
and were sought on trips or ordered through friends and river traders.
The narrow-mouthed tin pails, holding from four to eight litres of latexlO,
were also manufactured by tappers, with no soldering. Sacks holding from 20 to 30
litres were used to transport the latex when' a pail was full. The opening of these
plus' long~term
2~%
replacement cost (48 kgs.) remained around eight percent of the average annuat
product. These relatively low figures indicate that the tapper maintained a
reasonable technical independence (Table 9.2) under the 1982 prices.
sacks are tied by a strip of solid rubber, and they were strapped onto tappers' backs
using estopas (rude cloths fashioned as back-packs). Sacks were sewn together by
women and fully coated by men with caucho latex mixed with sulphur.
Knowledge
The foregoing description of the use of labour, of nature and of technical
The poronga was a kerosene lamp, which could be held in one's hand or
teeth by a wooden handle, with a protective device to protect the tapper's face,
procedure may be seen as a way of doing things - a recipe that includes material
allowing him to see the trail without being distracting by the glare. ll The forest or
ingredients, instruments and qualified labour, and which specifies what product is to
hunting knife usually was recycled from an old machete blade, equipped with a new
wooden handle as well as a wooden sheath. The rest of the equipment belonged to
the smoke house, whose principal item was an underground clay oven built by the
instruments and nature. This is a structure which, just as material equipment, may
tappers and periodically repaired by the women with the kitchen ashes. In the
smoke house, the only piece that had to be bought was the basin in which the latex
Hence, the same equipment, the same people and the same natural resources
was deposited from the sacks (though some houses substituted metal basins with a
piece of smoked cloth, the bangue). Tappers built their own rubber presses,
results and involving distinct qualities and intensities of labour. Under this
definition, technical procedures may differ because of the instruments used (for
Expenses with purchased equipment reveal how much rubber Was necessary
example, a smoker or a press), because of the persons involved (one worker could
prices, in, order to produce rubber "12 The total equipment cos t expresse d
,at current
,
tap an entire trail in one day, while a team of two might be employed to tap a
similar trail); and because of the particular use made of nature, or because of the
h ]l~d' Labtebx w(as l!leasur.ed in latas. One lata = 2 litres of latex = roughly 1 kg
res so I ru er losmg weight thereafter).
.
11. In the nearby Tarauaca Valley and elsewhere in Acre (but not in the
upper Jurua), such lamps are fixed on tappers' heads, much like miners' lanterns.
. 12. In. 19~3, a tapper a!rived at Riozinho Estate, bringing the followin
e<jblpment with him (expressed m rubber values): three incision blades (8 16 k
~
ru ber), one handle (4 kgs.), one pail (4 kgs.), one sack (2.5 kgs. for the ciothl~~e
the waning one, or the same tree might be milked twice or three times a day, or half
or one third of the circumference of the tree might be used in tapping; each of thesebasin (10 kgs.) and 730 tins (116 kgs.), adding up to 144 kilograms of rubber. The
gathenng tins had been brought by Brasil.
.
considering the resistance of the outer bark to the blade, the format of the trunk and
other characteristics that 'vary enormously from tree to tree, guided by his own
variations.
knowledge and manual dexterity. An incompetent rubber tapper wipes out rubber
Knowledge of the effect of moon on latex yield, on the one hand, and pacts
trails, by virtue of errors whether in the delicate incision process, in the division of
between the tapper and the "rubber-tree mother" on the other hand, point to
different sources of productivity. In the first case, changes in nature; in the second
result in annual outputs twice as great as those in use, but they destroy the tree after
case, changes in the tapper himself (analogous to the "lucky" condition of a hunter.
after similar pacts with game protectors). Such "mystical" productive forces (Santos
The rubber tapping example illustrates the general principle that one
in rubber tapping, and I could not see them in action in agricultural activities.
or non-action. It may appear strange that not acting upon a tree can be seen as a
technical procedure - ' but; much like hunting, fishing and collecting cannot be
within the range of possibilities offered by forest technology, that is, the forms of use
accomplished without game, fish and living plants, which are thus classified as
known to rubber tappers. The point to be stressed here is that there were no
processes which required the presence of specialists or which fell beyond the
predation and extraction, and thus such action may be classified as productive
investment. 15
The limits to tapping intensity, to the number of cutting days per week, and
to the number of months of activity per year all were defined clearly, but divergence
The basic question involved in the tapping process is that of short and long-term
could exist. For instance, there was no consensus 'opinion regarding the respective
sustainability. The key to a tapper's expertise resided in' his ability to execute
merits of the "half cut or "one-third cut" systems, which differed in the proportion of.
the tree's diameter used in tapping. The question was whether or not the "half cut"
was the cut that kept the tree healthy and maximized the amount of latex yielded
reduced the total productivity of the tree over the long run. I never found any doubts
over the long run. Another basic technical problem lay in the location, number and
frequency of incisions per tree (in a specific day and throughout weekly and yearly
periods). A competent rubber tapper will use procedures appropriate for each tree,
~3.
14 On the cultivated rubber plantations at the Sao Paulo State, the cost of
labour is high as a consequence of the need to maintain permanently employed
workers and of training costs.
15. To produce ("to bring forward", "to b~ing forth", <;>r "!O bring ab~ut") by
means of labour includes both predation/extractIOn, domesticatIOn and !abnca~?Il:,
as modes by means of which ' men/women "treats (nature) as belongmg to lID
(Marx, 1974:341. idem).
as to the two-day per week limit for tapping a tree, or to the need of four months'
The basic process of latex extraction consisted in revisiting the trees during
g~thering latex when it is still in a liquid state, and this phase is_
annual rest for trails - which means that any given tree could be tapped only 20% of
the year. 16
associated with the procedure of processing latex in the form of solid blocks of
In the Upper Jurua Valley of Acre, tappers continue to cut at least until _
rubber. Up until 1982, this operation took place through two serial sub-processes:
January, and waist-deep water is no obstacle to tapping. While the tappers face the
the initial production of himps of eight to ten kilograms of coagulated latex (from
risk of losing part of the latex in case of rains (water fills in the cups), this risk is
Hevea sp.) as a "ray~fish,,18, with the aid of the oie (also mown as caxinguba) tree
compensated in part by the greater productivity of rubber trails during the rainy
sap (Ficus sp.); this~as cut into strips and Tl1ade into a bundle on which successive
season. 17 By July, when polar cold fronts make the temperature drop abruptly
coatings of latex (mixed with a thickening vine, Philodendron sp.) were applied,
below 15 degrees Celsius, the rubber trees increase greatly in latex production but
they are not tapped on those days because this would harm them. By August, when
" - nu t (Orb\Junia
sn) fire until
:
being simultaneously smoked. :~,'over
'. th e , ;c-ocao
JO
'1'"
,
19
completing a pela (ball or bundle) weighing around 60 kilograms (Table 9.3) The
the trees renew leaves and blossom, and productivity is low, they are not tapped
proportions of coagulated in the core of the rubber ball to the smoked outer layers
either. It is explained that the trees need their whole latex to grow the new leaves.
varied from tapper to tapper. Some rubber tappers who wished to save on labour
This period coincides with the best period when rubber tappers can strike down the
increased the amount of coagulated rubber, although this was considered a inferior _
forest and burn the dried up vegetation in preparation of garden plots, after a
process (other tap~ers, although I did not see it in Riozinho rubber, could increase
weight adding foreign substances to the latex, including earth and manioc). Non-
The Riozinho tappers tend to complete at least half the tapping day before
dawn. It might appear that the reason for this is that the absolute length-of the work
In Process 11 (Table 9.3), the latex could coagulate in the tins (and thus does
day would require night labour; however, even when tappers work only a small part
of the trail, they begin work well before sunrise, oriented by the idea that trees cut
(using the coagulant agent); instead of being smoked, it is compressed in the press
during the night will yield more, and also because night work is preferred by some
to extract water and made into planks. Well dried planks make good quality rubber.
With this new technique, the second foray into the forest could be put off and
converted into greater tapping time. This technique could double the production
and it was introduced, notwithstanding the patron's initial resistance to it.
16. On plantations, trees can bee tapped on alternating days. The life
expectancy of the plantation rubber tree is calculated at thirty years, and dead trees
are used as fuel.
17. In the Purus-Acre Valley, rubber tappers suspend the tapping activity
comletely already- by December. There, this season comcides with the lucrative brazil nut harvesting season.
18. On account of the round, flat form acquired by the latex left to coagulate
on a metal basin (the river ray-fish is round).
19. Cf. Emperaire and Delavaux 1992 for scientif terms.
tappers commonly assert the number of days they intend to tap during the year; or
processes (sometimes on the same rubber, since smoking a plank protects the
how many kilograms .of .r ubber they jntend to produce. The assertions are for
rubber against bacterial decay; thus smoked rubber has a pleasing smell, in contrast
instance: "This year I am going to cut 110 days!" Or, "l want to see if make 800
to the acid smell of raw latex). I observed some tappers say that they missed the
kilos!" To be a tapper and to produce much rubber constitutes a source of pride for
waning hours of the day spent in the smoke house. In fact, these hours provided an
young men. Some young men between sixteen and twenty four spoke well of walking
opportunity for informal conversation with neighbours, associated with the pleasure
or running through the forest in ' the middle of the night, comparing this positively
of seeing the final result of one's labour transformed into a handsome ball.
with all-day agricultural work under the hot sun (particularly wage labour).
Beginning
However, the tenacity of work rhythms - including walking, running, going up steps,
Work Plans
Work routines and rules may be distinguished from the daily work activities
in real houses: in other words, the distinction may also be expressed as that between
the structure and the organization of labour. Idealized routines and statistical
averages capture the essentials of a technical language. But labour processes - the use of language through actions, so to speak - vary tremendously in intensity, the
composition of crews and technical variants. This variation is related to fluctuations
in the composition of the domestic work group as well as in the natural resources
and in individual technical styles; and it responds to prices, according to the house's
production plans. The description that follows presents idealized work routines as
themes upon which individual houses construct their own variants.
Daily work is quantified. Some rubber tappers mark the number of days they
tapped with a knife on a post in the house. Many tappers can recall without
hesitation the exact number of days they cut trees in each trail that season. 20 They
know the minimum and maximum latex yield of each trail, and of individual trees.
Though almost all are illiterate, most of them know how to calculate how much dry rubber will result from a given quantity of latex. In the case of house heads , the
20. Each tapping day leaves a single incision in each scra ed area of the
c:;rrent s7ason. Actual work can be thus checked tree by tree. This ~as not done b
y
t e matelro who only was concerned with the state of conservation of a tree.
crossing bridges, and making precise incisions at heights of one to two metres,
without rest until finished - must be accompanied by a resistance to hunger, and the
required capacity to sustained labour is much greater than in agricultural work. -A tapper with any pride is capable of working six hours tapping without eating,
proceeding with the gathering activities after eating only manioc flour mixed with
water, while agricultural labour is always punctuated by meals and often by rests. A
tapper does not expect that manioc flour mixed with river water and consumed on a
broad leaf in the middle of the forest be his only daily nourishment on work days:
but the capacity to endure work in the forest with so-little food guarantees that the
~ork plan is c~mpleted, when there had been no game meat, whose availability
essentially is uncertain the day before. Beyond the capacity to work much while
eating little, another virtue of the tapper resides in his resistance to sleep. On the
eve of tapping days, he must sleep early, since he begins work well before dawn.
Stories about the "rubber-tree mother" ("mae de seringueira"), a mystical being who
inhabits the woods, are common .throughout the -forest. Single rubber tappers
courageous enough to face her on Fridays at midnight in the middle of a storm were
said to increase their production tremendously - provided they remain single.
Daily, weekly and annual production processes are the objects of specific
planning and calculations. These processes give origin to individual houses' plans,
involving not only the quantification of production goals but also the distribution of
Based on these weekly and annual plans, we may reformulate the annual
plan for houses under the basic theme of "two trails, one worker" as follows,
Once the choice of technique is established and assuming the components of
the labour process described above are given, planning for a day's work requires the
, - . at most- a pro
- duc tIV!.ty---0 f elght kilograms of rubber per day on an ideal trail.
allowmg
allocation of house tappers and of auxiliary on trails. The basic or ideal plan cans
for the occupation of 120 trees (one trail) by one worker. Actual plans can vary
around this pattern. Thus, the number of trees ranges from 60 to 180, divided among
one to three workers in cutting, maybe more for gathering. During the work week, a
house distributes tappers throughout a number of work days. As examples, I will
show three different weekly production plans, the first one with a single worker on a-
Monday
Trail A
Tuesday
Trail B
Wednesday
Thursday
Trail A
Friday
Trail B
Saturday
Tuesday
Wednesday
Thursday
Trail A
Friday
=
=
Saturday
2 x 28 = 56
448 kgs.
Monday
Trail A
Tuesday
Trail B
Wednesday
Trail C
Thursday
Trail A
Friday
Trail B
Saturday
Trail C
144 days
rubber trails _ a necessity that is self-imposed. Once under way, the rubber work day
should ideally follow a rigid progression. A trail with 120 trees would take about six
hours to tap (20 trees per hour). This means an average of three minutes per tree.
Trails
123
Plan
Trails
123
This time includes the time required to move from one tree to the next (about 100
metres distant, taking one minute) and two minutes to get the tins, go up a ladder
Activity
Period
Preparation"
Dry Season
Preparation
Rains
Apr.
Apr.-July
Oct.
Oct.-Jan.
Labour Input
Rubber output
Apr.-Jan
preparation
6
12
18
12
18
12
24
36
Tapping days
when necessary, make the inCision, insert the tin (repeating this operation according
to the tree's ~apacity). When arriving at a tree, there are no choices to be made: the
30
56
78
26
48
72
56
448
*Rounded figures
areas to be tapped already have been scraped in a number which depends on the
tree's circumference, and the tins already have been placed on a tree fork nearby.
Steps made of trunks already have been set, affording access to tbe several scraped
areas of the tre~. This activity should be initiated at night, sometime between one_
and six A.M., though most plans got under way between three and five. Tapping was
to be concluding during the early morning bours, before nine, after continuous
activity. The use of more labour would reduce the total time of each phase.
The first rest came after the tapping. This pause could take place in the
forest or at home, depending on the location of the "closing" of the trail where tbe
constraints external to the productive arena, including especially prices, but also
loop is completed. This might be as far as half hour from home. A good tapper was
access to natural resources and known technology. Given these constraints, house-
measured by his ability to continue working after eating only jacuba (manioc flour
with stream water) at the end of the trail, far from his house, before starting again to
coercio~
sea~onal
Gathering on a 12q-tree trail took around four bours, or thirty trees per hour, two_
coercion adopts a special form in the case of rubber production: the demand that
minutes per tree, one to walk from one tree to the next, the other to scale the steps
work routines be executed in a continuous and complete way (whatever the size of
(if they exist), remove the tin, pour the latex into a pail, clear the remaining latex
the routine). This distinguishes the style of the extractive rubber labour process
with a finger, and replace the tin on the tree fork (repeating this operation as many
from the style of the agricultural labour process, and from collecting and hunting,
times as necessary). By the end of a work day, the product of 120 trees would fill a
none of which possess the continuous, sustained and necessary character of the
sack with 20 to 30 litres of latex, carried on one's back.22 This output could vary
balls, the globa1 sequence 'of activities "iiiduding tapping, gathering and smoking had
from 16 litres in the summer time (dry season) to 24 litres in the winter (rainy
to be completed by the end of the day. Yet, why was it not possible to reduce
season), or about 20 litres per trail on the average. Children and wives might
arbitrarily the length of a given work day? Here we must return to the issue of self-
participate at this stage to accelerate the gathering process, particularly under threat
of rain - gathering, not only a faster activity, also was considered less intensive,
except for the carrying of the heavy sack filled with latex.
The rubber tapper is motivated by the final return of his labour, and does not
consider a work day with only partial results as "lost". In some instances, when the
At the end of the gathering stage, the tapper would return home. There
rains come unexpectedly, instead of gathering fmm only part of the trail, the tapper
followed another brief rest period during which he ate, and perhaps bathed; but he
leaves the latex in the forest to be gathered later as sernambi (an inferior grade of
might have to look for food on rapid incursions into the forest (fish or nambu
rubber); a delay in starting out may lead to the abandon of the entire day; a
partridges, or tinamous), or gather fuel in the forest for the next phase of activity. By
"crossed" trail (one in which gathering was interrupted in the middle due, for
mid or late-afternoon, the next stage began with the smoking of latex, occupying two
more hours. Around five in the afternoon, one who is on top of a cleared hill or
who take pride in their work. Norms of the labour process - pace, productivity,
travels on canoe will be able to pick out houses in the middle of the forest by the
One woman expressed the compulsory character of labour thus: "The tapper
is like a pupil in school: he can't make up for a lost day ... if he doesn't produce so
concluding the smoking process, when six, eight, ten, or twelve kilograms of rubber
much, he can't buy the merchandise." Another tapper explained that he had to work
were added onto a rubber ball or bundle that was being developed; or this quantity
four days a week in tapping (two days for each trail) in order to complete the "120
was transformed into a plank. Tappers keep a specific volume in mind as a goal for
days by law", where "law" should be understoo~ as "duty" (a norm of what a tapper
There is more than one explanation for the necessary pace and consistency of
The tension increased towards the final months of the harvest - October,
the labour process in rubber production. One reason is of a technical nature: rubber
November and December - when the final volume produced would be determined .
tappers explained that the trees were "accustomed to certain schedules", and yielded
before the final animal balance. Though houses remained independent in their use
far more during the night. Hence it was necessary to leave home early and at regular
of resources and in the organization of production, debts and prices forced them to
adopt work norms, which although flexible from house to house and subject to dayto-day uncertainties, held the force of an imperative.
22. The average yield per tree was 170 millilitres per day, but there was great
variation between trees. Along one stretch of fourteen trees that I cut, one tree
yielded as much as ,900 rnI. oflatex from a single incision, gathered in a cooking oil.
can; other trees yielded as a whole less than 100 rnI.
The volume of debts to settle was calibrated not by the upper physical limit
of potential labour, but by the production anticipated according to the house's
246
productive norm. Once a norm or estimate was established, it became an obligation,
an internal production goal, but its actual application was a process necessarily
247
access to goods, and thmr it is 'felated to the strategic interations between tappers
and patron-traders.
subject to fluctuations. I emphasize that plans like "four days a week" and "120 days
b 1"
"
personnel
t.
Uncertainty also arises with sickness, with the duration of flour since the last
They were house targets'
time it was prepared, and with the availability of meat, which depends on the
around which each house organized its members and its resources (taking under
success of the latest hunt. This means that what really is to be done remains a
consideration the demands of other activities, such as hunting and agriculture). But
decision to be made each day and each week. In order to set out early to tap it remains necessary to define more precisely within what interval of variations the
between three and six A.M. - the members of a tapper's house sleep little. Durhg
freedom to establish amounts of consumption and production plans to pay for them
was exercised. What follows is a discussion of this issue.
the rainy season, final decisions are made at the last possible minute. On occasions
when the day's tasks are decided, the head often discusses the matter with his wife
Uncertainty
and children also voice their opinions; in addition, decisions involving work on the
garden plots and in the processing of manioc flour, as well as those pertaining to
the unexpected and is accompanied by anxiety and worry. While a tapper progresses
hunting, illness, have to take into consideration the existence of a cooperation group
in his annual production of rubber, he at the same time is accumulating debts whose
in the neighbourhood.
For example, one house included a son who was a tapper, along with other
.....
younger children; in the neighbouring house lived a married brother, who also was a
tapper. On the eve of the first cutting day, the single tapper of the first house
consumption patterns fall out of step with one another. In April, debts are high and
became ill and asked his married brother to cut the trail. But the brother had
the house is still preparing its trails; in August the product is at a peak due to four
planned to work on his own garden plot. The sick tapper's mother proposed that .the
months of work, but now commodities become scarce and extraction is about to be
members of her. household "give an afternoon" of labour to the married son's house.
interrupted for two months, August and September, in ord~r to take care of garden
(for work on the garden plot) in exchange for "a day" on his brother's trail. This was
plots. In October, the second preparation of trails begins under the tensions of
how the work plan for the following day was traced, exemplifying the cooperative
accumulated debts and of the need for a final effort to meet the year's goal: say,
action of a macro-house.
"two thousand pounds of rubber", but the January rains bring about yet another
During the "winter" months (from October to January) the toughest decision
lay in the risk of losing a good part of the latex in case of rain. On the night before
~ercantile are~a of prices and time lags separating the production of rubber and
latex gathering, it is common for house members in their hammocks to discuss the
probability of rain. Some tappers faced the risk (liThe cabra [goat, generic term for
man] who is afraid does not make rubber"), while others proved more cautious-
'The same reduction in daily intensity could be reached with the participation
of a higher number of worker in the labour process. Thus, in a house with three
mature tappers, the following schedule yield the same total weekly yield as in the
that they are all-or-nothing propositions - at three in the morning the decision is
case above, now with a weekly work intensity of 1/3 for each worker (and not only a
taken to risk a day's work at tapping, and the crucial point is whether or not the
rains will fall before the conclusion of the gathering period, lasting from nine to
noon. The alternative of reducing the size of the work day in order to reduce the
risk was not considered. This apparently rigid logic leads us to a discussion of
Monday
(1/3}x3
Tuesday
Wednesday
Thursday
(1/3}x3
'Friday
Saturday
spreading work intensity over individuals (second case) gives each house a great
What was the standard domestic work week really like? We look at this issue
having in mind
range of possibilities.
Suppose, one tapper covers a single trail.-
Then the following basic possibilities may occur (cf. example in Tables 8.6 to 8.6e):
(41 %) were dedicated to rubber trail labour. In other words, houses worked two or
24
three days a week on the trails at the beginning of the rainy season.
Tuesday
Wednesday
Thursday
1 Trail
Between November 18 and December 30, I registered the use of 187 work
23
days (excluding sundays) scattered in 32 tapper houses. In 187 work days, 77
Friday
Saturday
Each house -had more than one tapper on the average. When multiplied by
the number of "facas" or "blades", 187 work days in fact constituted 288 man-days
Monday
1/2
Tuesday
1/2
Wednesday
Thursday
1/2
Friday
1/2
Saturday
(1.5 worker per house). Now considering the number of actual man-days used in
Monday
1/3
Tuesday
1/3
Wednesday
1/3
Thursday
1/3
Friday
1/3
Saturday
1/3
23. The total number of actual work days for th~ 32 houses in the period was
1,280. The sample of 187 corresponds to 15% of the umverse.
In all of these weekly plans, though the tapper might work on the trail during
24. Taking th~ total number of days available as 187 days and the t<;1tal
number of tapping days a 77 days, we can calculate a rate R of non-used rubber tIme
(and therefore of potential increase in absolute surplus-value) as
two, four or six days, the total number of trees tapped is the same. However, the
R
daily intensity of labour is variable, ranging from over one (a standard trail per day)
to one-third (40 trees per day).
rubber production during the period, we come up with only 94 (2.9 man-days per
five trails were rented (five rented trails amount to ten trails technically available
per week), or the trail potential; (Ill) the weekly work plan adopted as a norm by
WORK INTENSITY
production during a given day, the figure under parentheses indicates the portion of
Week days
Tap days
Work days
%
Mon
Tue
Wed
Thu
Fri
17
32
53
Sat
18
32
56
11
32
34
15
32
47
8
30
27
8
29
28
Total I
77
187
41
Total 11
94
288
33
The table above shows that when we consider not just the house days
a trail that is tapped by a house during that day. Hence, 4(0.8) in the Wednesday
column means that four tappers worked on that day, each cutting an average of 0.8
trails; in other words, four tappers cut a total of approximately three trails that day.
Each "trail" is taken as a standard trail with 120 trees. 26
employed in rubber (Total I column), but also the house man-days employed in
DOMESTIC -WORK PLAN: " 5 TRAILS, 4 TAPPERS
rubber (Total 11 column), the total house work does increase, although labour
intensity declines per worker (per cent figures drop from 41 to 33). This may be
illustrated by the following example. One house had five rubber tappers, including
the head and four brothers, all over seventeen. Their father had the most productive
house on the Riozinho Estate. Seeking to work less intensively and to enjoy a
I
11
111
IV
V
Mon
Tue
Wed
Thu
Fri
Sat
4
4(.41)
4 (:7)
2 ( 6)
0
4
4(.41)
4 ( .8)
0
0
4
4 ( .41)
1 ( .9)
0
0
4
4(.41)
4 ( .5)
3 ( .8)
2 ( 6)
4
4(.41)
4 ( .8)
0
2 ( .7)
4
4(.41)
0(0)
0
2 ( .8)
Total
24
10
11.7
3.5
3.7
greater participation in the returns, these four sons moved together to the. house of a
newly married brother.
In the table below, we show: (I) the maximum amount of work that could be
performed along the week by the four tappers (or 4x6=24 men-days) assuming there
were enough trails available (or 24 trail-days), or the labour potential; (11) the
maximum amount of work that could be performed along the week considering that
During the weeks under observation, the real labour undertaken (less than
four days-trail per week) amounted to about one third of the projected goal (nearly
than 12 days-trail per wee~); this goal, in turn, was half the labour potential (about
24 days-trail per week), being nearly equal to the trail potential (10 days per week).
Note that the idea behind the weekly work plan was that four brothers would work "
the trails, leaving the married brother free for agricultural tasks; Wednesdays would
25. Since the total available man-days amounted to 288 man-days and the
total used added up to 94 man-days, we may calculate the ration R' as
be set aside for hunting by a group of three brothers, while a fourth would work a
" " 26 The number outside parentheses is the quantity of labourers .worki~g
under a given intensity; the number under parentheses may b~ seen as the mte~l~
of the work day for each labourer. The product of the two flgures mea}ures d ot
the number of standard-trails tap!>ed in a day, as well as the numb er 0 stan ard
work days realized as a whole withm a day.
"
253
trail; Saturday remained free for hunting or leisure. The work plan divided the trails
(628 available trees, over five trails) in such a way that each tapper only had to tap
initial investment Of around 105 kgs. of rubber, with an annual amortisation ~md
replacement cost of 64 kilograms.
While averages may be the most important indicator for outside observers,
of rubber would represent the minimum a house with a single tapper working one
individual hous~s pay more mind to the size of variations. If, over time, houses may
trail would have to produce only to meet rent and equipment expenses in the
be distributed among groups including one which produces around 400 kilograms of
"rubber sector" of the. do.mestic economy. This t:>tal could be produced in 11 work
rubber per year, another producing around 800 kgs. and a third group around 1,200
days by one man producing 8 kilograms per day. Adding another 12 days for the
kgs., although the average is 600 kgs., the important point is that an individual house
preparation of the trail, it would take 23 days (in seven work weeks).
can adjust its output within a margin of 800 kilograms. In terms of working days on
These costs do not include the consumption of the house. As we have seen, a
the Riozinho Estate, variation ranged from 40 to 150 days per year per worker.
house consumes "basics", "luxuries" and "valuables". In a house with six persons, two
According to the patron, a tapper who was not lazy should produce 1,000 kilograms
adult and four children (each representing a half-consumer), the minimum annual
"basics" of salt, soap, kerosene, ammunition, medicine and sugar, cost around 364
compliment, and this annual goal involved around 120 days of tapping per year
kilograms of rubber at 1983 trade post prices, after converted to rubber values. Now
the list of necessary rubber expenses is complete, including rent and payment of
tools and consumer goods. From the following tables, we may assess the minimum
Let us consider the lower limits. If a house rented a single trail (the
minimum), it would have to turn over at least 30 kilograms of rubber (without the
production required from houses With only one trail, and the maximum production
of houses with only one trail (assuming one tapper only).
tare) per trail. If it rented two, it would have to pay 60 kgs., or 90 kgs. for three, and
so on. Therefore these were variable costs, which depended on the house's
production plan, and which grew with the projected amount of rubber production.
,.
I calculated the average equipment cost per house at 140.2 kgs. of rubber
per year (Table 9.2). Now, since we are interested in minimum levels, we may
consider the case of a tapper who worked a single trail (and not two trails, as in the
"average" tapper described before). With this distinction, a tapper would need an
Costs
Prepar.
In Rubber
Tap Days
Total Days
12
Rent
Equipment
Consumption
30
4
64
8
364
46
46
Total
458 kgs.
58 days
70
254
Potential Surplus
In Labour
In Rubber
56 days
56 - 58
-2 days
448 - 458
-10 kgs
We conclude that, at the current trade post prices, a tapper renting only one
Prepar.
In Rubber
Tap Days
Total Days
Equipment
Rent
24
Consumption
60
8
128
16
364
45
16
45
Total
552 kgs.
69 days
93
trail and with a house to support would have to expend all his potential rubber
labour just to pay rent, equipment and basic consumption costs, still coming up with
27
With each work week employing two days for work on rubber
a deficit.
production, all the rubber produced would be used to pay for the house's expenses
paid in rubber. Note, however, that there would be four work days left over for
activities outside the rubber sector in the house economy.
Available Capacity
In Labour
108 days
Potential Surplus
108 - 69
39 days
In Rubber
864 kgs. rubber
864 - 552
312 kgs
Assume now that a house rented two trails at a time. Four work days of the
A house with one tapper and two trails had a potential surplus of 312
house would be technically available for rubber production. Rent would rise to 60
kilograms. This house might, however, choose between (a) not occupying the total
kilograms per year; equipment costs would double as well to 128 kilograms a year.
But the minimum consumption expenses would not rise in the same proportion,
since we assume that the number of workers (one) is constant, and also that the
the ' rubber production (and thus the consumption of commodities) could be-
number of consumers (two adults and four children under 10) is constant. Thus, the
replacing the use of time to produce manioc-flour, game or the use of time as
leisure.
an~
inc~ease
Let us now consider a house with one tapper and three trails. Following the
logic outlined above, we arrive at the following table:
27. From what we have seen before, this would also be the case of those houses that rented only one trail and had, say, three tappers each working at one
third of the full potential intensity.
257
256
-trails. The column 'i n the middle shows the surplus, expressed in the work days or in
the product that each hou'se could achieve over the amount found in a house
Costs
Prepar.
In Rubber
Tap Days
Total Days
Rent
Equipment
191. 85
24
90
12
36
12
Consumption
Total
364
46
24
646 kgs.
82 days
46
118
In Labour
In Rubber
150 ,days
high may these prices rise and still allow for the reproduction of all houses? The
answer lies in the first line of the table: to the point where the maximum output of
the least productive houses meets their minimum expenses, that is, where it is only
sufficient to pay for rent and imported goods.
Potential Surplus
150 - 82
= 68 days
1200 - 646
= 554 kgs
In other words, now we may formulate with greater precision the idea that
the trade post, in a conjuncture of high rubber prices, calibrated advances as a
function of the productive potential of individual houses. For less productive houses,
.
this calibration had to include at least the equipment and basic subsistence goods MINIMUM AND MAXIMUM WORKING DAYS AND RUBBER PRODUCTION
Technique
1 Trail
2 Trails
3 Trails
Minimum
Days Kgs
58
69
82
458
552
646
Surplus
Days Kgs
-2
39
68
-10
312
554
Maximum
Days
Kgs
56
108
150
448
864
1200
and their prices, allowing for the free manipulation ' of the trade post, would be
equal to the maximum amount produced by these houses and ' a small margin in
rubber that would remain as debt.
It should be noted that we are assuming a fixed productivity of eight
kilograms per day for all houses, and a fixed consumption for houses with six
residents. But trail productivity could vary from settlement to settlement, as could
The column on the left indicates the minimum (in days and in kilograms of
rubber) necessary for each house to produce with the corresponding number of
trails . .The column on the right indicates the maximum capacity (in days and in
kilograms of rubber) that each house may attain with the corresponding number of
the number of consumers per house. Hence, it is perfectly possible that some houses
following the pattern of one tapper and one trail could have a greater total output
and a lower volume of basic consumption, while others would invert this formula.
The upshot is that houses with low productivity also would oscillate between net
debts and net credits at the end of the year, though the total distribution was biased '
258
towards the debt side. This debt will appear as an average debt, because even in
specific categories (such as one tapper with one trail) there remains an inevitable
internal variation.
economy with aCGess to the forest and a simple technology, the notion of
indispensable consumer goo~s becomes elastic. 29 In other words, cutting back on
/
costs with essential goods (which were paid for in rubber produced) did not mean
One consequence of setting prices for necessary consumption by the measure
of the least productive houses is that houses with greater productivity _ those which
adopt strategies of tapping two or three trails - would benefit from surpluses in
commodities above and beyond the minimum reproduction cost. They earned a
surplus over
surplus merchandise in order to expand the gross production of the rubber estate
processing and storage of foodstuffs and for transport: in this category, one finds
and thus the gross profits -- although at the cost of reducing his average rate of
profits.
dogs, shotguns, the 3.5 HP gasoline engines used to power manioc graters in the
flour houses; 9 to 16 HP engines to power canoes made from tree trunks; shotguns;
As a result, some houses consumed coffee and sugar, and had wrist-watches
motors and record players - just as other houses bore the appearance of the sheer
poverty in imported consumption/equipment goods. This distinction was measUJ ,i
by the difference described above: tappers on the upper bound of domestic potent!::il
acquired as advances not only basic goods but also "lUXUries", "vices" and
'valuables".28
nylon for
fishin~
within each basic strategy was related both to the domestic consumption of
imported goods as well as to the productive inputs used in the domestic economy as
a whole. 30 This range of substitutions only was possible because relevant technical
decisions took place at the house level. This margin of technical freedom thus acted
as a barrier to the power of the commercial monopoly. Monopoly increased gross
Indeed, setting aside the simplified model of one worker per house and
(the case of
houses headed by widows), by saving on fixed equipment costs (producing homemade tins from recycled metal) and by cutting back on basic consumption. In an
28. Chagas possessed cattle, swine, an engine and a canoe' he run sometimes
a small local trade, ~n~ accum~lated m~ney in a savings account.' Juvenal, his client,
was reduced to a nnrumum, WIth all hIS modest material objects displayed on the
floor, or on the columns of his house without walls.
:'.
, ,
I1
,1I
, , \ 4'
, ...l ().
"'"
I \ I
s.~' I
. "
."
"
:
._
'0
, ...;
'
, I '
",
'\
,,'
'
, '\'
,\
.'
'.'
".,
',
... ..
.. ..
'(
.:. ;
,I
".'
~,
I'
:.
..
'
I"
"
""
,~ I
I
"
"
1'1II /', ,
. I,
'I
' ..
I'
(,'
I ,i
,,'
.' '."
;"
~.'
_ , _..
,.. ...'
;,
.......
"
'
'
~,'"',\/' /
J
. '
"-
"
"
,:.. '
Cfu~Jo"
~ ~" ' .
It
, ..... -
\.;.
'
"
"
... . . .
", I:.....--'
,'.
'"
',
'
-, -
:v:~,tciq,a.::'
.
' i ...
..
.. ..". "
.,; , f " vC..)'o.:
,; ','
I
--
'~'\
',.,.
:' .r.
- -- 'I" I~ ......." :..,1'"''
.
I
.". ""
" " , .'
., : h" \
,'"
, .'
,':'
~",
"
.J'' "".
"
-' I'~"~""
,
' .. ,.
/~'--"
" :i'>/I"
{ /
,,,
,..., .
, ,1
,.,' I
'I'
:
,.,' I'
I : . ;,::.' \ :':,
'
'I
'
I , , I'
'. :,
.. '
I ' "&I
L
.
I
"~
"
.... ,
'
'
I'
11
'
'. '
... 0.:
"
),
'J"'0" ',' ,
i I
I I
\ I , , 'I
I
I I'
I t ' ,", ,\ , "
'
~".
..
,I
L!' ~t:;;~~~~;-;;~::':~J1ili!j
"
I , I, - ,
1 , I ' ,'
fl8 ue
,. '-q'
' ,
"
I
'\
.' _
I,
'\,1
\.\........
___
,
, 'fii!11" I, ,
,
"
'"
't(..Vf(Jrr_(.,I--.,
'
',.."1' ii (.tV,
,\I \
I,
~.-='~.~
"
1, I
'\ I
,,:'
'\, .
"
I ,/11" , 'I
,' ,I ,
t"\' ~
",, :.:,'.',/
,
",
I
\ \'
,"
,
1 ,
"
I,
."
I,
f,
'.,
"
"
, . I, ,
, , , , . ',."
.
'
;.
, ./
I ,"
,
'
~,
.-
' , ,
I,
-\
'
;
,
."
l ," - , "
t.
r" ,
',' " )
,I
1 "I " 1 , ', -' , ' "
,',
,-f~'T;",'~~
", \
\
")
' ,\
;'
n
:
~ ~
1;'
\\ v
,V/ \\
7
261
Chapter 10
HOUSE ACTION PLANS
Introduction
The goal-of this chapter is to analyse house activities combinations of rubber- making, manioc cultivation and hunting. These activities, which together make up
what we now call house action plans, constitute a simplification of the full array of
houses' activities, which include both production and consumption according to
gender and age, both hunting and fishing, and both cattle-raising and handicrafts. 1
The analysis focuses on the freedom enjoyed by houses in establishing action plans,
given the constraints imposed by the known technology, by the particular stage of a
house's life-cycle, by the available natural resources and by the level of prices and
forest-rent. In other words, I describe the place of houses within a range of action
possibilities.
This approacp builds upon a previous chapter, where the freedom available
for different levels of produ ction (and corresponding levels of surplus) in the _
rubber-sector of the house economy (given the number of tappers, the number of
rented trails and the level of prices) was described. We can say that this freedom in
the production of rubber by houses constitutes a "space" of one single dimension,
and a step is now taken towards generalising this approach to a set of three house
activities, which can be thought of as forming a space of three dimensions. This is
not done with the intention of realism, but to suggest how a metaphor can convey a
sense of variation under constraints. I propose moreover that the amount of
262
.h f
ods land or labour
,
best in an autarkical economy with no markets elt er or go
A peasant house develops action plans around the labour time of its
producing/ consumption units facing a market for goods, takes into the account class
CBhaduri .983; cf. Ellis 1988:120; Sen 1984:37-72). A natural hypothesis on this last
both the use of time (thUS, fishing and hunting may be valued and compared as such,
situation is that, where peasant units are forced to the payment of land rents and are
and not only as means to an end) and of objects of consumption (not only produced
part of advance systems, suchas it is reportedly the case in India, the peasant houses
goods, but also the forest environment itself), here the description of action plans
are exploited to a maximum level, behaving thus to all purposes as if they were a
proletariat (Bhaduri 1983). Thus, the mix of class structure and of domestic
landowners' and traders' rents and profits, and minimisation of houses' income.
However, I have argued that this kind of maximisation does not hold necessarily in
the extractive forest economy, where the "interlocking of factor and output markets"
(the same patron "owns" the trails and buys the rubber) does not lead to a situation.
where a patron controls "resource decisions including the number, size, and crop
share of each tenancy" and" permits the extraction of every last ounce of surplus
The second approach captures not only a choice .between effort and
product above the bare survival needs of the tenant family" (Ellis 1988:157; cf.
Badhuri 1983). The reason is, first, that forest houses oppose resistance. against
consumption), but also a choice among different kinds of labour combinations and
arbitrary rents and to monopoly, benefiting from the scarcity of labour and from the
dispersive nature of the extractive technology. Also, the failure of patrons to enforce-
I now go into the details of the Chayanov approach, so as to make clear what
a maximum level of exploitation may be both cause and effect of the existence of
the basic ideas are. Each house has a numb~r a of workers (which is considered as
some degree of freedom for the action plans of houses. My own approach is
th~refore
ex-post), which represents how much each domestic labourer works (thUS, in a
rubber tapper's house, a is the number of males over 10 years of age, and t is the
amount of days per year spent by each tapper on standard trails); L is here a
number which measures the total labour put to use by the house (L
House Economies
= at).
The
266
product Q, divided equally among the consuming members b gives the (average) per
are working at maximum intensity t * (e.g., in the houses of tappers all males over 10
each tap three standard trails of 120 threes each, or six days a week). On the other
houses might first set the necessary level of per capita consumption w; then, since
the number b of consumers in the house is taken as given (the persons of the house,
consumption w* (say, in a tapper's house, all persons have the minimum of manioc
flour, game and clothes), and Q* requires a minimum labour of L*. Thus, *the
immediately production plans. The idea behind these Chayanov-type models is that
difference L * _
survives and there is no question of alternative plans; if it is positive, the house has a
87), but it cannot choose freely both t and w. There is then a single degree of
surplus of its own. 3 I suggest that in this last case the house plan remains
freedom: when only one variable is fixed, it determines the other one.
While in Chayanov's theory houses choose an "optimum" with a degree of
undetermined.
This suggestion is better expressed when we generalise the Chayanov model,
freedom, the above exposition was intended to suggest two ways of eliminating the
in the first, it is assumed that peasants minimise labour intensity per worker, and in
final products.5 Now the feasibility condition for houses is that the maximum house
the second, peasants are seen as minimising consumption. The first angle
labour (thUS L *, when all a workers are employed at the intensity t *, summing up all
Q (where Q
L* 'is negative, the house is not viable; if it is zero, the house barely
My own research did not support the hypothesis of a single peasant model.
Thus, I explore further Chayanov's idea that there are choices to be made within
certain bounds (Chayanov 1986:53). Given the number of house labourers a, the
upper limit L * for house labour is the total labour expended when all a members
2. Thus, "Chayanov's rule" (Sahlins 1974: 102, lorion 1984:74) sets the labour
intensity t as a function of the consumption w of family members. But Chayanov
himself assumes only that well-being is a function of labour utility U(w) and effort
disutility Vet). Since w itself depends on t, Chayanov's well-being function has only
one independent variable which is t on which both U and V depend.
.
nce t of su Ius is due to Gudeman (1978), following Sra~fa's
. IS oc o .caPrdo No~ that here both the house's structure (a,b) and pnces
.
3 Th
mterpre t atIOn f R1
.
affect the existence. of a surplus.
4 It will be noted that this is essentially a Leontief mO<;lefl of dPrlodductiOnt
.
. T
. g' models Leontle mo e sono
possibilities. In cont.rast. Wlt~ mearyr.ogrda(mrrun Buchl~r et alii 1986; lohnson &
assume that somethmg IS bemg maxlllllse as, e.g.,
Behrens 1982; but cf. Johnson 1980, 1983).
5 In a formal model we would write: L = (Ll,. L2 L3) This formal notation
will not be used here. Symbols are used only to make Ideas more clear.
269
268
each W*, where Q* is a vector). However, if there is a slack, then house plans may
(women's labour, men's labour and children's labour may not be interchangeable,
and thus may not be added together).7 Instead of choosing w, a single number
expressing, say, the amount of manioc per capita, houses may now choose a
commodities vector (game of different types and tastes; various manioc products,
such as cakes and beer; diverse wines made from forest fruits) which again may not
preferences; in the second formulation, once a rate is established, there is also the
be quantitatively comparable as greater and smaller. The spaces of labour plans and
issue of distributing the work among several activities, some of which may be
of consumption plans have each several dimensions. It should be noted that in this
model, the forest house produces its own work instruments, constituting a "sector"
not mentioned above (stone hatchets, bows and arrows).8
We now turn to a situation in which an economy of the sort described above
participates in a market for goods (they may purchase production goods such as axes
Employing the terms introduced above, let us now consider different cases.
The most basic .mo.del (of the Chayanov type) has a single product (manioc, let us .
say), in terms of which we also expresses consumption (manioc per capita w). In this
simple case, which we may call a "single-product, autarkic forest economy", a house
chooses its level of labour use (t per capita) and consumption (w per capita) within a
range of possibilities that is constrained only by technical factors, by nature, and by
the availability of persons. In the more general formulation, the house economy
produces a total "crop" Q with the house labour L, where Q is now a vector of
garden produce, manioc cakes and meat, and L is a vector of gardening, cake-
and guns, consumer goods such as glass beads and brandy; and they may sell rubber
or sarsaparilla or slaves). Now there are two effects. When there is not a direct
political domination of the forest economy by colonisers, we may follow Wallerstein
and call this situatio~ an "external arena" model. Then, first, part of the time may
now be employed to acquire imported goods; second, the introduction of production
goods (iron tools and guns) presumably increases earlier productivity, with the'
possible result' that now less time is needed to produce manioc itself, or to hunt. It is
even possible that ritual life increases (more "leisure"). Thus, in principle gross
product may rise while work intensity is reduced or remains constant (or is reduced
making and hunting, distributed among men and women. Now house plans lie in
points.. which are qualitatively different. Instead of a single figure t expressing a
homogeneous labour, houses may choose a vector of heterogeneous labour
("hunting", "agriculture", "food processing") which do not add up to a number
for men and increased for women): the movement of t depends on the rate of prices
.
.
rt Under an absolute
kind) is a figure expressing the tapper's capacIty to Impo .
between the exported product and the imported iterns. 9 This model also applies to
the caboclo economy of detribalised Indians in the Amazon. When the indigenous
.
.
(.It may purch ase equipment and minimum
VIable
"house" is still econoffilcal1y
(or caboclo) economies are subject to direct political control of colonisers, they are
consumption goods), but it does not have any choices to make (the only alternative
.
.
'ty) .11
possible is to work at a maximum
mtensI
(or beyond) and reduce w to a minimum (or even less), by ignoring the "subsistence"
Even during the boom this case may not have been general, for two main
needs (reproduction included). This is the reported effect of economic booms on the
reasons. First, i.:inerant traders competed with the patrons for the rubber, workers
were scarce and fled with frequency (Chapters 1 and 2). The degree of the
The rubber tappers' forest house during the boom may be seen as convergent
monopoly was not maximum, nor monopoly was homogeneous in all areas. Second,
to the above model, although having a non-indigenous origin. They are forest
producers within the same rubber estate; it might increase with distance while
rent. In one case (which we might call the single-product, dependent forest house
supervision costs also increased (the case of upper Tejo river), and local
model), a house produces a single export product (rubber), which is sold in order to
administration might not be capable of minimising the price ration down to the _
pay for all consumer and production goods (flour, dried beef and iron tools). This
presumably describes the boom years (1870-1912), where the "house" is reduced to a
optimum.
In the
(Chapter 2), the rubber tappers' house still purchases a bundle of commodities (as
variable (his own labour intensity t). The rubber tapper sought only to survive (by
debt). The house pays a rent (in kind and fixed) for the use of a number of rubber
costs). The relation between the price of rubber (paid to the rubber tapper) and the
trails chosen by the house itself. The house has now a variable number of workers.
Finally, the house runs three sectors - hunting, flour and rubber production. The
rubber sector generates a final product that is converted (through the trade post)
/'
9. See Gross (1969), Sweet (1974) and Farage (1991) for the early trade
across the frontier of "gift economy" and "commodity economy" in Amazonia; cf.
Hugh-Jones .1989 for the contemporary scene. Thus, several "exchange regimes" may
coexist within different "action regimes" (e.g. houses, firms). See Taussig (1980) and
Chevalier (1982) for an analysis of the contact between different regimes.
into imported commodities, which include both basic consumer goods ("estiva",
10. Both in the slave and wage-labour cases, one finds what Marx called the
"formal subsuming of labour under capital". We may further observe that what Marx
considered the "real subsuming of labour by capital" represents technical and
organisational changes: e.g., the substitution of house extraction of commodities by
plantations (cf. Marx 1987; 1976(1933)).
including hunting, fishing and rubber-tapping equipment, along with salt, soap and
273
272
kerosene) and non-basic goods ("valuables"). This model introduces two new
features. One is the variable number of house members, which leads to a variable
consumption.
It occurs that ther~ are three notional plans which together delimited the set
reduce labour intensity per worker t, while increasing total house labour Land
of possible house actions 12. One extreme plan concentrates on rubber production,
renting many trails, purchasing equipment, and focusing on the upper limits of
to the number Of mouths; or a house may increase labour intensity t and thus total .
technical potential. This production plan affords a consumption plan including both
house labour L merely to keep constant per capita consumption w, if the number of
basics and non-basic articles ("luxuries" such as coffee, cooking oil, perfume and
mouths is increasing in relation to the number of able hands. Thus, the trader may
shoes, or "valuables" such as an engine and others). Those houses which managed
observe an increasing L (and the resulting output Q) even though he is not able to
also to buy outside of the trade post could, under 1982 prices, accumulate
control t or w. The second new feature is that house labour (and, correspondingly,
house product) no longer is a number measuring the work days spent in rubber
production per year (or kgs. of rubber), but a vector of rubber, agricultural and
(including here the increased cost of inputs to flour making such as gasoline). This
sufficient to supply an expanded house with many dogs and backyard animals, and to
indeterminateness in the domestic economy. Suppose that .0-e price rate for a given
estate (for example, Riozinho) is set. Suppose that this estate operates with
third extremal plan consisted of minimizing both rubber and manioc production,
monopoly prices set in such a way as to maintain the total of houses as viable houses
using the surplus time for extensive hunting, fishing and collecting activities. Houses
(leaving aside the fact that the degree of monopoly is not maximum; see Chapters 5
actually identified with these plans were respectively houses of "rubber tappers", of
and 6). Then, after paying the rent for trails, those houses with a single labourer and
"ro<;;a people", and "lazy people". Most houses lay somewhere in the middle. There
a maximum number of dependents (otha wise, the hypothesis that all houses are
could be paths leading from one extreme to another, reflecting age changes (since
the ideal age for rubber collection was from 16 to 40, while a crop-tender's age
package with the maximum of a single worker (that is, all the time available during
~ould go well b~yo~d 40), and expressed in spatial movements from hinterland areas
the year is dedicated to acquire only the "estiva"). However, those houses with more
rich in latex, to good hunting zones, or to river banks appropriate for annual
than one worker and with fewer dependents could now use less work per labourer
planting activities.
and greater consumption for individual members of the house (Chapters 7 and 9), or
could diversify the use of the forest economy sectors, using the surplus time could
12. We do not assume that houses actually s~t out to fo~low a plan. Plans may
be the result of a sequence of actio~ which constram later actIOns (cf. Barlett 1980),
or may be accidental to a degree accldental (see later).
275
274
used in the "subsistence" sector (Livingstone & Ord 1981:115-17). How about
Chayanov's model? We st~ted above that it deals with production and with
distribute themselves in the space of production possibilities in the way they do. Can
we use the notion of maximising as an explanatory device? This notion (being the
basic tool of what we might call mechanical models) seeks to predict or deduce what
maximised 14, this model falls again within the class of maximisation models. 15 The
a house will do, perhaps introducing later ad hoc hypotheses in order to explain why
why individual houses fluctuate far away from the average. Such models try to mimic
the successful methods by means of which physical science obtains laws or principles
by assuming that some quantity is maximised or minimised (thus, e.g., optical laws
are deduced from the assumption that nature minirpises, so to speak, the time taken
by a light ray in its course across air and water).
A class model which ignores class struggle can be included under mechanical
models, or maximising models, as we have seen above. Thus, a forest house
specialised in a "single product, confronting a patron which combines extraction of
rents and advance of credit under monopoly clauses, is said to be forced to a unique
state which maximises physical the level of labour and minimises the consumption,
to their presumed physiological limits.13 Liberal maximisation models, in contrast
to marxist maximisation models assume that houses are trying to maximise profits,
14. The house is approached as a firm in which the manager-owner takes into
account the opportunity cost of his own labour)
15. "The peasants are guided in their allocational, effo~ts by the ai~ of
maximizing the happiness of the fa~ly"(~en 1984:38): Sen s artIcle wa~ publIshed
prior to the diffusion of Chayanov's Ideas III the West (it responds to Lew~s 1954 ~n~
Jorgenson 1961), but rediscovers and p~rfe~t~ Chayanov's m?~el: C~. wIth S~hlIlls
"Chayanov's rule" which is a greatly SImplIfIed (labour) rmmrmsatIOn verSIOn of
Chayanov's concept of utility.
16. The neo-classical Chayanovia~ h~~se maximises utility U -. V where .U is a
utility based on consumption and Va dISUtIhty based on labour. U IS ~ ~nctIOn of
per capita consumption w, which is equal to Q (the house's product) dIVIded by the
number of consumers b. U(w) = U(Q/b). On the other h,and, V depends upon
intensity t alone, V(t). The house head "assesses" this functIOn, and comes to the_
following progni.mrtle:
although allowing for the fact that they not be able to accomplish this goal due to
uncertainty (they protect themselves from price fluctuations in the "cash crop"
subject to:
sector),_or due to the impossibility of freely allocating labour or land, since male
labour may only be used in the "cash crop" sector, while female labour may only be
13. As it will be seen in more detail later on, this effect does not hold if
houses have different demographic compositions, In Chapter 3, we suggest that in
the "case of the "forest economy, the "real subsuming of labour under capItal" (e.g. substitution of forests by plantations) is a " previous condition fo r "formal
subordination" (wage labour) along these lines.
17. Observe that by choosing t the house produces Q with the labour L = tb.
This assumes a production function (and thus, a technology). Houses do not ma~e
technical choices (but if a tapper has a pact wIth the mother of rubber tre~s Q wI~1
presumably rise). Observe also that a (consumc:rs) and b (producers) are gIven: thIS
house has no foster-children, and does no~ hIre people. (Cf. Chapters 7 and 9).
Finally this model does not incorporate pnces. All of these features, on the other
hand, ~ould be added to the maximising theory of the house.
being taken as factors whi~h distort rational choice, are together summed up as a
have ordered all possible choices by increasing happiness, only in order to be able to
"deduce" that what they observe (actual plans, either stated or actualised in acts)
should observe that it alone does not provides an explanation (a kind of cultural
In simpler language, houses do what they like best. The proposition of treating a
house as analogous to a firm, thus, faces . a basic difficulty in that there lies a
house should be (a form of the house as part of culture), there is the actualised form
as it is seen in the real life of houses. Individual variability may be due both due to
strategic action toward specific goals (for instance to acquire a new motor ou
increase a house's role in the union), and to chance, environmental and other
~dopt a culturai ~odel for the house plans? Along this direction, one assumes a
sources of unpredictable change. These difficulties constitute the reason why I have
previously given image for the form of a house, extensive to a group of persons
used the notion of a form of the house always as part of a labour process. A "house
which share a cultural heritage. Notions such as "moral economy", but also the ideas
labour process, which is imprinted through the human action onto nature and things.
maximising peasants" (as a particular case), just as notions relating to the nature of
But a "house ior~" is also the actual result of the labour process. The form
forest (which may not be converted freely among uses or owners) and to the quality
anticipated by the house group should not be mistaken by the final product which
has form given to it by the labour process. If this identification were to be made,
then house forms would perpetuate themselves as an unchanging essence. Labour,
18. "... it requires something of an act of faith to assert that this [to construct
'curves' on the basis of a sample of individual, different houses] 'proves' that the
peasant farmers in the sample are efficie~t. .This ~ct of fai.th is that all farmers in the
sample are considered tc? have been. stnvmg, WIth varymg de~rees of success, t,o
reach [the optimal] pomt A" (BIllS 19~8:71~ .. C~mpare WIth Durrenberger. s
approach' (1984: 39-49) towards "operatlOnahzmg Chayanov. Cf. also OrtIz
(1973:271): "It is a" tautological system, but a very useful tautology"; and 10hnson
1980.
19. "Now it is clear that wealth-getting (krematistike) is not the same art as
household management (oikonomike), for the function of the former is to provide
and that of the latter to use ... "(Arlstotle, Politics, I, iii, 2). This distinction is
paralleled by Max Weber's contrast between Erwerbswirtschaft, acquisitive
management or "profit-making", and Haushalt, "management of a household"
(Weber 1947:187,191). Cf. Gudeman and Rivera 1990.
278
previous models may be affected by the way they succeed or fail in orienting action
towards an acceptable life.
279
growth in the number of labourers. This occurs because neither rubber trails nor
cultivable lands .are scarce, and therefore income does not decrease with an increase in the scale of production. But both in the case of rubber and of manioc, output per
cultivator decreases with the increase in the number of workers within a house.
Manioc and rubber production plans reveal a tendency of sparing labour per worker
as the number of labourers increases: in economic jargon, both rubber and manioc
invest heavily
labour are "inferior goods", since their use is decreased when the house can afford to
depriving themselves of game meat. Others invest heavily in hunting, in seeking land
turtles, in fishing and in forest collecting activities, while reducing their garden plots
III
cultivator/tapper must support, on the one hand, and the increase in effort by each -
output (and correspondingly low imports) and a minimum flour output (thUS unable
to attract visitors for festivities, for instance) may be viewed as a house with a high
rule" and various empirical studies. But the disperse pattern of houses along the
degree of time consumed directly by hunting, fishing or leisure. If the men of such a
hypothetical line correlating consumption per worker and labour per worker is
house value hunting over cultivating activities, the consumption of time in hunting
great. The appropriate conclusion is that there are lower and upper limits to the
(even with low productivity yields of meat) is not simply a question of "effort", but it-
is part of the house's well-being. Thus it is only possible to assert that of a given
each of these workers must support. There were houses where a tapper and his wife
stock of time in a house (also taking into account its consumption needs), there are
produced as much as 1,800 kgs of rubber in a single year; and other with a tapper,
minimums and maximums, between which the time of individuals can be distributed
his wife and several children who produced but 600 kgs of rubber (Diagram 10.4,
10.5).21
and foraging for local use. hence, for a given number of labourers, two houses may _
Each house is represented by a point within the space of possible production plans.
be employing all of them at a maximum intensity, while distributing the total output
Within this space, each dimension represents the intensity with which a house
employs its labour in one of the basic activities: rubber extraction, flour production,
manioc flour, units of deer hunted or fish caught. In this case, the hypothesis holding
or cynergetic fauna predation. For simplicity's sake, only two activities are
that labour intensity will increase with consumer demand does not lead to a
predictable outcome because there are several options for consumption as one
At one extreme (in the lower left-hand corner), we find houses that adopt a
draws away from the hypothetical point of minimum consumption. Some houses
minimum strategy. Their labour produces around 300 kgs of rubber to be turned
over to pay bills; they also produce a minimum amount of flour (3,000 covas).
The house with the greatest ratio of consumers per labourers in the
diagram included several persons ~ver 10, but by some demographic accident all
these people were women who dId not tap rubber. Hence, even with a high
consumer/worker ratio (seven to one), this house had only one rubber tapper,
whose annual output stood around 500 to 600 kgs.
21.
Garden plot sizes correspond to minimum subsistence of people. The rubber is that
which is indispensable (after trail-rent payments) for the acquisition of agricultural
implements and basic household items (pots, clothing, hammocks, soap, salt and
283
282
fuel), of tools for rubber extraction itself, as well as of guns and ammunition for
garden plots and hunting teams (Chapter 9). It is also possible to allocate one
hunting, at 1982/83 prices (cf. Chapter 9). At another extreme (in the upper corner),
labourer for the crops alone, and use teams to increase hunting efficiency. The
we find houses that may be producing much flour (8,000 covas), while also
producing around 700 kgs. of rubber (when the size of garden plots is maximised,
which act together at peak periods such as the preparation of trails or garden plots,
the minimum amount of labour in rubber also increases, in order to pay for more
and in the deer hunt; it also may be increased through adoption, or through the
fuel to process more flour). At a third extreme (the lower right-hand corner), we
find houses that employ "surplus" labour to increase rubber production (1,700 kgs.)
efficient technical procedures. The press technique both saves labour and allows for
and to acquire additional consumer goods, cattle or a second engine, while reducing
more free time either to produce more rubber, more game or more crops (thUS it
distribution of houses within this theoretical space, we also see that one group of
houses appears only on the horizontal rubber axis, since they have no garden plots
there is a stock ~f ~t least 300 hectares of forest surrounding a house. Territorial use
during the year under scrutiny. A second observation is that the houses tend to
balanced levels of predation and extraction: For example, this is the case when
flour production as a function of the increase in labour. We may also point out a "
patoa and bacaba palms are not stricken down during the collection of fruits, or
greater density of houses within the triangle formed by the ranges of 400 to 800 kgs.
when specialised "Paulista" hounds are not used in hunting. What one house
A house with a single worker covering a single trail (thus with a potential for
two work days a week in the rubber sector) produced the lowest amount of rubber
polyhedrons - something like a pyramid. At one of the vertices, we would find a zero
feasible; it would dedicate the remainder of its time to maintaining a garden plot
(tended twice a week) and to hunting and collecting (once a week). In the patron's
manioc and game to sustain not only house consumption, but also the basic needs of
estimation, these were the "lazy" houses. They were chronic debtors. It should be
the other productive sectors, both direct and indirect. It is at this vertex that we find
noted~
however, that houses producing between 500 and 1,000 kgs of rubber enjoyed
the possibility of a "maximum use of time for leisure. At the other three vertices of
the highest rate in deer hunting (over 14 per year), while the most productive houses
the pyramid we would find production plans where leisure is reduced to a minimum:
(over one ton) on the average captured fewer than six deer per year. When houses
begin to add on- more labour, their field of options grows correspondingly. Now it is "
possible to occupy three rubber trails and maintain the ideal routine of six tapping
days a week, though rarely observed in practice, and at the same time maintain large
variable in the model. This is the same as postulating the existence of a degree of
freedom in the agents' behaviour.
That which
.
.
(from e.g. mlm
. .mum and maximum use
houses and given a range of possIble
actIOns
,
of labour)
all possible actions are equally prob abl e. 23 We might treat this
.
h t eally determines behaviour
as to war
"
hypothesis as an expression of our Ignorance
..
b u t we Ignore ItS
within the limits already known to us (there is determImsm,
. of the unpredictable course of actions, as in the
mechanism), or as an. expreSSIOn
chaos metaphor (small changes lead to great effects). In the last alternative, we may
. an d unpre d'IC table processes . Thus, it remains
reason that there are cumulatlve
impossible to predict the evolution of any given house over t h e Iong run, 24 although
statistical averages for the whole set of houses may be amenable to prediction.
Th~s
try to maximise profits or rather total produce, but may not quite suceed), and
.
. a homogeneous
community, it
average of sixteen p~rsons apIece.
Instea d 0 f formIng
.
leaves the cultural form of houses open to a degree of variation. If one house has a
makes more sense that the individual houses fluctuate along different points; that
22. This is a Leontieff model developed during field research in 1983. Similar
studies have been published (Buchler et aI, lohnson 1982 e outros). These are
constrained maximisation models: one may direct the same objections raised against
simple maximisation models. The approach suggested here, by contrast, seeks only
to Identify ranges within which individual choices can be made (when prices,
technology, resources and persons are given), and not to predict individual choices
on the basis of simple or constrained maximisation hypotheses.
., (
h G k ergon, translated as 'work')
23. The 'ergodic hypotheSIS from t ~ibl:e;;ith the constraints to which it is
states
a SJ:stem
all possible states (cf. Kemeny et alii
sub
'ectthat
Thus
m the may
long adoP!
run It all
maystatesth~::tgh
go
1959, chap. 6:2; Shannon 1964:45). See next footnote.
24. Teodor Sbanin's an~lysis o~ social mobility .among hea~ant houses (1972) _
may be considered from the pomt of VIew of the ergodlc hypot eSlS.
. tb t they are non-linear. Nono! ~omplex
ph enom~~~/sma~e
prediction intractable in
A characte~istic
mteract ID
ways
linearity25.means
that c~uses
individual cases (cf. Stem 1989).
Political economy
We go back again to the role of the prices and rents paid by houses to a
.,
b h
as monopolists of their own'
they organise themselves m umons, houses may e ave
ining power which derives from the possibility of
produce. They possess a barga
.h
h
h the parallel market or by
withdrawing frqm ~he patron's game - elt er t roug
.
d . th
1 of rubber and increasing
- "fleeing to the forest", that IS to say, by re uc10g e sa e
the production of other products. When there are two monopolies we speak of a
the terms of trade are not determined and there is no
patron. Suppose there is a drop in rubber prices (or an increase of rents). In this _
duopoly. In a duopoly,
new situation, the range of plans that may be adopted shrinks. Hence, the effect of
equilibrium even in theory - bargaining and conflict substitute the elegant points of
1
.,
d' equilibrium which
equilibrium. The typical situation most likely wil rema10 m a IS
..
h' h
th t cunning cheating and
is both permanent and of pnnclples, w IC means a
,
limited
1 r'
violence will enjoy a free course. This also suggests that patrona po lCles are
f
.
the part of the houses to fixing parameters - restricting the space for ree actlOn on
national price policies. In the first case, a maximum level of exploitation (and
wage-labour system.
This conclusion, which is valid for other peasant situations, also is consistent _
with a marxist
discriminatory monopoly (prices adjusted house by house), thus maximising both the
Marx did not suppose that capitalism would tend towards a maximum degree of
total product while _at the same time maximising profits. Not only discriminatory, the
monopoly would have to be complete. We may now argue that as houses resist
against monopoly they in fact retain a high degree of information in the forest
economy, one implication being that they experiment with and discover species,
techniques and uses and thus are able to evolve their technical knowledge. This
knowl~dge
27. For the sake of simplicity, we do not pursue the distinction between a
monopoly (a single seller of merchandise) and a monopsony (a single buyer of
rubber).
J:
-288
exploitation determined by some natural law. In his own English prose, and with his
289
Chapter 11
own emphases,
CONCLUSION
... although we can fix the minimum of wages, we cannot fix their maximum.
We can only say that, the limits of th~ working day being given, the maximum
of profit corresponds to the physical minimum of wages; and that wages
illustrate a generic model for how the capitalist world system acts upon a
working day as is compatible with the physical forces of the labourer. The
periphery. 1 In this model, the world system consists of a centre with a free labour
market and a periphery with forced labour, as in the case of debt slavery;2 a centre
and physical maximum of the working day. It is evident that between the two
periphery which loses resources and labour (Bunker 1985; Hornborg 1992; Alier &
possible. The fixation of its actual degree is only settled by the continuous
author's emphasis).
Conclusion
(1870-1912), the rubber estates provide an example of how a periphery is formed,
This chapter had two objectives in mind. The first was to exhibit the range of
variations within the micro-economy of the forest houses. In houses, the presence of
the forest and of several technical procedures play a crucial role. They possess a
thriving technology precisely as a consequence of the range of their technical niche,
though leaving several questions that must be addressed through a more detailed
analysis. Why have the rubber estates remained active during the eight decades
ensuing the crisis? Why did the world system not 'transform this periphery? What
local history remains hidden behind the simplified negative concept of
2. Rosa Luxemburg employed the rubber boom example and that of 'the
"primitive systems of exploitation" used on the Putumayo (Luxemburg 1970
[1912]:chap. 23, footnote 5). Rudolf Hilferding emphasised the effects of the world
market (Weltmarkt) as an economic arena (Wirtschaftgebiet) that generates the
coercion of labour in systems like that of the "contract labour" of coolies in the
British colonies of Asia (1968 [1909]: chapter 22, pp. 421,431-34, 447-48).
291
290
indigenous economy and the capitalist market. One of his conclusions is that this
this thesis.
relation results in an "ambiguous economy", where objects "are now gifts, now
Beginning with the end of the boom (1912), rubber tappers within the
commodities". Like Meillassoux and others, Gregory insists on the idea that a "gift
capitalists and with a status similar to that of indentured servants, instead becoming
argument provides ' an example of the theory postulating the functional role of a-
plantation economies because of this peasantry (between 1912 and 1943), and was
1982:117). The post-boom Amazon example, however, demonstrates that this sort of
1985). The extractive economy moved from a phase dominated by firms with a
labour from the forest peasant sector to the capitalist sector (with salaries subsidised
dynamic response to the world market and with investment of capital in the
industries were not subsidised by the exportation of goods produced below their
and with forest peasant houses providing the local reproduction of labour and of
basic goods (1912-1992). During the second phase, we find stagnated firms (which
do not reinvest profits and do not act competitively) and peasant houses (which have
relations between a forest peasantry and forest patron-traders has not been
This formulation calls for the need to refine concepts. Gregory applied
between subsistence production and production for the market. Instead, I have
focused upon the differences between economic actors -- the commercial firms of
patrons and the houses of rubber tappers, which integrate the same economy --, who
appropriate and produce goods within several regimes of circulation. Thus, forest
houses operate under local gift regimes (between neighbours), simple commodities
exchange regimes (exchanges "for their value" between houses) and debt-
commodities circulation regimes (between houses and patrons). Houses use means
of production (which include commodities and non-commodities) to produce goods
converted into objects of consumption (acquired as debt-commodities, received as
gifts or obtained through simple exchange). Houses may make investments and
grow, in the process increasing their prestige and influence, while accumulating
some wealth. The patrons' firms experience phases of growth and of crisis without
any technical change, in effect simply reacting to external price fluctuations. The
expansion and contraction of patrons' firms also occurs as a reflex in relation to
government policies? In short, this discussion leads to the causal role of the State in
The role of the State thus leads to the role of politics and classes in the local
economy. Merchant-businessmen (firms) and forest peasants (houses) are not onlypart of a market. In the Upper Jurua situation, they also constitute dominant and
dominated classes, one subordinated to the other by virtue of the fact that the
patron-traders monopolise the forest (there is no land market; and there is a
monopoly over services of the forest trails) and they monopolise trade (debt-rubber
is the counterpart of State loans). Neither of these monopolies is perfect: there is
class struggle in the forest, and there is a middle class of small itinerant traders and
of free cultivators.
the economy.8
There is an intimate relationship between the stability of rubber trading
destruction), on the one hand, and the role of State bureaucracy in regulating the
This analysis must be set against the theory according to which the patrons
10
form a class of capitalist-producers, and the rubber tapper form a proletariat
which, instead of selling its labour, is forced to a maximum degree of labour (that is,
economy formed by houses and firms. Hence, local history involves not only regimes
the debt system. In assuming that "capital" thus astutely subordinates labour on the
circulation) and gifts: there also are tax-objects. In the Amazonian exa'11ple, taxes
periphery, such an analysis forms part of a wider teleological theory of history of the
come in the form of subsidies transferred by the State to the patrons (thus
periphery. Marx's ideas on the "subsumption of labour under capital" are useful in
"redistributing" industrial and consumer taxes), as well as in the form of rubber trail
explaining why the rubber tappers are not proletarians. 11 Marx distinguished two
forms of this process: formal and real. Formal subsumption involves the
7. According to Furtado (1963), the same mechanism that allows for growth
("vent for surplus", Hart 1982: 12) during market booms is also responsible for
involution dunn~ market slumps. This mechanism is "structural dualism" where a
non-export actiVIty plays the role of labour storage. This is not the "two economies"
kind of dualism found m Boeke (1953).
Northeast, and rubber. estate patrons in .the Am.azon) align~d ~hemse~ves ~th
industry, under the aegIS of the State, leadmg Braz!l to a combmatIOn of mdustn~l ,
growth (for example, the production of automobIles and tyres .for the ~omest1c
market) with subsidies for stagnated rural firms and the importatIOn of agncultural
inputs (e.g., extractive rubber is subsidised through import tariffs on plantat~on
rubber). At the same time, the State impeded agrarian reform (Berno de AlmeIda
1991:239-41).
8. National states are not passive in the face of the world market (Cardoso,
1979; 1976:191-204), contrary to what Wallerstein's and Frank's approaches imply
'
with which Cardoso at times IS mistaken (as in Hart 1983: 121-22).
10. Pradb Jr. 1979 [1960]:67-71; Frank 1971 [1969]. Applied to the rubber estates in Silva 1982; Pinto de Oliveira 1984; Bakx 1986; Geffray 1991 and others.
transformation of workers into wage labourers: labour becomes part of "capital" and
distinct causal sequences: according to Warren Dean, there was a technical barrier
is placed at its command. This command allows the owner of capital (who is
(the domestication of the Amazonian seringueira tree) to that which I call here the
temporarily owner of the wage labourer) to control the labour process, to prolong
the work period and, with the relative abundance of labour, to maintain wages at a
Barbara Weinstein, it was the social barrier (the resistance of local merchants, the
fixed rate. Formal subordination thus permits the increase of surplus value in
absolute terms. But with the formation of Asian rubber plantations, the capital -
technical process by capitalists. 15 In any case, in terms of the world market and in
terms of cap~talist accumulation at the beginning the century, the real and formal
subsumption of the rubber production process occurred in Asia and not in the
Amazon.
(temporary slavery legitimised by debts) and wage labour all have supervision costs,
which, when adopted, are much greater than those of plantation economies with
similar regimes. Indeed, I would argue that these systems never were established
asserted that forest peasant houses constitute a class (insofar as they are m
firmly and efficiently in the Amazon, even during the boom, because such costs face
opposition to a class of patrons who mOl}opolise land or trade), but also that they
limits in a forest extraction economy. 12
The method of real subsumption 13 is based on technological change, which
have social mobility. One implication of this model lies in explaining how forest
houses exist even in the absence of patrons; another holds that the advance
places the labour process out of reach for individual producers. This may be
trade/debt system is not synonymous to the trade post system. The trade-debt
illustrated by the replacement of the forest with plantations, the adoption of new .
regime is one of demonetised commerce, with long intervals between the reception
inputs (clones, pesticides and fertilizers) and administration. However, though
and supply of goods, dictated by the fractal dispersion of the extractive forest
attempted by Henry Ford, neither real nor formal subsumption occurred within the
economy. In this commercial circuit, goods assume the form of debts. This form may
Amazon rubber economy.14 In explaining this, different authors have adopted
be seen as a variant of commodities-form, demonetised and personalised through
d
2. "As to the limits of the value of labour, its actual settlement always
ep.en s upon supply and demand, I mean the demand for labour on the art of
capItal, and the supply of labour by the working men. In colonial countries &e law
of supply and demand favours the working man" (Marx 1950 [1898]:402).
long-term diadic contracts. This form operates over great distances and time .
edition of Capital (Marx 1969 [1872-75]) does not include the aforementioned
passages.
15. The most direct form of developing a supply of labour, freeing up land
and selling the product is to expel the rubber tappers and allow immigrants in - and
not await their endogenous stratification. Thus the Brazilian counterpart of the
Russian debates on the disintegration of peasantry as a condition for the emergency
of an "internal market" for "national industry") IS irrelevant for tappers (cf.Lenin
1979 [1899]).
intervals in the forest economy, alongside the gift-form that operates between
neighbours. The debt-form of commerce may take place between equals (itinerant _
traders, small retailers in the forest) as well as between unequal parts (patrontraders and rubber tappers). In the latter case, violence (supported by the State) and
monopoly play roles; thus, a debt may be collected through the use of violence by
patrons, but not by regatoes or marreteiros. This distinction is necessary if we are to
understand why the debt regime is so persistent in the Amazon, with or without
patrons backed by forced monopoly or by violence. Debt is not the cause for
labour's immobility; but the immobility of labour through coercion allows for debtcapital (that is, merchant capital that exists under the form of debt-rubber) to obtain
extraordinary monopoly profits.
Perhaps it is necessary to justify both the idea that patrons and tappers
constitute
classes as well as the conflict model subordinated to this analysis. Patron- _
.
.
traders appropriate in a legal sense two components of the labour process (the
forest and production or consumption inputs), which finds support in the State and
which is legitimised by an ideology, and through these monopolies they obtain rents
on natural resources (tribute-goods, rubber to pay rent) and extraordi1J.ary profits
based on the commercial monopoly (debt-goods, rubber to pay debts).16 Control
over the State (police, judges) ensures the flow of these returns. Forest peasants
constitute a class because they are not only peasants but also "debtors" (of monopoly
rubber) and tenants. 17 Thus, the class conflict between patron-traders and peasantdebtors who owe rent and goods is expressed in the refusal to pay rent or to pay to
the monopoly (by turning over rubber to an itinerant trader).18 The patrons' ,
hegemony is made up of two components: political force (control over State
agencies) and ideological
conflict within the hegemonic order to the contest over hegemony itself when they
form alliances with other groups, altering the correlation of political forces (for
example, in joining labour organisations,
representation in the public sphere) as well
18. It should be noted the role of the components o~ the labour proc~ss in
this view of class. Exploiting classes appropriate persons, thm~s, nature, and Ideas:
this results in slave labour, capitalist profits, landed rent, royaltIes. I observe that the
theory of the labour process was also used by Marx to analyse techmcal change
(Elster 1983:164-68, 1985:143-154).
transformed into land and appropriated privately? We shall return to this point
(between houses and patrons), where the notion of distributive injustice
below.
IS
(gift exchange between neighbours), where injustice may take the abusive form of
over consumption by other houses.
question why labour-value is relegated to a secondary role, while the central subject
Gudeman (1978) has revived the role of classical political economy in the
of the present thesis is labour. In Marxist theory, the concept of labour in fact has
two different roles, one quantitative and the other qualitative. The first is
disseminated a neo-classical style based on Chayanov for the same objective. Both
approaches take labour as the base and measure for wealth. In the political
(Roemer 1981) holds that profits exist only where there is exploitation (the net
labour through the exchanges between peasant houses and other social classes; in
transfer of labour value from one class to another). But this theorem has met well-
labour and leisure within the house in order to find the point of equilibrium
Morishima and Catephores: 22-23; Roemer 1982), and in order for it to remain
between effort and satisfaction (Gudeman 1978; Sahlins 1974). To be sure, both
valid, it has been argued that it is necessary to abandon the notion that goods are
approaches should be combined when studying houses that are related to a market.
traded "at their value", or even that micro-values can be attributed to individual
But while both Gudeman and Sahlins seem to converge on the matter of the
allocates (as in Robinson Crusoe's economy, Marx n.d. [1887]: 76-79) among its
Another function of the theory of value was moral: that is, it serves as an
This disequilibrium appears in the relations between houses and merchants; it also
ethical parameter ' for the existence of inequality. In exchange between non- '
shows up internally in house plans. In the first case, while patrons seek a monopoly,
neighbour houses (simple commodities exchange), rubber tappers frequently use the
idea of labour-value to regulate, for example, the exchange of flour for rubber. But
between supply and demand of labour and goods. There is thus a distributive border
this procedure does not work when imported goods are traded for local products
between houses and patrons (or a Ricardian frontier). This border moves between
extremes: minimum subsistence and maximum labour, zero profits and maximum .
profits. But it is not fixed, and it is up to class struggle in the forest to determine the
exact level of exploitation (an idea consistent with Sraffa 1960).
margin of freedom in the houses' economy that allows for the existence of pockets
of "thrifty" houses, "labour aveI:se" houses, "spendthrift" houses, and so on.
Houses
within each, and where consumption was homogeneous and paid for - would allow
The notion of forest houses and their extension as macro-houses and macro-
maximum profits to be extracted uniformly from all houses. Only an average rate
house networks provides a conceptual tool for understanding the political economy
can be established, but individual houses can then move around the average.
and the domestic economy of the rubber tappers. 20 Houses control forest
Let us turn our focus to the inside of the domestic economy. Each house can
settlements
(coloca~6es),
make choices about labour use and consumption, within the limits imposed by price
squatters (without individual property rights over the colocas;ao, as in the Jurua of -
levels and by the payment of rent, which regulates the relations between houses and
Acre), or they may have permanent rights over tapping areas, which in turn they
traders or patrons. This is the sphere of house production and consumption plans.
may sub-let to other houses (as in Eastern Acre, under the influence of a market for
We have shown that within this sphere houses are dispersed, much like particles
land). In the Upper Jurua region, the control over resources was not associated to
occupying the volume within the walls of a cylinder. In this metaphor, the piston
the permanent ownership of the forest; the forest was not a commodity. Houses
defines the borders of exploitation; the greater the resistance of the houses to the
production plans. This point is at the root of my refusal to accept the approach
make budgets for the acquisition of goods (including "valuables" and goods for
based on optimisation for each individual house. The state of the houses as a whole
is not the same as the sum of atomised actions of isolated houses: were this true, we
would be able to predict the behaviour of individual houses through the application
riverside areas. Houses can accumulate a certain wealth (people, things), which I
of micro-economic models. Instead, such behaviour first fluctuates and second is not
have not described -as capital accumulation but as the accumulation of a patrimony. -
between blocks of houses. Indeed, the piston metaphor makes sense because in the
relations between houses (seen as a whole) and the trade post, the role of the trade
post does not involve the control over individual labour and consumption processes
(as would happen if there was the "formal subsumption of labour under capital"),
but it does involve setting only global variables like prices and rents. There is thus a
20. Gudeman and Rivera (1990) describe in detail a "house form" as part of a
cultural tradition. A terminological note: these authors employ the term house
instead of household to express that which Weber called Haushalt and Aristotle
oikonomia. a. Goody's (1990:21) "house" (between quotation marks) with an
essentially similar connotation. a. French maison (Descola, 1978), and Portuguese
casa (Pina Cabral, 1983). Unfortunately, the concept of household has become
spe~iali~ed in ~he literature as meaning the domestic group, as in the French
maIsonee (Nettmg et al. 1984: xiv; xx).
1989); a way of doing things. Hence, a house's actions are not turned only to the
with one another in flexible extensions that I call macro-houses 21 , which exploit a
final product resulting from those actions; the notion of maximisation may not be
contiguous territory under a common leadership and articulate their production and
applied here, as it could in the case of a firm oriented towards the accumulation of
capital (Gudeman and Rivera 1990). The distinction between homes ("consumers")
unions, during elections, in marriages, in religion and in business, and they may
homes choose "consumption plans" and firms choose "production plans", each
compete with one another. Since houses do not need to have a property stake in the
subject to comtraints (Debreu 1959: 37, 50). Peasant houses (not unlike artisans')
forest (as in the case of the highly mobile upper Jurua tappers), the continuity of
choose action plans that simultaneously act as consumption and production plans.
houses essentially resides in the unit of labour processes over time. People grow, a
Using this language, houses develop and execute dual "action plans" oriented
territory may undergo changes or be substituted for another when a house moves,
towards a quality of life standard (Sen 1987: 89), but which may turn out to be
objects are used up and replaced, and ideas are transmitted, invented and
agonistic plans (as when houses expand and develop feuds WIth other houses).
abandoned. A form persists, embodied in real houses that emerge, expand into -
macro-houses and die, through repeated acts of labour.22 In the tappers' case, the
technical procedures), by prevailing technology, by people (since there are new and
house's form includes a forest with a stock of niches and living things; a group of
old houses), by objects (since gasoline engines and manual wheel presses result in
people with a social composition of women, children and men; things like flour,
game meat and imported objects, necessary not only for the technical functioning of
the house but also to guarantee a decent life for each category of persons (Gow
becoming part of real processes. They are more than mere representations, made
evident by the difficulty of adopting a technique, the mysteries offered by the forest,
21. Compare with Morris (1982:56, 135, 165); Tanner (1979 passim); Murphy
1978:1-22,27-50, 80, 153-193; Murphy & Murphy 1974:179-203; Steward & Murphy
1977(1956). Cf. Wagley 1964:90-102.
.
22. The following concepts may be traced to Aristotle: the distinction
between use and exchange values, the theory of status-regulated exchange (not
market-regulated), the distinction between management aimed at providing and
profit-oriented management aimed at profit (Aristotle 1977a:39 1982:267-71
1977a:33). These distinctions are known more widely because of' their moder~ _
versions (with the due acknowledgements) by Marx, Polanyi and Weber
respectively. It shoul.d come as no s.urprise! then, that .the theory of labour processe~
also belo.ngs to ArIstotle (see Glannottl 1983). HIS theory on the on gins and
preservatIOn of a substance was formulated in the language of production actions
exemplified in the construction of a house (Aristotle 1980: 105), the making of a
statue or in the actions of a doctor (1980:211-15). Art anthropological focus on the
labour process originates with Malinowski 1978 [1935]; cf. Richards 1969 [1939]
Firth 1966 [1946], Salisbury 1962, Gudeman 1978, Wallman 1979 and Lee 1979.
'
and the birth a~d death of persons. Commodities form an integral part of the
houses' form, whether as means of production or as things to be desired and
consumed (Hugh-Jones 1989), but prices mediate the access to them, and establish
limits for houses' production plans, just as ecological constraints set limits to the
density of houses surrounding a given territory as well as to the size of macrohouses. In sum, the study of causal relations focusing on technology, nature and
23. Note that in Aristotle's description of house management, the hous~ head
must manage "instruments" (slaves and inanimate tools, Aristotl.e 1977:.1?), III two
senses: in organising the labour and consumption process, and. III acqmrmg ~oods
through exchange, necessary (and sufficient) for a "good life" (Anstotle 1977:39).
-
pnces, interfaces with the recognition of a human order that treats these as
a peasant gains squatter's rights over land by proving agricultural activity, while it
establishes an upper limit of 100 hectares (for the Amazon) for titles thus acquired.
Therefore, there is no way to recognise the legal rights of rubber tappers over the
human labour faces the action of time in the form of corruption. Just as buildings
forest settlements (coloca~6es), which include 400 hectares of forest per house, used
crumble while memory distorts the purity of old ways, the very form of forest houses
for extraction and hunting. Around 1984, some rural unions in the Amazon involving
may cease to exist. The rupture in form may result from the excessive use of
rubber tappers began to push for lots ranging from 300 to 500 hectares. But there
material resources, from natural disaster, from death or ruin, or from inevitable
was another problem involved, different from the one represented by the dispute
between land buyers and occupants. All it took was for one house to agree to sell its
Lenin called the ."disintegration of the peasantry" (Lenin 1979 [1899]) may come to
pass with expulsion from the land. Where there were forest houses, plantations and
ranches then would sprout up, or maybe they would give way to wasteland. But
reason was that the buyer immediately would obtain legal permission to strike down
the area acquired, -affecting both hunting and extraction in neighbouring areas. No -
(Popkin 1986:199), a version of the "prisoners' dilemma" (Sen 1987:80-88) that here
emerges in the form of a "tragedy of privatisation".25 Thus in 1985, a meeting of
facing the new "owners" was to expel its inhabitants. The only sure method for this institutionalisation strategies: individual property holdings in the forest settlements,
sort of primitive accumulation was to strike down the forest, thus destroying the
allowing individual land sales, property in condominium, allowing land sales by
houses' material base, literally uprooting the forest peasantry itself. The occupants,
with support from the recently formed rural unions, began to develop a strategy of
empates. 24 In principle, Brazilian legislation for the agrarian sphere recognises that
24. An empate consists of (a) avoiding the act of striking down the forest by
peons, (b) protesting against a rubber tapper who sold his coloca~ao to a rancher.
Linguistic note: empatar means to prevent (someone) (from doing something), to
block. Empate is the action of preventing another action. It is not to be confused
with the dictionary meaning, which is "to even a match, to obtain a draw" (a common
mistake committed -by non-Amazonian authors).
25. The econo~st Peter May has coined the term "tragedy of non-common.s"
to illustrate a similar effect in Babassu palm areas. Unfortunately, I could not obtam
his Ph.D. thesis on the subject.
26. Silberling (1991) forcefully defends the argument that social movements
constitute an important factor in the success of commons management. Here I lend
support to this VIew.
307
306
majority decision, and public property conceded to peasants, blocking land sales (cf.
Brornley 1991:22). The final decision was reached after a morning-long session in
concessions made through a local association also blocked another effect of the
the absence of advisers (who recommended the second alternative): the extractive
"prisoner's dilemma" (cf. Dasgupta and Heal 1979: 18). One resident (or even a
reserves were to turn the forest into public property to be exploited exclusively by
resident rubber tappers. In 1977, a group within the National Agrarian Reform
purposes, making other houses face hunger. In this case too, the Extractive Reserve
restricts individuals from disposing privately of a public good (which means the
integrity of. the forest), thus distinguishing the sum of individual interests recognised
"families" and limiting the area of forest that could be stricken down. In 1989, a task
through the practice of concessions, from collective interests, which must be-
force of tappers and advisers (within IBAMA) introduced in law a formal definition,
dwellers (and not only to individuals), subject to a use plan. 28 Local regimes
("traditional" or not) were to enforce rules of access to, exit from and transmission
of forest, at the same time blocking the commoditisation of the forest and hence the
effects of the market over this peculiarly indivisible good (Dasgupta and Heal
1979:3-8, 472-73). The concept resulted from the objective of blocking the
privatisation tragedy.
reserves thus provided the response of rubber tappers' political organisation to the
prisoner's dilemma (in the form of the privatisation tragedy), rather than
representing a mere case of the survival of a traditional system of access to
commons.
29. This mechanism is being applied recently to traditional fishing areas (on
the Atlantic coast) and in babassu palm areas in Maranhao.
30. Tariff protection and subsidies for rubber production cOIl;tinue to b.e
demanded (similar to the protection of small farmers in the ~EC): ThIS. defence IS
corporativist and not universalist, using the terms of the foregomg dISCUSSIOn.
the above processes (agro-forestry). Palms, brazil nut trees, rubber trees, cupuassu
consider a short-term strategy in which forest houses maintain their prior niche
(a wild species similar to cocoa) and others are planted during part of the
diversity (they hunt, cultivate and extract commercially), while at the same time
integrating themselves into ~ variety of new markets for extractive products. These
domesticated forest with a high flow value. 31 This strategy requires time spans of up
'
.
markets exert different effects over different products. 32 Pred
atlOn
tech ruques
possess an equilibrium level (when tapirs are preyed on too much, they disappear
territories coexist with hunting and extraction zones, and with agro-pastoral patches
(as in the current pattern), under conditions of low population density. A third
trees die). The system of extracting latex is sustainable, for instance, because it
increasing population density significantly, say from 1-2 inhabitants per square
grouped along trails, subject to the supervision of mateiros: the forest's character as
service fund is preserved) as well as a cultural one (there are consensus norms
consumption increases as does the technical fund), forest houses in the second case
panema). Now let ~s imagine unrestricted markets for copaIba oil and wild game. _
becomes a secondary economic niche. The number of inhabitants increases and the
Supervision may certify the preservation of a forest fund (for example the state of
forest gives way to an agro-forest. The three strategies described above do not
trees tapped for latex or copaIba), but the supervision of game reserves is much
represent phases. Global and local political factors, as in the past, can alter the
more difficult, and the effects are different according to species. The Rubber
growth.
31. Below, I return to the distinction between forest use as a flow and as a
stock.
fruits (Almeida 1992b). All of these strategies apparently come into conflict with the
idea that the best way to "preserve" the forest is to value the economic flow
withdrawn from the forest, that is, commoditise a fund (Robinson and Redford
311
310
1991). Rubber tappers are not moved by their love of animals and the environment,
but they are defending a majority against the action of "free riders". For example,
Indeed, over the long run, this scenario may lead to a prisoner's dilemma syndrome.
hunting with Paulista hounds (of high value) in order to commercialise game meat
There is no guarantee that"the forest will ensure a stable pattern of livelihood for
was begun on the edge of the reserve by a small group that neither cultivated garden
plots nor produced rubber; the effect was to ruin the neighbourhood as a hunting
area. 33 The effect of a market for game meat is comparable to the introduction of a
land market, and the solution here also involves a struggle to limit the market's
action.
introduction of domesticated forests within the settlements). Here the main problem
intensive agro-forestry (and later on intensive agro-ranching), and in the short run
putting a premium on depletion-oriented extraction and predation.
is whether or not there is an inevitable transition from extensive use of forests (with
of land and corresponding high demographic density. Lower prices for extractive
poor (Anderson, 1989; Browder 1992), and second, it is the only scenario that is
products and higher prices for agricultural products will be a stimulus to replace
economically competitive without depending upon state subsidies. As for the first
extensive-use forests (generating a "flow of copaiba, rubber and oils, along with
point, I argue that there is no absolute scarcity of land in the Brazilian Amazon, and
1989; cf. Lipton 1990; McNicoll 1990); it is a case of the defeat of the agrarian
reform movement (Berno de Almeida 1991). The poor are blocked from access to
that the mechanism leading to demographic growth involves the competitive interest
",
of individual houses in always increasing their piece of the forest flow value pie
(either by seeking to enter in the area, or by increasing the number of houses in a
33. ~ blacksmith ";,,hose shop is,in the forest was be affected in the same way
as a tapper s house (preCisely because If he moved to town, he would have to give up
the niche diversity that subsidises him).
land or to food, which nonetheless exist in abundance. Rather than preventing land
34. The Malthusian cycle: population growth, pressure upo~ resour~es,
increased surplus and further po~ulatlOn growth. The attractor for thIS cycle I~ a
maximum population under SubSIstence levels ("misery checks'). T~e Bo~~rupIan _
cycle: population growth, pressure upon resources, technologIc.al mtensIfIcatlOn,
increased surplus, demographic growth, having as attractor SubSIstence levels ~nd
maximum sustainable populations at a higher level than before. A ChayanovIan
cycle: family growth, mcrease in resource use, labour intensification, incre';\sed
surplus per labourer, balance in family in~oI?e at subsist~nce levels (~CNIcoll
1990). In all three cases, the attractor is a rrnmmum of SubSIstence, a maxImum of
labour intensity, a maximum of environmental degradation. Such determinist~c
scenarios hinge, therefore, upon the inevitability of Hardin-type or demographIC
"tragedy of commons".
312
313
has been made that while in the short-term agricultural products may be of high value,
in the long-term extraction may yield a sustainable income at constant costs while
Schwartzman 1988). I' will not pursue this line of reasoning which assume constant
distributed with a low density. This scenario may be avoided. Under current
relative prices, because, due to technical change in the overall economy, relative prices
conditions, it is avoided because children in excess leave the settlements, either
change. In the future, the price of any product which is extracted today may decline
spontaneously or moved by paternal authority. Systems of access and transmission
relative to the price of products that the extractive economy imports (just as happened
control the proportion of persons to resources; thus the essencial variable is not the
to rubber). In fact, this argument applies not only to extractive, but also to any
absolute population, but the parcel of which enjoys land rights (received through use
agroforestry or agricultural product. Thus, the intensification of population growth in
concessions) according to rules in force. In the absence of such systems, which along
such areas, even with a transition from extensive forest use to intensive agriculture,
with the formal mechanism of use concession guarantee the stability of resource
only increases the chances that such peripheries will assume the role of "poverty
exploitation, each house becomes economically unfeasible, whatever the technical
procedure in us~, and whatever the initial density. Indeed, each house has a stake in _
maintaining resources indivisible, thus blocking this variant of the "prisoner's
dilemma".
In assuming ( erroneously) that extractive reserves are "specialised" in
extraction (within the first of three scenarios outlined above), critics argue extractive
economies inherently are doomed to be replaced by domesticated economies and
later by industrial economies (Homma, 1989; Torres and Martine 1991). Thus, if a
vegetable oil is first extracted and there is a growing market for it, prices will rise
and signal to investment in the cultivation of the wild plant (usually in a region
maintained within the system, the flows originate from a forest stock, the value of
which then may be calculated in terms of the capitalisation of the flows' value. The
stock may be converted into flows at greater or lesser rates, and it is possible to
calculate the "optimum rate of depletion" of the forest stock in
314
terms of market prices; hence, at one extreme there is a flow level that preserves the
forest indefinitely, while at the other there is a level at which the forest will be used
up in a single year, after which land (and not forest) will be available for other uses
(which is the case of the immediate conversion of all the timber and animals into
commodities). Is this an appropriate way of assessing a forest's value? The forest
may be seen as a fund which generates services (just as living persons do), and which
as such is not assessed by market value~. There is no futures market for the forest as
a fund; nor is there for the services it affords in the present and ' which do not take
the form of commodity-flows (or local use-value flows). Can the value of a forest
such as that of the Upper Jurua be assessed correctly' in market terms through a
method of capitalisjng product flows and then estimating the optimum rate of stock
depletion? This method can lead us to consider as efficient the use of 1,000 hectares
of degraded pasture by a rancher (or by twenty impoverished families who cultivate
beans), while considering as inefficient the use of the same area by a group of three
houses generating a flow and, in addition, maintaining 955 hectares of forest with a
high degree of biological diversity (Wilson 1988: 193-226). This diversity involves
"materials" (such as latex, or a slain animal), but also includes information, such as
the chemical processes associated with biological systems (Brown 1989; 1982; 1991).
The limitations of the market in adequately signalling what natural resources are to
be used from a service fund (such as the air, water and, in this case, the natural
fore.~t)
and Heal
1979~ P~arce
~he
absence of future,
unborn participants from current market transactions (cf. also Alier & Schlumpann1991). Extractive reserves, which represent an institutional innovation insofar as
they block the prisoner's dilemma in its form as the tragedy of land privatisation,
also may check the same dilemma as it appears in the form of a demographic rush,
as in the rush to maximise present value flows in response to market signs. With
315
for the right to become free of rubber trail rents and for the right to sell their
product free of .debt (Macedo & Almeida 1988; O'Dwyer 1989; Brown & Cardoso _
1989, 1991; Embrapa 1989; Ministerio Publico 1989). Hence, contrary to the theory
316
which predicted that Western Acre possessed only debt slaves incapable of
mobilisation since they were subjected to all-powerful patrons, the effect of the
Council was to catalise existing struggles directed against rent payments and the
commercial monopoly. In . Eastern
fought the invasion of lands by ranchers from the south; in Arnapa, they mustered
forces against the Jari project empire, while in Rondonia they challenged the wave
?f imigrants that pushed them into marginal areas. Analogously, in different
locations and under different political circumstances, different scenarios of technical change and of social pacts emerge to control components of the labour process of
the houses: the forest, the information, the persons, the objects.
A Maxwell Demon
317
nature, persons and things by means of a technique and aiming towards an end. This
What is of interest is the way he acts. The stage set for this metaphor is a system (a
box) linked to the rest of the world by a door. At the beginning, there are green
management of a body politic, and seen as a system, which involves flows that cross
things inside the box (the box has colour, so to speak). Outside the box, there are
mostly grey things (many different colours mixed together perhaps). Things flow in
and ideas), persons (migrants), and commodities that arrive; things (forest and
both directions. It is enough that some randomness exists in the movement of things
to and fro. Each thing moves on its own, as individuals. Then, in the long term, there
chemical substances and other procedures) and persons -- all this crosses the
is a final homogeneity between the interior of the box and the rest of the world, both
regulated by a price system. Houses and their coalitions form decision units (atoms
b
so to speak) within the system. On the other hand, above the individual houses at
the system's border, there are associations, bills of rights and responsibilities,
supervision institutions, leaders and ideologies.
Forms are created and reproduced. They also die. Under human use, there i~
no inertia of forms: a forest is not perpetuated by custom, information is not
318
This brings to mind the role of norms, institutions and other mechanisms means to use information on flows of things and ideas to make a form endure and as
a means to fight against corruption of differences. But the original metaphor also
leads to another conclusion. The Maxwell demon cannot exist forever. His existence
is of a local nature. Like a modern Nemi priest-king mounting guard at the doors of
the temple of Diana, he is subject himself to death, threatened by the continued
onslaught of objects and ideas that in the end affect his own ability to discern and to
take decisions, and degrade his own form. The metaphor asserts that such a demon .
may only overcome the increase of entropy locally as long as he is fed from without
with a subsidy of energy and information, so to speak. Thus, a dialectic of local
permanence and global processes emerges, combined with the possibility that local
history may develop its own course as the result of fluctuations that are amplified
but which cannot be foreseen as the inevitable outcome of determinism.
TABLES
"
.
,
'
'
I
-
---
321
TABLE 1.1. BRAZILIAN RUBBER EXPORTS 1827 - 1940: AMOUNT AND VALUE
year
Amount
in tons
1827-30
1831-40
1841-50
1851- 60
1861-70
1871-80
329
2,314
4,693
19,383
37,166
60,225
82
231
460
1,938
3,717
6,022
4
12
24
100
192
311
17
168
214
2,282
4,649
10,957
0.05
0.07
0.05
0.12
0.13
0.18
42
58
42
100
108
150
1880-90* 110,048
1891- 00 213,755
1901-10 345,070
10;004
21,376
34,507
518
1 102
1 780
16,519
43,666
134,394
0.16
0.20
0.39
133
167
325
Boom
1911-20 328,754
1921-30 185,222
1931-39# 103,722
1941-47
32,875
18,522
11,524
15,753
1 696
956
595
800
83,036
22,631
3,306
0.25
0.12
0.03
208
100
25
Decline
4,393
3,542
5,374
2,647
29
227
182
277
136
2
1948 - 50
1951- 60
1961- 70
1971-80
1981- 87
Peak year
1912
Yearly
average
Amount
(Index)
Price
Value
Gold st. aver.
1,000
tons rubber
42,286
Price
Index
War
U$
0 . 53
0.72
0.80
1. 30
value index (100=1850s)
0.38 317
Rubber Index
ton
Value
1,000St.jton
1911- 20
1921-30
1932 - 39
32,875 100
18,522
56
11,524 35
0.25
0.12
0.03
Index
100
48
12
323
322
--------------------------------------------------------------------------TABLE 2.2
Year
Population
1908
1910
1911
1912
1920
1940
65,000
69,457
80,175
86,638
92,374
79,768
Source:
1909-1940
TABLE 5.2
Rubber
Estate
1,987
Tejo Estates
93
Boca do Tejo
738
Hor/Fort/Bag
Restaurayao 1,156
200
Maranguape
175
Iracema
10.466
11.544
5.711
5.179
COELHO 1982:46-65.
Totals
Middle Estate
Minor Estate
105
205
25
31
Trails
total
rented
657
11
Pop.
1,195
147
170
451
74
60
2,343*
150
2,632
282
2,644
Major Estate
Settl.
2,362
Indigenous Area
Total
TABLE 5.1
Houses
Localization
TEJO ESTATES
BOCA DO TEJO
HORIZONTE
FORTALEZA
BAGE
RESTAURACAO
OTHER ESTATES
Source: field research 1982/83
(mouth)
(low course)
(middle)
(affluent)
BAGE
(mouth)
SERINGUEIRINHA (middle)
DIVISAO
(headwaters)
(upper course)
RESTAURACAO
(banks)
DOURADO
(headwaters)
MANTEIGA
(headwaters)
RIOZINHO
(headwaters)
CAMALEAO
(headwate
BOA HORA
(headwaters)
MACHADINHO
(headwaters)
IRACEMA
MARANGUAPE
(low course)
(middle)
TABLE 5.2A
Estates
Restaura y ao (Hdqrts)
Dourado
Riozinho
Manteiga
Camaleao
Boa Hora
Machadinho
Total
Settlements
Houses
Trails
Rented
Trails
Idle
Trails
Total
4
17
27
21
8
14
14
14
41
60
21
16
36
17
25
91
130
62
37
71
35
5
39
59
50
13
16
24
30
130
189
112
50
87
59
105
205
451
206
657
324
TABLE 5.3
325
Estate
km2
Houses
Settl.
Total
2,362
491
166
Tejo Estates
Boca do Tejo
Hor/Fort/Bage
RestauraC;2lo
Maranguape
Iracema
Total
-1 ,987
93
738
1,156
200
175
2,362
396
23
157
216
54
41
491
145
7
63
75
12
9
166
Indigenous Area
Total Tejo River
282
2,644
Trails
Pop.
total rented
1,414 982
3,432
TABLE 5.6
Sourc e~
Rent Due
1,243
782
Total
3,905
2,190
TABLE 5.7
Estate
TABLE 5.4
Population
Area in sq km
Persons/sq km
Source:
Thaumaturgo
(2)
5,983
5,062
1.18
Tejo River
(2)
3,432
2,362
1. 45
RestauraC;2lo
(1)
(2)
1,292 1,195
1,156
1.12 1. 03
Riozinho
(1)
(2)
390
320
230
1.7
1 . .39
IRACEMA
MARANGUAPE
SUB-TOTAL
218 . 0
89.2
307.2
TEJO ESTATES
FOZ DO TEJO
HORIZONTE
FORTALEZA
BAGE
Families
34
36
1,052.96
67.0
74.0
252.1
171. 0
50
15
37
15
179.3
309.5
21
32
Titles Recognized
0. 0
89.2
89.2
561. 90
43.0
74.0
84.0
137.0
RESTAURA~AO(sede)
TABLE 5.5
RESTAURA~AO
Houses
Settlements
Population
1981
205
105
1,195
1992
216
75
1,292
Area
per house
Area
per sett'lement
Houses per settlement
5.64 km2
11.01 km2
1. 95
5.35
15.41
2.88
Sources: Tables 5.2 and 5.3 above. The 1991 census had the hous e a ~ its
unit and settlement names (recorded at the trade-post in 1982) may lluve
been dropped in 1991 when trade - post did not exist anymore.
DOURADO(S.Franc.)
MACHADINHO(B.Vista)
BOA HORA(Aracati)
TOTAL Tejo River
1,360.12
240
179.0
44.9
0
651.06
TABLE 5.8
Date
15 June
5 July
26 July
31 Augst
326
10
19
13
12
5
3
4
4
Estivas
(non-durable)
Yearly consumption
Montly consumption
Item
Value in Rubber
Amount
cooking oil 1
liter 1.6 kg
sugar
1.3
0.8 kg
rice
1.0
1.3 kg
salt
0.9
1.3 kg
soap
0.7
0.5 kg
nescafe
0.25 unit 0.4
medicine
0.1
kerosene
0.7
0.5 1
gunshot
1.3
0.12 kg
gunpowder
2.6
0.12 kg
Sub-Total
10.6 kg
127.2
utensilios
(Durable)
Duration
(Wear)
axe (blade)
0.8
7.8 kg 10 years
machetes(")
4.6 kgs rubber
2 years
1 9.2
shotgun
33.0
1 131. 0
4 years
pails
2.6
1-2 years
1 3.9
knives
4.4
1-2 years
1 6.6
sack
3.5
1-2 years
1 5.3
small sacks 1 6.6
1-2 years
4.4
tin cups
600
10 years
Sub-Total
52.7
Farinha
Amount Monthly Value
O.~ ~aneiros)
5.3 kgs rubber
63.6 kgs
.
(1 paneiro=30 kgs)
Total Baslcs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243.5 kgs
Non-basics (luxury,miudezas, vices, cloth, valuables) ...... 97
kgs
Rent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
kgs
Tota 1 rubber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 400
kgs
Source: field research, 1982/83.
TABLE 5.10
Trader
2,681
2,329
1,465
1,366
2,336
2,513
361
898
13,949
Ribeiro
Diniz
silva
praxedes
silva
silva
castelo
Timoteo
TOTALS
AVERAGES
Average
Rubb/House
Rubber
kgs
Number Houses
Total
**
1,190
1,034
650
606
1,037
1,115
160
398
6,190
774
23
4
10
0
4
14
6
19
29
15
0
8
27
13
4
18
52
19
10
8
31
27
10
37
23
54
65
75
33
41
16
11
10
14
24.5
40
At Riozinho Estate
** At Restauracao Estate less Riozinho
U$1.00 = 222 cruzeiros (end of 1982)
Rubber prices = 500 cruzeiros/kg = US$2.25/kg
Source: Field Research
TABLE 5.11
Delivered to
Based in
Role
Town
Marreteiro/tapper
Faria
Town
Regatao
Magalhaes
Town
Regatao
Tim6teo
Iracema
Regatao
silva
Manteiga
Marreteiro
Pompilio
Manteiga
Trade-union/tapper
Ginu
Riozinho
Marreteiro/tapper
Faria
Sub-total to regatoes/marreteiros
Restaurac;:ao
Correa
Correa
"
Sub-total to patron
Total rubber
Patron/accounts
Patron/rent
Rubber (kgs)
35
68
20
62
35
12
65
297 (60%)
169
33
202 (40%)
499
100%
TABLE 5.12
RESTAURA~AO
MAIN MARRETEIROS,
(1982-1992)
TABLE 6 . 2A
~
o
Estate
Name
status
Item
Riozinho
Manteiga
Camaleao
Boa Hora
Machadinho
Timoteo
E. Ribeiro
Bills 82 june-aug.
49,925
"Commission" on goods
14,977
Bills 82 aug-dec.
111,305
Juvenal's bill to patron 3,815
Bills 83
13,974
purchases at trade-post
3,000
payment orders
12,666
It. peddlers/Chagas
It.Peddlers to Juvenal ' 80,415
55,848
Interest for the above
purchases not at trade post
193,996
56%
151,929
44%
345,925
345,925
100%
z.
Ribeiro
Ze de Luna
TABLE 6.1
Total
Date
Value
Currency
Goods
Cr$
*
*2
Interests
Rubber (kgs)
15%
*
0
*1
0
*1
*1
*1
*1
*1
*1
*3
*2
*2
*1
*2
*2
*2,*4
*3
*3
*2
*3
0
0
30%
30%
60%
60%
*1
*1
*2
*2
*2
*1,*3
*3
*3
0
0
0
*1
*1
*2
0
0
*(2)
TABLE 6.2B
The numbers show the order in which currencies are entered and
converted into , another. A comissao is charged for merchandise advanced
when rubber is paid "at final price".
Obs.:
1,061 kgs
100%
kgs
kgs
kgs
kgs
kgs
8%
34%
38%
26%
3%
-100 kgs
109%
90
360
480
272
30
TACARATU)
Heads
Account I: Chagas
Rubber . account
output
1,061
100%
90
8%
408
38%
(COLOCA~A-O
y
360
34%
272*
26%
30
3%
debt
net
100 kgs
9% over output
account
177
35%
7%
output
499 kgs
100%
33
o
0%
278
56%
debt
net
274 kgs
55% over output
( summary)
Rubber
output .
account
700
100%
340
49%
60
9%
300
43%
0
0%
debt
net
300
43% over output
Balance 82
Variables
I II
III
+106'
+ 54
+ 21
+ 79
+ 3
+ 98
+182
+ 31
+ 23
+200'
+ 6
-283
-262
-898
-495
-467
-100'
-320'
-47
-173
-124
-765
-174
- 81
-121
-111
- 88
-120'
-646
- 45
-153
R.Farias(Nabe)
3 3
Rui Santana
2 1
Altevir Vitorino 2 1
osmarino Cunha
1 1
Jose Sabino
2 1
Alfredo Avelino 1.5 1
Joao Cunha Filho 3 2
Hilario Benevenu 4 4
Aldenizio Loureiro2 2
Jose Candido
2 3
Manuel Vitorino
2 1
simeao Apolonio
3 1
Cario e Sansao
3 1
Francisco Roberto 6
Ferreira Freitas 1 1
Valdeir Vitorino 3 2
Esterlito Cunha
2 1
Claudio Elias
1 1
Alberci Alves
1.5 2
Franc. Teodoro 1.5 4
Evilasio Santana 2 1
Assis Farias
2 1
Raimundo Isaura
2 2
Joao Nascimento
2 1
Aurino Batista
3 2
Chagas Farias
3 3
Adelino Leao
4 5
Manuel .Apolonio
2 1
Joao Cunha pai
2 2
Narciso Lima
1 1
Joao Teixeira
1 2
Total
70.5 54
Cases
31
Averages
2.3 1.8
30
-4,670
31
-150 kgs
Balance Aug-83
Rubber
Debt
IV
V
151
175
175
180
213
391
606
662
666
748
60
110
115
122
145
160
165
203
250
270
326
350
395
395
454
519
545
61,8
9,169
-438
173
-213
-211
-105
-742
-1,170
-149
Balance
VI
-263
+ 2
- 33
+ 2
+286
-136
-508
+517
-282
251
-345
-887
-309
-558
-558
-315
-258
-449
. -195
-636
-374
-156
-748
-111
-502
-282
-191
-235
-772
-187
-413
-398
-150
- 55
-199
+ 75
-310
- 24
+239
-353
+343
+ 17
-1,126
296
311
-508
-296
-311
-11,020
28
28
327
393
-4,143
28
-
148
333
332
TABLE 6.4b
Houses with
cases
a net debt:
a net credit:
total sample
20 65%
11 35%
31 100%
TABLE 7.2a
average balance
- 273.65 kgs
+ 73.00 kgs
-150.65 kgs
census
total
5.74
Houses
53
68
1.7
TABLE 6.5
TABLE 7.3
Houses per
Settlement
1
2
3
4
5
Totals
Balance
Min
kgs
Value
-437
-656
-68
-103
Average
Max
1,662
1,180
TABLE 7.4
E=Ego W=wife
Category
Women
Men
Census
Averages/house
Averages/settl
Estimates
85
1.6
4.47
109
74
1.4
3.89
95
Boys
Girls
Hired
Total
70
1. 32
3.68
90
69
1.3
3.63
88
6
0.11
0.32
8
304
5.74
16
390
53
House patterns
19
68
TABLE 7.2
Settl.
Census 19
Total 26
Pop. Houses
304
390
53
68
Area
sq km
230
230
Houses
p/settl
2.79
2.62
Persons
p/ settl
16
15
Area
Area
p/settl.p/h.
8.85
3.38
h = hired
~
o
Nuclear Family:
30
(E, W, S) or (E, W, D), (,E, W, S, D)
Extended Nuclear Family:
15
(E,W,S,O), (E,W,D,O), (E,W,S,D,O,e)
Recombined, extended nuclear families:
8
(E,W, Ego'Son) , (E,W, Wife's daughter)
(E, W, S, WS, WDH).
Others
(E)
4
Working-teams
(E,B) or E,e)
3
50%
Total
100%
60
25%
13%
7%
5%
TABLE 7.5
Patterns
One macro-house
(x+S+S+DH+Z), (X+S+S+DH=BS),
(X+WS)
(X+B+B+ZH), (X+B) , (X+B)
25
(X+S+S),
3.1 3
Velha Rita
Flo r es t a
B
26
3.71
Flore sta
Sao Salvador
ZS,ZDH
ZH
Isaura
SS'
S
Two macro-houses
(X+DH / Y+DH)
Y)
his
(X+WS+WS)
(X /
Santana
Degredo
ZH
FB
DH
,DH
DH,S
ZH
ZSW
ZH,ZH,ZS
13
10
4.00
1.30
Mainteiga
Os Pires
Nazaro
Roberto
DH
DH
DH,S
F
ZH
ZH
(X)
WM
Totals
68
26
2.62
Santana
Val
Santana
Nascimento
Ginu, Bispo
Luiz Barbosa
Aldenizio
Santana
Roberto
Batista
Batista
Roberto
Apolonio
Nabe
Isaura
Settlement
family
link
Settlement
Tacaratu
Farias
Bananal
Castelo
Bom Jardim
Cunha
Barraquinha
Santana
HB
S, SS
S,DH
DH
DH
DH
S, DH
S,S
DH
S
DH
DH
DH
Z
S
DH
DH
WF
WB
F,WB
MS
Maloca
Bom Jardim
Inferno
Bom Jardim
V.Q.Quer
Patoa
Inferno
V. Q.Quer
Barraquinha
Cachoeirinha
V.Quem Quer
Sao Salvador
Cafe Xique
Degredo
Degredo
Tacaratu
Cachoeirinha
Bananal
Inferno
Bananal
Camaleao
Cachoeirinha
WB
WB
Machadinho
Barraquinha
cocal
Os piqui
Camaleao
Barraquinha
Degredo
Manteiga
Boa Hora
Os piqui
Barraquinha
C Xique
c.xique
C dos Baixos
Os Pires
Guariuba
Fadario
Manteiga
Morada Nova
Jarana
S.Salvador
Cafe Xique
C.dos Baixos
Nascimento
Patoa
Carmina
Inferno
Feitosa
Linked family
TABLE 7.7
MARRIAGES
Houses
Cunha
Cunha
Cunha
Carmina
Castelo
Cast./Santana
Santana
Nascimento
Cunha
Isaura
Isaura
Santana
Farias
Santana
Castelo
Castelo
Castelo
Castelo
Cassiano
Santana
Santana
Ginu
Ginu
Farias
Farias
Farias
Farias
Santana
Leao
Cunha
Cassiano
Cunha
Cunha
Nascimento
Nascimento
Cacundo
Cunha
Castelo
*Nascimento
**Fonten'e le
**Val
Isaura
Number of marriages in
Generation
III
11
I
(1)
2
4
(1 )
1
(1)
3
3
2
2
1
1
1
3
3
2
TABLE 7.7a
Santana
Isaura
Peba
Valdeci
Castelo
Castelo
=
=
=
=
=
=
Santana
Santana
Zulmira
Maria
Cunha
Cunha
(FBD)
(MZD)
(MZSD)
(MZSD)
(ZD)
(BD)
TABLE 7.9
33%
14
67%
Houses involved
0
2
7
16 in, 13 out
2
6
6
checked
Total of settlements
.
21
100%
Name of
Settlement
Deposito
Tacaratu
C.Jarana
Cachoeirinha
Bom Jardim
Inferno
V.Q.Quer
Degredo
Os Piqui
Cocal
Divisao
Guariuba
Cafe Xique
Fadario
Os Pires
Morada Nova
Duels Bocas
Floresta
Barraquinha
4
4
1
3
3
3
2
4
4
1
3
1
2
4
2
2
2
4
4
7
7
3
3
4
5
3
9
5
1
5
2
3
6
4
2
2
10
4
7 4
4 2
2 2
5 1
5 1
6 11
3 6
5 1
5 4
2 2
3 6
1 2
3 2
3 3
2 6
2 4
1 2
10 6
5 5
1
6
3
2
4
5
4 .
2
3
2
3
4
4
4
0
4
2
6
10
1
0
0
0
0
0
0
0
0
0
3
0
0
0
0
0
2
0
0
85 74 70 69 6
Total
53
m f h
19 settlements -House M F
Source: field research, 1982/84
20
19
10
11
14
27
16
17
17
7
20
9
12
16
12
12
9
32
24
304
Total
TABLE 8.1
Age
Adults
Over 60
37 to 48
15 to 18
10 to 14
Children
3 to 5
under 1
Total
Male
Female
Total
0
1
3
1
1
1
0
2
1
2
3
3
2
0
1
1
3
1
13
H
A,B,C
e,f
4
1
1
2
4
11
339
338
days
team
Fell trees
Gather the palm
Raise the structure
Cover with palm
Place the walls
Total
2
1
2
1
1
man days
4
2
4
4
4?
18
Walls, floor and ceiling materials are made from the paxiuba palm-tree
piles and the house framework are made from different hard woods; stro~g
tree-barks (enviras such as "iron envira") or vines (cip6s) replace nails
to fix together the structure. Poles and framework last 10 years; roof
floor and walls last half this time.
'
Activity
Garden 'Plot
preparation
Weeding
Sub-total
Area
Product
69
0.3 ha
3000 covas
42
27
60 paneiros
48
Flour making
~
Product
Man-Days '
0.3
17
Month
Working team
Days
Man days
4
1.5
0.5
4
4
20
6
1
8
6
14.0
41
August
August
September
September
September
Total
Aug.-Sep. 9 persons
Tobacco
Head,
Head,
Whole
Head,
Head,
3 sons+son-in-Iaw
1 mature son
macro-house
3 daughterss
wife
Soil
Mata bruta
Manioc
Maize
Rice
Banana '
Tobacco
Area
unit
1,000
1,5
6
25
100
pits (stems)
litres(seeds)
litres(seeds)
bushes
plants (lost to pigs)
TABLE 8.3B WORK AND' MAIN CROPS, MANIOC GARDEN 11, CURRENT YEAR
Crops
Soil
Area
2,000
2,5
10
unit
Days
Man days
21 dias
pits (stems)
litres (seeds)
bushes
I'
341
340
Activity
Month
Working team
Day
team days
c<;itch
LimQa
LimQa
?
?
10
6.5
Total
5 persons
16.5
Friday
Saturday
Sunday
Total of week
A,H,B,C,D
H
A,H,B,C,D
3 days
7
2 armadillos
nothing
2 peccaries (T.tajacu) 35
42 kgs 2.47* 3.0**
4 individuals
Days
Man days
mass
p/person
total
Soil
Current
Mandioca
Milho
Banana
mata bruta
500 pits
o litres
Many
unit
(stems)
(seeds)
(bushes)
Left from
3,100 pits
4 litres
60 bushes
TABLE 8.4
Consuming unit
Calculated
Persons
Adults Children
House
House
Total
9
2
4
2
11
Amount
kgs/month
kgs/month
p/person
TABLE 8.6
Trail Names
Rubber trees
Latex*
Pao
Braba
Estradinha
De Baixo
225
130
114
111
18
12
8
12
1/2
1/2
1/2
Total
Average
Average**
580
(4.83)
145
1. 21
50
12.5
10.33
3/2
0.5
0.61
1/4
1/4
1/2
1/2
B
17
1/2
1/2
3/4
0.75
0.91
1
3/4
0.38 0.5
0.46 0.61
*latex yield per day, measured in latas (one lata=2 liters=l kg smoked
rubber, not dry).
** trails converted in standard sizes of 120 trees.
H=head; A~B,C male sons, ages given in parenthese~.
Source: field research. Riozinho, settlement Floresta, 1982.
TABLE 8.6B
Pao
Braba
Estradinha
De Baixo
Tuesday
Wednesd
Thursday
A+H+C
B(1/2)
H+C(1/2)
A+B
H+C
A+B
Friday
H+C
A+B
I,
TABLE 8.6E
TABLE 8.6C
Pao
Braba
EstradinhaA+C
De Baixo
Tuesday
Wednesd
A+C
B+C(1/2)
Thursday
A+H
A+H
A+B A+B+C
all+D
C+D
I
TABLE 8.7 PURCHASES (DRY SEASON HARVEST ONLY; TO AUGUST)
Purchases at
A 15HP motor
Total at the
Purchases at
Trail
Pao
Braba
Estradinha
De Baixo
Total
kgs
kgs
kgs
kgs
Purchasing/leisure
Hunting
TABLE 8.6D
254
1,130
50
1,434
clearing
Work
team
Days Man-days
A,C,D
A,C,D
A,C,D
A,C,D
4
2
4
5
15
12
6
12
15
45
Flags,poles
Tapping
Work
team
Days Man-days
Days~
H+C
H+C
H+C
H+C
2
1
4
2
1
1
2
2
10
50
12
50
52
164
TABLE
Item
Amount
oil
Kerosene
Salt
Soap
sugar
Coffee
powder milk
Ammunition
Gunpowder
Gun shot
Detonator
Metal caps
4
3
1
2
2
3
3
caixas
latas
saca
caixas
sacas
kgs
caixas
43.2 litres
54 litres
30 kgs
60 kgs
36 cans
1,5 kgs
4 kgs
3 boxes
10
I
\
TABLE
3
3
1
2
1
1
1
4
1
2
1
8.7B
owner
dogs (good for armadillos and paca)
old shotguns
A, B, C
fish net
motors (16HP, 14HP)
A
singer machine
radio
disc-player
long-plays
ca~oe (1 ton of capacity)
wrlst-watches (Citizen orient) A,
C
broken wrist-wat.c h . (SeYkO)
1 ax
4 machetes
2
4
1
1
1
4
hoes
knifes (forest knives, not rubber knives)
hammer
saw
pliers
battery-powered torchs
TABLE 8.7C
PRODUCTION PLAN
setor
Days
Man - days
Product
Rubber
Manioc-flour
Gardens
184
28
21
184*
56
58.5
1,434 kgs
1,260 kgs
Total
Total days
233
547.5
846**
*Note that teams ~hared a single trail. Thus two or three men would
contribute together with one man-day. source;field work, 1982. Not
plants (nonkinds) ,
fruit-trees
course)
Batatas (Roots)
Manioc
Yellow (4 named kinds)
White (5 named kinds)
Other (5 named kinds)
Braba ("wild", slightly poisonous)
Sweet potato, Inhame
Maize (2 main kinds)
Beans (3 kinds as bush, 2 kinds as creeper)
sugar-cane, bananas (several kinds), pinn~aple,
Papaya, avocado, oaca) , banana (sev~ral.klnds),
pineapple, papaya, avocado, Oaca (tlngul
,
fishing)
CAPOEIRA (Second-growth)
PLANTAS (plants)
At least 12 herbs (medicine, wrapping
manioc, tea)
At least 15 different common pests in first
PRAGAS (pests)
year.
Source: field research. Riozinho, 1987.
347
346
TABLE 8.8b
Environments
MARGEM margin
VARGEM
varzea (covered in floods)
PRAIA
beach
LOMBO (top beach)
PRACA (lower beach)
BARRANCO (river banks)
IGARAPES
PARANAS
IGAPOS
POCO
GROTAS
CACIMBA
(small rivers)
(flooded streams)
(isolated lakes)
(deep places)
(small streams)
(wells)
fishing
fishing
fishing places
water sources
fishing
CACHOEIRA (rapids)
REMANSO (whir pools, pits)
(oxbow lake)
SANGRADOURO (lake channel to river)
TERRA FIRME
(non- flooded banks)
CAPOEIRA (second growth)
TABOCAL (bamboo forest)
MATA BRUTA (mature forest)
CAMPINA (open forest)
RESTINGAS (marshy areas)
TERRAS
(hills)
Vegetation
PRAIA
PRA9A
LOMBO
Soil
Crops
White sand
Dark sand
Black clay
Weak Tobacco
Sandy clay
Red clay
Dark sand
VARZEA
BREJOS
TERRA FIRME
MATA BRUTA
CARANAI
BAMBURRAL
CAPOEIRA
..
TABLE 8.8c
348
No apparent
349
CACA game
VEADO
Deer ~
Veado vermelho/capoeiro
Veado roxo
PORCO DO MATO Peccaries
Por quinho, caititu
Queixada
PACA Paca
Paca - moga (small)
Paca - concha (big)
Paca - de - rabo
edible
Mazama americana
?Mazama gouazoubira reimoso
reimoso
reimoso *
Tayassu tajacu
reimoso **
Tayassu pecari
Agouti paca
Dinomys branickii
reimoso
reimoso
not eaten
ANTA Tapir
CAPlVARA Capybara
Tapirus terrestris
H. hydrochaeris
edible, *
not eaten**
EMBlARA
QUATIPURU Squirrels
Sciuridae
Q. - mandigueiro (edible), Q. - roxo (brings panema) , Q. - ver melho
COTlA Agouti
Dasyproctidae
edible
cotia - preta,c. - rosia(D. fuliginosa?}, Cotiara(Myoprocta pratii}
MACACO* Monkeys
Cebidae
edible
Macaco- prego (Cebus apella), cairara (Cebus albifrons),
Guariba - vermelha(Capelao) (Allouata seniculus), Guariba - preta
(Allouata caraya?), Paruacu (Pithecia monachus), Zogue- zogue
(Callicebus molloch), Macaco - de-cheiro(Saimiri sciureus},
Macaco-da - noite (Aotus sp.), Macaco- barrigudo (Lagothrix
Lagothricha), Uacari or Cara - de-sola (Cacajao calvus
rubicundus, only left bank of Jurua river), Macaco- preto
(Ateles paniscus)
SOlM
Tamarins
Callitrichidae
S. - Bigodeiro (Saguinus imperator,pet), S .-Branco,S.- Preto
'l'ATUS Armadillos
Dasypodidae
T. - verdadeiro (Dasypus septemcinctus) , T.-china (Dasypus sp?),
T. - agu (Dasypus kappleri, panema), T.-rabo-de-couro (Euphractus
sexcinctus), T.-da- grota (Dasypus novemcinctus), T. - canastra (
Priodontes maximus, tabu).
QUANDU (Porcupines)
Coendou prehensilis Rarely eaten
QUATl
Nasua nasua
Rarely eaten
351
350
EMBIARAS (birds)
NAMBUS (Tinamous)
Tinamidae
edible
N.-galinha (Tinamus sp), Sulurinda (cripturellus soui?) N.pr t
(Crypturellus cinereus), N.-relogio (Crypturellus variegatus; a
N.-papo-de-fora (?), N.-azul (Tinamus tao?), N.-serra (?),
'
Macucal
(Tinamus solitarius?), Nambuzinha-vermelha
(Crypturellus erythropus?) .
JURITI (Doves)
Colombidae
Leptotila verreaux
BICHOS-DE-PENA ("Plumed Animals")
Jacu (Guan) (Penelope jacquacu, edible,pet*), Aracua
. (Chac:halacas) (Ortalis gul tata ?), Jacamim (Trumpeter) (Crax
fasc~olata, pet*), Mutum (Curassow) (Mitu mitu
pet edible*)
Cujubim (Piping guan) (Pipile pipile, pet, edible*) ,
,
PAPAGAIO Parrots
rarely eaten
P.-estrela (Amazona sp.), P.-urubu (Pionopsita vulturina?) etc . .
Toucans and Macaws appreciated for beaks and plumage (meat too hard)
(TUCANOS) (Toucans)
T.-grande (Ramphastus toco), T.-~iudo (Ramphastus tucanus),
Arayari (pteroglossus sp), Mayarlco (?)
ARARA (Macaws)
Psyttacidae
rarely eaten
Jandaia (Aratinga guarouba), Pedres (Ara manilata), Arara (Ara
chloroptera), Caninde (Ara sp), Maracana (Ara sp)
(Too small and have a hard meat)
Aratinga aurea etc. not eaten
PERIQUITOS (Parakeet)
not eaten
CURICAS (small Parrots)
Curica (Amazona vinaca), Curica-de-barreiro (Toiuite surda?),
C.Marianita (Pionites l.?)
EMBIARAS (reptiles)
BICHOS DE CASCO ("Shelled Animals")
edible, eggs
Tartaruga (river turtle), Tracaja (river turtle,
male:Capitari), Iaya (small river turtle, male:Ze Prego)
JABOTIS
Jaboti comum, Jaboti-ayu (land turtle)
TEIJU-ACU
edible
with the cUltivated oaca and put in shallow streams to catch small fish.
Harpoons and bow-and-.a rrow used.
(Tarrafa)
(Bicheiro)
(Harpoon)
(Tingui)
FISH
Fish is secondary as food in the Tejo headwaters. Fish is classified
in reimoso and not reimoso. The most common fish is the bottom-feeder
bode. ,sample kinds of fish are grouped according to the main fishing
technlques used. The tarrafa is the cast net. The bicheiro is a hook
which a diver inserts in a fish under fallen trees. Tingui may be made
DOLPHINS
Dolphins (Inia geoffrensis, sotalia fluviatilis) are
abundant on the Jurua, and are not killed or eaten.
GAL<;A (snowy egret), CIGANA ("gypsy woman") (Opisthocomus
hoazin), JABURU (Jabiru stork), URUBU (Vultures).
352
Small birds are only hunted by children. Some are protected, as the JAP
IIM (the yellow-rumped cacique, Cacicus c. cela) , MARIA-DE-BARRO
(ov
en-bird, Furnarius r. rufus) , BACURAU (Night-hawk), UlRAPURU (hunting
charm)
(Repul~iv~ animal~)
50
21
28
0.88
0.37
0.49
0.88
0.37
0.25
0.88
0.37
99
1. 74
1. 50
1. 25
Totals
auxiliaries 1/2
Workers per h ouse, l'ncluding
'liaries
counted ' as ,
Workers per house, auxl
including
auxillarles
not
III Workers per house,
8)
7 house heads did not work in
Data from 57 houses (~rom a total
~'eid research, Riozinho, 1982/83.
rubber (including 2 wldows). Source.
1
I
11
0:
TABLE 9.2
Item
Used by
Amount/
Duration
House
527
300/trail
Tins
1.5
l/tapper
Tap Knife
3
2/tapper
Blades
l/gatherer 2
Pail/Latex
l/ga therer 2.7
Sack/Latex
2.7
Carrying Sling l/ga th.
1.5
Lamp
. .
l/tapper
1.5
Common Knife 1jtapper
1
1
Basin
Total Value
10
10
1
1
1
1
10
10
10
yrs
yrs
yr
yr
yr
yr
yrs
yrs
yrs
Wear
Rubber
Value
cost/
Year
53
.15
3
2
2.7
2.7
.15
.15
.1
80.3
6
8.2
8
10.6
10.6
1.5
2
13
8.0
0.6
8.2
8
10.6
10.6
0.2
0.2
1.3
140.2
47.7
355
354
TABLE 10 . 1
Age of
Head
tapping
6 hrs.
5 am
pause
1 hr.
11 am
collection
4 hrs.
12pm
preparation
1/2 hr.
4 pm
Total
11hrs
5 am4:30 pm
tapping
6 hrs .
5 am
pause
1 hr.
11 am
colI.
4 hrs.
12pm
pause
1 hr.
4 pm
smoking
2 hrs.
5 pm
Total
14hrs
5 am6 pm
tapping/collection
pause
Time:
7 hrs.
Hour:
5 am
1 pm
hr.
pressing
Total
1/2 hr.
8.5 hrs
2pm
5 am2:30 pm
W/cons
1
2
0
0
0
2
2
10 - 19
1. 37
41
1
10
0
30
30
20-29
2
14
0
0
8
6
7
30 - 39
3
21
1
0
13
7
7
40 - 49
1. 25
5
0
0
1
4
4
50 - 59
1
1
0
0
1
1
1
60 - 69
----------------- -------------------------- - ----------- ---84
2
10
23
50
51
Totais
iozinho, 1982. Only male persons over 10 years
Source: field research, R
old included. 51 houses surveyed in 68.
35~
DIAGRAMS
Diagram 5.1
~-----------
CRUZEIRO DO SUL.-------------------+--------------------------,
$ 30,000 (10% OUTPUT)
RUBBER TRADE
COMPANY
IN
RUBBER
.----------------TEJO
RIVER----------------------+-----------------------~
(10 % AVERAGE
TOTAL OUTPUT)
ttlRACEMA
llMARANGUAPE
IN
RUBBER
00
361
Diagram 5.2
SERINGAL STRUCfURE: THE FLOW OF MERCHANDISES
Diagram 6.1
70..0.0.0.
NATURAL
RUBBER
IMPDRTS
(ASIA)
SYNTHETIC
RUBBER
(B RAZIL)
PIRELLl
GDDDYEAR
ETC.
12
NATU RAL RUBBE R,
<J.LI.:..c..,u.~ AMA ZDN WILD
10.000.
11
10
1.20.0.
CAMELI
RUBBER
.,
X
MILL
RUBBER
MILL
"~
0
:r:
'0
...
"E
.Cl
:J
5
.3
2
REGATAD
(ITINERANT
TRADERS)
30.0.
b TEJO
10.0.
MARRETEIRD
(FDREST TRADER)
200
400
363
362
Diagram 6.2
Diagram 6.3
FINAL DEBTS
10
17
16
15
13
14
....
~
:l
In
::l
Z
11
10
:l
0
J:
'0
lJ
....0
12
.,t'
U
il
E
::l
o~~~~~~~~~~~~~~~~~-J
100
300
400
500
700
800
900
1300
1 400
1 500
O~~~~~~LULf~~-.~~~~~LL-.--~-'~~
-800 -600 -400 -200
2 00
400
600
800
1000
1200
1400 1 600
1700
Diagram 9.1
lE)
\D
84 days
= -120
48 days
102 days
36 days
MILK
"BREAK"
15C
days
FLOODS
1
"
: PREPARE
/,
,I
./
'
"
...
Q.~
.' , ,.'....,J;
l<;t~
,.
. ' /'
./
RAINS
DRY
//0/~
RAINS
SEASON
:RIVER
V~ GARDENS/ /
: GARDENS ~ /
/ / / / / / /
MANIOC + MAIZE
: BEANS:
HUNTING
,,"
'RIVER
GARDENS
HARVEST
MAIZE
FISHING
HARVEST
HUNT IN G
(FAT GAME)
NEW
HOUSE
INCREASING
DE BT
MOVING
\D
<J .'
."
.'
c
U)
r-.
p.,
E
~
U)
.-;
bh
"~
Cl
'~
0:::
J:
DELIVER
BALANCE
THE
PURCHASES
RUBBER
TRIPS
367
366
Diagram 1O.lb
Diagram 1O.la
Rkninho 1983
11
10
r:::..t
u
ill
:J
.,..,
Ii
:x:
-'"
..
en
Cl
.Ll
:>
0
E
:J
'0
'-
I)
.lJ
:3
:J
(5
Ki ll s in
5.-----~-
,--T_________~~~~1~~~8~J--------------------------~
~~:
4.~ .
r~
.~.
!,;
s
::l
0
'0
I
Ii
t /'
'I
t:;;;
rf
I
I
"
;::
100
200
JOO
400
000
9
0
10
11
12
13
Year In t he Somple
14
15
16
17
18
19
20
369
368
Diagram 10.2
Diagram lO.3a
"
1.8
1.2
1.7
1 .~
1. 1
1.5
1.4
~
0
:t:.-.,
..
+I
.x.~g
1 . .3
III
:l
3;~
.. "J
0.7
v c
0.0
0.9
0>
O.E!
0.8
1:
1.1
a:: '-"
1.2
.0
::>1-
0.9
0
0
0
0.5
0.4
0.3
:J
0.$
2~
:J ' - '
05
0>
0.4
I:]
"
a::
0.7
06
..
0 .3
Cl
0
0.2
01
rp
.3
Work<)rs p<)r House
:5
1
1,
I
I:]
0
I:]
t
1
8
0
.3
O-lfl
.r
371
370
Diagram lO.3b
Diagram 10.4
. CONSUMER PER HOUSE AND OUTPUT PER HOUSE (IN KGS RUBBER)
1. 2
1.1
Cl
5
0.9
...
"o
3=~
...
"
" c
~g
... :l
ll:~
ob
0
c.,
~
'1
21
j
0 .8
.:>:
... r-.
(;
III
~ ~
"0..;:1
tu
... Q
VL
0
0
.g'-'
...
n::
0
0
.,
f-
0 .7
0
0
o. ~
0. 5
0 .3
Cl
0 .4
C
0
0
Cl
0.2
Cl
Cl
0 .1
r
2
::;
Wo rkers p<')r Ho u se ( Mole ove r 1 0 )
4
Consumers per Hous e
0
Houees
(5
373
372
Diagram 10.5b
Diagram 1O.5a
1.2
1.1
1
I
0 .9
~ ,.,.
"V
o c
0.8
~ ::I
0.8
0.7
.IJ~
~'-'
/11
'IJ>
:!<:
0
0
8
0
O.J
...
'.
:J
III
'c
2 -
DD
0
0.2
r::fl
0.1
oG
0
0
0
3-
irl
0
Cl.Q
.. 'U
13
0.4
~Ifi'"
0.5
4 -
-'L
0. 6
.0 0
....,
0
5-
1 -f
CD
0
0
0
0
.}
5
Consumer/Worker Rote
375
374
Diagram lO.6b
Diagram 1O.6a
Ruot-~r
7 Cl
6 -
Cl
6 -
Cl
Cl
Cl
:l
~~
O'l
.... :l
Cl
~1
10
.. .
5Cl
4 -
Cl
Cl
.~
Cl
Cl
"~o
c
\ ' f-
C
::;;
....
"'V
n:~
0'-"
4 -
3-
CJ
IIJI
"
., 411
.. :l
CJ
CJ
Cl
2 -
Cl
Cl
2 -
Cl
Cl
0.7
0.9
1.1
.
(Thouoon<is)
K9S Rue,o.,r per House
1.3
T
1.5
CD
Cl
Cl
0.5
Cl
Cl
C
Cl 0
0.3
Cl
Cl
Cl
t -
Cl
0
1 - Cl
Cl
Cl
"'P
1r~
vC
0
C
[P
Cl
Cl
Cl
Cl
Cl
1.7
0.2
0.4
T
O.~
0.8
1.2
(Th<:lusan<:!s)
KQ5 RUer per Worker
'\
376
Diagram 1O.6c
HOUSE PRODUCfION PLANS (RUBBER, DEER KILLS)
~rif)9Q1
'-
20
'~,
1~ 1 !!I -
"-
"- '-
,7 -
16 15 H~
RI'nJnno 1~J
',,-
1.:5 12 -
0
'-
'-
"
"
""~
11 10!J-
'-" "
8-
""
7-
e5-
_1
0
"
,
0
o co
18
0 .1
4.:5- 0
2-
o.~
.~
Cl
, 0
aoo
0.0
1..:5
0.9
0.7
(Tho.UlOoOnd~)
R~r
Cl
1.1
"
1.0
.,
,(
1.7
,/
Ol"ll:!;ut (k<J5)
He'uGoeoG!
,
/
\
/0
\
\
/.
'"
/.
..
/
/
/
T-
,
/
/.
,,
:r
/
/.
,"
/
"
"/
'b.....0' ,
/.
"
"
\tI /
,
/.
()
.f .
.::;.-
to
-32383
23187
Totals
-675 483.062
Averages
419 316.553
Stdeviatl
48
47
Cases
-2035
65
Mln
-100
Max
1?85
2790
59
l"l8
48
0
8"l7
cl
28024
584
455
48
52
2109
-225
-2862 -605
-4967
-4.7872
-103 -130 .09 -12.8
390 292.151 45.03 15.8099
48
48
"l8
"l8
-60
-387 -260
-656
0
0
11 08
1662
RIOZINHO
Rubbe~
84
320 - 212
207
:300
362
81.5
600
420
23
22
:36
46
72
400
600
200
;;.!00
400
24
40
1.42
20
260
t60
~!4
72
295
1.13
18
1.5
:37
152
Pat~on
Rubbe~
Rubber p/worker
Total (half)
576
75.3
700
154
1600
520
320
324
1400
1600
212
500
207
490
i ~:;0
760
120.7
1400
1315
1000
:::i30
616
300
1000
420
570
1.200
1310
133.3
900
530
600
520
200
600
200
631
400
490
215
124
400
454
735
B37
1246
71
700
315
420
260
462
53.3
480
50 ;1
570
595
36
654
700
1750
295
564
1. "t;.,.C"
600
631
400
490
124
454
~~7
126
15
1 ;;~6
40
20
12095
'1490<1
7276
20808
2 '. 9
7868
20933
2.7
1 .2
132
150
2737.1
56795
1.1
2. 1
1300
650
900
300
700
600
380
700
600
380
3
1
350
31 ~:i
420
46;':'!
160
B00
360
;2 61
50-100
0-50
Number
Area
Avg.Area
800
360
200-500
500 - 1000
~r
2219
1<18338
66.8
955
25906
27.1
5213
671588
128.8
4139
1371090
331 .3
736
480353
652.7
219<155
2239.3
2697275
203.<1
---------------------------------------------------------------------100"10
1"10
6"fo
650
3 "r.:"-...J
700
233.3
17"10
7"10
1"10 .
Number
Area
D. Working Hours
5"fo
3 2 ~5
31"10
51"10
8"10
18"10
100"10
Total
650
39"10
25"10
------------------------------------------------------------------------
50:3
570
595
3 27
233.:3
350 .
564
-Area
- ------------------------- - ---------- --- ---- -- ---------------- - ~ -----13262
98
Hours
Number
0-15
152
23053
15-29
243
30-39
3244
40-48
11103
+49
No data
8107
354
-----------------------------------------------------------------------500
250
5 00
500
250
500
~.
o
.- }
"-
9 \.1 0
'1
t:...c' 1'::'
. \J
16
1
800
600
600
;.:.!;.~0
2~:.~0
~500
~3 00
71.0.9
77'1
Number
Area
Avg.Area
400
7
-4
5 66.1.
E.
Women
Men
Over 14
Over 14
Workers
38366
24273
14093 .
18280
10132
")
(:04
311.~5
:3 3L ;-:;
~2 ~1 :3. 3
212.3
1.68,,5
43 .. 3
75
Yield
Hunt ing
(Deer)
3 67 ~:j
418.5
63
49
16
520
12 2
317
849
:337
:t :30
j.50
26~-5
11.3
39
40
500
570
1200
436.7
300
152
270
;.:.!70
1044
663
7 00
RIOZINH O
1 983
Rubber
Maniac
Man i acl
p Iwor- k el~ Current
Wor ke r
(tot a l)
10 pits
144
800
200
700
320
400
80
104
324
300
300
700
400
500
600
600
490
300
300
400
3130
200
466.7
70 0
233.3
1000
740
740
530
- - - - -- - - - - 4 .. 6
BIBLIOGRAPHY
112
831
5033
5810
3474
2619
--------- -136
______________
_ _____________________
1215
Total; 19230 work e r s wit h a n ' n
-W
- --------------------- - ------ - -- I com e.
ork e r s without I ncome;4145.
TABL E
ACRE 198 2
__________________ _ _ _
ALLEGRETTI, Mary H., MAY, P.H. and RUEDA, Juan C. 1990. Avaliar;ii.o e
Redirecionamento da PoUtica para a Borracha na Amazonia. Curitiba:
Instituto de Estudos Amazonicos/Conselho Nacional dos Seringueiros,
manuscrito.
ALMEIDA, Mauro William Barbosa de. 1984. "Labour Control in Contemporary
Amazon: the Case of Rubber Tappers". Paper presented at the Sermnar of
Latin-American History, London School of Economics.
~ ~ = ~~~
~ ~~ =: ~~
5 00 ~ 1 000
25906
148338
671 5 88 13 710 9 0
4 803 53
_____________ : ~ ~~-- ___ ~~ ~~ ____ ~ 28 . 8 '
331 . 3
652 .7
2 239.3
2 03 .4
Numbe r
7~
17~
-; ~~------;~~------- ~~----------------~--
Ar ea
Num b e r
:~: ~ Ar..
- 955
r~
22 1 9
5~
~---------- -- ---------
52 1 3 "
4 13 9
25\
5 1'Jo
736
18 \
2 194~~ 26~~~~~
~~
~~~~
378
1990. "AS Coloca~6es como Forma Social, Sistema Tecno16gico e Unidade '
de Recursos Nat~rais". T~"a Indfg~na, Ano 7, n.S4, pp. 29-39. (Originally a
paper presented m the FIrst Meetmg on Environment, Rio Brancoj Acre
January 1988.)
,
~~:2.
ANDRADE, 9nofre de. 1937. Amazonia: Esbor;o Historico, Geographia Physica Geographza Humana e Ethnographia do Rio Jurua. Macei6: Officina Graphica
da Casa Ramalho.
BARLETT, Peggy F.
APP ADU~, Arjun. 1986. The Social Life of Things: Commodities in Cultural Perspectzve. Cambridge: Cambridge University Press. .
BARLOW, Colin. 1978. The natural rubber industry: its ~evelppment, technology, and
economy in Malaysia. Kuala Lumpur: Oxford UruversIty Press.
AMAZONAS,
Cap.
Louren~o
da Silva.
1852. Diccionario
TC!pograph~co, Historico, Descriptivo Da Comarca do Alto-Amazonas. Recife:
Tlpographla Commercial de Meira Henriques.
BATES, H.W. 1895 (1863). The Naturalist on the River Amazon. London, John
Murray.
BAUMOL, William. 1979. "Preface" toAn Introduction to Karl arx's Das Kapital: A
Study in Formalisation, by Gerard Maarek. Oxford: Martm Robertson.
ARAUJO
ARISTO~E ..1980.
Richard E., FA..~G, Tula G. and mANE.Z, Luis M. 1988. "Primates and
Ungulates: A Comparison of Susceptibility to Hunting". Primate Conservation
9(1988):79-80.
California Press.
BOEKE, J.H. 1953 Economics and Economic Policy of Dual Societies. Haarlem:
. H.D.Tjeenk "Willink & Zoon N.V.
CHANDLESS, W. 1867. "Notes on the River Aq~iry, the. principal Afflu~nt of the
River Purus". Journal of the Royal Geographzcal Soczety, vol.XXXVI. 119-28.
1867. "Ascent of the River Purus". Journal of the Royal Geographical Society,
vol. XXXVI:86-118.
1869.
BUNKER, Stephen G. 1985. Underdeveloping the Amazon: Extraction, Unequal Exchange, and the Failure of the Modem State. Urbana and Chicago: The
University of Illinois Press. .
CHAY ANOV,A V. 1986. On The Theory of Peasant Economy. Wisconsin: The University of Wisconsis Press.
CHAYES Avelino de M. 1913. Explorac;ii.o da Hevea no Territorio Federql do
Ac;e/Monographia Apresentada a Exposir;ii.o Nacional de Bo"acha. RlO de
Janeiro, n.d.
CHEVALIER, Jacques M. 1982. Civilization and the Stolen Gift: Capital, Kin, and
Cult in Eastern Peru. Toronto: University of Toronto Press.
p:.,
COATES, Austin. 1987. The Commerce in Rubber: The First 250 Years. Oxford and
Singapore: Oxford University Press.
DOUGLAS Mary & ISHERWOOD, Baron. 1978. The World of GOffs: Towards an
Anth;opology of Consumption. Penguin Books: Harmondswort .
Ro Branco Casa da
A
DUARTE, Elio G. 1987. Conflitos pela Terra no cre. 1
.
Amazonia.
.
h
Peasants, and Economic
DURRENBERGER, E. Paul (Ed.) .1984. C ayanov,
Anthropology. London: Acadermc Press.
(1903-1945).
1986 (1906). "Relat6rio da Comissao Mista Brasileiro-Pemana de Reconhecimento do Alto Pums." In: Um Parafso Perdido: Ensaios, Estudos e
Pronunciamentos sobre a Amazonia. Rio de Janeiro:Jose Olympio Editora e
Rio Branco:Funda<;ao de Desenvolvimento de Desenvolvimento de Recursos
Humanos.
DALY, Herman (ed.). 1980. Economics, EcoloffY' Ethics. Essays Toward a Steady. . State Econo'!lY. New York and San FranCISCO: W.H.Freeman.
DASGUPTA, P.S. & HEAL, G.M. 1979. Economic Theory and Exhaustible
Resources. Cambridge: Cambridge University Press.
DEAN, Warren. 1987. Brazil and the Struggle for Rubber: A Study in Environmental
History. Cambridge: Cambridge University Press.
DEBREU, G. 1959. A Theory of Value. New Haven and London: Yale University
Press.
DENEVAN, W. 1984. "Ecological Heterogeneity and Horizontal Zonation of
Agriculture in the Amazon Floodplain". In Frontier Expansion in Amazonia,
M.Schmink and C. H. Wood (orgs.). Gainesville: University of Florida
Press:~ 11-336.
ELSTER, Jon. 1983. Explaining. Tec~nical Change. Cambridge and Oslo: Cambridge
University Press and Uruversltetsforlaget.
1985. Making Sense of Marx. Cambri~ge and P~ris: Cambridge University
Press and Editions de la Maison des SCIences de 1 Homme.
1986. (ed.) Rational Choice. New York: New York University Press.
0t.m~iM~~~~u~:~~~el~~~e~~ri~: C2~ci~~~d~~
.
h'
f th e Bolivian rubber boom and
FIFER, V. 1.970. f"Th eHempirefbsUlaldree~~: J~u~~rXI' 1atin American Studies 2: 113-46.
the nse 0 t h e ouse 0 u .
'j
FIRTH,Routledge
Raymond.& 1966(1946)
Kegan Pa~l.U aI
ay '
Fzshennen: Thelr. Peasant Economy. London:
1970. Time
and Social
Stru tup~~s~.n d 0 ther Essays. London: The Athlone
Press,
New York:
Humanitie~
AndreBo~ks.
G 1971(1969) . Caplta
. rlsm and Underdevelopment in Latin America.
FRANK,
Penguin
.
Press.
GROSS, S. 1969. The Economic Life of the Estado do Maranhiio & Griio Para. Ph. D.
Thesis, Tulane University. Ann Arbor: University Microfilms . .
GUDEMAN, Stephen. 1978. The Demise of a Rural Economy: From Subsistence to
Capitalism in a Latin American Village. London: Routledge & Kegan Paul.
GUDEMAN, Stephen & RIVERA, Alberto. 1990. Conversations in Colombia. The
Domestic Economy in Life and Text. Cambridge: Cambridge University
Press.
.
GUERRA, Antanio T. 1955. Estudo Geografico do Territ6rio do Acre. Rio de
Janeiro: Instituto Brasileiro de Geografia e Estatlstica.
GUIOMARD DOS SANTOS. 1946. Papers. Centro de Hist6ria da Universidade
Federal do Acre, Rio Branco.
HALL, Arlthony & GOODMAN, David (eds.). 1990. The Future of Amazonia:
Destruction or Sustainable Development? London: M acMill an.
HAMES, R. B. and VICKERS, W.T. 1983. Adaptive Responses of Native
Amazonians. New York: Academic Press.
HARDENBURG, WaIter E. 1912. The putumayo: The Devil's Paradise. London: T.
Fisher Unwin.
HARDESTY,D.E. 1975. "The Niche Concept: Suggestions for its use in Studies of
Human Ecology". Human Ecology 3(1975):71-85.
HART, J. Keith. 1974. "Migration and the Opportunity Structure: a Ghanaian Casestudy". In Samir Amin (ed.), Modern Migrations in Western Africa,
International African Institute and Oxford University Press.
GOW, economy".
Peter. 1989.'''The
Perverse Child'. De'"
. Subsistence '
Man, 24(4):567-82.
SIre m a NatIve Amazoruan
GRAMSCI,
Antonio. 1975 . Quaderni Del Carcere. 4 volumes. Torino: Einaudi
Editore.
HECHT Susanna and SCHWARTZMAN, Stephen. 1988. "The Good, the Bad and
the Ugly: Amazonian Extraction, Colonist Agriculture, and Livestock in
Comparative Perspective." Unpublished Manuscript.
HECHT Susanna NOORGARD, Richard, and POSSIO, Giorgio. 1988. "The
E~onomics 'of Cattle Ranching in Eastern Amazonia". Interciencia, Vol. 13,
No. 5. Caracas.
HELMS, Mary W. 1969. "The Purchase Society: Adaptation to Economic Frontiers".
Anthropological Quarlerly, Vol. 42 (4): 325-342.
HERDON, W. and Gibbon, L. 2 vols. 1853-1854. Exploration of the Valley of the
Amazon. Washington.
HERRERA, R., JORDAN, c.F., KLINGE, H: a~d M~DINA,. E. 1978. "Am.azon
Ecosystems. Their Structure and FunctIorung WIth PartIcular EmphasIs on
Nutnents." Interciencia, 3(4):223-323, jul-aug.1978.
HILFERDING; Rudolf. 1968 (1909). Das Finanzkapital, 2 volumes. Frankfurt am Main: Europaeische Verlagsanstalt.
HILL, Polly. 1970. Studies in Rural Capitalism in West Africa. Cambridge:
Cambridge University Press.
HILL, R.D. 1982. Agriculture in the Malaysian Region. Budapest: Akademiai Kiad6.
HIRSCHMAN Albert O. 1977. "A Generalized Linkage Approach to
Development, with Special Reference to Staples". In M.arming Nash (ed.),
Essays on Economic Development and Cultural Changes m Honor of Bert F.
Hoselitz. Chicago: University of Chicago Press.
HOMMA, AIfredo K.O. 1990. "A Sustentabilidade do Sistema Extrativista .na
Floresta Amazonica". Bel~m (Brazil): EMBRAPA e Centro de PesqUlsa
Agropecuaria do Tr6pico Omido.
HORNBORG AIf. 1992. "Machine Fetishism, Value, and the Image of Unlimited
Good: Towards a Thermodynamics of Imperialism". Man (N.S.), 27, pp.1-18.
HUGH-JONES, Stephen. 1989. "Yesterday's Luxuries, Tomorrow's Necessities;
Business and Barter in Northwest Amazonia.". Manuscript. Paper for a
volume of essays on barter to be edited by Caroline Humphrey anri Stephen Hugh-Jones.
IBAMA. 1988:Anuario Estatfstico-Mercado da Borracha. Brasflia: IBAMA.
IBDF. 1988. Alterm;iio da CoberlUra Vegetal Natural do Estado do Acre/ Relat6rio
Tecnico. Brasflia: Instituto Brasileiro de Desenvolvimento Florestal (IBDF)Departamento de Econornia Florestal.
IBGE. 1981. Sinopse Estatistica da Regiiio. N,.Orl~. Rio de Ja~eiro: Rio ~e Jan~iro:
Secretaria de Planejamento da Presldencla da Repubhca/Funda~ao Instltuto
Brasileiro de Geografia e Estatfstica - IBGE.
'1 1980/Volume l/Tomo 5/Numero 3 1983. IX Recensamento Geral do Brasl - R'o de Janeiro: Secretaria de
Censo Demografico-'1iic:-d~-Odbr~ ~bli'ca/~unda~ao Instituto Brasileiro de
Planejamento da ~r~sldencta a epu
Geografia e -EstatIstlca .- lBGE.
'l 1980/volume 2/Tomo 3/Numero 3
1983b. IX Recensa;n.e~to Geral.dod~rfa~;iro: Secretaria de PlaneJament? da
Instituto Brasileiro de Geografla e
- Censo Agropecuano: A~re/. FRlOd
Presidencia da RepublIca un a~ao
.
Estatistica -IBGE
.
.
e 10-1982. Rio de Janeiro:Secr~ta~Ia
1984. Produc;iio ExtratIva ,V~get,al-ro~m ublica/Funda~ao Instituto Brasllelfo
de Planejamento da ~r~sldencta a ep
de Geografia e EstatIstIca.
,
- VI II Repert6rio Estatistico do
1986. Series Estatfsticas, Retrospec,tlVfiom ~~ 'arata do Anuario E.st~tfstico
Brasil/QLfadros Retrospectlvos (R)ep~fot de J an~iro: Instituto Brasllelfo de
do Brasll, Ano V, 1939-1940.
Geografia e Estatfstica.
, '
. "
d B '1' Series Economicas, Demograficas. e
1987. Estatfsticas HLSton(cas ~'t' raslr'evised from lBGE 1986, Vol. 3.) RIO
d
Sociais qe 1550.a 1988. ~lnl'rOe Jel~~ografia ~ Estatistica.
B
de JaneIro: InstItuto raSl e
das Potencialidades dos Recursos Naturais. da
1990. Projeto Zonea:nento Janeiro: Instituto Brasileiro de Geografla e
Amazonia Legal. RIO de
Estatistica. .
N,
M nchester Manchester
INGOLD, Tim. 1986. The AppropriatLOn of ature.
a
.
University Press.
--'I he Sertiio' Economics and Dependence
JOHNSON'BAll~I~
Wp' 1197t1atSI'o~r~~fag:~a~ S~anford Uclversity Press.
on a razl zan an
, ' A ' It ral Decision Research". In Peggy
1980. "The Limits of Formahs.~ m ~J,rtdu .uAnthropologicaI Contributions to
BARLETI. Agricultural DeclsLOn 1Yla .ng.
P 19-44
.
Rural Development. New York: Acaderruc Press. p . ,
.
" . R H
and W Vickers (eds.), AdaptIve
1983. "Machiguel!ga Garde~s . IIINe~ ;:::~~ Acaderclc Press. Pp. 29-64.
Responses of NatlveAmazonzans"
.
.
1982 "Nutritional Criteria in Machlguenga
JOHNSON, A. W., ~EHRDEN,S! c. .AA Li~ear Programming Analysis". Human
Food ProductIOn eClSIOns.
Ecology, 10(2):167-89.
.
omy" . Economlc
D IE
JORGENSON,D.W
, 1961. "The Development of a ua con
Journal, 71 (June).
.
.
. h.
t'n Chayanov's Rule III a
JORION, PauL ~98~. "Chayanov .sh,?u1fn bec~I0g ;~o~s Ip~asants, an~ Economic .
French FIshmg commurutyb'
d New York Acaderruc Press. pp.
Anthropology (E. Paul Durren erger, e . ,
'
71-96.
KAHN, Joel S. 1980. Minangkabau Social Formations: Indonesian Peasants and the
. World-Economy. Cambridge: Cambridge University Press.
,
.
nvier 1897-Dec,1898", Bulletin de la
1899. "Missions
~e la ~azfs~9
Congregation
du Saznt-Spnt,
Fev.,J~ome VI , Numero 145, pp.531-552.
KEMENY, John G., MIRKlL, H., SNELL, J.L. and THOMPSON, G.L. 1967. Finite
Mathematical Structures. New Jersey: Prentice-Hall.
~~ura~fo~agg:'~il:'g~~~dUni~:sityp~~~s,
.
LIVING STONE, I. and
0 RD , H '. W. 1981. Agricultural Economics for Tropzca
Africa. London & Ibadan: Hememan.
. .
, Rosa, 1970(1912), A acumulac;iio do capzta
, I, R'10 de JaneIro: Zahar
LUXEMBURG,
Editores.
W B 1988. "Levantamento da
MACEDO,
Antonio
e ,AL~EIDdA,
R~a~
0" in 'Ministerio Publico Federal
Situac;ao
Atual L.B.
dos Sennguelfos
0
10 eJ .
.
1989a, vol. 1(2-32).
, ,
,
C unitario-Reserva Extratlvlsta d? Rw
1988. Desenvolvimento Economz~ . om I dos Seringueiros/Secretana do
Tejo . . Rio BrdancGo:
~ons~~~sta~~l~~~cre. Manuscript.
Plane]amento
0
overno
.
A
1979. The IKung San: Men, Women and Work in a Foraging Society.
LEONTIEFF, WassiIy W. 1966(1941). The Structure of American Economy, 1911939. New York: Oxford University Press. (Second Edition.)
LEVI-STRAUSS,
Claude. 1976(1955). Tristes Tropiques. Harmondsworth: Penguin
Books.
1979. La voie des masques. Paris: PIon.
'
"
MAGALHAE,
- S J uraCI.
'P ...... 1977. A Discnmznac;ao
de T.1 erras na Amazonia (0 Acre).
,,
Brasilia: Senado Federal.
.
A onauts of the Western Pacific: 1n
MALINOWSKI, Bronislaw. 19.61(1922). rg . the Archipelagoes of Melaneszan
Account of Native Enterpnse andAdv~~e zn
New Guinea. New York: E.P.Dutton
o.
, ,
.
, . A. nd
Stud
of th New
Methods
of Dover
Tlllzng
1978(1935). Coral Gardens a11f1 t~ezrhMaFcb
I~ands.
York:
the Soil and of Agricultural Rltes zn t e ro na
Books.
.
Segunda
Guerra
Mundial e
MARTINELLO Pedro.
1988. A Batalha da Borracha naRio
Branco:
Universidade
suas Consequenclas para 0 Tv.fale Amazonzco.
Federal do Acre.
.
A'
'
A'
'
(Bresil),
}~ie~t~~8W~~~,,;!g~~oP~~7u~~
~~~~eI KLoarlM
da~Laand
Frederick E!lgels,
Ltd.
, .
non.
wrence and Wlshart
"I K
1974(1875). "Critique of the Gotha Pro
International and After (ed. David Fernbf~~)mmNe . yn k. aVr~ Marx, The First
339-359.
' ew or. mtage Books, pp.
1976(1933)
of th Imme dlate P rocess of Production", in Capital,
Vol.!.,
Pelican"Results
Marx Libra;
MIY AZAKI, Nobue and ONO, Morio. 1959. "0 Aviamento na Amazania (Estudo
S6cio-Econamico sabre a Produ~ao de Juta)". Sociologia,
MAUSS, Marcel. 1954. The Gift. London: Routledge & Kegan Paul.
MAXWELL,
March Kenneth.
7,pp.24-29.1991 . "The Tragedy of th e Am azon"Th
. .L J e New York Review,
MAY,. Extrativistas"
Peter H. 1989.P "Direitos
deI .Propriedad e e a S0 b reVlVenCla
.
~D
das Economias
. ara esenvo vlmento, vol.25:65-71.
A
McCAY,
Bonnie
& ACHESON
Tucson.
Universlty
of Ar.' Jameps ( e ds. ) 19 90. The Question of the Commons
.
lzona ress.
.
McNICgLL, Geo~~rey. 1990. "Social Organization and Ecological Stability U d
emograp IC Stress." In Geoffrey McNicoll and Mean Cain ( ) n er
Development and Population: Institutions and P r
N
Y ork eds , Rural
The Population Council, Oxford University Pres~,l~. 1:;67.
and Oxford:
MEGGERS
.. .A Ilusao
- de um Para{so (Translated from
Ama~ Betty
. . . 1977. A mazoma.
.:
. 1 om_a. Ma~ a.nd Culture m a Counterfeit Paradise) Rio de J
.
anelro.
CIVI lza~ao BrasIlelra.
McCALLUM
C 1l~i 19Jr tGend~, Perso,!hood and Social Organization amonst the
Cashin~h~~}
Economics.
es em
mazoma. Ph.D. Thesis, London School of
1904-1908" R
Editora
393
MURRAY,
"The Rubb
I'
. Indochina: The Colonial
State M.
and1979
the Class
Stru er P
antatlOns
of Colomal
In The World-System Ofg~:p7t~r:;':ne.npWage-Labor and Capital, 1910-1940".
Beverly Hills and London:Sage Publicat~~s~nd Present, ed. W.L.Goldfrank,
NETTING,R.McC.,
WILK,
R Rand
Households:Comparative and' R"
J\RNOULD,
E.J.
1984.
Berkeley: University of CalifornialS~~~~:1 Studles of the Domestic Group,
PARKER, Edward P.(ed.) 1985. The Amazon Caboclo: Historical and Contemporary
Perspectives. Studies in Third World Societies, vo1.32.
PARRY, J. and BLOCH, M. 1989. Money and the Morality of Exchange. Cambridge:
Cambridge University Press.
P ASINETTI, L. 1987. Lectures on the Theory of Production. Cambridge: Cambridge
University Press.
PEARCE, D.W. 1976. Environmental Economics. London: Longman.
NORDENSKlOLD
E 1922(191)
.
Stuttgart.
,.
1. Indzaner
und Weisse in Nordostbolivien.
NORGAARD,
B 1980 C"AI- orgamza~ao
' - da produ~ao na industria da
borracha Richard
natural" C:
. zencza e u tura, 32(6):790-793.
A
O'DWYER,
Eliane
1989Rio
Parecer
sob re T rab alh? Escravo na Reserva Extrativista
do Alto
Rio Tejo
de Janeiro
_
.
, manuscnpt.
ODUM, Eugene
POPKIN, Samuel L. 1979. The Rational Peasant: The Political Economy of rural
Society in Vietnam. Berkeley: University of California Press.
1986. "The political economy of peasant society". In Jon EIster (ed.), Rational
Choice, New York, New York University Press, pp. 197-247.
."
.
1898a. "Le principal produit de la silva am
Generales de le Congregation du St. Esprit. azomenne. Manuscnpt. Archives
PRADO JR, Caio~ 1979(1960). A questiio agraria no Brasil. Sao Paulo: Editora Brasiliense.
-
1922. "Unen.4,tournee
XXXVIII,
113-114.dans le Bas Ju ru a".
190~. Primeiro
NaclOnal.
Relatorio Semestral
SALISBURY, R.F. 1962. From Stone to Steel. Melbourne and Cambridge: The
University Presses.
1970.
Vunamami: Economic Transfonnation in a Traditional Society.
Melbourne: Melbourne University Press.
:k
RIC~~~0~~~~~~d~9:1t~1:1i)~f:~~b~t;~d~;do~~~jnu~iv~~:i;y ~~~.esia:
RIVIERUE!
Pet.er. Press.
1984. Individual and Society in Guyana. Cambridge: Cambridge
mverslty
RYLANDS, Anthony B. 1990. Evaluation of the C
J.V.T.,
LOPES,
J.R.B.,
MOREIRA,
RJ.,
NAKANO,
Y.
101).
"
SCOTI', James C. 1976. The Moral Economy of the Peasant: Rebellion and
Subsistence in Southeast Asia. New Haven and London: Yale University
Press.
E.x-
VERNANT, Jean-Pierre. 1965. My the et Pensee chez les Grecs: Etudes de Psychologie
Historique. Paris: Fran~ois Maspero.
VICKERS
~92T8.
III
g43. The Midd~e Amazon: Its People and Geography: Eleven Articles by
vYnstq.nt Tastevm. Translated by The Strategic Index of the Americans
ashIllg~on:
anuscnpt.
OffIce
for
Emergency Management/Research
VON IHERING, H. 1904. "0 Rio Jurua". Revista do Museu Paulista, V1:385-460.
Division
'
Brasilia.
VON NEUMANN, J. 1971(1938). "A Model of General Economic Equpibrium". In
Readings in the Theory of Growth (ed. F.H.Hahn), London:MacMlllan, pp.1-9.
TAUSSIG, Mich.ael. 1980.. The. Devil and Commodity Fetishism in South America.
Chapel HIll: The UruversIty of North Carolina Press.
Sham~nism, Coionialism, and the Wild Man - A Study in Terror and ealmg. ChIgaco and London: the University of Chicago Press.
M87:
Press.
1989. The Modern--- World-System III: The Seco,!-d Era of Grea.t Expansion of the
Capitalist World-Economy, 1730-1840s. San DIego: AcademIC Press.
THAU~TURGO
DE AZEVEDO. 1905. Prefeitura do Alto Jurua. Primeiro Relatono Semestral (1904). Rio de Janeiro Imprensa Nacional (cf
PREFEITURA DO ALTO JURUA 1905).
'
.
.
THE SOU!H
"
TOCANTI~S, Le~nd:o.
TAX, Sol. 1972 (1953). Penny Capitalism: A Guatemalan Indian Economy. New
York: Octagon Books.
R10
WAGLEY, Charles. 1964 (1953). Amazon Town: A Study of Man in the Tropics. New
York: Alfred A. Knopf.
WlLSON, E.
TROTSKY, Leon. 1967(1934). The History of the Russian Revolution. Vol. I. London: Sphere Books.
California Press.
WOODROFFE, J.F. and SMI'FH, H.H. 1916. The Rubber Industry of the Amazon
and How its Supremacy can be Maitained. London.
XENOPHON. 1990. The Estate-manager. Penguin Books.
CAMBRIDGE
UNIVERSITY LIBRARY
.~f