Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Territorial use-rights in fishing (TURFs) and the management of small-scale fisheries : the case of Lake… Levieil, Dominique P. 1987

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1987_A1 L48.pdf [ 13.8MB ]
Metadata
JSON: 831-1.0097449.json
JSON-LD: 831-1.0097449-ld.json
RDF/XML (Pretty): 831-1.0097449-rdf.xml
RDF/JSON: 831-1.0097449-rdf.json
Turtle: 831-1.0097449-turtle.txt
N-Triples: 831-1.0097449-rdf-ntriples.txt
Original Record: 831-1.0097449-source.json
Full Text
831-1.0097449-fulltext.txt
Citation
831-1.0097449.ris

Full Text

TERRITORIAL USE-RIGHTS IN FISHING (TURFS) AND THE MANAGEMENT OF SMALL-SCALE FISHERIES: THE CASE OF LAKE TITICACA (PERU). By DOMINIQUE P. LEVIEIL DEUG, I n s t i t u t National Agronomique Par is -Gr ignon, 1974 Ing. Agro -Hal ieute , I n s t i t u t National Agronomique, Par is -Gr ignon, 1977 M.A. (Marine A f f a i r ) , Univers i ty of Washington, 1979 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (School of Community and Regional Planning) We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA June 1987 © Dominique P. L e v i e i l , 1987 4 6 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date DE-6(3/81) ABSTRACT The purpose of t h i s thesis i s to evaluate whether the T e r r i t o r i a l Use-Rights in F ishing (TURFs) of Lake T i t i e a c a , Peru, are e f f e c t i v e in overcoming the common property problem of t yp ica l f i s h e r i e s and therefore whether TURFs may prove va luable as part of a more formal management system. I t has recent ly been argued that TURFs should be incorporated into s m a l l - s c a l e f i s h e r i e s management schemes since they should be e f f e c t i v e in c o n t r o l l i n g f i sh ing e f f o r t , in promoting a more equi tab le d i s t r i b u t i o n of the benef i ts from f i s h i n g and in reducing admin is t ra t i ve i n e f f i c i e n c i e s . To determine whether TURFs are in f a c t e f f e c t i v e in c o n t r o l l i n g f i s h i n g e f f o r t , I examine Lake T i t i eaca f i s h e r i e s in Peru. F i r s t , I demonstrate the widespread existence of Lake T i t i caca 's TURFs and t h e i r control over the ent i re shore l ine , most of the l i t t o r a l area and even part of the pe lag ic area. Second, I document how, in sp i te of TURFs' i l l e g a l i t y , Lake T i t ieaca shore dwel le rs are able to combine lega l and i l l e g a l means to enforce t h e i r t r a d i t i o n a l ri-ghts over the i r f i s h i n g areas. Th i rd , by showing that the r e l a t i v e difference between the returns to labour from f i sh ing with those from a l t e r n a t i v e a c t i v i t i e s ranges from 50 to more than 100%, I demonstrate that l o c a l fishermen capture substant ia l f i s h i n g rents. I f one takes in to account that most f i sh ing a c t i v i t i e s are car r ied out when there i s l i t t l e e l s e to do, t h i s range increases to 90-180%. I thus conclude that Lake T i t i eaca f i s h e r i e s have not reached the i r bioeconomic equ i l i b r ium yet and that the predict ions of the common property theory do not apply to them. And fourth, I demonstrate that the or ig ins of these rents can be traced to fishermen's membership in TURF-holding communities, t h e i r a b i l i t y to r e s t r i c t physical access to the shore l ine , and the ob l igat ions associated with t h i s membership. Among these ob l iga t ions are the p a r t i c i p a t i o n in communal projects and ce lebrat ions , the i i f u l f i l l m e n t of administrat ive or ceremonial r e s p o n s i b i l i t i e s , and the undertaking of a g r i c u l t u r a l a c t i v i t i e s , a l l of which constrain the amount of household labour a v a i l a b l e for f i s h i n g . In the concluding sect ion , I consider the potent ia l r o l e of TURFs in a formal management context. I show that , in the long term, even formal ly recognized TURFs would not be s u f f i c i e n t in themselves to prevent overf ish ing. I therefore propose that Lake T i t i e a c a TURFs be incorporated into a broader, decentra l i zed management strategy which would c a p i t a l i z e on the i r strengths and promote cooperation between members of shore communities, f i sher ies s c i e n t i s t s and administrators. i i i TABLE OF CONTENTS page CHAPTER 1: INTRODUCTION 1 1.1. Objectives and r a t i o n a l e 1 1 .2 . Need for appropriate forms of f i sher ies management 2 1.3. T e r r i t o r i a l Use-Rights in F ishing (TURFs) 6 1.4. Effect iveness of e x i s t i n g TURFs. 13 CHAPTER 2: CONTEXT, MATERIAL AND METHODS OF THE STUDY 19 2.1. Natural and human environment of Lake T i t ieaca 19 2.2. Lake T i t ieaca f i s h e r i e s 23 2.3. Material and methods 30 CHAPTER 3: EXISTENCE, DISTRIBUTION AND DIMENSIONS OF TURFS 35 3.1. Existence and d i s t r i b u t i o n of Lake T i t ieaca TURFs 35 3 .2 . Spatial dimension 44 3.3. Resources dimension , 62 3.4. Demographic dimension 72 3.5. Cul tura l dimension 85 CHAPTER 4: ENFORCEMENT OF TURFS 99 4 . 1 . TURFs Legal and admin is t ra t i ve environment 99 4 . 2 . TURF-related forms of ownership 108 4 . 3 . Enforcement of TURFs 114 CHAPTER 5: EVALUATION OF FISHING RENTS 121 5.1. Fishing costs and f i s h i n g returns 121 5 . 2 . Fishing rents 123 5.3. Discussion 130 CHAPTER 6: TURFS AND THE CAPTURE OF FISHING RENTS 138 6 . 1 . Exclusion of i n l a n d dwel le rs from Lake T i t ieaca f i s h e r i e s 138 6 . 2 . Communal constra ints on i n d i v i d u a l f i sh ing e f fo r t 139 6.3. Discussion of a l t e r n a t i v e constraints 152 CHAPTER 7: CONCLUSIONS AND RECOMMENDATIONS 156 7.1. Conclusions 156 7.2. Interpretation of the ef fect iveness of TURFs 159 7.3. TURFs and the management of Lake T i t ieaca f i sher ies 161 BIBLIOGRAPHY 169 APPENDIX A: Orestias species of Lake T i t ieaca watershed 187 APPENDIX B: History of Lake T i t i e a c a f i sher ies 188 APPENDIX C: Research methods - 194 APPENDIX D: Calculat ions of f i s h i n g costs and revenues 200 i v LIST OF TABLES page Table 1. Types of F ishing Economic Units (FEUs) encountered on Lake T i t i c a c a 26 Table 2. D i s t r ibu t ion of answers to quest ion: "Do you encounter d i f f i c u l t i e s when operating in other f i s h i n g areas?" 40 Table 3. D i s t r ibu t ion of r e l a t i v e frequencies for the number of f i sh ing s i tes by group of fishermen 40 Table 4. Shore communities of Lake T i t i c a c a for which there i s evidence of the existence of TURFs 42 Table 5. Frequency d i s t r i b u t i o n of proport ion of shore length to perimeter for Lake T i t i caca shore communities 49 Table 6. D i s t r ibu t ion of frequencies for the number of f i s h i n g s i tes per interviewed fisherman from Lake T i t i c a c a 84 Table 7. Mater ia l and labour costs for Lake T i t i c a c a small sca le f i s h e r i e s 122 Table 8. Returns to f i sh ing and f i s h i n g r e l ated labour per FEU type in Lake T i t i caca s m a l l - s c a l e f i s h e r i e s 123 Table 9. P r o f i t a b i b l i t y and a v a i l a b i l i t y of cashgenerating a c t i v i t i e s for the shore dwel le rs of Lake T i t i caca 125 Table 10. R e l a t i v e difference [%) between hourly returns to labour for f i s h i n g on Lake T i t i caca and a l t e r n a t i v e a c t i v i t i e s a v a i l a b l e to shore dwel lers according to FEU type 130 Table 11. R e l a t i v e difference (%) between the return to f i s h i n g labour and i t s opportunity cost , with or without granting a lower opportunity cost to f i s h i n g - r e l a t e d a c t i v i t i e s 133 Table 12. Labour requirements for a g r i c u l t u r a l production in various shore communities of Lake T i t i c a c a 144 v LIST OF FIGURES page F i g . 1 . Basic economic model of a s ingle species f ishery 3 F i g . 2 . D i s t r i b u t i o n of Lake T i t ieaca shore communities 20 F i g . 3 . Sources of evidence of the existence of TURFs on Lake T i t i eaca 43 F i g . 4 . Facsimi le of the map of the community of Ramis, Taraco 46 F i g . 5 . Facsimi le of the map of the community of Requena, Taraco 47 F i g . 6 . Facsimi le of the map of the community of Sajo, Pomata 48 F i g . 7 . Shore communities of the Lago Sur and Lago Pequeno 50 F ig . 8. D i s t r i b u t i o n of the three types of TURFs aroung Lake T i t i eaca 52 F i g . 9 . TURF types I, I I , and III 53 F i g . 10. Facs imi le of the map of the totora reed beds of the communities of Huerta and M i l l o j a c h i Huaraya 58 F ig . 11. Facs imi le of the map of the community of LLachon 61 F ig . 12. Resource dimension of Lake T i t ieaca TURFs 64 F i g . 13. Access to Lake T i t ieaca TURFs by b i r th or by marriage 82 F i g . 14. Map of the "Liga de defensa de l a totora de Catura Pampa" 104 F i g . 15. Changes in annual pr ice indices of f i sh and f i sh ing gear on the Peruvian A l t i p l a n o between 1975 and 1982 137 F i g . 16. A g r i c u l t u r a l work requirements in a shore community of Lake T i t i e a c a , Peru 145 F i g . 17. Organization of f i s h e r i e s management on Lake T i t ieaca 167 v i ACKNOWLEDGEMENT This d isser tat ion study sprung from my i n i t i a l involvement with Lake T i t ieaca f i sher ies whi le working i n Peru, between 1979 and 1981, for the Food and Agr icu l ture Organization of the Unfted Nations. I wish to acknowledge the f inanc ia l support of the Un ivers i t y of B r i t i s h Columbia through i t s Summer Fel lowship and Graduate Student Fe l lowsh ip programs, of the World Un ive rs i t y Services of Canada through i t s Franco-Canadian Fel lowship program and of the Donner Foundation. I a l so wish to acknowledge the support of Project FAO-PER/76/022, the National Science Foundation, the CIDA/UBC/UNTA project and the Canadian Association for La t in American and Caribbean Studies for the undertaking of my f ieldwork during the summer of 1984. I wish to acknowledge the support and co l labora t ion of the many i n d i v i d u a l s without whose help the completion of t h i s project would not have been possib le . I p a r t i c u l a r l y wish to thank the f o l l o w i n g i n d i v i d u a l s : Homer Campbell , leader of Project FAO/PER-76/022, for introducing me to Lake T i t i e a c a ; Rene A l f a r o , Eduardo Beni tes , Eufracio Bustamante, Charo Roncal , J u l i a n Torres and Hugo Trev ino, my Peruvian counterparts from the Peruvian Marine Research Ins t i tu te in Puno (IMARPE-Puno), for the i r to lerance for my strange "gringo" ways; C i r i l o Cut ipa, G i l mar Goyzueta, Percy Paz and Pedro Quispe, my interpreters and research ass is tants , for sharing with me t h e i r profound knowledge of the Peruvian Al t i p i a n o and for leading to many of the ins ights of th i s study; Tony Dorcey, Gordon Munro, Tom Northcote and B i l l Rees, members of my supervisory committee, for providing the i n t e l l e c t u a l guidance necessary for the completion of my program; and my f r iend Ben Or love for f ree l y sharing his knowledge of the Andes and for making a v a i l a b l e to me the data he had gathered on Lake T i t i e a c a f i s h e r i e s . I a lso want to acknowledge the c o l l a b o r a t i o n of Lake T i t i eaca shore dwel lers who answered my questions and gave me h o s p i t a l i t y . I chose t h i s p a r t i c u l a r thesis topic in the hope that my work might one day contr ibute to an improvement of the i r standards of l i v i n g . Although pr imar i l y an academic exerc ise , th i s project resu l ted in a continuous chal lenge for personal growth. I wish to acknowledge here the tremendous help received during t r y ing times from my f r iends . May they a l l f i nd here the expression of my s incere appreciation and profound grat i tude for sharing both the enjoyable and the l e s s enjoyable times with me. I dedicate th i s thes is to my parents who by the i r t rus t and a f fec t ion have demonstrated to me t h e i r continuous support and the i r conv ic t ion that I wouldeventua l l y complete a meaningful and worthwhile project. CHAPTER 1 INTRODUCTION 1.1 OBJECTIVES AND RATIONALE The overa l l objective of th is study i s to determine whether T e r r i t o r i a l Use-Rights in Fishing (TURFs), which I define as exclusive and l imi ted rights held by l o c a l l y or c u l t u r a l l y defined communities of shore dwellers over f i sh ing resources found in spec i f ic parts of thei r aquatic environment, are e f fect i ve in overcoming the common property problem of typical f i sher ies . Taking the TURFs system of Lake T i t icaca f i sher ies in Peru as a case study, I consider whether TURFs could be integrated into a formal management regime. According to common property theory, f i sh ing ef for t within an unregulated fishery tends to increase unt i l the corresponding resource rents are dissipated (Gordon 1954). To prevent the diss ipat ion of these rents, f isher ies management in the form of control of f ish ing ef for t i s needed. TURFs have recently attracted a growing amount of attention because of thei r potential for mit igating the problems resu l t ing from the common property status of most f i sher ies resources (Christy 1982; Panayotou 1984). While there remain serious doubts as to whether th is theoretical potential i s t ranslated into practice (Lawson 1984), the effectiveness of TURFs can be evaluated in terms of their a b i l i t y to control f ish ing e f fo r t and to prevent the diss ipat ion of the rents from a f ishery. In the case of Lake T i t i caca , Peru, such an evaluation e n t a i l s : 1. Documenting the existence and dimensions of Lake T i t icaca TURFs; 2. Demonstrating the i r act ive enforcement a l l along the shorel ine; 3. Determining whether f ish ing rents are captured by loca l fishermen; 1 4. Evaluating the contribution of TURFs to the generation of these rents. The research addresses the general need for appropriate forms of f i sher ies management by documenting the potential of TURFs for the management of s m a l l - s c a l e f i s h e r i e s . The se lect ion of Lake T i t icaca TURFs for study i s s p e c i a l l y useful because of thei r potential contribution to the economy of the Peruv ian .A l t ip lano surrounding Lake T i t i caca . 1.2. NEED FOR APPROPRIATE FORMS OF FISHERIES MANAGEMENT 1.2.1. Need for f i sher ies management Fisher ies resources have to be managed because, although renewable, they are vulnerable. They can be destroyed by p o l l u t i o n p a r t i c u l a r l y in shallow coastal areas and inland waters where most s m a l l - s c a l e f ish ing takes place (Scudder and Conel ly 1985). And, because of the i r common property status, they can be can be overexploited and eventual ly decimated by the very people whose welfare depends upon them (Gordon 1954), a phenomenon popularly known as the tragedy of the commons (Hardin 1968). According to common property theory, f ish ing e f fo r t within an unregulated fishery tends to increase unt i l average f i sh ing revenue becomes equal to average f ish ing cost (Gordon 1954). As long as the expected revenues from part ic ipat ing in th is f ishery (AR) are greater than the costs faced by an average f ish ing unit (AC), new f ish ing units have an economic incentive to enter th is f ishery , unless phys ica l l y or l e g a l l y prevented from doing so (Figure 1). Once average revenue equals average cost, th is incentive dissappears, the resource rents represented by the difference between average revenue and average cost are said to have been dissipated, and the f ishery i s said to have reached i t s bionomic equi l ibr ium (Ibid.). 2 FIGURE h BASIC ECONOMIC MODEL OF A SINGLE SPECIES F ISHERY ( f rom Gordon 1954). An unregulated f ishery which has reached i t s bionomic equi l ibr ium i s said to be exploited beyond i t s social optimum (i.e. overfished) because some of the resources a l located to f i sh ing could have generated higher returns in other sectors of the economy (Gordon 1954). Within such a f ishery regulation i s needed to control f i sh ing e f fo r t in order to prevent the diss ipat ion of f ish ing rents and therefore to promote a more e f f i c i e n t a l l o c a t i o n of society's resources ( Ib id . ) . The regulation of f ish ing ef for t i s only one form of f i sher ies management, because f i sher ies management includes a l l a c t i v i t i e s designed to increase the u t i l i t y that accrues to society from i t s f i sher ies resources (Carlander 1969; Rothschild and Forney 1979). However, I use the terms regulation and management interchangeably, fo l lowing a practice common in the l i t e r a t u r e of f ishery economics which i s more s p e c i f i c a l l y concerned with l e g i s l a t i v e or regulatory techniques of f i sher ies management (Anderson 1977; Panayotou 1982). Given the predictions of common property theory, the only way to determine whether a regulation or management technique i s e f fec t i ve in c o n t r o l l i n g f i sh ing e f fo r t within a par t i cu la r f ishery i s to demonstrate that th is technique i s responsible for the capture of f i sh ing rents within th is f ishery. Fisheries management techniques could also be evaluated in terms of other c r i t e r i a such as the i r a b i l i t y to promote an equitable d is t r ibut ion of the benefits from a f ishery (Christy 1982). However, I have not considered these addit ional c r i t e r i a here, because they may be promoted independently of e f fo r t control and without an increase of the economic y i e l d from the f ishery considered. Furthermore, t rade-offs may have to be made between them and economic ef f ic iency , as in the case of social equity (Ibid.). Fol lowing Gordon's (1954) seminal analys is , the occurrence of overf ishing 4 has been documented in many la rge -sca le commercial and indust r ia l f i sher ies in the developed world (Crutchf ield and Ze l lner 1962; Crutchf ie ld and Pontecorvo 1969; Scott and Neher 1981). More recent ly , the occurrence of overf ishing has been demonstrated in many smal l - sca le f i sher ies of the Less Developed Countries (LDCs) as wel l (Smith 1981; McCay 1981; Bai ley 1982; Neal 1982). Furthermore, the f i sher ies regulations used so far have proved inef fect i ve in c o n t r o l l i n g f i sh ing e f fo r t and p a r t i c u l a r l y inappropriate for the l imi ted technical and administrative a b i l i t i e s of most LDCs (Scudder and Conelly 1985; Pol lnac 1986) such as Peru. 1.2.2. Shortcomings of past f i sher ies management practices A wide array of regulations has been proposed and implemented for the purpose of c o n t r o l l i n g f i sh ing e f fo r t . These regulations have often proved ine f fec t i ve , because fishermen repeatedly f ind new ways to circumvent them and to increase the i r f i sh ing e f fo r t (Pearse 1979; Wilen 1979). Moreover thei r implementation has proved expensive because, to counteract fishermen's evasive t a c t i c s , f i sher ies regulations have to become more complex, which makes the i r administrative ef f ic iency dwindle (Scott and Neher 1981; Panayotou 1984). The administrative costs of sma l l - sca le f i sher ies management have outgrown the f inanc ia l a b i l i t i e s of many LDCs, and loca l f i sher ies agencies are hardly able to keep thei r a c t i v i t i e s at even a minimum l e v e l (Scudder and Conelly 1985; Pol lnac 1986). In addition to administrative i n e f f i c i e n c i e s , the implementation of f i sher ies regulations often leads to higher f i sh ing costs by inducing an increase in redundant f i sh ing capacity (Munro 1982; Munro and Scott-1984). The shortcomings of the management techniques presently used have spurred a continuous search for a l te rnat ives which would be both more appropriate for LDCs1 administrative a b i l i t i e s (Pol lnac 1986) and more e f fec t i ve . I d e a l l y , such a l te rnat i ves should el iminate the common property 5 problem without leading to an increase in redundant f i sh ing capi ta l and to the administrative inef f ic iency resu l t ing from fishermen's evasive tac t i cs . 1.2.3. A l te rnat ive forms of f i sher ies management The search for more appropriate forms of f i sher ies management has led to the ident i f i ca t ion of two major sets of a l te rnat ives invo l v ing ei ther the imposition of taxes or the conferring of property r ights (Clark 1979). Taxes are payments of part of the f i sh ing revenues to the state. They correspond to an a r t i f i c a l increase in f i sh ing costs, which can be incurred either before f i sh ing takes place in the case of l icence fees, or af ter i t has taken place in the case of taxes on landings. Property r ights correspond to f i sh ing r ights granted to indiv idual or groups of fishermen, ei ther over portions of the aquatic space, or over portions of the f ish ing resources regardless of the i r locat ion , or over portions of the f i sh ing resources in spec i f i c locat ions. The creation of l imi ted property r ights has been proposed as a solut ion to the common property problem (Demsetz 1967; Eckert 1977; Pearse 1981). Such r ights may be translated into ind iv idual catch quotas which are granted to each fisherman. A l t e r n a t i v e l y , these r ights may be constrained by such c r i t e r i a as time or f ish ing method. Individual fishermen may thus hold temporary access r ights to spec i f ic f i sh ing grounds, providing they use spec i f i c f i sh ing methods. The speci f icat ion of property r ights i s referred to as s t in t ing (Ciriacy-Wantrup 1975; Pearse 1980). I t may i n d i r e c t l y resu l t in a quanti tat ive a l l o c a t i o n of f i sher ies resources between users. Theoretical comparisons between property r ights and taxes indicate that they are roughly equivalent in terms, of c o n t r o l l i n g f i sh ing e f fo r t and promoting economic ef f ic iency within a f ishery (Clark 1979; Crutchf ie ld 1979). 6 However, property r ights are l i k e l y to invo lve lower administrative costs than taxes (Maloney and Pearse 1979; Scott 1979). One should thus try to determine which of the various systems of property r ights would be most e f fec t i ve . In his theoret ical comparison of various forms of property r ights , Pearse (1980) shows that the more exclusive f i sh ing r ights such as sole property are l i k e l y to be more e f fec t i ve in c o n t r o l l i n g f ish ing e f fo r t , but that the less exclusive ones such as group property would minimize administrative costs. A regime of property r ights of intermediate e x c l u s i v i t y between the extremes of no property and sole property would therefore be optimal, because i t would give the best t rade-off between economic ef f ic iency and administrative costs (Pearse 1980; Runge 1986). Ex ist ing systems of water tenure and customary f ish ing ru les , including T e r r i t o r i a l Use-Rights in Fishing (TURFs) are among the forms that such a regime could take. 1.3. TERRITORIAL USE-RIGHTS IN FISHING (TURFS) 1.3.1. Def in i t ion of TURFs In the l a s t decade, a large number of studies have documented the existence of t e r r i t o r i a l f i sh ing r ights . These r ights have a l te rnate ly been referred to as native possessory r ights (Goldschmidt and Haas 1946), t rad i t iona l or customary f i sh ing r ights (Lawson 1984), t e r r i t o r i a l f i sh ing rights (Acheson 1975; 1979; Panayotou 1982), sea tenure (Alexander 1977), sea ownership (Peti t -Skinner 1983) or sea t e r r i t o r i a l i t y (Pal lson 1982; Levine 1984), and t rad i t iona l f i sher ies conservation or management practices (Kapetsky 1981), un t i l the term of T e r r i t o r i a l Use-Rights in Fishing (TURFs) was coined by Christy (1982), and gained some currency (Panayotou 1982; 1984; McGoodwin 1984). For the purpose of th is study, I define T e r r i t o r i a l Use-Rights in Fishing 7 (TURFs) as systems of aquatic tenure which invo lve the holding of exclusive and l imi ted rights by s p a t i a l l y or c u l t u r a l l y defined communities of shore dwellers over f ish ing resources found in spec i f ic parts of their aquatic environment. This def in i t ion d i s q u a l i f i e s both nat ional l y held Exclusive Economic Zones and p r i va te l y owned f ish ing areas (Panayotou 1983). It expands on previous def in i t ions (Christy 1982; Panayotou 1984) because i t does not r e s t r i c t the types of r ights involved to exclusive use r ights . F i n a l l y , i t underscores the need for caution in a survey of the l i t e r a t u r e in which the existence of TURFs is documented. Studies in which the existence of TURFs i s documented have often focussed upon par t i cu la r f i sh ing resources, for example lobster (Acheson 1975; 1979), Crayfish (Levine 1984) or cod (Andersen and S t i l e s 1973). However, TURFs are general ly more s i t e - s p e c i f i c than resource-specif ic (Christy 1982). They can invo lve any type or combination of aquatic resources including f i s h , algae (Arzel 1984), s h e l l f i s h e s and crustaceans (Asada et a l . 1983), i r respect ive of whether they are used for food, medicinal or decorative purposes (Pet i t -Skinner 1983). Contrary to resource r ights which cover the whole range of a par t i cu la r resource, they may cover only part of the range or habitat of the resources they invo l ve , as in the case of highly mobile or migratory species. Such studies have also often focused upon par t i cu la r f ish ing methods, such as f ixed f i sh ing traps (Martin 1979) or beach seines (Alexander 1977), although TURFs are not method-specific. They often concern a range of f ish ing methods used by loca l fishermen (Johannes 1977; 1978), even i f they do not affect a l l of them to the same extent. For instance less e f fec t i ve f ish ing methods, such as the small gear used by foot fishermen (Andersen 1979; Labby 1976: 86) or hand c o l l e c t i o n by women and chi ldren may not be subject to the res t r i c t ions of TURFs (A l lan et a l . 1948: 118). 8 F i n a l l y , many of these studies have dealt with spec i f ic ecological or f ish ing zones, such as brackish coastal lagoons (Cattarinussi 1973; Demestre et a l . 1977) or r i v e r estuaries (Cordel1 1974; 1978), although TURFs can affect any part of the aquatic environment surrounding a f i sh ing community (Berkes 1977; Davis 1984). Many f i sh ing communities a c t u a l l y hold a series of f ish ing r ights over various parts of thei r aquatic environment defined by such c r i t e r i a as water depth, distance from shore (Andersen 1979; Pet i t -Skinner 1983) or distance from the community (Johannes 1977; 1978). 1.3.2. Dimensions of TURFs Because some authors have focussed on the spat ial dimension of TURFs, we tend to think of TURF as a unidimensional concept. However, as a system of water tenure, TURF is a multidimensional social concept which, l i k e a land tenure system, can be characterized in terms of four major dimensions: s p a t i a l , temporal, demographic and c u l t u r a l . The spat ia l dimension refers to the physical boundaries of the te r r i to ry to which a TURF system appl ies . The temporal dimension refers to the period during which i t i s in force. Impl i c i t in the spat ia l and temporal dimensions i s a resource dimension, which corresponds to the proportion of the resource involved. The presence of resources within a TURF area may be temporary, as with migratory resources, or permanent but only a small proportion of the resources concerned i s found within any given TURF area. The demographic dimension classes ind iv idua ls and groups according to the i r inc lus ion or exclusion from a pa r t i cu la r set of r ights and obl igat ions. I t indicates how exclusive the r ights involved are by describing the referent group. F i n a l l y , the cu l tu ra l dimension refers to the legal (formal) or customary (informal) conditions under which the d i s t r ibu t ion , transfer and exercise of r ights apply, thus to the kind of r ights TURF holders can exercise 9 over loca l resources and to the duties these r ights e n t a i l . As for land tenure systems, these di f ferent rights are grouped into six broad categories: d i rect use-r ights , r ights of indi rect economic gains, control r ights , transfer r ights , residual r ights and symbolic r ights (Crocombe 1974). Direct use-r ights include the rights to harvest or to husband the resources within a TURF area. Rights of indi rect economic gain invo lve TURF holders' a b i l i t y as ind iv idua ls or as a group to get an economic compensation from outsiders whom they al low to use thei r resources temporarily. Control r ights refer to the types of controls and regulations that can be implemented by TURF holders. Transfer r ights s t ipu la te the conditions under which TURF holders can transfer thei r r ights permanently, or those under which outsiders can gain membership into a TURF holding community. Residual r ights specify what happen to r ights which cannot be transfered by the i r holders for lack of an acceptable rec ip ient , or because these r ights have been withdrawn from the i r holders. F i n a l l y , symbolic r ights represent the p o l i t i c a l , c u l t u r a l , or sp i r i tua l s ignif icance of the resources and areas of a TURF for the i r holders. Not only are TURFs multidimensional, but they may also be reinforced by other forms of ownership, for example, the ownership of such intangibles as f i sh ing techniques, f i sh ing songs and f ish ing r i t u a l s (Johannes 1977; 1978; Pet i t -Skinner 1983). They can also be reinforced by exclusive nav igat ional , landing, berthing or mooring rights (McCay 1980; Fernando et a l . 1985; Ruddle and Johannes 1985: passim) and by exclusive r ights over portions of the shore used for f i sh processing, p a r t i c u l a r l y f i sh drying (Petit -Skinner 1983) or f i sh smoking (Suttles 1960). F i n a l l y , TURFs can be found in combination with other informal or s e l f regulatory mechanisms (McGoodwin 1984), sometimes referred to as customary rules (Alexander 1980) or as social practices (Berkes 1977), which various authors have t r i ed to c l a s s i f y (Pol lnac and L i t t l e f i e l d 10 1983; McGoodwin 1984; Scudder and Conelly 1985). 1.3.4. Potential benefits of TURFs For decades, f ish ing t rad i t ions , including TURFs, have been considered as impediments to f i sher ies development and to the improvement of fishermen's welfare because f isher ies development was assimilated with an unconstrained increase in f i sh ing e f fo r t (Kirby and Sczepanik 1957; Sabri 1977; FA0/RAS/77/P10 1979). Although government regulations have frequently superseded t rad i t iona l TURFs, the l a t t e r are s t i l l customarily enforced in many regions of the world (Pol lnac and L i t t l e f i e l d 1983; MacGoodwin 1984; Scudder and Conel ly 1985; Ruddle and Akamichi 1985; Ruddle and Johannes 1985; Berkes 1986). Most recently , thei r potential contribution to f i sher ies management has warranted them a growing amount of interest . The potential contribution of TURFs to f i sher ies management ac tua l l y invo lves f i v e major elements. F i r s t , i t has been claimed that TURFs could reduce i f not el iminate the common property condition of f i sher ies resources and el iminate the problems associated with i t (Pol lnac and L i t t l e f i e l d 1983: 235). These problems, or e x t e r n a l i t i e s , resu l t from the negative effects of each fisherman's actions on the catches and returns of other fishermen. Fishermen do not perceive these impacts, nor do they have any economic incentive to take them into account. Because TURFs incorporate s e l f - s e r v i n g ind iv idua ls into small and autonomous communities whose members know each other, they create social or moral incentives for these ind iv idua ls to take ex te rna l i t i es into account, thus imposing on them "the larger r a t i o n a l i t y of community s u r v i v a l " (Emmerson 1980: 32). However, both the exclusion of outsiders (Ciriacy-Wantrup 1975) and the control of the use of labour and capi ta l within TURF holding communities must be combined to achieve the 11 desired resu l t (Panayotou 1982: 155). The implementation of f ish ing rules within these communities may a lso be necessary to reduce interferences and c o n f l i c t s between fishermen. Second, TURFs could promote administrative ef f ic iency in at least three d i f ferent ways. They would a l low for adjustments to social or economic change by introducing f l e x i b i l i t y into the control of f i sh ing e f fo r t since TURF holders could adapt the i r management objectives to new social or economic circumstances almost instantaneously (Panayotou 1982). They would also a l low for the introduction of necessary o n - l i n e revis ions of management objectives to adapt to environmental v a r i a b i l i t y (Scott and Neher 1981). F i n a l l y , they would transfer a large share of organizational costs to TURF holders themselves (Pearse 1980). In p a r t i c u l a r , they would reduce implementation costs wherever TURFs already ex ist (Kapetsky 1981), simply because formalized management regimes based on them would be more readi ly acceptable than a l te rnat i ve forms of f i sher ies management. And since TURFs are so common in s m a l l - s c a l e f i sher ies as to be almost ubiquitous, such strategies could be used in many parts of the world. Third, TURFs could promote an increase in the physical productiv i ty of exploited f i s h resources, because they would give the i r holders the incentive to invest in the i r resources for increased future returns. They would also a l low for the c o l l e c t i o n of more appropriate and more r e l i a b l e technical information, since under a TURF-based management regime fishermen are both the pr inc ipal producers and benef ic iar ies of such information. This could in turn a l low for a reduction in the security margins of catch l i m i t a t i o n s , which would permit the u t i l i z a t i o n of f i sher ies resources at a l e v e l c loser to thei r actual potent ia l . This could be further reinforced by the p o s s i b i l i t y of engaging in an act ive process of adaptive management (Walters and Hi 1 born 12 1975; 1978) since harvesting experiments could be undertaken in the discrete management units corresponding to d i f ferent TURF areas. Fourth, TURFs could promote socia l equity because t rad i t iona l sea tenure and customary f ish ing practices often lead to an equitable d is t r ibut ion of f i s h catches (Alexander 1980). The corresponding j o i n t r ights of access could also promote a more equitable d is t r ibut ion of aquatic resources among users than a system of pr ivate r ights (Runge 1986). And a TURF-based management strategy would also reduce the bureaucratic complexities associated with centra l i zed f i sher ies management (Neal 1982) which tend to discriminate against the disadvantaged, the least educated or the poorest (Ward 1982). F i n a l l y , i t would i n s t i l l greater accountabi l i ty into the present system of f i sher ies management which has already lent i t s e l f to accusations of despotism (Scott and Neher 1981) and i n s e n s i t i v i t y to fishermen's needs (Davis 1984). Because of these benefits some authors do not hesitate to advocate the granting of a formal status to t rad i t iona l sea tenure and customary f i sh ing rules (Johannes 1982), or the i r incorporation into f i sher ies management practices (Pol lnac and L i t t l e f i e l d 1983; Davis 1984). Most contributors, however, are more cautious in the i r recommendations. They acknowledge the potential benefits of TURFs, but they a lso recognize the paucity of relevant information, and suggest careful examination of t rad i t iona l TURFs and of thei r appropriateness for f i sher ies management purposes (Acheson 1981; Pearse 1980; ICLARM 1981; Kapetsky 1981; Christy 1982; Panayotou 1982; 1983; Lamson and Cohen 1984; Lawson 1984). A l l th is confirms that i t i s indeed c r i t i c a l to determine whether the "many kinds of norms and ins t i tu t ions fishermen (have) invented to control access and f ish ing procedures could be used as a basis for successful resource management" (Acheson 1981: 307-308). Such an invest igat ion would lead to the 13 elaboration of new ins t i tu t ions which would c a p i t a l i z e upon the advantages of t rad i t iona l f i sher ies management practices (Alexander 1980). I t would thus contribute to the emergence of a new model of resource management s i m i l a r to those proposed by various contributors: resource self-management (Berkes 1985), community-based resource management (Lamson and Cohen 1984) and f ishery co-management (Pinkerton 1986). 1.4. EFFECTIVENESS OF EXISTING TURFS 1.4.1. Limitations on the effectiveness of TURFs I t has been argued that the theoret ical effectiveness of TURFs in c o n t r o l l i n g f ish ing e f fo r t i s not t ranslated into pract ice. For example, these systems resu l t in an a l l o c a t i o n of space, rather than of resources (Andersen and S t i l e s 1973; McCay 1980; 1981), and they provide rules for the conduct of f i s h i n g , rather than for the conservation of f i sher ies resources (Acheson 1981; Pa l l sson 1982). Although these objections do not necessari ly inva l idate TURFs for f i sher ies management purposes, they raise the issue of whether ex ist ing TURFs a c t u a l l y lead to a control of f ish ing e f fo r t . The effectiveness of TURFs in c o n t r o l l i n g f i sh ing e f fo r t depends in part -on the a b i l i t y of thei r holders to coordinate thei r f i sh ing pract ices, both within and between TURF holding communities. Coordination within TURF holding communities i s necessary, because the creation of incentives for indiv idual fishermen to take the external effects of the i r f i sh ing a c t i v i t i e s into account is not an automatic consequence of the existence of TURFs. Internal mechanisms c o n t r o l l i n g TURF holders' f i sh ing a c t i v i t i e s have to be devised to induce them to incorporate these ex te rna l i t i es into the i r decision-making processes. As Gordon himself stated: "In cases of group tenure where the numbers of the group are large, there i s s t i l l the necessity of coordinating 14 the practices of explo i tat ion" (Gordon 1954: 134). Coordination between TURF holding communities i s a lso necessary, because highly mobile or migratory f i sh resources do not remain within a TURF area for the i r entire l i f e cyc le . Moreover, neighbouring TURFs often share the same transboundary stocks, since indiv idual TURFs are rarely large enough to encompass the whole range of a stock. In ei ther case, TURF areas rarely correspond to the ranges of f i sher ies resources. In short, one cannot argue that ex ist ing TURFs should be used for f i sher ies management purposes simply because they already ex ist or because of thei r theoret ical effect iveness. Their actual effectiveness has to be tested in each case. 1.4.2. Evaluation of the control of TURFs on access to f ish ing A review of the relevant l i t e r a t u r e demonstrates that few scholars who have documented TURFs evaluate thei r effectiveness on the basis of empirical data. Some by-pass the d i f f i c u l t y of th is assessment by claiming that the assumptions of the common property model do not apply (Pol lnac and L i t t l e f i e l d 1983: 235). Others do c laim that ex ist ing TURFs prevent outsiders from gaining access to loca l f i s h e r i e s , thus resu l t ing in a control of f ish ing e f fo r t through a control of the number of part ic ipants (Acheson 1975; Berkes 1977; 1985; 1986; Davis 1984). Berkes (1977) observes that the lack of incentive to create a surplus in a subsistence economy may resu l t in a high y i e l d per unit of e f fo r t while c o n t r o l l i n g tota l e f for t and therefore impact on stocks. The evidence presented by these authors i s insu f f i c ien t to prove the effectiveness of TURFs for three reasons. F i r s t , the c laim by insiders that they exclude outsiders does not prove that the l a t t e r would want to enter the 15 f ishery anyway. For instance, insiders ' control of access to a f ishery may be of l i t t l e relevance to outsiders i f the bionomic equi l ibr ium of th is f ishery has already been reached. Secondly, the exclusion of outsiders i s a form of l imi ted entry which al lows for the control of only one dimension of f i sh ing ef for t . I t does not imply a w i l l ingness or an a b i l i t y on the part of the l i censed - in or insiders to l i m i t the i r own f i sh ing a c t i v i t i e s (Rettig and Ginter 1978). TURF holders may s t i l l increase thei r respective f i sh ing ef for ts t i l l the bionomic equi l ibr ium of the f ishery i s reached. And t h i r d l y , contrary to Berkes's (1977) conclusion, a high b io log ica l y i e l d does not prove that either TURFs, or the social practices associated with them control tota l f i sh ing e f fo r t . The f ishery studied may have a high b io log ica l y i e l d , even at i t s bionomic equi l ibr ium, simply because the cost of f i sh ing happens to be high r e l a t i v e to the value of the f i s h caught, as could occur with any l i g h t l y exploited f ishery (Troadec 1983: 10). If average f ish ing cost was twice as high as depicted in Figure 1, a much higher b io log ica l y i e l d would be obtained for the f ishery considered, with a much lower l e v e l of f ish ing e f fo r t , but s t i l l at bionomic equi l ibr ium. TURFs and associated social practices would not have to intervene in the explanation of th is high y i e l d . 1.4.3. Evaluations of the control of TURFs on f ish ing ef for t Only a handful of scholars have a c t u a l l y considered the ro le of TURFs in the promotion of economic ef f ic iency to demonstrate the i r effectiveness in c o n t r o l l i n g f i sh ing e f for t . Gordon (1954: 134) does c la im that in a few places along the Canadian A t lan t i c coast, lobster fishermen manage to capture f i sher ies rents, but he does not provide empirical evidence to support his c la im. Acheson and Wilson confirm Gordon's e a r l i e r c laim by demonstrating that both b io log ica l and economic productiv i ty are higher in the cont ro l led areas of the Maine lobster f ishery (Acheson 1975; Wilson 1977). However, they end up demonstrating that p r i v a t e l y - h e l d f i sh ing r ights are more e f fec t i ve than communally-held TURFs. Fernando, Munasinghe, Panayotou and the i r col laborators demonstrate that s m a l l - s c a l e fishermen in Sr i Lanka are able to capture substantial economic rents, and argue that loca l TURFs are responsible for th is (Fernando 1984a; 1984b; Fernando et a l . 1984; Munasinghe 1984; Panayotou 1984). Unfortunately, they provide no information on the dimensions, d is t r ibut ion or enforcement of these TURFs about which, with one exception (Alexander 1975; 1977), l i t t l e i s known. F i n a l l y , Smith and Panayotou (1984) demonstrate that the system of municipal concessions in the m i l k f i s h fry gathering f i sher ies of the Phi l ipp ines a l low the TURF-holding munic ipa l i t ies to capture the corresponding rents through competitive bidding over annual concessions. Only those studies which determine whether a f ishery i s operated at an economically e f f i c i e n t l e v e l , and a lso assess the ro le of TURFs in preventing th is f ishery from reaching i t s bionomic equi l ibr ium, ac tua l l y assess the effectiveness of TURFs in c o n t r o l l i n g tota l f i sh ing e f for t . Therefore, to be complete, an empirical evaluation of the effectiveness of TURFs should include four successive steps. F i r s t , i t should document the existence of loca l TURFs; second, i t should demonstrate that TURF holders are ac tua l l y able to enforce the i r TURFs, in spite of any o f f i c i a l opposition resu l t ing from the informal and eventual ly i l l e g a l character of TURFs; t h i r d , i t should show that substantial rents are being captured by TURF holders ( i .e. that loca l f i sher ies have not reached thei r bionomic equi l ibr ium yet) ; and fourth, i t should provide evidence that these rents resu l t from the act ive enforcement of loca l TURFs rather than from some technological or economic l imi ta t ions on tota l f i sh ing e f fo r t . 1.4.4. Evaluation of the effectiveness of Lake Ti t ieaca TURFs There were several advantages to se lect ing the communal f i sh ing r ights t r a d i t i o n a l l y enforced by Lake T i t ieaca shore dwellers as a case study. F i r s t , there was preliminary evidence that the t e r r i t o r i a l f i sh ing rights of Lake T i t ieaca f i t the def in i t ion of TURFs (Lev ie i l 1986). Second, i t was c lear that one could ca lcu la te the economic value of Lake T i t ieaca fishermen's catch, since they exchange about 83% of i t through barter and trade (Orlove 1986), which i s not always possible with subsistence-oriented f i sher ies . Third, a large amount of information on the social and economic structures of loca l populations was already a v a i l a b l e , which i s not usual ly the case for s m a l l - s c a l e f i sh ing populations of the Third World. A l l th is information was necessary for the completion of the tasks invo lved in the assessment of loca l the effectiveness of TURFs. F i n a l l y , there i s ample evidence that Lake T i t ieaca f i sher ies did contribute to the economy of the Peruvian A l t i p i a n o surrounding Lake T i t ieaca , and that the i r e f fec t i ve management might preserve and eventual ly increase th is contribution over the long term. Lake T i t ieaca f i sher ies represent a major source of income and protein for more than 3,000 shore-dwel 1 ing f a m i l i e s (Bustamante and Trevino 1976; A l fa ro et a l . 1981) and a supplementary source of protein for the population of the rest of the Al t ip iano (Ferroni 1980). Appropriate management of these f i sher ies i s thus highly necessary given the poverty of the A l t i p l a n o r e l a t i v e to the rest of Peru (Amat y Leon 1981), the f a i l u r e of numerous loca l development projects (Sanchez 1983), the dramatic depletion of some of the loca l f i sh resources (Appendix B), and the i n a b i l i t y of the loca l f i sher ies administration to assume i t s management r e s p o n s i b i l i t i e s (Laba 1979). 18 CHAPTER 2 CONTEXT, MATERIAL AND METHODS OF THE STUDY 2.1 NATURAL AND HUMAN ENVIRONMENT OF LAKE TITICACA 2.1.1. Natural environment Lake T i t icaca (Figure 2) i s a large (8,559 km2), high a l t i tude (3,808m), t ropical (15 degrees south) lake which l i e s on the border between Peru and B o l i v i a . According to the o f f i c a l map of the Peruano-Bol i v ian hydrographic comission, i t s maximum length i s approximately 250km, i t s maximum width 60km and i t s perimeter 1,850 km. It includes three connected basins: the Large Lake or Lago Grande (6,542 km2, 284 m deep), the Puno Bay or Bahia de Puno (589 km2, 25 m deep), and the Small Lake or Lago Pequeno (1,367 km2, 40 m deep) (Boulange and Aquize 1981). Two smal ler lakes, Arapa (195 km2) and Umayo (46 km2) are intermit tent ly connected with Lake T i t icaca through the f loodpla ins of the Ramis and 111 pa Rivers respect ively (Figure 2). The Peruvian portion of Lake T i t icaca represents about 60% of i t s tota l area. I t includes about 65% of the tota l shorel ine, f i v e of the seven main t r ibutar ies to the lake, and ha l f of i t s only out le t , the Rio Desaguadero, which marks the boundary between Peru and B o l i v i a (Figure 2). Most of the 57,340km2 (Ibid.) of Lake T i t icaca's watershed corresponds to a high a l t i t u d e plateau, the A l t i p l a n o , which l i e s between the western and the eastern Andean ranges. The A l t i p l a n o i s covered with r o l l i n g h i l l s of a l l u v i a l and lacustr ine o r ig in . Low average temperatures (6 to 12 degrees Celsius) with frequen t m - g h t f rosts , and low annual prec ip i tat ion (500 to 1,000 mm in the cirumlacustrine area, 200 to 500 mm further inland) are responsible for i t s semi-arid mountain cl imate (Ibid.). Prec ip i tat ion i s 19 FIGURE 2 : DISTRIBUTION OF L A K E TITICACA SHORE COMMUNITIES WC COLOMBIA BRAZIL Ariquipa P A C I F I C O C C A M CHILE ' 69° r i ^\ D " , a fl u a d , r o D i s t r i c t of CONIMA 1. T l l a l l - P a t a s c a c h i 2. Pucuraya -Mi l i laya 3 . Sucunl 4 . CayanajonI 5. Is la Soto D i s t r i c t of WHO 6. Chujucuyo-Naquercota 7. lacasanl 3 . Car iqu i ta 9 . Umuchi 10. Iscaj ja 11. Jachaparu 12. Quellojane 13. Jacantaya O l s t r i c t of YILQUECHICO 14. l i a c h a j a t a 15. A l l r u n i 16. V i l q u e c h i c o O l s t r i c t of HUANCANE 17. gt iel lahuyo 18. CHujachl 19. Huarisani 20. Junsani 21. J i c incoya 22. P iata 23. Cohasla 24. Balsapata 25. Yucahue D i s t r i c t of TARACO 26. Rirais 27. Tuni Requena ' 28. Requena 29. Huancollusco O l s t r i c t of PUS I 30. Corpa 31. Pus1 32. Cerabuco - l lapas D i s t r i c t Of CAPACHICA 33. E s c a l l a n l 34. C l i i l l o r a 35. Chif Ion 36. Ccotos 37. S ia le 38. Llachon 39. Yapura 40. Capano 41. S i l a c a c h l 42. J l l a t a D i s t r i c t of AMANTANI 43. Araantani 44. T a q u i l i O l s t r i c t of COATA 45. L luco Coata 46. Carata O i s t r i c t of HUATTA 47. Faon 48. Yasin 49. Moro 50. Matare 51. Pampa 111 pa O i s t r i c t of PAUCARCOLLA 52. Yanaricu 53. Col l ana D i s t r i c t of PUNO 54. Huaraya 55. J l r a t a Vlzcachunl 56. Chul lunl 57. T1t1n1 58. Huacahuacani 59. Totoranl 60. Puno Cludad 61. Chimu 62. Ojerani 63 . Ichu O l s t r i c t of CHUCUITO 64. Cuslpata 65. Barco-Chuculto 66. Chinchera 67. Concachl 68. Potojani 69. Canacanl 70. I s la Quipata 71. I s la C h l l a t a 72. Cochiraya 73. Ccarana 74. Parina 75. Tacasaya 76. Pucani 77. Luquina Grande 78. Luquina Chico 79. Car ina 80. Churo 81 . Perca D i s t r i c t of PLATERIA 82. Huataraque 83. Ccota 84. P a l l a l l a 85. Laccone 86. Huermeamaya 87. Sihuecanl 88. Anu Ca l le jon 89. Hu inca l la 90. T i t i l a c a O i s t r i c t of ACORA 91 . I s la Socca 92. Sta Rosa Yanaque 93. Ccoccane 94. Tunuhuaya 95. Ccoccosani 96. I s la Iscata 97. Amaya O l s t r i c t of HAVE 98. S u l l c a c a t u r a 99. Camicachi 100. Sta Rosa Huayl l a t a D i s t r i c t of PILCUYO 101. Huayl lata 102. Cachlpucara 103. Acaso 104. Queti 105. Vilcamaquera 106. Quispemaquera D i s t r i c t of JULI 107. Santiago Mucho 108. Suahacota 109. I rupalca 110. Ju l i Pueblo 111. Chucasuyo Caj je 112. Sihuayro 113. Challapampa O l s t r i c t of POMATA 114. Huacanl 115. Pomata Pueblo 116. Sajo 117. Santiago Ccama 118. Chatuma 119. Chimpu 120. Cuturapi D i s t r i c t of YUNGUYO 121. Quenuani 122. Acari 123. loricate 124. Yunguyo Pueblo 125. Tinacahl 126. Unicachi 127. Is la Iscaya 128. I s la Caana 129. I s la Yuspique 130. I s la Anapla 131. Is la Suana 132. V i lu rcun i 133. Pajjano 134. Tapoje 135. Calampune 136. Sta Cruz Chambl 137. Copapujo 138. Chlnomani 139. Yanapata 140. Tahuaco 141. Sanquira 142. Aaaqul l la 143. Calacoto 144. Chal 1 apampa O l s t r i c t of ZEPITA 145. Tacaplze 146. Copani 147. Isani 148. Zepita 149. Huilacaya 150. Ayrihua D i s t r i c t of OESAGUAOERO 151. Sta Cruz Cunl » Communities numbered sequentially along the shoreline highly seasonal on the Al t ip iano where the rainy months (December to March) account for 72% of the tota l r a i n f a l l , the t rans i t ion months ( A p r i l , September to November) for 22% and the dry months (May to August) for 6% (Ibid.). Variat ions in prec ip i tat ion are la rge ly responsible for the seasonal f luctuat ions of the lake l e v e l which average 1 m per year (Co l lo t 1980). Long term changes in prec ip i tat ion also generate year to year f luctuat ions of the lake l e v e l , with a difference of 5.2 m between the recorded extremes of 1940 and 1980 (Ibid.), which has been superseded since the rainy season of 1986, when the lake reached a l e v e l higher than i t s recorded maximum. Because much of Lake T i t ieaca i s surrounded by f l a t shore areas where a change of the water l e v e l of 1 or 2 meters means a displacement of the shorel ine of a few ki lometers, f luctuat ions of the lake l e v e l often enta i l drast ic changes in the loca l supply of c u l t i v a b l e land with dramatic impacts on the loca l economy (Claverias and Manrique 1983). 2.1.2. Administrative o rgan i za t ion Peru includes 24 Departments. The Department of Puno represents 5.6% of the country's to ta l area, and 4.6% of i t s to ta l population. The Peruvian Al t ip iano accounts for 30% of the area of the Department, and more than 90% of i t s population of 871,000 (INE 1981). The population comprises two l i n g u i s t i c groups, the Quechuas and the Aymaras, and i s concentrated around Lake T i t ieaca (Figure 2) with a density 37 inhabitants/km2 close to the lake, compared to a density of only 5 inhabitants/km2 further inland (INE 1981). This concentration may be explained as the resu l t of the favourable conditions for agr icu l tu ra l production in the circumlacustre area and of the history of the A l t i p l a n o (Appendix B). The Department of Puno includes 9 Provinces, three of which encompass the whole lake shore area: Puno, Chucuito and Huancane. These Provinces are 21 divided into D i s t r i c t s , 21 of which are adjacent to the lake (Figure 2). F i n a l l y , these D i s t r i c t s comprise a number of human settlements with various administrative statutes: agrarian cooperatives, munic ipa l i t ies , communities and parcial i t i e s . Only a minor part of the shoreline belongs to c o l l e c t i v e l y owned agrarian cooperatives, or to munic ipa l i t ies such as the City of Puno. The majority belong to peasant communities (Comum'dades Campesinas) and to p a r c i a l i t i e s (parc ia l idades) . Peasant communities and parcial i t i e s are t e r r i t o r i a l groups of fami l ies exp lo i t ing both p r i va te l y owned or usufructed lands for continuous production, and col 1 e c t i v e l y owned lands for f a l 1 ow c u l t i v a t i o n . They are not only characterized by the sharing of a common residence and te r r i to r y , but a lso by the i r s o c i a l , p o l i t i c a l and re l ig ious organization, a l l of which resu l t in the i r s o l i d a r i t y and maintenance (Casaverde 1978; F io ravant i -Mo l in ie 1978). The major difference between them i s that land i s by law c o l l e c t i v e l y owned in communities and p r i va te l y in p a r c i a l i t i e s . In pract ice, however, the d is t inc t ion i s blurred. In peasant communities, ind iv idua ls a c t u a l l y own the land customarily i nher i t ing , exchanging or s e l l i n g i t , even though the l a t t e r i s i l l e g a l . In p a r c i a l i t i e s , the group of residents retains t rad i t iona l r ights of control over i t s members' agr icu l tura l a c t i v i t i e s (Mayer 1985). Since p a r c i a l i t i e s and peasant communities operate in the same way as s o c i a l , p o l i t i c a l and administrative i n s t i t u t i o n s , I t reat them as equivalents for the purpose of th is study. Unless spec i f ic reference to o f f i c i a l status i s required, I w i l l refer to them both as communities, and I w i l l indicate the d i s t r i c t to which each belongs because they are often d i f f i c u l t to locate. 22 2.2. LAKE TITICACA FISHERIES 2.2.1. Aquatic resources a) Fish resources Lake T i t ieaca f i sh resources include native species and exotic species. The former are endemic to the lake and include the cyprinodont genus Orestias, which represented 67% of the f i sh harvested from Lake Ti t ieaca in 1980 (Alfaro et a l . 1982), and the cat f i sh genus Trichomycterus which represented 6.7% of th is harvest (Ibid.). Exotic species include the rainbow trout (Salmo  gairdneri) introduced to the lake in the ear ly 1940s, which contributed 14.2% of the tota l catch in 1980 (Ibid.), and the s i l v e r s i d e or pejerrey (Basi l ichthys bonariensis) introduced in the mid 1950s and which contributed 15.2% of the t o t a l 1980 catch ( Ib id . ) . According to the l a t e s t taxonomic rev is ion (Parenti 1984), the genus Orestias consists of 43 species grouped in four complexes, 29 of which can be found in Lake T i t ieaca i t s e l f , and 33 in i t s watershed (Appendix A). Local fishermen commonly ident i fy four major types of Orestias, the umanto, the boga, the i s p i s and the carachis. They a lso dist inguish f i v e types of carachis, according to the i r s i ze , shape and colorat ion (Appendix A), but often give the same names to various species, or d i f ferent names to the same species. For example, they refer ind i f fe ren t l y to a l l small Orestias as i s p i s , and to a l l ye l lowish carachis as carachis amar i l los . Conversely, they refer to 0. agassii as f i s h (Ch'aul 1 a), white, grey or black carachi (carachi  blanco, g r i s , negro). The remaining two types of carachis, the dwarf (carchi  enano) and the greenish one (carachi gringo), are only of marginal economic importance. As for the genus Trichomycterus, i t i s represented by only one species on the A l t i p l a n o (Tchernavin 1944b), although fishermen commonly 23 dist inguish the mauri, a small l i t t o r a l form, from a larger one usual ly found in deeper waters, the suche. b) Other lacustr ine resources Apart from the f i sh resources, shore-dwellers extract two other types of aquatic resources from Lake T i t i caca : waterfowl and lake -p lants (or aquatic macrophytes). About 14 species of migratory waterfowl are harvested when they come to the lake for nesting (CENFOR-Puno 1979). Aquatic macrophytes are represented by a variety of plants found in associations, depending upon depth and other l imnological parameters (Col l o t 1981). Two associations are of pa r t i cu la r economic importance, the totora (Scirpus or Schoenoplectus sp) and the l l a c h u , an association of three genera of aquatic plants (Myriophyl1 urn, El odea and Potamogeton). F i n a l l y , aquatic toads from the Telmatobius genus are sometimes acc identa l l y caught in g i l l nets and may be used in the elaboration of t rad i t iona l remedies. 2.2.2. F ish ing operat ions Fishing on Lake T i t i caca invo lves s m a l l - s c a l e operations, with simple technology and small f i sh ing c ra f t s , often single-handed. Fishing operations are short range, rarely las t ing for more than an overnight t r i p , although some f ish ing invo lves dayl ight operations. G i l l nets are set la te in the afternoon and checked or retr ieved ear ly the fo l lowing morning. Ind iv iduals , usual ly from the male gender, go f i sh ing with the i r c lose r e l a t i v e s , sometimes with those of the i r wives ( i .e. the i r aff ines) and share the catch. Fishing infrastructures are minimal, although a few t iny harbours have been b u i l t on exposed parts of the shorel ine. About ha l f of the fishermen operate from a balsa reed boat (48%), and the other h a l f from a wooden boat propel led with oars (48.5%). A few use a motorized wooden boat (3%) with a greater range of 24 operation, or a small f lat -bottom boat {7%) which would be unsafe to operate offshore (Bustamante and Trevino 1976). The 1976 census demonstrates that Lake T i t ieaca f i sher ies are not s p a t i a l l y concentrated. Fishing i s practiced by ind iv idua ls from each shore community (Bustamante and Trevino 1976), and because fishermen leave the i r c ra f t at a landing spot within easy reach of thei r home compound, and because houses are scattered among the f i e l d s rather than gathered in v i l l a g e s , there is l i t t l e c luster ing of f ish ing c raf t into f l e e t s operating from the same harbour. Only in the case of the i sp i trawl f ishery do fishermen tend to congregate on the i sp i schools. Lake T i t ieaca f i sher ies are a lso characterized by a lack of occupational spec ia l i za t ion . Fishing i s usual ly carr ied out by men, although there i s no prohibit ion against the part ic ipat ion of women in f i sh ing a c t i v i t i e s . Most engage in f i sh ing as an a c t i v i t y secondary to agr icu l ture , and often in combination with a host of addit ional a c t i v i t i e s such as handicrafts, c a t t l e ra i s ing , and migration for wage labor. They rare ly undertake more than a few f ish ing t r ips per week, once agr icu l tu ra l a c t i v i t i e s have been completed. Shore dwellers often dist inguish between occasional , seasonal and permanent fishermen. Occasional fishermen f i sh sporadica l ly with a pushnet in shallow waters, with a small balsa and a couple of g i l l nets near shore, or as crew members for a r e l a t i v e . Seasonal fishermen operate a few months per year, usual ly during the rainy season when temperatures do not go below freezing at night. And permanent fishermen f i sh year round, although they may do so only a few days a week. F i n a l l y , Lake Ti t ieaca f i sher ies are characterized by low c a p i t a l i z a t i o n . Local fishermen use a variety of simple and r e l a t i v e l y inexpensive f i sh ing methods. They carry out most of thei r f i sh ing with g i l l n e t s made of purchased 25 mesh panels (Ibid.) but a lso re l y upon pushnets, small harpoons, or beach seines along the shorel ine, and various kinds of g i l l nets and trawls further from shore. To reduce f ish ing costs they use loca l materials as much as possible and undertake, with the help of the members of thei r household, as many f i sh ing - re la ted operations as poss ib le , including the construction and maintenance of thei r f ish ing c ra f t and gear, and the r e t a i l i n g of thei r catch. 2.2.3. F isheries s t r a t i f i c a t i o n The combination of f ish ing c raf t and gear, and of the crew to operate them represents a Fishing Economic Unit, or FEU (Bazigos 1974a). The terms "fisherman" and "FEU" are thus not s t r i c t l y equivalent except when a fisherman operates alone or with a junior helper. For the purpose of th is study I have ident i f ied 10 types of FEUs according to f ish ing c ra f t , f i sh ing gear and the mesh sizes (measured along the stretched diagonal) of the g i l l nets used, according to d is t inct ions made by loca l fishermen themsel ves (Table 1). However, because f i sher ies are defined as including a l l FEUs which target the i r f i sh ing operations on the same group of species in the same aquatic areas, and because di f ferent types of FEUs may be used in the same f ishery , I have ident i f ied only f i v e d i f ferent f i sher ies on Lake T i t i caca . TABLE 1: TYPES OF FISHING ECONOMIC UNITS (FEUs) ENCOUNTERED ON LAKE TITICACA Gear G i l 1 nets: Mesh sizes (stretched diagonal) Huayunaccana <63mm >=63mm Both ( local trawl) Craft Balsa reed boats 1 2 3 Rowing boat 4 5 6 Motorized boat 7 8 9 10 26 a) Demersal g i l l n e t f ishery for native species The demersal g i l l n e t f ishery for native species targets on the black carachi (0. agass i i ) , although ye l low carachis [0^ 1 uteus) and a few mauris (Trichomycterus) may also be caught. Fishermen set mult i f i lament nylon g i l l nets with mesh sizes between 38 and 63 mm along the lake bottom, not deeper than 30 meters ( l i t t o r a l zone) in the la te afternoon, and check them at dawn the fo l lowing morning. Those who use balsa reed-boats for th is f ishery (FEU type 1) own an average of 6.9 nets, thus somewhat less than the 9.7 nets owned by those who use wooden boats (FEU type 4). b) Trawl (huayunaccana) f ishery for native species The trawl (huayunaccana)) f ishery for native species previously targeted the boga (0. pent landi i ) , but nowadays y i e l d s mostly black carachis (0. agassi i ) . I t i s practiced in the l i t t o r a l zone of the Small Lake where the conditions required for safe and e f f i c i e n t operations are met, i .e. a f l a t bottom with l i t t l e or no macrophytes and protected waters. I t invo lves two c ra f t , i d e a l l y two l i g h t rowing boats with one or two rowers each, though a r e l a t i v e l y heavy s a i l boat may be operated in conjunction with a small row boat, or even a balsa. The corresponding FEU (type 10) thus includes one t rawl , two f i sh ing vessels , and two to four fishermen. Trawling usual ly takes place from dawn to midday, for an average of twenty tows, each las t ing f i f teen to twenty minutes (Trevino et a l . 1980; Franc et a l . 1986). c) Pelagic g i l l n e t f ishery for introduced species Fishermen from the northern and southern ends of the Lago Grande practice a pelagic g i l l n e t f ishery for introduced species using nylon g i l l nets of meshsizes between 63 and 152 mm. Those from type 5 FEUs use rowing boats, while those from type 8 FEUs use boats powered by outboard engines. A l l set the i r nets overnight in the pelagic zone of the lake, and most spend the night in the i r boats d r i f t i n g with the i r nets. During the rainy season, they set the i r g i l l n e t s c loser to , or d i r e c t l y into the mouth of the major t r ibutar ies of the lake, to catch the adult trout migrating upstream for spawning. d) Mixed species g i l l n e t f ishery A number of fishermen part ic ipate simultaneously in the demersal f ishery for native species and in the pelagic f ishery for introduced species, because they use a combination of g i l l n e t s of meshsizes both smal ler and larger than 63 mm. Although they could use a balsa (FEU type 3), almost a l l of them use a wooden boat (FEU type 6). e) Ispi f ishery Fishermen involved in the i sp i f ishery , harvest i s p i s at night with beach seines or with small trawls (Bustamante and Trevino 1976) when the f i s h come close to shore to spawn or to the surface to feed (Nunez 1982). Few i sp i fishermen f i s h for other species, and most come from only four shore communities: Jacantaya (Moho) in the Lago Norte, Llachon (Capachica) in the Bahia de Puno, Cachi Pucara (Pilcuyo) in the Lago Sur and V i lurcuni (Yunguyo) in the Lago Pequeno. f) Explo i tat ion of other lacustr ine resources Lake T i t ieaca shore dwellers frequently hunt waterfowl, c o l l e c t thei r eggs, and harvest aquatic macrophytes. These a c t i v i t i e s are most important in the economy of shore communities with access to large areas of totora beds ( i .e. totorales) and l lachu beds (i .e. l l a c h a l e s ) . Sometimes they catch waterfowl acc identa l l y in the i r f i sh ing nets, but they usual ly hunt them and c o l l e c t thei r eggs on the way back from f ish ing t r ips . They harvest aquatic macrophytes when s t i l l green as c a t t l e fodder, and as material for roofing, 28 balsa making, or for handicrafts when dry and ye l low (Lev ie i l et a l . 1985). 2.2.4. Status of Lake T i t ieaca f i sher ies Since World War II some major changes have affected Lake T i t ieaca f i sher ies which h i s t o r i c a l l y had been quite simple un t i l then (Appendix B). Exotic f i s h species were introduced into the lake: the trout in the ear ly 1940s and the s i l v e r s i d e in the la te 1950s. Subsequently, a successful rainbow trout f ishery developed in the 1960's. This f ishery supported up to f i v e art isanal canneries and reached a peak production of 400 mt in 1965. Overfishing apparently brought about i t s co l lapse and forced the canneries to close down in 1969 (Everett 1971a), however, recent evidence indicates that th is f ishery has recovered (Alfaro et a l . 1982). In the ear ly 1970s, the s i l v e r s i d e became the object of a l imi ted fishery (Wurtsbaugh 1974) which has increased ever since (Alfaro et a l . 1982; A v i l a et a l . 1985). In the same period Lake T i t ieaca fishermen have adopted a number of technical innovations. They have abandonned cotton g i l l n e t s and f i s h traps made of totora reeds in favor of nylon g i l l n e t s . Many have replaced the i r balsa reed boats with wooden boats, and some have even adopted outboard engines to propel the i r boats (Appendix B). F i n a l l y , f i s h marketing has changed considerably during th is period. Urbanization, transportation improvements and the m u l t i p l i c a t i o n of markets on the A l t i p l a n o have opened new opportunities for f i s h traders, with a consequent increase in the i r number, and in the amount of f i sh they trade (Ibid.). Data c o l l e c t e d for th i s study indicate a to ta l catch in the Peruvian portion of Lake T i t ieaca of 8,160 mt in 1980 {95% confidence i n t e r v a l : 6,490 to 9,830 mt), which i s much higher than o f f i c i a l catch figures of about 1,000 mt for Peru (MIPE 1980), and of 500 mt for the B o l i v i a n portion (Vergara 29 1980). But even after doubling to take B o l i v i a n f i sh ing into account, th is f igure is s t i l l much lower than the lake's annual productiv i ty estimated to be between 20,000 and 160,000 mt (Richerson et a l . 1977). This observation i s further reinforced by the estimation of the lake's standing biomass at 120,000 mt (Johannesson et a l . 1981) which suggests that an annual y i e l d of more than 10,000 mt could t h e o r e t i c a l l y be sustained. This has lead to the hypothesis that loca l f i s h resources were underfished or u t i l ized bel ow the i r maximum b io log ica l potential (Orlove 1979). An analysis of the res t r i c t ions on f i sh ing resu l t ing from the enforcement of communal TURFs by Lake T i t ieaca shore dwellers should indicate whether and to what extent th is underfishing does occur. 2.3. MATERIAL AND METHODS 2.3.1. Existence and structures of TURFs Demonstrating the existence of TURFs requires documenting insiders' control of access to the i r TURF area, which invo lves a number of d i f f i c u l t i e s , because few TURF holders are w i l l i n g to acknowledge the i r informal and often i l l e g a l enforcement a c t i v i t i e s (Acheson 1975). Interviewed fishermen who c la im that outsiders are to lerated in the i r area often conveniently forget to mention that they require the l a t t e r to meet certain preliminary conditions. Fishermen are thus more w i l l i n g to report on TURFs when the l a t t e r are already vanishing (Johannes 1978) or gone (Andersen 1984), and when they do not have to fear rep r i sa l s for thei r i l l e g a l enforcement a c t i v i t i e s . To these d i f f i c u l t i e s one must add the l o g i s t i c a l problems of research in geographically i so la ted regions where f i sh ing communities are widely dispersed and d i f f i c u l t to locate (Pollnac 1984). These problems are further compounded by informants' suspiciousness of enquiries in view of the ubiquity of smuggling and of drug production a c t i v i t i e s , and of the r i se of p o l i t i c a l radical ism on the Peruvian A l t i p l a n o . To demonstrate the existence of TURFs and the i r widespread d is t r ibut ion around Lake T i t ieaca , I have used three types of sources described and analyzed in a methodological appendix (Appendix C): a series of b ib l iographical references, interviews conducted during fieldwork in 1980-81 and in 1984, and a survey of the f ish ing a c t i v i t i e s of 251 fishermen in 1976. The l a t t e r r e l i e d upon a questionnaire which avoided se l f - i nc r iminat ion since informants were requested to report the problems they had encountered when try ing to operate in other f i sh ing areas rather than the i r own i l l e g a l enforcement a c t i v i t i e s . 2.3.2. Enforcement of TURFs Since both exc lus ive use and defense characterize t e r r i t o r i a l i t y (Dyson-Hudson and Smith 1978), a complete demonstration of the existence of TURFs should demonstrate both the control of TURF holders over access to f i sh ing and the repulsion of outsiders. Enforcement or defense must be documented to dist inguish TURFs from cases where exc lus ive use resu l ts from the wide dispersion of a v a i l a b l e resources or the i r a v a i l a b i l i t y in quantit ies far beyond the catching or consumption a b i l i t i e s of loca l populations (Ibid.). To document the legal arguments used by TURF holders to protect the i r resources, I have r e l i e d upon o f f i c i a l texts , and to document TURFs enforcement I have used f i e l d interviews in which I have asked informants about the d i f f i c u l t i e s they had encountered when operating outside the i r own f i sh ing areas (Appendix C). However, I have systematical ly cross-checked trespassers' accounts with those of the vict ims of the i r trespassing a c t i v i t i e s . This approach proved successful because trespassors were prompt 31 to accuse TURF holders of s tea l ing nets, and because the l a t t e r were equal ly w i l l i n g to complain of the former's misbehaviour when they did not have to mention the i r own i l l e g a l enforcement a c t i v i t i e s . 2.3.3. R«nt Capture To demonstrate that rents were captured within Lake Ti t ieaca f i she r ies , I had to demonstrate that f i sh ing revenues per FEU were greater than f i sh ing costs, which include material costs and the opportunity cost of fishermen's labor (see inequal i ty below). The l a t t e r cost i s defined as the income that fishermen could have earned in the best employment a l te rnat i ve a v a i l a b l e to them. For th i s purpose I had to se lect an appropriate time period. Because employment a l te rnat i ves cover periods of a day to a few months, I had to base my comparison on a shorter time period. I have thus evaluated economic returns to f i sh ing labour on an hourly basis , and compared the l a t t e r with the income per hour that fishermen could have earned in the most f i n a n c i a l l y rewarding occupation a v a i l a b l e to them ( i r respect ive of the l e g a l i t y of th is occupation). An addit ional advantage of th i s method i s that i t al lows me to account for the labour invested by fishermen and the i r households in the construction and maintenance of the i r c r a f t and gear, or in f i s h r e t a i l i n g (which I gener ica l l y refer to as f i sh ing - re la ted a c t i v i t i e s ) without assuming, a p r i o r i , an opportunity cost for th is labor. The inequal i ty I therefore demonstrated for an average FEU on an annual basis i s : FR > MC + (L x OC) or i t s equivalent: (FR - MC)/ L > OC where: FR = Fishing revenue MC = Material costs of f i sh ing c ra f t and gear per year L = Total labour involved in f i sh ing and re lated a c t i v i t i e s OC - Hourly opportunity cost of f i sh ing labour [(FR - MC)/ L] = Hourly return to f i sh ing 32 Because most a v a i l a b l e s t a t i s t i c s were either unre l iab le or i n v a l i d , to evaluate f i sh ing cost and revenues I had to c o l l e c t data on f ish ing y i e l d s and on the time and resources invested in f i sh ing and f i sh ing - re la ted a c t i v i t i e s . For th is purpose I used three d i f ferent surveys, a l l described in the methodological appendix (Appendix C): a Catch Assessment Survey, a Coverage Check Survey, and a socio-economic survey. To estimate f ish ing revenues I needed f i sh price figures and catch estimates. For the former, I have used the o f f i c i a l f i s h price l i s t of the Ministry of F isher ies (hereafter MIPE) because i t provides a conservative estimate of f i s h prices (Appendix C). For the l a t t e r I have used a Catch Assessment Survey (CAS) for which 50 co l laborat ing fishermen reported the i r respective catches and da i l y f i sh ing a c t i v i t i e s from August 1979 to Ju ly 1980 for a small fee (Ibid.). In addition to th is CAS, I have used a Coverage Check Survey (CCS) which was conducted twice during th is same year, approximately at s ix month in te rva l s to test for misreporting and for biases (Ibid.). To estimate f ish ing costs such as investments in f i sh ing c ra f t and gear, I have used a socio-economic survey (Ibid.). For var iable f i sh ing costs, I have used data from the CAS survey which, in addition to information on catch l e v e l s , provided information on f ish ing e f fo r t l e v e l s (e.g.: duration of f i sh ing t r i p s , number of nets used, number of crew members involved) . 2.3.4. Contribution of TURFs to rent capture To demonstrate the contribution of TURFs to rent capture, I had to show that the i r control over f i sh ing and landing space i s transformed into a control over f i sh ing e f fo r t . This required demonstrating that TURF holders are not only able to exclude outsiders, but also to control the i r own f ish ing e f fo r t . I have done the former by demonstrating that shore dwellers were successful in preventing inland dwellers from f ish ing on the lake , and the l a t t e r by demonstrating that the obl igat ions enta i led by community membership constrained the amount of household labour that shore dwellers could dedicate to f i sh ing . For th is l a t t e r purpose I have evaluated how much labour was required by communal obl igat ions including agr i cu l tu ra l production, using information a v a i l a b l e from the above-mentioned questionnaires, from a number of monographs on shore communities (Hickman 1963; Lewellen 1977; Brown 1978; V e r l i a t 1978; C o l l i n s 1982; Painter 1982) and from studies of the economy of agr icul ture in various communities (Ccama 1981; Lescano et a l . 1982; Figueroa 1984; Gonzales 1984). 34 CHAPTER 3 EXISTENCE, DISTRIBUTION AND STRUCTURE OF TURFS In th is chapter I demonstrate the existence of Lake T i t ieaca TURFs and thei r widespread d is t r ibut ion . I describe which areas of the aquatic space they encompass, the aquatic resources they include, who i s l i k e l y to benefit from them and in what ways, thus analyzing the i r s p a t i a l , temporal, demographic and cu l tu ra l dimensions (Crocombe 1974). The area dimension applies to the physical boundaries of the te r r i to ry to which the TURF system considered appl ies . The time dimension refers to the period during which i t has force. Imp l i c i t in these two dimensions i s that of a resource dimension, which corresponds to the proportion of the resource invo lved. As for the population dimension, i t refers to the s p e c i f i c i t y of the ownership. I t classes ind i v idua ls and groups according to thei r inc lus ion or exclusion from a par t i cu la r set of r ights and obl igat ions. F i n a l l y , the cu l tu ra l dimension refers to the legal (formal), or customary (informal) conditions under which the d i s t r ibu t ion , transfer and exercise of r ights apply, thus to the kind of r ights that can be exercised by TURF holders over the resources found within thei r area. 3.1. EXISTENCE AND DISTRIBUTION OF TURFS ALONG THE SHORES OF LAKE TITICACA 3.1.1. Existence of TURFs a) H is tor ica l evidence During the Inca times and the ear ly co lon ia l period which fol lowed i t , c o l l e c t i v e property r ights , not only over beaches and landing spots, but also over f ish ing zones, were common in Peru (Antunez 1981; Buse 1981; F lo res -Gal indo 1981; Ramirez-Horton 1981; Rostworowski 1977; 1981). According to Buse, the Inca administrators had even combined th is a l l o c a t i o n of f i sh ing space with an adequate control of fishermen's density, using such drast ic measures as the forced transfer of f i sh ing populations to new f ish ing zones along the P a c i f i c Coast and eventual ly , on inland water bodies (Buse 1981). In the case of Lake T i t ieaca , the Inca emperor Huayna Capac granted exclusive r ights over aquatic resources to the Uros ethnic group, now extinguished, at the beginning of the 16th century: "[He] determined how the Uros had to l i v e on that lake, and divided i t s shores where they had to f i s h " (determino l a manera en l a que 1 os Urus  debian v i v i r en ese lago, y d i v i d i o los bordes donde elTos tenian que  pescar), (CabeTT o~B"a I boa Llbbttj, quo'Eeci in Manells 1 9 / 3 : 1 4 / , tootnofe 2). Evidence that the Inca empire, or Tahuantinsuyo, was a centra l ized organization (Pease 1985), and ear ly chroniclers ' c la im that the same organizational structures were repl icated everywhere (Ramirez 1985: 420-30) confirm th is testimony. Within the Tahuantinsuyo, l o c a l i z e d kinship groups, or ayl 1 us, were granted formal recognit ion, and exclus ive r ights over the resources within thei r t e r r i t o r i e s (Murua in Ramirez 1985: 430). There are strong reasons to be l ieve , in p a r t i c u l a r , that the c o l l e c t i v e f i sh ing r ights found a l l along the P a c i f i c shores of the Tahuantinsuyo (Rostworowski 1977) were p a r a l l e l e d on those of inland water bodies. Antunez provides the example of the d i v i s ion of the Rimac River in two f ish ing zones: two di f ferent f i sh ing groups had exclus ive f i sh ing r ights , each in only one of these two zones respect ively (Antunez 1981: 33). In addit ion, forced transfers (mitimaes) of f ish ing populations from the P a c i f i c coast, to the shores of inland water bodies, such as the transfer of a 36 Moche group of fishermen from Northern Peru to the shores of Lake T i t ieaca and to the Maranon River (Buse 1981) are l i k e l y to have resulted in the establishment of f ish ing zones, either through the granting of f i sh ing r ights by the Inca administrators, or through the transfer of the ensuing custom by the transferred populations themselves. A l t e r n a t i v e l y , the existence of A l t i p l a n o colonies on the Southern P a c i f i c Coast of Peru, since the Tiahuanuco period preceeding the Tahuantinsuyo, and the extensive contacts between these colonies and the A l t i p l a n o (Hyslop 1976; Mujica 1985) lead to the hypothesis that centr i fugal influences for the organization of f i sh ing may have come from the A l t i p l a n o to the coastal area, instead of the reverse. During the co lon ia l period, the Spanish administration reorganized the t e r r i t o r i e s of the ayl lus . I t stripped many of them of thei r most distant sections. I t a lso forced them to regroup in s ing le settlements, or reduceiones. However, the number of human settlements along the shores of Lake T i t ieaca was not modified then, as the correspondance between t rad i t iona l ayl 1 us and modern administrative units demonstrates (Martinez 1981). This suggests that the boundaries of loca l aquatic t e r r i t o r i e s were not d r a s t i c a l l y modified. The Spanish ru lers also t r ied to open f ish ing to a l l , i r respect ive of ethnic or geographical or ig ins , and to promote the development of commercial f i sher ies . Apparently they met with l i t t l e success at f i r s t . In 1566, the Spanish administrator Gregorio Gonzales de Cuenca, for example, had to revoke his i n i t i a l decision to open ocean f i sher ies access to a l l , when confronted with the strong opposition of indigenous fishermen of the northern coast of Peru (Rostworowski 1981). I t was not un t i l the early 1770's, that repeated c o n f l i c t s between neighboring f i sh ing communities eventual ly leading to j u d i c i a l actions 37 provided the Spanish co lon ia l government with a j u s t i f i c a t i o n to open access to landing spots (caletas) to a l l , and to a l low everyone to f i sh (F lores-Gal indo 1981:160-1; Rostworowski 1981:85-86). Customary access r ights to landing spots and f ish ing zones must have persisted for decades after that, since in 1786 the Gobierno Superior del Peru had to reafirm the "1 ibertad de  pescar" decreed by the V is i tador Escobedo (Flores-Galindo 1981:160-161), and since lawsuits opposed various f i sh ing communities at least unt i l the la te 1790s (Rostworowski 1981: 85). A l l th is suggests that t e r r i t o r i a l f ish ing r ights were formal ly recognized on Lake Ti t icaca's shores for more than 250 years (1500's to 1770's). Evidence that Uros people were the only one required to pay thei r t r ibute to the Spanish crown in dried f i s h suggests that informal respect of these rights may have persisted unt i l the beginning of th is century, as i t did on the P a c i f i c coast (F iedler 1944). To determine whether they have persisted unt i l the present requires addit ional information. b) B ib l iographical evidence. The existence of TURFs over exclusive f i sh ing areas along Lake Ti t ieaca shores has been mentioned by various contemporaneous authors. Most of them have claimed that f i sh ing on Lake Tit ieaca and i t s t r ibutar ies was an a c t i v i t y undertaken by shore dwellers exc lus i ve l y (Romero 1925: 438; Martinez 1962: 55; Gal do 1962a: 92; 1962b: 66; 1967: 188; Poe 1979: 213; C o l l i n s 1982: 106). Some of them have referred to the ownership of landing spots (Pacori 1976: 15) and of f ish ing weirs by ind iv idua ls (Tschopik 1946: 525; Sole 1969: 51) or by extended fami l ies (Ve l la rd 1949: 180). Others have indicated that f i sh ing places (Wegener 1934, quoted in La Barre 1948: 184-185), or f ish ing zones (Tschopik 1946: 521; Gal do 1962a: 92; Sole 1969: 53) were owned by the immediately adjacent f ish ing communities. 38 However, l i t t l e information i s a v a i l a b l e regarding the extension, d is t r ibut ion and character ist ics of the corresponding TURFs. After indicat ing the existence of f i sh ing t e r r i t o r i e s within one par t i cu la r community, such as Chucuito in Peru (Tschopik 1946: 521) or Suriqui in B o l i v i a (Sole 1969), or within a group of communities unevenly distr ibuted around the Lake (Galdo 1962a), most contributors have claimed that TURFs were enforced a l l along Lake Ti t icaca's shores (Tschopik 1946, Sole 1969) and those of i t s major t r ibutar ies (Galdo 1962a). Even i f cumulatively the number of communities surveyed gives some c r e d i b i l i t y to such a statement, these authors f a i l to provide su f f i c ien t evidence to support thei r c la im. Fortunately, information from a survey of f i sh ing settlements carr ied out in 1976 along more than 90% of Lake T i t icaca's Peruvian shores confirms the TURFs' ubiquity. c) Empirical evidence. In 1976, the sc ient is ts from the Peruvian Marine Inst i tute (IMARPE) carr ied a comprehensive survey of Lake T i t ieaca f i sh ing settlements. They enumerated 3,040 fishermen and personally registered more than 60% of them, a l l residents of onshore communities (E. Bustamante and H. Trevino 1984: pers. com.). During th is survey, they used a structured questionnaire to interview 8.2% of the fishermen censused. They asked each informant whether he had encountered d i f f i c u l t i e s when operating in other areas, and recorded the type of problems reported i f any. I have summarized informants' answers in Table 2 and 3 (IMARPE-Puno 1976: Unpublished data). In Table 2, I have c l a s s i f i e d the 251 interviewed fishermen according to whether they claim to encounter d i f f i c u l t i e s when operating in other areas or not. The 40.6% informants (Group I) who claim to have problems both confirm TURFs existence and thei r act ive enforcement. 39 TABLE 2: DISTRIBUTION OF ANSWERS TO QUESTION "DO YOU ENCOUNTER DIFFICULTIES WHEN OPERATING IN OTHER FISHING AREAS?" (DATA SOURCE: IMARPE-PUNO 1976) Yes Interference and theft of nets Not specif ied 32.3% 8.3% No Do not operate in other areas Frequent 1 f i sh ing s i te Frequent 2 f ish ing s i tes Frequent 3 s i tes or more 36.7% 8.0% 5.6% 6.7% No answer 2.4% TURF INDEX Confirm that TURFs ex is t Do not confirm nor deny i t 90.9% 9.1% TABLE 3: DISTRIBUTION OF RELATIVE FREQUENCIES FOR THE NUMBER OF FISHING SITES BY GROUP OF FISHERMEN (DATA SOURCE: IMARPE-PUNO 1976) Number of s i tes Group I Group II Group IIA Group IIB 1 2 3 4 to 17.6 25.5 28.4 17.6 10.8 42 23.8 19.6 7.0 7.7 43.5 21.7 15.2 7.6 11.9 39.2 27.5 27.5 5.9 0.0 Total 100 100 100 100 Group I = trespass but encounter d i f f i c u l t i e s Group II = do not encounter d i f f i c u l t i e s Group IIA = do not encounter d i f f i c u l t i e s because do not trespass Group I IB = do not indicate why do not encounter d i f f i c u l t i e s 40 Their answers imply that they trespass into other TURF areas and have to endure the brunt of TURF holders' r e t a l i a t i o n . The 57.0% fishermen who claim not to have such problems can be divided into two groups (group I la and l i b ) . Those in the f i r s t group ( Ila) indicate that they do not experience any problems, because they do not operate in other areas, that i s because they do not trespass. Those in the second group ( l i b ) do not indicate whether they operate outside the i r own area or not. This second group ( l i b ) can be subdivided according to the number of f ish ing s i tes indicated by each fisherman (Table 3) since a l l informants were asked about the f ish ing areas they frequented in d i f ferent f ish ing seasons. Most of the names of f i sh ing s i tes are e a s i l y i d e n t i f i a b l e and can be related to those of the f ish ing settlements and of the various parts of the i r t e r r i t o r i e s . An analys is of these names reveals that the fishermen who indicate only one f ish ing s i te (39.2% of group l i b , or 8% of the informants) and those who indicate two f ish ing s i tes (27.5 of group l i b , or 5.6% of the informants) operate within the f ish ing areas of thei r respective communities. As for the fishermen who indicate three or more f ish ing s i t e s , although some of them do remain within the f i sh ing areas of the i r communities, i t i s not c lear how many ac tua l l y do so (Table 3). Summing up the percentages of fishermen corresponding to the groups and subgroups which do confirm TURFs existence y i e l d s an estimate of the proportion of interviewed fishermen who ac tua l l y confirm TURFs existence, which I have c a l l e d a TURF index (Table 2). This index sums up to 91% approximately. The analysis and interpretation of informants' answers thus leads to the conclusion that at least 91% of the fishermen interviewed in 1976 confirmed the existence of TURFs, and that at least 40.6% of them confirmed TURFs act ive enforcement, while the remainder (9%) simply did not provide 41 evidence either way. Not a s ingle one of the 251 informants denied the existence of loca l TURFs, i r respect ive of his own attitude towards them. An analysis of the same data in terms of f ish ing communities rather than fishermen, also demonstrates that for a l l but two communities (94%), at least one fisherman confirms TURFs existence. As for the remaining two, other sources indicate that for the f i r s t one, Jasincoya (Huancane), loca l fishermen do r e s t r i c t f i sh ing to thei r own bay (Bustamante and Trevino 1976: f ie ldnotes) , and that for the second one, Isani (Huancane), TURFs were ac tua l l y enforced in previous years (Galdo 1962a). 3.1.2. Distr ibut ion of TURFs along the shoreline A sample of 80% of the shore communities shows that v i r t u a l l y a l l of them hold TURFs over the portion of the lake immediately adjacent to the i r shoreline (Table 4). Furthermore, by locat ing these communities along the shoreline (Figure 3), I was able to ver i fy that there i s no gap in thei r d is t r ibut ion . Each shore section corresponds to a TURF area, and no gap i s l e f t open for in land dwellers to get access to aquatic resources (even i f in some areas TURF holders rarely f i sh) . TABLE 4: SHORE COMMUNITIES OF LAKE TITICACA FOR WHICH THERE IS EVIDENCE OF THE EXISTENCE OF TURFS Lake area Number of shore Communities sampled communities Number Proportion Bahia de Puno 50 44 88 Lago Norte 43 35 81.4 Lago Sur 43 28 65.1 Lago Pequeno 29 25 86.2 Total 165 132 80 42 FIGURE 3 : SOURCES OF EVIDENCE OF THE EXISTENCE OF TURFS ON L A K E TITICACA. 1980-81 a 1984 fieldwork 4 3 3.2. SPATIAL DIMENSIONS So far the evidence provided suggests that most of Lake Ti t ieaca shore dwellers are members of t e r r i t o r i a l communities who hold TURFs over part of the aquatic space adjacent to the i r land te r r i to ry . These communities have wel l defined t e r r e s t r i a l boundaries formally recognized by the administration. Their t e r r e s t r i a l boundaries extend into the lake , either perpendicularly to the shore, or fo l lowing some i r regu lar but c l e a r l y defined trajectory , with the same types of v isual markers often being used to delineate both aquatic and t e r r e s t r i a l boundaries. The physical extension of the aquatic te r r i to ry delineated by these aquatic boundaries corresponds to TURFs spat ial dimension. To describe the spat ial dimensions of Lake Ti t ieaca TURFs, I use evidence gathered during f i e l d work in 1979-81 and 1984, data co l lec ted in the f i e l d by sc ient is ts from the IMARPE, documents from the Ministry of Agr icul ture in Puno such as maps drawn by shore dwellers of the te r r i to ry of the i r communities, and a few b ib l iographical references (Bustamante and Trevino 1976; Nunez 1982). I consider f i r s t the l a t e r a l or intercommunal boundaries which separate the TURF areas of neighbouring communities and determine the width of the shoreline corresponding to each shore community. Later, I consider the posit ion of the offshore boundaries which determine the distance from shore to which TURFs apply (which I c a l l the length of the TURF areas) and the few cases where offshore boundaries turn into intercommunal boundaries. F i n a l l y , I turn to the v isual markers used to delineate aquatic boundaries, because of thei r influence on the amount of overlap between neighbouring TURF areas. 3 .2.1. L a t e r a l boundaries. Members of Lake Tit ieaca shore communities consider the aquatic space immediately adjacent to thei r land as thei r c o l l e c t i v e aquatic property. The 44 l a t e r a l boundaries of these communal t e r r i t o r i e s are extended in direct prolongation of the t e r r e s t r i a l boundaries which separate neighbouring communities, as i l l u s t r a t e d by community members in graphical depictions of thei r community such as those of Ramis (Figure 4) and of Requena in the Lago Norte (Figure 5). In some cases, the l a t e r a l boundaries may change direct ion r e l a t i v e l y to the t e r r e s t r i a l ones, as those between Sajo and the neighbouring community of V i l l a Santiago de Ccama in the Lago Sur (Figure 6). The or ig in of the intercommunal boundaries along the shores of Lake T i t icaca can be interpreted in terms of the h i s to r i ca l attempt by Andean communities to control a maximum number of ecological l e v e l s distr ibuted along a ve r t i ca l gradient (Murra 1985a; 1985b). At least since Tiwanaku times (1500 BC to 1000 AD), Andean communities have s t r iven for s e l f - s u f f i c i e n c y , rather than r e l i e d upon the exchange of goods between specia l i zed and independant producers (Browman 1981; Masuda et a l . 1985). An important character is t ic of the v e r t i c a l i t y p r inc ip le i s i t s compatibi l i ty with a phys ica l l y discontinuous te r r i to r y , often described as a ve r t i ca l archipelago whose is lands are characterized by s t r i k i n g l y d i f ferent ecological conditions. To use the land and resources of these is lands , Andean people have thus had to migrate p e r i o d i c a l l y from one to the other, taking advantage of the lags between asynchronous production cycles in d i f ferent ecological l e v e l s . The system of forced migration and colonizat ion widely practiced by the Incas is thought to have contributed to or reinforced the p r inc ip le of v e r t i c a l i t y common to Andean communities, while the Spanish colonizat ion and administration i s considered to have weakened i t considerably (Saignes 1978; Masuda et al 1985). The appl icat ion of the v e r t i c a l i t y p r inc ip le to Lake T i t icaca shore communities implies that the l a t t e r should stretch the i r t e r r i t o r i e s across as 45 FIGURE4: FACSIMILE OF THE MAP OF THE COMMUNITY OF RAMIS, T A R A C O * (1974, Direccion de Comunidades Campesinas, MINA-Puno) * No s c a l e g i v e n on o r i g i n a l . Redrawn by the author b o u n d a r y 4 6 FIGURE 5 : FACSIMILE OF THE MAP OF THE COMMUNI -TY OF REQUENA, TARACO (Direccion de Comunidades CampesinaS, M I N A - P U N O ) . * No scale on o r i g i n a l . • h o u s e • b o u n d a r y ma rke r bound a ry 50km FIGURE 6 : FACSIMILE OF THE MAP OF THE C O M M U -NITY OF SAJO, POMATA (Direccion de Comunidades C a m p e s i n a S , MINA-Puno). * No sca le given on original Redrawn by the author - ^ ^ » B o u n d a r y 4 8 many ecological l e v e l s as possible , from the lakeshore to the top of a nearby mountain range, a physical imposs ib i l i t y for the communities of the vast p la ins of H a v e , P i l cuyo , Rami's or Taraco. This al lows community members to combine a variety of agr icu l tura l zones and pract ices, thus spreading the r isks of poor environmental conditions. A g r i c u l t u r a l i s t s with widely dispersed p lots of land are less l i k e l y to suffer a catastrophic crop f a i l u r e , than those whose p lots are concentrated in one ecological area (Brush 1977). In a region where f rost i s a serious r isk (Morion 1978; Figueroa 1984) th is i s an important consideration for agr icu l tura l practice. The t e r r i t o r i e s of a sample of 32 communities distr ibuted along the shores of the Lago Sur and Lago Pequeno ac tua l l y ver i fy th is prediction. They look l i k e elongated t r iangles whose bases correspond to the shorel ine, and apices to the summit of the Cerro Ccapia (Figure 7). Computing the rat io of the length of their shoreline to that of thei r boundary perimeter (Table 5) shows that for a majority [78%) of them, i t i s smal ler than 30%, and for ha l f of them smaller than 15%. TABLE 5: FREQUENCY DISTRIBUTION OF PROPORTION OF SHORE LENGTH TO PERIMETER FOR LAKE TITICACA SHORE COMMUNITIES. (SOURCE: PLANO CADASTRAL, MINA-PUNO, 1973) Range of values Absol ute Relat ive Cumulati ve frequency frequency frequency 0 to 5 3 9.4 9.4 5 to 10 7 21.9 31.2 10 to 15 6 18.8 50 15 to 20 3 9.4 59.4 20 to 25 3 9.4 67.8 25 to 30 3 9.4 78.1 over 30 7 21.9 100 Total 32 100 49 FIGURE 7 : SHORE COMMUNITIES OF THE LAGO SUR AND LAGO PEQUENO (Direccion de Comunidades Campesinas, MINA-Puno 1973). This confirms that the shores of Lake T i t icaca are divided into r e l a t i v e l y narrow st r ips between neighbouring communities whose ter r i to ry often stretches far inland over a variety of ecological l e v e l s . How far offshore the aquatic portion of th is te r r i to ry stretches i s the topic of the fo l lowing section. 3.2.2. Offshore Boundaries According to the evidence a v a i l a b l e , physical and ecological factors determine to a large extent the distance from shore to which aquatic boundaries extend (Lev ie i l and Orlove 1987). As a general r u l e , communal f ish ing zones include a shallow water area, often demarcated by the presence of aquatic macrophytes, and an area of open water, both of var iable width. Since i t i s the slope of the bottom which determines the distance to which TURFs apply, I have distinguished between three d i f ferent types of TURFs which grade continuously into one another, for the purpose of th is analys is . a) Type I TURFs TURFs of the f i r s t type are found where the shallow water area extends to a great distance from the shorel ine, because of a gentle bottom slope. In Puno Bay and the Ramis del ta area for example, th is area i s greater than 5 km. These areas are characterized by the presence of aquatic macrophytes, more p a r t i c u l a r l y by that of dense beds of totora reeds. Communities with type I TURFs extend the i r boundaries far into these totora les , and usual ly claim some open water space, not more than a couple of hundred meters wide, on the i r outer edge. I t i s d i f f i c u l t to associate the outside boundary of type I TURFs with any par t i cu la r depth contour, though i t rarely goes beyond 3 meters and almost never beyond 5 (Figure 8). 51 FIGURE 8 : T U R F T Y P E S : I, H & HE FIGURE 9 : DISTRIBUTION OF THE THREE TYPES OF T U R F S A R O U N D L A K E TITICACA, P E R U . TYPE m 5 3 Type I TURFs are common in the southeast part of Laguna Arapa, in the northwest of the Lago Grande between the Ramis del ta and Capachica peninsula, in the northern ha l f of Puno Bay, and a l l along the western shores of the Lago Grande between Socca is land and J u l i (Figure 9). In the l a t t e r case, however, the outer edge of the totora reed beds i s demarcated by a l i t t o r a l dune or shicata which i s cu l t i va ted during the rainy season. Local community members s t i l l c laim the open water space on the outer edge of the dune as part of their exclusive f i sh ing zone, even i f they rarely exp lo i t i t , since they would need two craf t for that purpose: one to cross the lagoon between the main land and the dune, and the other one to go f ish ing on the lake. A l te rnat i ve solut ions could be used too, such as carrying the f i sh ing craf t over the dune, as done by the fishermen from Balsapata in the Lago Norte (Bustamante and Trevino, f i e l d notes 1976) or a small ferry service for fishermen to cross the lagoon. But th is requires some organization and cooperation on the part of the fishermen (between at least s ix of them in the case of Balsapata). Another solut ion t r ied in 1983 by some fishermen from the community of Santa Rosa de Yanaque (Acora), i s to move into a temporary dwell ing on the dune. However, th is experiment came to a rapid end, because i t required too much spec ia l i za t ion on the part of the fishermen involved who, lacking f l e x i b i l i t y , could not continue thei r normal agr icu l tu ra l practices on the mainland. b) Type II TURFs For the second type of TURFs, a steeper bottom slope brings the outer edge of the totora reed beds within a few hundred meters from shore. In such cases, loca l community members claim the totora beds in the shallow waters and an area of open and deeper water at least a few hundred meters wide, as thei rs . Given that totora reeds rarely grow in water deeper than 4 meters (Tutin 1940; Col l o t 1981), the outer l i m i t of type II TURFs i s often close to the 10 meters depth contour (Figure 8). However, when th is depth occurs f a i r l y c lose to shore, the outer edge of type II TURFs can be pushed out as far as the 20 meter depth contour. Type II TURFs are found along the northwest and southeast shores of Laguna Arapa, in the bays of M i l i l a y a , Moho and Vi lquechico on the Northeast shores of the Lago Grande, on the western sides of the Capachica and Chucuito peninsulas in Puno Bay, and of the Lago Pequeno (Figure 9). c) Type III TURFs For the th i rd type of TURFs, the shallow water area and the totora reed beds are only a few meters wide, because of a very steep bottom slope. The l a t t e r may even disappear altogether, because of a rocky substrate or exposure to wave action. In these areas, a l te rnat i ve c r i t e r i a of demarcation are used, and the width of the corresponding area i s highly var iab le , depending on the f i sh resource and harvesting methodology invo lved. In the case of g i l l n e t f ish ing for demersal (bottom) native species, the 50 meters depth contour provides a reasonably good approximation of outer edge of type III TURFs, even i f g i l l n e t s for demersal f i sh are rarely anchored deeper than 25 meters (Figure 8). This can be observed along the shores of the Lago Norte between T i l a l i (Huancane) and Jacantaya (Moho), around the Jonsani peninsula, on the eastern sides of Capachica and Chucuito peninsulas, of Socca and Iscata is lands , and along the shores of the Lago Sur from Jul i to Yunguyo (Figure 9). A l t e r n a t i v e l y , in the case of the trawl f ishery for i s p i , no depth contour seems to correspond to the outer edge of type III TURFs. Distance from shore and competition with fishermen from other communities appear to be the major determinants for the width of type III TURFs in th is case, as can be 55 observed in the s t r a i t separating the communities on Amantani and Taqui l i is lands and that of Llachon at the southern t ip of Capachica peninsula. In the few cases where i s p i s are fished with beach seines, communal ownership of the beaches is determined by the posit ion of intercommunal t e r r e s t r i a l boundaries, as in the case of Jacantaya (Moho) and i t s neighbours, in the Lago Norte. Beach seining is therefore of l i t t l e relevance for the determination of the outer edge of TURFs. Whenever communities are s i t t i n g on opposite sides of narrow bodies of water, they have to div ide the corresponding water space between themselves. The general practice then i s for an imaginary l i n e to be drawn, approximately equidistant from both sides. This can occur in r e l a t i v e l y small bays whose shores are divided between a number of neighbouring communities, such as the bay of Moho whose shores are divided between Lacasani and Muel le Cariquita in the Lago Norte, in s t r a i t s between is lands such as in the one between Taqui l i and Amantani i s lands , or in s t r a i t s between is lands and peninsulas such the one between I s l a Iscaya and V i lurcuni (Yunguyo). This can a lso occur along r i v e r mouths, such as that of the Ramis River which i s divided between the communities of Balsapata and Ramis, or that of the Have River which i s divided between the communities of Huayl lata , and Santa Rosa de Huayl lata (Bustamante and Trevino 1976: 53-54). Natural features when they occur are a lso taken advantage of, i r respect ive of the i r equidistance from the shores of both communities, as for the Rio Tujsa-Jahuira in the part of the Puno bay c a l l e d Huayl lata Ccota, which div ides the totoral of S i l l a m u r i between the communities of Ccarana, Cuchiraya and I s l a Quipata to the northeast, and those of Chucuito-Si l 1 amuri-Attoja and Chincheras to the southwest. That natural features may determine TURFs boundaries underscores the importance of v isual markers in thei r de l ineat ion , which i s the topic of the next section. 56 3.2.3. Visual markers According to shore dwel lers , the l a t e r a l boundaries of Lake Ti t ieaca TURFs are extended in d i rect prolongation of the t e r r e s t r i a l boundaries which separate neighbouring communities and, i f necessary, v isual markers or hitos are used to delineate them. The l a t t e r are r e l a t i v e l y simple natural or man-made features of the landscape, which are used to delineate TURFs l a t e r a l boundaries rather than thei r outer edges. The only type of v isual markers used to delineate TURFs outer edge are the totora beds and the natural channels d iv id ing them, which can be found in type I and type II TURFs. Environmental conditions determine to a large degree what type of v isual markers can be used in spec i f ic circumstances, either because i t i s best suited for a given area or TURF type, or because appropriate material i s 1 ocal l yava i1ab le . For type I TURFs, the presence of totora reeds in shallow waters greatly s impl i fy the problem of boundary marking. The natural channels cr isscrossing the totora beds can be used and, i f necessary a r t i f i c i a l l y enlarged, both for l a t e r a l and offshore boundaries. In the case of Huerta and M i l l o j a c h i Huaraya in the Bahia de Puno, for example, the Rio Hui le , a large natural channel, delineates TURFs outer edge, while the smaller channels of the Ranja Mayo and Sanja Mayo separate the areas of th is community from those of the neighbouring communities of J i r a t a Huaraya and Chincheros respect ively . A few i s l e t s l i k e that of Lampa K'ara are a lso used as boundary markers along these channels (Figure 10). Closer to shore, in very shallow waters, man-made markers pers ist ing from the times in which the lake l e v e l was Tower, such as lanes, ridges, trenches and wide furrows or sanjas are used. Stones are rarely used in those areas, simply because they are not l o c a l l y a v a i l a b l e in su f f i c ien t quantit ies. 57 FIGURE 10: F A C S I M I L E OF THE M A P OF THE TOTORA R E E D B E D S OF THE COMMUNIT IES OF H U E R T A A N D MILLOJACHI HUARAYA*(CENFOR-Puno) PA&CtALlC* OP'S. -Z>£~ HOcTO-T* - ^ • * i M/ V/ v/ V/ v/ vv ^ ^ \n V / / / N / " J/ T*^ (/ V v ' V s^-^ V ^ «]/ \l' y (/ ^ (/ i/' w 1/ y V / s i v v J v u i/ y V ij V W ^ \/ ^ ^ ' f 4/ . V (/ ^ Ir' ^ f * No s c a l e g i v e n on o r i g i n a l R e d r a w n by t h e a u t h o r B o u n d a r y 5 8 As for natural markers, they are rarely v i s i b l e on a low shore l i n e hidden behind t a l l reeds. When they are high enough to be v i s i b l e , whether in the form of h i l l s , i s lands , as in Huerta and M i l l o j a c h i Huaraya, or in the form sand dunes as in Ramis and Requena in the Lago Norte, people incorporate them into the i r system of boundary marking (Figure 10, 4 and 5). In the case of type II TURFs, a steeper shoreline and r e l i e f make land marks much more readi ly a v a i l a b l e for boundary iden t i f i ca t ion . Natural features such as prominent rocks, h i l l s or promontories on shore, and small is lands offshore are often used for that purpose. In various communities of the southeastern shores of the Peninsula of Chucuito, such as Perca, Sihuicani and Huincal la for example, loca l fishermen consider that the i r respective f ish ing areas correspond to the open bays facing thei r communities which are l imi ted by small peninsulas on both sides. Although i t may occur much more frequently in TURFs of type II than in those of type I, simply because of shore r e l i e f , onshore promontories may a lso play a ro le in the del ineation of type I TURFs. In Huerta and Mi l lohachi Huaraya, for example, TURFs l a t e r a l boundaries fo l low natural channels within the totora beds, but they also correspond to the onshore promontories of Cerro Ch'uri and J a l l o Pata (Figure 10). When natural features are missing, TURFs holders re ly upon material a v a i l a b l e l o c a l l y to delineate the i r boundaries. Rocks and stones for example are found in such abundance in areas corresponding to type II TURFs that i t i s often a problem for loca l farmers to get r i d of those they p u l l out of the i r f i e l d s . Rocks are p i l e d up on the side of agr icu l tu ra l p l o t s , or used to bu i ld terraces and small w a l l s to delineate pr ivate p lots and communal boundaries. Submerged wa l l s reminiscent of times of lower lake l e v e l s can be seen in many areas of type II TURFs . Most corrrespond to pr ivate rather than 59 communal boundaries. In a few instances, however, stone constructions b u i l t on shore and la te r submerged provide a c lear del imitat ion of aquatic t e r r i t o r i e s , as can be observed in between the various communities between Chimu and Cusipata on the Western shores of the Puno Bay, or in the Northeastern part of Laguna Arapa. When rocks and stones are less abundant, trenches up to a couple of meters wide and a ha l f meter deep, cal led sanjas or rayas, are often dug through near shore totora beds and in the mud of the l i t t o r a l . These trenches extend t e r r e s t r i a l boundaries to the aquatic space, as in between Pucani and Tacasaya, or between Anu Cal le jon and P a l l a l l a on Chucuito Peninsula, and a l l along the shores of the Lago Sur from I s l a Socca to Santiago Mucho. F i n a l l y , in the case of type III TURFs, a l te rnat i ve c r i t e r i a of demarcation have to be used, because totora beds are almost non-existent and water depth prevents the construction of wa l l s further than a few meters from shores. Physical features such as small v a l l e y s , outstanding rock formations, or i r r e g u l a r i t i e s of the shoreline are convenient, because they are eas i l y noticed from offshore. In the community of Llachon, at the southern t i p of Capachica peninsula, for example, fishermen refer the canyon of the K'o l lpa Mayo (an intermittent stream) as thei r boundary with the community of Yapura. Onshore man-made markers, such as lanes, wa l l s or houses are also commonly used since a rapid shore e levat ion makes these markers v i s i b l e from quite a distance. In Llachon, people also refer to the dock of C h u c a r i l l a , which l i e s at the mouth of the K 'o l lpa Mayo, as the i r boundary with Yapura (Figure 11). In conclusion, i t may be said that the outer edge boundary of type I TURFs i s in most cases better defined than that of type I I , and always better defined than that of type III TURFs. The reason for th is i s that v isual markers in the form of totora beds are nearly always a v a i l a b l e for outer edge 60 FIGURE II: FACSIMILE OF THE MAP OF THE COMMUNITY OF L L A C H O N , CAPACHICA (NUNEZ 1982) * n o s c a l e g i v e n o n o r i g i n a l del ineation in TURFs of type I and I I , while they are almost always absent in those of type II I . Lateral boundaries are also often better defined in TURFs of the f i r s t two types than in those of the th i rd type, for the same reason. Because of t h i s , var iat ions may be expected in the degree of overlap between neighbouring TURFs according to the i r type. F i n a l l y , despite the i r c ruc ia l ro le as boundary markers, totora beds are more properly considered as resources. As informal questioning during fieldwork revealed, outsiders are wel l aware that they have to stay at least a couple of hundred meters away from them, not to become suspect of any wrongdoing. This leads me to consider the aquatic resources included within TURFs areas, thus TURFs resource dimension. 3.3. RESOURCES DIMENSION Lake T i t icaca shore dwellers bel ieve that aquatic boundaries apply to the aquatic space and a l l the resources therein, jus t as t e r r e s t r i a l boundaries do to land space and resources. Aquatic resources are of two types: b io log ica l and phys ical . B io log ica l resources of economic interest include f i s h , aquatic macrophytes, and both waterfowl and thei r eggs. Physical resources include the land and shallow waters of the shorel ine, both of which are used by shore dwellers for c raf t storage, navigation, f i sh ing , macrophyte harvesting, l i v e f i s h storage, waterfowl hunting and egg c o l l e c t i o n . The objective of th is section i s to determine which aquatic resources are l i k e l y to be affected by TURFs1 existence. Since these resources are heterogenously distr ibuted along and across the l i t t o r a l zone, and since the outer edge of most TURF area coincides roughly with the 50 m isobath or with a distance of 1 km from shore (whichever comes f i r s t ) , the question is what proportion of each type of aquatic resources i s found within the corresponding l i t t o r a l zone. This raises the question of f i s h horizontal migrations whether perpendicular to shore from the TURF-control1 ed l i t t o r a l fr inge to the "no man's land" beyond i t , or p a r a l l e l to i t from the f ish ing zone of one community to that of another one. 3 .3.1. F i s h Resources. For th is sect ion, I have had to re ly upon fishermen's accounts whenever published information was unavai lab le , or when the l a t t e r was of doubtful s c i e n t i f i c value. The relevant information i s summarized in graphical form in Figure 12. For th is f igure I have used the ecological zonation of Loubens, Osorio and Sarmiento (1984) because i t uses the same c r i t e r i a as those upon which my d is t inc t ion between TURF types i s based: depth, topography and the presence of aquatic macrophytes. The s ix zones considered are: the deep zone (zone 1: 50 to 284 m deep), the medium depth zone (zone 2: 10 to 50 m deep) with no bottom vegetation, the Chara zone which bottom is blanketed by Chara and and other lake plants c o l l e c t i v e l y referred to as l lachu by the shore dwellers (zone 3: about 2 to 10 m deep), the totora zone characterized by the presence of beds of Scirpus c a l l e d totora by the shoe dwellers (zone 4), the l i t t o r a l zone characterized by i t s shallowness, muddy bottom and the presence of 11 achu (zone 5), and the rocky zone which corresponds to steep rocky shores with l i t t l e or no macrophytic vegetation (zone 6) (Loubens, et a l . 1984: 155). A simple comparison of the character is t ics of these zones and those of the three types of TURFs demonstrates a close correspondance. Areas of type III TURFs coincide with zone 6, while those of type II encompass zone 2 to 5, and those of type I encompass zone 5 and either part of zone 4, or the whole of zone 4 and part of zone 3 (Figure 8 and 12). 63 FIGURE 12: R E S O U R C E DIMENSION OF L A K E TITICACA T U R F S Fig . 13a: Orest ias a g a s s i i T y p e I T y p e U juveni le form p e l a g i c fo rm E E 3 d e m e r s a l f o r m Fig . 13b: 0. luteus, o l i vaceus ,mul le r i Type I 5 4 f lo ods Type I 3 2 F ig . 13c: O r e s t i a s p e n 11 a n d i i F ig . 13d: 0. i_spj , 0. f o r g e t i S p a w n i n g o n " L L a c h u " S p a w n i n g on s h o r e 2 8 4 -F i g . l 3 e : B a s i l i c h t h y s b o n a r i e n s i s T y p e I T y p e I B f i n g e r l i n g s EMJ j u v e n i I e s a d u l t s F i g . 13 f S. g a i r d n e r i , Tri c h o m y c t . 6 4 m 6 . g a i r d - v ^ n e r i T r i c h o m y c ^ " " e r u s a) Carachis . The carachi bianco or negro (0. agassii) i s the species for which most information i s a v a i l a b l e . I t i s a very polymorphic species, capable of coloniz ing the most diverse habitat in the lake (Lauzanne 1982). Loubens and his col laborators (1984; 1985) indicate that adults are found either in a benthic or in a pelagic form in zone 2 and 3, juven i les in a l i t t o r a l form in zone 4 (Figure 12a), and that spawning occurs year round in zone 3 to 5 with a s l i g h t maximum during the dry season. They also indicate that 0. agassii i s d e f i n i t e l y a perimacrophytic species (Loubens et a l . 1984; Loubens and Sarmiento 1985). This shows that the majority of 0. agassi i 's habitat i s included within exc lus ive f i sh ing areas where type I TURFs ex is t , and that i t i s t o t a l l y included in them where type II TURFs ex is t . I t a lso suggests that the majority of 0. agassii populations remain within TURF-control 1 ed shore area during thei r entire l i f e cycle (Figure 12a). Most carachis are caught at night and i t has been shown that they fo l low a diurnal pattern of migration, staying in deeper waters by dayl ight and coming c loser to shore at night to feed or to spawn (Trevino et a l . 1984: 14). Fishermen claim that carachis a lso fo l low a seasonal pattern of migration which brings them c loser to shore during the rainy season, and sends them back to greater depths during the dry season. However, i t i s thought that carachis migrate only through short distances, whether perpendicular or p a r a l l e l to the shoreline (Bustamante and Trevino 1984: pers. com.). Three types of evidence suggest that carachis have a l imi ted range of migration. F i r s t , because of the i r body shape carachis can only swim slowly (T. Northcote 1987: pers. com.) and presumably not very far . Second, because they stay close to the bottom or within protected macrophyte beds, they are un l ike l y to be transported alongshore by l a t e r a l currents (Peter Richerson 65 1986: pers. comm.). And th i rd , d i s t i n c t populations of 0. agassii have been ident i f ied in the Lago Grande and in the Lago Pequeno, thus indicat ing that only l imi ted exchange takes place between populations separated by distance or deep water areas (Loubens and Sarmiento 1985). This suggests that l a t e r a l d i f fus ion between neighbouring TURFs i s bound to be slow and l i m i t e d , at least for the adult carachi. Although far less information i s a v a i l a b l e on the biology of the two other carachi species of commercial in terest , the carachi amar i l lo (0. luteus) and the carachi gringo (0. o l ivaceus) , the same major conclusions seem to apply. While infrequent and only of small sizes in zone 2, thei r d is t r ibut ion overlaps with that of 0. agassii in other zones (Figure 12b), and the i r ecology seems to be quite s i m i l a r (Bustamante and Trevino 1976; Loubens et a l . 1984; Loubens and Sarmiento 1985). 0. m u l l e r i i s mostly found in the benthic part of zone 1 and 2 (Figure 12b), (Loubens et a l . 1984), but experimental f i sh ing suggests that i t i s not ent i re l y benthic (Vaux et a l . 1986: 15). It seems to remain within these two zones for most of i t s l i f e cyc le , with the possible exception of spawning migrations to zone 3 and 4 where loca l fisheremen capture i t occasional ly . Lauzanne (1982) never found the species at depth less than 15 m. It i s thus un l ike l y to be much affected i f at a l l , by ex ist ing TURFs, and in any case, i s only of marginal commercial interest . b) Boga. Most of the information a v a i l a b l e on the boga (0. pentlandii ) re lates to Laguna Arapa, a small lake at the northwest t i p of Lake T i t i c a c a , where i t was r e l a t i v e l y abundant unt i l la te 1983 (Lev ie i l and Paz 1984). I t i s a pianktivorous f i sh which i s found in the pelagic part of zone 2 to 6, and on the shallower fr inge of zone 1, between 10 and 20 m of depth (DIREPE 1981). 66 This suggests that the boga spends most of i t s l i f e cycle outside of the TURF-contro l led l i t t o r a l area (Figure 12c), except for spawning migrations which bring adult bogas to the macrophytes of zone 3 and 4 (Ibid.). c) Suche and Mauri The mauri and the suche are two forms of the same benthic species (Trichomycterus dispar) whose ecology i s as poorly understood as that of the carachis. Mauris and carachis are often captured in the same nets, in zone 3 to 5 and in the lake's t r ibu ta r ies , while suches are caught at greater depths, in zone 1 and 2, and in the lake's largest t r ibutar ies (Figure 12f). Suches and mauris are caught at night and thought to spawn at the end of the dry season in zone 3 (Bustamante and Trevino 1976; DIREPE 1981), but there i s no indicat ion that spawning migrations occur. I t has been claimed that the mauri i s very sedentary, and that the suche roams much larger areas (DIREPE 1981) though l i t t l e i s a c t u a l l y known of the movements of e i ther type of ca t f i sh . Given that suches are found further than the outer edge of the TURF-control1 ed l i t t o r a l area, i t may be suggested that they are u n l i k e l y to be much affected by the existence of TURFs, while mauris are as l i k e l y to be as the carachis. d) I s p i . The commercially exploited i sp i s are small pelagic f i s h (0. i sp i and 0. forget i ) , which l i v e in r e l a t i v e l y large schools in zone 1 or 2, and come to spawn on the macrophytes of zone 3 to 5 (Figure 12d) throughout the year, but mostly during the dry season (Bustamante and Trevino 1976; Loubens et a l . 1984; Trevino et a l . 1984). Echo-traces indicate large concentrations of f i sh thought to be i s p i , at 30m depth (Johannesson et a l . 1981), and as deep as 50m (Vaux et a l . 1986: 8). Most i sp i captures in zone 1 and 2 occur at night when, according to fishermen, i s p i s come c loser to the surface (Nunez 1982). This has led to the suggestion that i sp i s fo l low a diurnal pattern of ve r t i ca l migration which has not yet been confirmed by experimental f ish ing or echo-sounding (Vaux et a l . 1986). Large quantit ies of i sp i s are a lso harvested at night with beach seines and col 1 anchas in zone 4 and 5, when they come c loser to shore (Bustamante and Trevino 1976; L e v i e i l 1981: f ie ldnotes ; Trevino et a l . 1984). Ispi fishermen often underscore the unpredictabi l i ty of i sp i a v a i l a b i l i t y . In August 1984, for example, fishermen from Llachon were complaining that for the past two years, the i sp i s had stayed out of thei r reach around the is lands of Taqui l i and Amantani. S i m i l a r l y , those from Vi lurcuni in the Lago Pequeno were complaining that in the previous six years, the i sp i s had come only twice to the i r area. The common b e l i e f among them i s that i sp i s represent a large but e lus i ve and unpredictable resource which should be taken advantage of whenever a v a i l a b l e , l e s t they disappear, or worse, l e s t they move into someone else 's f i sh ing area. Because i sp i s spend most of the i r l i f e cyc le outside of the TURF-contro l led l i t t o r a l area, one could think that they are u n l i k e l y to be much affected by ex ist ing TURFs. However, the opposite i s a c t u a l l y more l i k e l y to occur because the i sp i f ish ing locations known so fa r , whether spawning beaches ( in zone 4 or 5) or open water areas ( in zone 1, 2 or 6) , are few and because most of them are ac tua l l y found within TURF-control1 ed areas. Only in those locations do i s p i s appear often enough for loca l fishermen to invest in gear appropriate for i sp i f i sh ing . e) Si 1 verside The s i l v e r s i d e or pejerrey (B. bonariensis) i s found at less than 5 m from the surface of a l l s ix ident i f ied zones (Figure 12e). The adults are 68 found predominantly in zone 6, 1 and 2, and may come to zone 3 to spawn, the juven i le forms are found in zone 2 to 4, and the smallest juven i les in zone 5 (Wurtsbaugh 1974; Loubens et a l . 1984; Trevino et a l . 1984). Fishermen catch adult s i l v e r s i d e s offshore with surface g i l l n e t s less than two meters deep, and juven i les inshore with surface g i l l n e t s , beach seines or hooks. The s i l v e r s i d e i s known to migrate both hor i zonta l l y and v e r t i c a l l y on da i l y and seasonal time scales. The pattern of these migrations i s not yet wel l understood, although i t has been shown that the s i l v e r s i d e moves inshore in the afternoons (Wurtsbaugh 1974; Trevino et a l . 1984). Considering the speed and distance of the s i l v e r s i d e ' s migrations, the l a t t e r i s un l ike l y to remain within the TURF^control 1 ed l i t t o r a l area during much of i t s l i f e cyc le . Moreover, and contrary to the i s p i , the s i l v e r s i d e can be caught almost anywhere in the lake, inc luding some highly po l lu ted areas (Trevino et a l . 1984). I t i s thus very un l ike l y that any group of fishermen could reserve for i t s e l f any r ight of access to the s i 1 verside. f) Rainbow trout The rainbow trout (Salmo gairdneri) i s found in most parts of the lake, but more p a r t i c u l a r l y in zone 1 to 3, and in waters shallower than 25 m of zone 6 (Figure 12f), (Everett 1973: 439). The large adults are harvested year round in these 4 zones (Alfaro et a l . 1982), and in the mouths of Lake T i t icaca's major t r ibu ta r ies , when they migrate upstream for spawning at the end of the rainy season (Everett 1971; 1973). Because the trout spends most of i t s adult l i f e roaming in r e l a t i v e l y deep offshore waters, i t i s most vulnerable at the time of i t s upstream spawning migration when i t passes through narrow r i v e r mouthes. Access to the corresponding TURF areas i s thus very valuable and severely rest r ic ted by the i r holders (Bustamante and Trevino 1976). 69 In conclusion, TURFs affect loca l f i sh resources according to the la t te r ' s d is t r ibut ion and migrations. Widely roaming species l i k e the rainbow trout and e lus i ve species l i k e the i s p i s , which become a v a i l a b l e to loca l fishermen only p e r i o d i c a l l y and in a l imi ted number of locat ions, are u n l i k e l y to be affected in the same way as more sedentary species. Yet access to these resources can be res t r i c ted , either because access to a l l f i sh ing locations in the case of the i s p i s , or because the best f i sh ing locations in the case of the trout, i s the p r i v i l e g e of a few f ish ing communities. Sedentary species l i k e the carachis and the mauris which remain within the TURF-control led l i t t o r a l area during most of thei r l i f e cyc les , are l i k e l y to be most affected by the existence of TURFs. 3.3.2. Aquatic macrophytes: Totora and l lachu The aquatic macrophytes category includes two sub groups: submerged aquatic p lants , or l l a c h u , and emergent aquatic reeds, or totora. Totora and l lachu harvesting i s an important economic a c t i v i t y for most shore dwellers (Lev ie i l et a l . 1986). In the i r mind TURFs and totora ownership are re lated , not only because the carachis l i v e and hide in the totora beds (Galdo 1962a: 92), but also because they are part of the same organic whole (Bustamante and Trevino 1976), a notion of great importance in Andean cosmology and sense of t e r r i t o r i a l i t y (Bastien 1978; 1985). The same can be said of the re lat ionship between carachis, mauris and l l achu . The l lachu i s exploited both in the shallow waters of zone 5 where c a t t l e graze i t , or further from shore in zone 3 and 4 where people c o l l e c t i t themselves as fodder (Lev ie i l et a l . 1986). However, l lachu weeds are not subject to the same type of pr ivate ownership as the totora reeds, because they are never planted whereas the l a t t e r usual ly are, and because planting confers special r ights to the indiv idual who undertakes i t independent of his 70 r ights to the land. 3.3.3. Waterfowl: eggs and birds Waterfowl are exploited for human consumption, both as eggs and as f u l l y grown birds. Juveni le birds are sometimes captured a l i v e and raised much l i k e domestic fowl . E c o l o g i c a l l y , waterfowl are c l o s e l y associated with the totora beds of zone 4, because the reeds provide them with a sheltered nesting area close to the muddy shallows of zone 5 where they feed. V i r t u a l l y a l l the i r nesting and feeding area i s included in zone 4 and 5 thus within the TURF-control led l i t t o r a l area. Most of th is area i s divided between shore communities, although the Reserva Nacional del T i t ieaca in the Bahia de Puno is theore t i ca l l y cont ro l led by a state agency. The conclusion from th is analys is i s that the type of TURF encountered in a par t i cu la r l i t t o r a l area i s bound to affect aquatic resources in very d i f ferent ways according to the i r ecology. The proportion of the habitat of these resources included within the TURF-control1 ed area provides an indicat ion of the l i k e l y impact of TURFs existence on access to these resources. Type I TURFs can encompass a large proportion of the habitat of the demersal and benthic species such as the carachis and the mauris (not to mention a large part of ex ist ing totora and l lachu beds). Type II TURFs, however, can encompass the whole habitat of these same species. Both types al low for the control of access to the spawning areas of most f i sh resources, and to the nesting areas of waterfowl. Conversely, type III TURFs may only a l low for the r e s t r i c t i o n of access to a very l imi ted part of the habitat of pelagic f i sh resources. The l a t t e r , however, may be the only one where the resource i s vulnerable to f i sh ing , thus the most important for the exp lo i tat ion of th is resource as in the case of the i s p i s . 71 3.4. DEMOGRAPHIC DIMENSION In th is section, based on fieldwork carr ied out in 1980, 1981 and 1984, I determine who belongs to the TURF holding category. Since only members of a TURF holding community have access to aquatic resources, I consider the c r i t e r i a for community membership, according to which I dist inguish between three categories of TURF holders: those who have acquired membership in a TURF holding community by b i r t h , those who have acquired i t by ass imi la t ion , and those who have established r i t u a l or f i c t i v e kinship t i es with members of a TURF holding community. 3.4.1. TURF holding communities In the previous sections, I have demonstrated that a l l the shore communities of Lake T i t ieaca hold TURF over the area of the lake immediately adjacent to thei r portion of the shorel ine. I have also shown that these communities are wel1 defined, l o c a l l y based s o c i o - p o l i t i c a l i n s t i t u t i o n s , many of which have the special administrative status of peasant communities. However, the question arises as to whether in land communities could not a lso hold some TURFs. In th is section I demonstrate that Lake T i t ieaca TURFs are held exc lus i ve l y by shore communities. The appl icat ion of the v e r t i c a l i t y p r inc ip le to the communities of the A l t i p l a n o e n t a i l s that shore communities could have retained access over distant lands and resources, and conversely, that inland communities could have done so over Lake T i t ieaca shore areas. I could f ind various onshore communities with access to distant in land resources. The members of the community of Maquercota and of Santa Rosa de Yanaque in the Lago Sur, for example, have retained thei r t rad i t iona l access r ights to small lakes about 100 kilometers further in land (Velasco 1978; Chirapo 1982; F lores 1983). 72 Those of various communities of the Moho and Conima d i s t r i c t s s t i l l maintain thei r t rad i t iona l connections with the Amazonian slopes of the Andes, often combining agr icu l tu ra l practices in both areas (Co l l i ns 1982; Painter 1982). However, contrary to the predictions of the v e r t i c a l i t y model, I could not f ind a s ing le example of an inland community whose members had any r ights , even temporary ones, to Lake T i t icaca's resources, nor could I f ind any reference of inland dwellers establ ish ing temporary f i sh ing settlements on the lake's shores or on those of a neighbouring lagoon. Apparently, inland dwellers never came to the lake shore to f i s h but to exchange the i r products with those of shore dwel lers , including dried f i sh from the lake. Members of distant communities, such as the llama and alpaca herders of Paratia and V i l a for example, s t i l l come once a year to the shores of the Lago Norte for th is purpose (Flores 1977), as confirmed in 1984 by the fishermen of Pusi (Taraco). This resu l t can probably be explained by the predominance of the A l t i p l a n o mode of economic integration over the archipelago one, around Lake T i t i caca , during most of the Tiahuanaku era, from 200 B.C. to 1000 A.D. approximately (Browman 1981). In the a l t i p l a n o model, access to goods from other ecological zones i s achieved through trade networks between specia l ized communities rather than di rect exp lo i tat ion by members of a s i n g l e , unspecialized community. Thus members of communities specia l ized in cameloid herding obtained f i s h from the lake through trading with shore communities, rather than through f i sh ing . Having demonstrated that membership in shore communities determines access to aquatic resources, I can now consider the c r i t e r i a which govern the l a t t e r : b i r t h , marriage, and f i c t i v e kinship. 3.4.2. Access by b i r t h Community membership i s granted to the chi ldren of community members, who reside within the community te r r i to r y , although only periodic residence may be required. Membership in o f f i c i a l l y recognized peasant communities i s theore t i ca l l y granted to the heirs of former members of e i ther sex, who have reached legal age or who have establ ished a fami ly , who l i v e within the community, but who do not own agr icu l tu ra l premises within or outside the community, nor earn s ign i f i cant income from outside sources, nor belong to another peasant community (Decreto Supremo 37-70A: A r t i cu lo 23a to 23g). Membership can a lso be acquired by ass imi la t ion , or common law marriage with a community member, providing approval i s granted by a majority of the community members present at a General Assembly meeting (Art icu lo 24a to 24c). However, in pract ice, the kinship and residency requirements appear to be the only decisive ones. On the other hand community membership may not always be su f f i c ien t for an indiv idual to get access to communal TURFs, as when the community i s divided into sectors, and the ind iv idual considered happens to be from the inland sector. This i s i l l u s t r a t e d by a c o n f l i c t in the B o l i v i a n community of Compi on the northern shores of the Lago Pequeno, in which the residents of the coastal sector t r i ed to deny t rad i t iona l r ights of access to the shoreline to those from the inland one (Buechler and Buechler 1971: 52-53). However, examples of such c o n f l i c t s are rare. Instead, informants recurrently confirmed the r ights of access to the shoreline of the inhabitants of the inland sectors, on the grounds that the l a t t e r owned land within the onshore sectors, and that they had kinship t i es with members of these sectors. Fishermen from Esca l lan i (Capachica), for example, acknowledge that those from the landlocked community of Isanura, a former sector of E s c a l l a n i , could and did operate within E s c a l l a n i ' s area. Male community members who no longer l i v e within the i r community may s t i l l f i sh in the TURF area of the i r or ig inal community. This i s true i f they 74 reside in the community of the i r wife (ie uxori 1 o c a l l y ) , as for a fisherman of Santiago Mucho who could s t i l l f i s h in the area of J u l i where he was from, or as for a fisherman who has se t t led on the outskir ts of Puno c i t y who could s t i l l f i sh in the area of Ch'ul luni where he was from. It i s a lso true i f they reside neither in the i r or ig inal community, nor in that of the i r wife, ( i .e. neo loca l l y ) , as for various fishermen l i v i n g in Barco (Chucuito), or along the shores of the urbanized part of Inner Puno Bay, who could s t i l l operate in the areas of thei r or ig ina l community. Access r ights can a lso be inherited by the chi ldren and grandchildren of community members who do not l i v e within the community themselves, providing thei r kindred within the community confirm the i r family t i e s . One informant, for example, indicated that he could go f ish ing in the Uros area of Kapi in the middle of Puno Bay where his father and grandfather were from, although he was himself l i v i n g in Barco. .3.4.3. Access by marriage Access to the aquatic resources of a shore community may be granted to a young man who has married a woman from th is community after she se t t les in his own community i.e. v i r i l o c a l l y . V i r i l o c a l residence i s by far the most common pattern of residence in the Andes, though i t i s not exceptional for the-groom to s e t t l e in the bride's community i.e. uxori l o c a l l y (Bolton and Mayer 1977). Because of the b i l a t e r a l i t y of Peruvian laws of inheritance, women are l e g a l l y e n t i t l e d to the same share of thei r parents' inheritance as the i r brothers. Although numerous exceptions to th is ru le have been reported, i t i s nowadays common practice in the Andes and around Lake T i t icaca in par t i cu la r (Lambert 1977: 12-14). Submerged plots of land on which totora i s planted, for example, are commonly inherited by women, and a woman's household i s 75 e n t i t l e d to harvest her totora, whether she resides u x o r i l o c a l l y or v i r i l o c a l l y . For f i sh ing , however, the s i tuat ion varies according to the choice of residence of the couple considered, as deta i led below. a) Uxor i local residence A young man who se t t les u x o r i l o c a l l y i s e n t i t l e d to operate within the f ish ing area of his wife's community, in addition of that of his or ig inal community. According to the O f f i c i a l Statute of the Peasant Community, he can become a legit imate community member (Decreto Supremo No 37-70A). In practice though, men in his s i tuat ion rarely acquire the same status as members by b i r th (Casaverde 1978: 32; Hickman and Stuart 1977: 32). For example, they may not be able to a l low thei r consanguinal r e l a t i v e s to operate within the f ish ing zone fo the i r wife's community, unless some kind of a deal i s struck with other loca l fishermen as w e l l . They may also become the target of envy and mockery, as one fisherman of Ju l i who had set t led u x o r i l o c a l l y in Santiago Mucho, and who repeatedly complained of harassment by other fishermen from the community. The l a t t e r was, however, allowed to f i s h and did so constantly. Conversely, a young man who had recently set t led in Laccone (Acora), the community of his wife, could f i s h there without any problem, and various informants used s i m i l a r cases as arguments to deny that TURFs were enforced in their respective communities. b) V i r i l o c a l residence A young man who marries a woman from a shore community and keeps l i v i n g in his own community, i .e. v i r i l o c a l l y , never acquires the legal status of member in his wife's or ig ina l community. Nevertheless, he i s usual ly e n t i t l e d to go f i sh ing in the TURF area of the l a t t e r , but according to informants consulted during f ieldwork, two di f ferent s i tuat ions may ar i se , depending on 76 whether the groom himself belongs to a shore or an inland community. i - A young man from a shore community who marries a woman from another shore community i s general ly e n t i t l e d to operate in the TURF area of his wife's or ig inal community. Other fishermen are u n l i k e l y to deny him access, although they may resent his competition. One of the co l laborat ing fishermen from Ichu Raya (Puno), for example, reported that he could go f ish ing in the area of P a l l a l l a (Acora), his wife's or ig inal community, but that he f e l t more comfortable when she would come f ish ing with him. This r ight of access i s not simply transmitted by a woman to her husband, but also to her household, thus chi ldren and grand ch i ldren . Many fishermen have access to the TURF zone of a neighbouring community, because i t was the or ig inal community of thei r mother or grand mother, as demonstrated by fishermen's common indicat ion that "they can f i sh in the area of [another community], because the i r mother was from there" ("puede pescar en l a zona de [otra comunidad], porque su mama es de a l i a " ) , or "because they have r e l a t i v e s there" ("porque tiene sus parientes -o su parentesco- al1 a"). Another common expression, though not as s e l f explanatory, to confirm t h i s , i s that "they can operate in the f ish ing area of [another community], because they have guarantees" ("puede i r a_ pescar en 1 a zona de [otra comunidad] porque hay  garantias"), or "because they are (well) known" ("porque es conocido). Fishermen do not usual ly have to t rave l very far from thei r community's f ish ing zone to that of the community from which the i r mother or grandmother came. They are frequently c lose together i f not almost contiguous. Some fishermen, however, are required to t rave l over f a i r l y long distances. A fisherman from Barco-Chucuito, for example, had to t rave l out of the Puno Bay and around the Chucuito Peninsula to reach the i s land of Socca where he had r e l a t i v e s , and where he could f i s h . i i - A young man from an inland community who marries a woman from a shore community, i s a lso e n t i t l e d to go f ish ing in the TURF area of his wife's or ig inal community. S i m i l a r l y , th is access r ight i s transmitted to th is woman's heirs , even i f they l i v e in a landlocked community. Indiv iduals from inland communities, whose mother or grand mother o r i g i n a l l y came from a shore community, a lso have access to the la t te r ' s f ish ing zone. If they do not own a c ra f t themselves, they may go f ish ing with the i r wife's father or one of her brothers, as crew members, and receive a share of the tota l catch, or the catch of the i r own nets i f they bring any. Since i t may be time consuming for them to commute on foot from their own community to the shore every time they want to go f i s h i n g , they may move temporarily with some of thei r shore dwell ing r e l a t i v e s . One fisherman from Barco (Chucuito), for example, reported that three of his r e l a t i v e s from the inland community of Laykoma used to come and l i v e in Barco to go f i sh ing almost d a i l y , for up to three months at a time. He a lso indicated that they would have to purchase a small f i sh ing balsa which would l a s t approximately that long, since they would be not be allowed to harvest totora reeds from the communal beds of S i l l a m u r i . 3.4.4. Access by f i c t i v e kinship The t ies of f i c t i v e kinship between a fisherman and a family from another shore community when the fisherman becomes co-parent of one of th is family's chi ldren may a l low th is fisherman to operate within the TURF area of th is family's community. By sponsoring one of th is family 's chi ldren for some of the r i t u a l s which the c h i l d has to undergo as part of his or her growing up, the co-parent (i .e. compadre) creates a p r i v i leged re lat ionship with th is family (Michaud 1973; Lambert 1977). F i c t i v e kinship e n t a i l s a number of reciprocal ob l igat ions , just as true 78 kinship does. For example, people are expected to offer the goods they have to exchange or to s e l l , to thei r co-parents f i r s t , or to reserve some of these goods to meet the l a t t e r ' s needs, and eventual ly to give them a preferential rate of exchange or a better pr ice. Mutual help for tasks requiring a large amount of labour can a lso be requested from co-parents in c r i t i c a l periods of the agr icu l tu ra l cyc le , and so can loans of small amounts of money (Brush 1977; Brown 1978; Hickman and Stuart 1977). Although co-parenthood does not necessari ly guarantee access to the TURF area of the co-parents' community, i t does so in some of the communities with type III TURFs. Various fishermen from the community of Llachon (Capachica), for example, could operate in the f i sh ing zones of Amantani i s land and of some communities of the Chucuito peninsula because of thei r f i c t i v e kinship t ies with members of those communites. In addition co-parenthood had the advantage of insuring them of a place to stay, should the weather preclude them from s a i l i n g back home. Having ident i f ied the c r i t e r i a for community membership and access to TURF area, I can now consider the short to long term deals which can be worked out between ind iv idua ls and the members of the communities in whose TURF areas they want to operate. 3.4.5. Strategies to gain access to TURFs a) "Making friends" Short term agreements are easier to enter, because they do not imply a commitment to repeat s i m i l a r reciprocal exchanges in the future. They can be reached simply, when outsiders offer some coca leaves, alcohol or food, for immediate or 1ater consumption, to the fishermen of the areas where they want to operate for a while (a few nights to a few weeks). Fish or agr icu l tura l products such as cereals and tubers can a lso be given to loca l shore dwel lers. Fishermen often refer to th is as "making fr iends" (hacer amistades). Such a practices have been reported in most parts of the lake, though i t may have been more systematic in some areas, such as the Lago Pequeno. The fishermen of V i lu rcuni (Yunguyo), for example, used to compensate the shore dwellers of various communities such as Pajano, Tapoje, Calampune, systematical ly , with coca, alcohol or f i sh (Bustamante and Trevino 1976). S i m i l a r l y , those from the southern shores of the Puno Bay and of the Chucuito Peninsula exchanged agr icu l tura l products for the r ight to f i sh in the area of K'api cont ro l led by the residents of the Uros i s lands , who lacked these products. b) "Making Co-parents" Another common type of arrangement i s that by which f i c t i v e kinship t i es are established between fishermen and in land dwel lers. This i s commonly referred to as "making godparents" (hacer compadres). The creation of co-parenthood t ies (compadrazgo), as strategy to gain access to scarce resources, i s wel l documented for the Peruvian A l t i p l a n o (Michaud 1973; Bolton 1977; Brown 1978) and in the Andes in general (Lambert 1977: 22-24). The use of th is strategy by Llachon (Capachica) fishermen to gain access to aquatic resources i s mentioned above. Another common example of such t i e s i s given by shore dwellers of the Capachica and Chucuito peninsulas, who gain a more secure access to the aquatic resources of the totora beds of K'api by becoming co-parents with residents of the Uros is lands (Mario Nunez 1984: pers. comm.). c) Getting married Establ ishing true kinship t ies by arranging marriages between offsprings of d i f ferent communities i s a lso a strategy to gain access to aquatic resources over the long term. Its use has been documented for various parts of the Andes (Brush 1977: 137; Sherbondy 1982: 24) and for the Peruvian 80 A l t i p l a n o (Bolton 1977; Hickman and Stuart 1977). As mentioned e a r l i e r marriage i s commonly used along the shores of Lake T i t icaca to get access to a l l the aquatic resources of another community, not just i t s f i s h resources. d) Crewing for an in - law Because of a t rad i t ion for Andean people to col laborate more frequently with the i r s is ters ' husbands or wives' brothers than with the i r own cousins (Lambert 1977: 20), a would-be fisherman from a landlocked community whose s i s t e r has moved in with her shore dwel l ing husband, may engage in a f i sh ing partnership with his brother - in - law. A man from the landlocked community of Camacani (Chucuito), for example, could t rave l to the r e l a t i v e l y distant community of P a l l a l l a (Acora) to f i sh with his s is ter 's husband. A s imi la r arrangement with a fisherman from the closer -by community of Ccota would not only have shortened his t r i p considerably, but would a lso have given him access to a larger and apparently r icher f i sh ing ground. However, he did not have any r e l a t i v e with whom to go f ish ing in the l a t t e r community. e) Exchanging f i sh ing r ights The l a s t type of arrangement by which fishermen from neighbouring shore communities would trade f i sh ing r ights for d i f ferent species in the i r respective TURF areas, i s the least commonly encountered. The fishermen of Llachon (Capachica), for example, c laim that they can f i s h for i sp i in the TURF areas of the neighbouring communities of Capano, Yapura and Si a l e , because they a l low the fishermen from the l a t t e r communities to f i s h for carachi in their own area. 3.4.6. Synthes is A fisherman may be granted access to the exclusive f i sh ing areas of a 81 FIGURE 13: A C C E S S TO LAKE T I T I C A C A T U R F A R E A S BY BIRTH AND BY M A R R I A G E . FIG. 13A: T H E O R E T I C A L A C C E S S Cl = Ego 's community C 2 = E g o ' s mother 's c. C 3 = Ego 's w i f e ' s c. C 4= E g o ' s f a t h e r ' s mother ' s commun. C5= Ego 's m o t h e r ' s m o t h e r ' s c o m m u n . A E go O F e m a l e r e l a t i v e A M a l e r e l a t i v e FIG. I3B: E X A M P L E OF INFORMANT FROM B A R C O A In f o r m a n t A M a l e r e l a t i v e O F e m a l e r e l a t i v e 82 series of nearby communities in addition to his own, as a consequence of his true kinship t ies (Figure 13) and eventual ly of his f i c t i v e kinship t ies too. He may be allowed to operate within the TURF areas of the communities from which his mother and eventual ly those from which the mother of his mother and the mother of his father o r i g i n a l l y came from (C2, C4 and C5 respect ively on Figure 13). F i n a l l y , he may be allowed to operate within the area of the community from which his wife came from (C3 on Figure 13). I t i s un l ike l y though not impossible that a fisherman could be lucky enough to get access to f i v e TURF zones. One informant from Barco (Chucuito), for example, had access to Karina on the Chucuito Peninsula where his paternal grandmother was from, to the Uros is lands of K'api where his paternal grandfather and his father were from, and of course to Barco where his mother and maternal grandfather were from. Because his maternal grandmother was from the landlocked community of Laykoma, he did not have access to a fourth TURF area, but instead had to f a c i l i t a t e access to the lake for some r e l a t i v e s on his mother's side. Although one could think that the patterns of access to TURF areas are regulated by a wel l establ ished, and homogeneous set of norms common to a l l shore communities, a number of exceptions demonstrate the l i m i t s of these norms. Access to the TURF area of a woman's community may not be granted to the consanguinal r e l a t i v e s of a fisherman who resides uxor i local l y , for example. Yet an informant from Esca l lan i (Capachica) could f i s h in the waters of Taqui l i i s land where his brother had set t led with his wife. S i m i l a r l y , by s e t t l i n g in her husband's shore community a woman from an in land community does not normally confer access to the la t te r ' s TURF area to her blood r e l a t i v e s . As mentioned e a r l i e r , the informant from Camacani whose s i s t e r had set t led in the shore community of P a l l a l l a would not have been able to f i sh there on his own, but by crewing for his s is ter 's husband he got access to the 83 aquatic resources of the Tatter's community. I did ident i fy during fieldwork a number of exceptions to general pract ices, where ind i v idua ls who normally would not have been e n t i t l e d to i t , had been granted access to TURF areas because of some special circumstances. The reverse a lso occurs, and ind i v idua ls may be denied access to which the i r consanguinal or a f f i n a l t ies e n t i t l e s them. But ident i f i ca t ion of the l a t t e r would have required a d i f ferent approach such as a survey of in land communities. Documenting these exceptions would show how much v a r i a b i l i t y there may be in the way shore dwellers enforce the i r TURFs. I t would be very d i f f i c u l t to estimate the exact proportion of fishermen with access to more than one TURF area because of true or of f i c t i c e kinship t i e s . The number of f i sh ing s i tes indicated by the fishermen interviewed in 1976 (Table 6) provides some indicat ion of th is proportion. TABLE 6: DISTRIBUTION OF FREQUENCIES FOR THE NUMBERS OF FISHING SITES PER INTERVIEWED FISHERMAN FROM LAKE TITICACA (DATA SOURCE: IMARPE-PUNO 1976) Number of s i tes Absolute Relat ive 1 78 31.8 56.3 2 60 24.5 3 57 23.3 4 28 11.4 43.6 5 to 7 22 9.0 Total 245 100 100 As underscored e a r l i e r (cf. section 3.1.1.), the names of these f ish ing s i tes indicate whether fishermen operate within the TURF areas of more than one shore community. An analysis of these names, for a l l interviewed 84 fishermen, reveals that at least 56.3% of them operate within the TURF area of thei r own community. I t may be assumed that the other 43.6%, who operate regular ly within more than one of these TURF areas, do so because of thei r special t i es to loca l TURF holders. The size of the TURF areas, the proportion of the aquatic resources these areas include, and the ident i ty of the people who have access to these resources determine TURFs' concrete dimensions. The kind of r ights TURFs enta i l determine the i r abstract dimensions. 3.5. CULTURAL DIMENSION Nowadays, f ish ing in Peruvian waters i s l e g a l l y open to a l l Peruvian c i t i zens duly registered as fishermen. The only res t r i c t ions that can apply to their f ish ing a c t i v i t i e s are those emanating from national laws and o f f i c i a l regulations. Yet, as demonstrated above, Lake T i t icaca shore dwellers are able to maintain thei r t rad i t iona l r ights over the f i sh ing areas immediately adjacent to the shores of the i r communal t e r r i t o r i e s . These informal or customary r ights correspond to the cu l tu ra l dimension of loca l TURFs. Customary r ights refers not only to the kinds of r ights c o l l e c t i v e l y held by shore dwel lers , but also to the conditions under which the d i s t r ibu t ion , transfer and exercise of these r ights apply (Crocombe 1974; Christy 1982). I have used Crocombe's (1974) c l a s s i f i c a t i o n of customary land r ights in s ix broad categories to organize the fo l lowing section: d i rect use, indi rect economic gain, cont ro l , t ransfer , residual and symbolic r ights . Since none of these r ights i s presently recognized or protected by Peruvian laws, what I summarize in th is section are shore dwellers' concepts of customary r ights over aquatic resources. 85 3.5.1. Rights of d i r e c t use Lake T i t ieaca shore dwellers consider that they hold a number of direct use rights over the resources encountered within the i r TURF areas. They bel ieve that the i r membership in hsore communities give them the r ight to harvest and to husband aquatic resources, whether f i s h , macrophytes or birds. a) Fish resources Lake T i t ieaca shore dwellers consider that a l l community members hold the r ight to f i s h within the TURF area of the i r community, without r e s t r i c t i o n on the species caught. There are some res t r i c t ions on the locat ion of capture, as one may not set his nets within p r i va te l y owned patches of totora. One co l laborat ing fisherman from Ichu Raya (Chucuito), for example, indicated that some of the nets he had l o s t , had probably been stolen because he had set them in the patch of a loca l land owner (calados en el totora! de un vecino). However, the rat ionale behind th is r e s t r i c t i o n appears to be the protection of reeds rather than that of f i s h . Fish captured a l i v e can sometimes be kept in small pens submerged in secret locat ions , in pr ivate patches of totora, or occasional ly in small f i s h holding tanks b u i l t onshore. I t i s then considered the pr ivate property of the owner of the container. This a lso appl ies to rainbow trout raised in f l oa t ing pens for experimental or commercial ventures (Lev ie i l and Paz 1984). b) Totora reeds Depending upon the community considered and the corresponding type of TURF, the r ight to husband and harvest totora may be shared equal ly between community members or not. In TURF areas of type II and in most of those of type I, the totora beds are usual ly completely divided into i n d i v i d u a l l y owned patches (Gal do 1962b: 138; V e l l a r d 1963: 191; Sole 1969: 29; Hyslop 1976: 61; 86 Chirapo 1982: 11). The same pattern of indiv idual ownership a lso occurs in the rare parts of type III TURF areas where totora can be found (eg small bays). However, in some TURFs areas of type I, the totora beds are divided into two sections, the c losest to shore being comprised of i n d i v i d u a l l y owned patches, and the farthest one being accessible to a l l community members (Martinez 1962: 22; Hickman 1963:3). The explanation commonly provided for the p r i va t i za t ion of part of the totora beds, i s that the i n d i v i d u a l l y owned patches correspond to the private p lots of land on which owners have planted totora reeds when these p lots were submerged by the r i s i n g lake l e v e l (Hickman 1963: 3; Gonzales 1979: 87; Chirapo 1982: 11) c) Llachu weeds Any member of the TURF holding community can harvest l l achu weeds without r e s t r i c t i o n on the locat ion of the harvest, even on submerged plots of land which are i n d i v i d u a l l y owned. Informants j u s t i f y th is by indicat ing that contrary to the totora reeds, the 11achu weeds cannot be pi anted. S t i 1 1 , l lachu weeds do contribute to reinforce shore dwellers' sense of t e r r i t o r i a l i t y over nearshore waters, as i l l u s t r a t e d by some c o n f l i c t s between neighbouring communities. During the drought of 1983, for example, when competition for c a t t l e fodder was acute, members of the community of Sajo appealed to the Ministry of Agr iculture in Puno, to repel 1 trespassers from the neighbouring communities of Chatuma and V i l l a Santiago de Ccama (Letter of the community of Sajo to the Regional Director of the MINA, Ju ly 11, 1983). d) Waterfowl and t h e i r eggs Waterfowl hunting and egg c o l l e c t i n g are accessible to a l l community members without any r e s t r i c t i o n of species. Some res t r i c t ion on where eggs 87 can be co l lec ted may occur, because waterfowl nest in the thickest of the totora beds, and eggs seekers avoid the p r i va te l y owned patches. Juveni le waterfowl captured a l i v e may be raised with domestic fowl , thus becoming the property of the family who raises them. e) Aquatic space Shore dwellers consider both the water column and i t s bottom as resources. Direct use r ights over the bottom are held by the community as a whole or i n d i v i d u a l l y by i t s members, depending on depth and distance from shore. Outsiders would not be allowed to plant totora reeds on any part of the bottom of a TURF area. The same holds true for the water column even i f t rans i t by outsiders across TURF-controlled areas i s to lerated in the open water parts of these areas where totora reeds cannot be damaged or s to len. Outsiders are cannot use i t for trout aquaculture without compensating the community in whose TURF area the f l oa t ing pens are located. Fa i lu re to do so explains some of the problems which have plagued trout aquaculture on Lake T i t ieaca. Because such a c t i v i t i e s were often imposed on loca l communities, they were deeply resented by TURF holders except for the few who benefitted d i rec t l y from them, and sabotage occurred more than once (Lev ie i l and Paz 1984). In conclusion, members of shore communities hold direct use r ights over the aquatic resources encountered within TURF areas. These include the r ights to harvest and to husband aquatic resources in any locat ion within a TURF area, except for the p r i va te l y owned patches of totora reeds where these rights belong to the patch owner. I found no s i tuat ion , theore t i ca l l y possible (Crocombe 1974), in which r ights of d i rect use over d i f ferent aquatic resources within the same portion of aquatic space were held by d i f ferent actors. And the only example of subsidiary r ights of d i rect use I could 88 ident i fy was that of t rans i t through the open waters of TURF areas. 3.5.2. Rights of indi rect economic gain Shore dwellers hold a r ight to ind i rect economic gain for the explo i tat ion of some of the aquatic resources encountered in the i r c o l l e c t i v e TURF area. There i s a l o t of variety depending of the resources and of the communities considered. a) Fish resources The r ight of ind i rect economic gain from the f i s h resources of a TURF holding i s held by the community as a whole. However, i t i s considered acceptable practice for community members to receive a compensation for a l lowing outsiders to f i s h in the TURF area of the i r community. As indicated e a r l i e r , a common expression used by outsiders to refer to th is practice i s "to make friends" with TURF holders (hacen amistades) by g iv ing them something, either a r i t u a l g i f t of coca leaves and a l c o h o l , or a more pragmatic g i f t of f i s h or agr icu l tu ra l products. TURF holders may also request compensation in the form of a preferent ial treatment when purchasing or bartering f i s h from known trespassors. Fishermen from Socca i s l a n d , for example, indicated that customers from Santa Rosa de Yanaque always demanded a preferential pr ice , or a more generous overweight, or yapa, for the f i s h which they claimes had been caught in the i r TURF area. S i m i l a r l y , an e l d e r l y lady from the outskir ts of Puno indicated that she required the fishermen from Chimu to give her a better price for the f i sh which had been caught in the f i sh ing zone of her community. As she said "we t e l l the fishermen from these communities (who trespass) to lower thei r pr ices, because the f i s h they s e l l comes from our (area of the) lake" (a esos  Pescadores [quienes incursionan] 1e decimos de bajar el precio, porque el 89 pescado es de nuestro 1 ago). Shore dwellers can also get f i sh from the fishermen of thei r own community for a lower pr ice , or for "goodwil l" (de voluntad). In most cases, they receive th is preferential treatment because they are r e l a t i v e s of the fishermen (famil iares) who can expect some unspecified favor in return. But there is a lso a moral obl igat ion for fishermen to s e l l or to barter thei r f i s h with other community members (par t i cu la r l y the poorest ones) at a preferential rate for the l a t t e r . The TURF holding communities have general ly been unable to c o l l e c t the ind i rect economic gains to which they are customarily e n t i t l e d , because of TURFs i l l e g a l status. This could explain why indiv idual members may c o l l e c t access fees from outsiders in the name of the community. Some o f f i c i a l l y recognized communities, however, have required outside fishermen to pay a lump sum in cash to the i r treasurer, for the r ight to f i s h in the i r area. One community from the Southern part of the Bahia de Puno, in p a r t i c u l a r , has required outsiders to pay a lump sum (equivalent to 25 k i l o s of s i l v e r s i d e ) as a contribution to i t s school (cuota para el colegio) , for the r ight to f i sh in i t s area during a whole year. Some communities from the Mantaro V a l l e y , a few hundred kilometers north of the A l t i p l a n o , a lso charge outsiders a fee everytime they want to f i s h for trout in the i r r i v e r (Mayer 1985: 75). b) Totora reeds TURF holders do not usual ly a l low outsiders to harvest totora reeds from thei r communal totora beds, as confirmed by a majority (72%) of the col 1 aborating fishermen (18.6% of whom did not provide a def in i te answer). The reason invoked i s that few communities have enough totora for the needs of the i r own members. Only in the very large totora beds of Moro and K'api in 90 the Bahia de Puno, are outsiders allowed to harvest the reeds themselves for the payment of a small rental fee or arriendo. Members of the community of Uros-Ch'ul 1uni , for example, have t r a d i t i o n a l l y obtained agr icu l tu ra l products and cash nowadays, from th is practice (Calancho 1984; L e v i e i l et a l . 1986). In conclusion, shore dwellers are e n t i t l e d to ind i rect economic gains from the exp lo i tat ion of the f i sh resources within the TURF area of thei r community. Residents of the Uros is lands of K'api in the Puno Bay may a lso hold a s imi la r r ight for totora resources. But I encountered no evidence of such a r ight for l l achu weeds or waterfowl. There i s probably some l i m i t to these gains, as demonstrated by the c r i t i c i sms l e v i e d against some of the co l laborat ing fishermen who, being paid a modest sum for the i r co l laborat ion , were accused of s e l l i n g the lake of thei r fe l l ow fishermen to the foreigners (venden nuestro 1 ago £ 1 os gringos!). 3.5.3. Rights of control Lega l l y , the loca l administration holds the r ight of control over a c t i v i t i e s re lated to the exp lo i tat ion of aquatic resources and decides how these resources may be used. In pract ice , however, community members are normally involved in decisions re lated to matters of communal interest and thei r membership e n t i t l e s them to a share of communal resources and to the r ight to part ic ipate in the control of these resources. Numerous examples of communal control over agr icu l tu ra l practices and water rat ioning have been documented a l l over the Andes (Mayer 1985). They demonstrate Andean a g r i c u l t u r a l i s t s ' a b i l i t y to organize productive a c t i v i t i e s at the community l e v e l by coordinating practices among member households. Around Lake T i t i caca , th is a b i l i t y i s t ranslated in practice by the rules which various communities have implemented for the protection of the i r aquatic resources or for thei r orderly exp lo i ta t ion . 91 a) Fish resources. Fishermen from some communities have promulgated rules for the orderly conduct of f i sh ing . In the ear ly 1970s, for example, fishermen from the Associacion de Pescadores Menores de Consumo Humano de Escal1 ani set f ish ing schedules and turns to avoid net theft and interferences in the area of Sejnachi (Bustamante and Trevino 1976: 78). S i m i l a r l y , those from the "Comite  de Pescadores Jose Olaya de Arapa" organized f ish ing a c t i v i t i e s for the boga in the western ha l f of Laguna Arapa. They prohibited nocturnal f ish ing a c t i v i t i e s , to avoid f i sh ing accidents and net theft under the cover of darkness. They also agreed on res t r i c t ing to two the number of nets per fisherman and prohibit ing meshsizes under 41 mm (A. Gonzales 1984: pers. com.). F i n a l l y , those from Ccota (Acora) in the Puno Bay s t ipu lated a minimum size of capture for the carachi , and those from Socca (Acora) implemented a closed season (veda) without much success in either case. T r a d i t i o n a l l y , some r ights of control over the practices of f i sh ing were held by loca l master fishermen. In Laguna Arapa, for example, an experienced fisherman, the 11 i c a l 1 oc, was responsible for the direct ion of c o l l e c t i v e f ish ing operations, often because he was the owner of the large trap net, or 1 1 i c a , used. The 1 1 i c a l 1 oc could decide when to use his net and who could part ic ipate in an operation which required two dozen fishermen and ha l f as many balsas. However, the 11 icas are not used anymore, and the 11 i c a l 1 oc have l o s t thei r l imi ted powers. b) Totora reeds Many more communities have implemented regulations for the harvesting of totora reeds than for f i sh ing . The major objective of these regulations is to minimize the theft of reeds. Because of the minute size of most totora patches, and because one i s hidden among the reeds when harvesting them off a low l y ing shore, i t i s tempting to harvest the totora of one's neighbour, when few people are around. To mitigate th is problem and the ensuing brawls, most communities holding type I TURFs (eg Huayl lata , Cachi Pucara), and some holding type II TURFs (eg Barco, Socca) have implemented a cutt ing season, and a schedule of cutt ing days, which apply to a l l totora within the community's TURF area, whether i n d i v i d u a l l y owned or not. Some of these regulations a lso promote greater reed production. The cutt ing season, for example, has the advantage of al lowing the reeds to grow during the rainy season when a l te rnat i ve c a t t l e fodder i s p l e n t i f u l . In addit ion, many communities have prohibited p u l l i n g out the edible root, or chul1 o, without which the totora reeds cannot regenerate. They have a lso prohibited cutt ing the reeds too far from the surface of the water, or too " c lose from the bottom, which would hamper the i r regeneration. In conclusion, the r ight of control over TURF resources i s held by community as a whole. The peer pressure exerted by community members upon ind iv idua ls who appear to take more than the i r share of the communal resources may be assimilated to a form of communal cont ro l . This form of control resul ts from a shared understanding among community members, of what constitutes an acceptable l e v e l of exp lo i tat ion of communal resources, for barter or sa le . A l t e r n a t i v e l l y , the community may delegate the respons ib i l i t y of i t s control r ights to ind iv idua ls appointed on an annual s h i f t basis. In the case of the totora reeds, the rescat istas or v ig i lan tes de totora are responsible for the respect of the corresponding regulat ions, and thei r f a i l u r e to do so may lead to them being f ined. This underscores the potential for the appointment by the shore communities of an indiv idual in charge of the control of f i sh ing . The communities of the Mantaro Va l ley mentionned e a r l i e r , for example, created a special charge, or cargo, for the spec i f i c purpose of c o n t r o l l i n g trout f i sh ing in the i r creeks and of c o l l e c t i n g outsiders' f i sh ing fee (Mayer 1985:75). 3.5.4. Rights of transfer and residual r ights As demonstrated e a r l i e r , TURF holders transfer the i r f i sh ing , macrophyte harvesting and bi rd hunting r ights to the i r heirs , with thei r community membership. They can a lso transfer the i r r ights over i n d i v i d u a l l y owned patches of totora by sale to other community members, but th is i s less common. Among a sample of 53 totora patches owned by 38 shore dwel lers , only 11% had been acquired through purchase by 10.5% of these shore dwel lers. The shares of the co l l e c t i v e r ights of a community member who dies without chi ldren are reverted with the community, when his/her r ights to i n d i v i d u a l l y owned patches of totora reeds are returned to his/her c losest consanguinal k ins , rather than to a surv iv ing spouse (Bolton 1977: 223), a pattern commonly observed in the Andes (Lambert 1977: 12). TURFs residual r ights thus belong to the community as a whole, except for i n d i v i d u a l l y owned patches of totora beds which belong to the consanguinal kins of the deceased. 3.5.5. Symbolic r ights Perhaps more important to shore dwellers than a l l the material resources included within the te r r i to ry of the i r community and a l l the corresponding r ights , i s the deep re l ig ious s ignif icance of th is very te r r i to ry . The l a t t e r ' s most s a l i e n t features are usual ly associated with s p i r i t s and dei t ies (Bolton and Mayer 1977: passim; Sherbondy 1982) of great s ignif icance in the Andean cosmology. In every mountain peak l i v e s a Wamani, or Apu, to whom r i t u a l offer ings, or "payments" (pagos) have to performed p e r i o d i c a l l y , usua l l y where natural features known as huaca ex is t , or where small shrines, or capi 11 as, have been erected ( Isbel l 1977). S i m i l a r l y , fishermen have to 94 "pay" the lake goddess and owner of the f i s h , Mamacocha, at least once a year (Tschopik 1946; La Barre 1948; Nunez 1982). The de i t ies of the Andean communities are integrated into a whole cosmological order, with hierarchies and functionnal re lat ionships (Cordero 1966; V a l l e e 1972). Lake T i t i caca i t s e l f used to be the highest lake in the Inca cosmological hierarchy, because i t was the one from which the founder of the Inca dynasty, Viracocha (or "Lake of fat" ) , had o r i g i n a l l y come from. And the Wamani are ranked according to the size of the i r mountain peaks. They are connected to the earth, Pacha Mama, by the Amaru who represents the force and fecundation power of water, and who i s often symbolized by the ra in , a r i v e r , an i r r i g a t i o n channel or a large snake. Every major section of a community's te r r i to ry including the shrines of one or more s p i r i t s , i s thus part of the residents' cosmological v is ion of the i r own world. In many cases, th is v is ion i s integrated into an organic analogy. The intimate correspondance between the physical te r r i to ry and the cosmology of Andean people indicates how int imately p o l i t i c a l concepts and re l ig ious symbols are connected (Gow 1981: 221-255; Sherbondy 1982). The loss of part of the community's te r r i to ry i s deeply f e l t because i t implies the imposs ib i l i t y to perform the corresponding r i t u a l s . In cases where the corresponding cosmological v i s ion assimi lates th is te r r i to ry to the body of a human being, or that of an animal (Gow 1981: 235; Bastien 1985), such losses have been compared to a loss of limbs. Rec iprocal ly , the features of the physical landscape are often portrayed as l i v i n g characters in the myths which serve to record t rad i t iona l claims to land (Bastien 1985) or to water (Sherbondy 1982). I recorded a few myths associated with Lake T i t icaca aquatic resources during fieldwork. Most of them deal t with a giant c a t f i s h , or Catar i , who 95 would drawn imprudent or greedy fishermen by dragging them and the i r nets into water holes. Fishermen from Laguna Umayo and Laguna Arapa reported the existence of loca l shrines, or huacas, associated with f i sh ing . I t was indicated that the huaca of I s i a Arapa, had been blown up with dynamite by a metis land owner (misti) who wanted to prevent fishermen from paying the i r dues to the Cocha Mama, and who paid dearly for his sacr i lege when he died paralyzed a few years la te r . The r i t u a l celebrations and "payments" to the mountain, earth and water s p i r i t s are performed on spec i f ic dates, usual ly associated with the Cathol ic calendar of saints (La Barre 1948: 184; Gal do 1962; 1967;. Nunez 1982). Because they determine the agr icu l tu ra l calendar, they can be considered as agr icu l tu ra l (Sherbondy 1982: 111-2) and f ish ing technologies (Tschopik 1951: 200; Cordero 1966:63). But most importantly for the implementation of TURF areas, r i t u a l s contribute to reassert t rad i t iona l r ights (Sherbondy 1982: 25), and to reinforce the sense of t e r r i t o r i a l i t y shared by community members (Labarre 1948: 184; Tschopik 1951: 280). In many Andean communities, for example, the r i t u a l s of the 1st of January or of carn iva l "celebrate the i n v i o l a b i l i t y of boundaries and prescribe the r ight means to cross them" ( Gow 1981: 242). In Luquina Grande (Acora) for example, on San Sebastian day (January 20th), a band of musicians leads the whole community as i t dances along the boundaries of i t s te r r i to r y , thus teaching the younger generation to recognize them (Lescano et a l . 1982: 146). It i s in th is way that the contribution of symbols to the del ineation of aquatic t e r r i t o r i e s i s made most obvious nowadays. 3.5.6. Obligations of community members Within a tenure system "each r ight i s associated with reciprocal duties, 96 and with a tota l network of re lat ionships" (Crocombe 1974: 7). This network can be "a rank system (which) includes i n s t i t u t i o n a l i z e d differences between r ights and the i r re lated obl igat ions" (Ibid.). In the case of Lake T i t i caca shore communities, no such i n s t i t u t i o n a l i z e d differences ex is t between members with respect to TURFs. Each and every member i s e n t i t l e d to the same share of the communal aquatic resources, excluding pr ivate patches of totora. Consequently, a l l members are expected to f u l f i l l in the same way the obl igat ions entai led by thei r membership. These obl igat ions include "the giv ing of labour, and part ic ipat ion in the rule-making, rule-breaking and rule-enforcing mechanisms" (Mayer 1985: 64). The g iv ing of labour, or faena, i s required for the construction and maintenance of the inf rastructura l features necessary for the community (roads, i r r i g a t i o n channels and ditches, school or communal bui ldings) . A contribution of 100 mud br icks , or adobes, i s a standard contribution to the construction of a school for a community member and his household. In some shore communities, the faenas may also include the construction and maintenance of docks to protect small harbours. In addition to the performance of the necessary r i t u a l s , the repair ing and cleaning of i r r i ga t ion channels, for example, i s necessary, for the annual reassertion of communal r ights to the channels and the i r water (Sherbondy 1982: 23). S i m i l a r l y , indiv idual r ights are gained by par t ic ipat ion in the construction of communal infrastructures, and maintained by regular faena part ic ipat ion (Mayer 1985: 61). Some fishermen from Llachon, for example, were allowed to f i s h in the area of a community on Chucuito peninsula because they had part ic ipated in the construction of the loca l school. Par t ic ipat ion in the implementation of ru les for the control of the community over indiv idual pract ices, e n t a i l s f i l l i n g the various posit ions, or 97 cargos, necessary for the p o l i t i c a l administration of the community, and the organization of i t s major celebrat ions, or f i e s t a s , including the performance of the associated r i t u a l s (Carter and Mamani 1982). F i n a l l y , a f inanc ia l contribution may be required from community members, for the use of communal resources, such as pasture lands and fa l lows . I recorded only two such examples during f i e l d work, one in Chaul la Camani, where fishermen were required to pay "a small annual contribution to help the community" (una quotita al ano, como ayuda para 1 a comunidad), the other in Arapa, where they were required to contribute to the purchase of a typewriter for the municipal i ty . Because a l l the r ights ident i f ied are held by we l l -def ined and r e l a t i v e l y autonomous communities (except for the resources within pr ivate patches of totora) one may speak of shore communities' c o l l e c t i v e property over aquatic resources. However, these r ights are only informal , since TURFs are i l l e g a l in Peru, as in most countries. One may thus wonder how Lake T i t ieaca shore dwellers manage to enforce the i r TURFs and how e f fec t i ve th is enforcement i s . Documenting th is enforcement in the next section demonstrates that TURFs do not r e s u l t from an absence of competition for resources within fishermen's reach, from a lack of interest on the part of potential competitors, or from some form of social prejudice against f i sh ing . 98 CHAPTER 4 ENFORCEMENT OF TURFS Despite the opposition of the administration, shore dwellers are able to enforce the i r informal TURFs. This enforcement may be through various mechanisms, including advertisement and overt or covert defense, a l l of which communicate to outsiders the i r exclusion. I analyze these mechanisms in the three sections of th is chapter. F i r s t , I describe the legal and administrative environment in which Lake T i t icaca shore dwellers have to operate to enforce the i r TURFs. Second, I consider the a l te rnat i ve forms of ownership of the aquatic space and i t s resources (whether pr ivate , semi pr ivate or c o l l e c t i v e ) . And t h i r d , I show how these a l te rnat i ve forms of ownership and the laws and regulations for the use of aquatic resources contribute to the enforcement of Lake T i t icaca TURFs. 4.1. TURFS LEGAL AND ADMINISTRATIVE ENVIRONMENT In Peru as in many other countries, i t i s i l l e g a l for a group of fishermen or shore dwel 1 ers to enforce t e r r i t o r i a l f i sh ing r ights . Government o f f i c i a l s strongly c r i t i c i z e the l a t t e r for "not knowing the law" (son  ignorantes de l a ley) . Yet shore dwellers are able to take advantage of various laws and regulations to protect some of thei r customary r ights . Among these are the Legal Statute of the Peasant Communitiess, a 1947 law dealing with the ownership of the bottom in the shallows of the lake and various o f f i c i a l texts dealing with the conservation of totora reeds and f i sher ies organization, a l l of which I consider here. 99 4.1.1.. Land and water laws The Peruvian Law of Agrarian Reform (Decreto Ley No. 17716 de Reforma Agraria 1969) and the Legal Statute of the Peasant Communities (Decreto Supremo No. 37-70A 1970) formal ly recognize pr ivate , cooperative and communal land ownership. According to the Water Law (Ley General de Aguas, Decreto Ley No. 17752, 1969) and to the Fisheries Law (Ley General de Pesqueria, Decreto Ley No. 18810, 1971), aquatic space and resources belong to the state. The Water Law also includes in the publ ic domain a f i f t y meters wide s t r ip of land above the highest t ide l i n e (Art icu lo 5a) and a l l land gained on r i v e r s , lakes and oceans, by natural or by a r t i f i c i a l means (Art iculo 5g). However, a special law grants ownership of the submerged l i t t o r a l fr inge of Lake T i t ieaca to i t s shore dwellers. The fo l lowing sections of these laws and of the corresponding regulations are relevant to TURFs enforcement. a) O f f i c i a l Statute of Peasant Communities Various sections of the O f f i c i a l Statute of Peasant Communities (Decreto Supremo No. 37-70A) deserve attention because they correspond to some of TURFs1 dimensions. The legal de f in i t ion of the peasant community corresponds to thei r human dimension. The ident i f i ca t ion of i t s te r r i to ry and of i t s communal resources corresponds to the i r resource dimension. And the section defining the powers formal ly granted to the community for the management of i t s resources, corresponds to the i r legal dimension. According to the i r O f f i c i a l Statute, Peasant Communities are t e r r i t o r i a l groups of fami l ies l inked by common social and cu l tu ra l features (A r t i c le 2 and 13 to 16). A peasant community has a j u r i d i c a l personal i ty , exc lus ive ownership of i t s te r r i to ry and the r ight to elaborate and enforce an internal set of rules (A r t i c le 4 to 6). I t i s e n t i t l e d to have rules for the "appropriate use of i t s natural resources" (A r t i c le 8). The statute 100 st ipu lates the conditions for membership ( a r t i c l e 23 and 24), the r ights and duties entai led by th is membership (A r t i c le 25 and 26) and the penalt ies incurred for lack of compliance (A r t i c le 27 to 30). The o f f i c i a l statute also deals extensively with the administrative organization of the peasant community and with the creation of i t s three administrative bodies, each consisting of a l imi ted number of elected representatives: the General Assembly composed of a l l community members (Ar t i c le 33 to 39), the Administrative Council (A r t i c le 40 to 50) and the Committee of V ig i lance (A r t i c le 51 to 55). In addition to these three essential bodies, ad-hoc or special committees can be created to ass i s t the Administrative Council on spec i f i c issues, p a r t i c u l a r l y technical ones, such as the bui ld ing of physical infrastructures and issues re lated to the exp lo i tat ion of communal resources (A r t i c le 58 to 62). S i m i l a r l y , Local Administrative Committees (Juntas Locales) can be created to ass i s t the Administrative Council in the administration of certain parts of the communal te r r i to ry ( ca l led Annexos) (A r t i c le 63 to 66). F i n a l l y , the statute specif ies the elements of the peasant community's patrimony (A r t i c le 89), i t s sources of labour (A r t i c le 106 to 108), and of income. The l a t t e r may include among others "the fees paid for the usufruct of community property or communally-owned services" (A r t i c le 90). I t also indicates that the peasant community i s responsible for the ident i f i ca t ion and inventory of i t s communal resources and of the accounting of the i r rents, under the supervision of various national agencies (A r t i c le 93 to 102). b) Ownership of the lake shore During the mid 1940s, after a long period of drought, the l e v e l of Lake Ti t ieaca was as much as 5 meters below i t s highest recorded l e v e l (Col l o t 101 1980: 11). After much debate a law was passed in March of 1947, declar ing as Indian property a l l newly exposed shore lands (Davies 1970: 140-141). This law gave preferent ial treatment to shore dwel lers , p a r t i c u l a r l y landless ones, and l imi ted adjudications to ten hectares per family (Law No. 10842, 1947: a r t i c u l o 2). I t a lso lega l i zed the transfer of these lands by inheritance, but prohibited their sale or embargo (ar t icu lo 3). Since then the lake l e v e l has been r i s i n g unt i l i t reached i t s maximum • present l e v e l . The p lots of land distr ibuted have been submerged again and planted with totora by the i r or ig ina l owners or by the i r heirs (Chirapo 1982; Sur 1979a: 17-18). Even i f these indiv idual p lots have been submerged for more than two decades, the i r exact locat ion has not been forgotten. However, more recent laws and regulations have either modified law No 10842, or i n d i r e c t i y affected i t s appl icat ion . c) Laws of forestry and w i l d l i f e The Forestry and W i l d l i f e Law (Decreto Ley No. 21147, 1975) defines the terms under which state-owned forest and w i l d l i f e resources can be used (Art iculo 1) and the appropriate measures for the conservation of these resources (Art icu lo 14 to 20). I t defines subsistence, s c i e n t i f i c and commercial exp lo i ta t ion , and st ipu lates the conditions under which each can be undertaken under the control of the Ministry of Agr icul ture (hereafter referred to as MINA). But i t does not s t ipu late the conditions under which one can extract these resources for subsistence purposes. Since the promulgation of th is law, concerns over conservation issues in Peru and over the future of the large totora beds of Lake T i t icaca have led to the creation of Lake T i t icaca National Reserve, in la te 1978 (D.S. No 185-78-AA, 1978). This reserve was establ ished "to conserve natural resources and esthetic values, while contributing to the socio-economic development of loca l 102 populations through a rat ional use of animal and vegetal resources and the promotion of loca l tourism" (D.S. No 185-78-AA 1978: 1, my t rans lat ion) . The creation of th is reserve has brought a considerable amount of confusion for people who had t r a d i t i o n a l l y considered the totora as theirs because they or the i r forefathers have planted i t or otherwise contributed to i t s dissemination on submerged land to which they have legal t i t l e . They feared that with the creation of the reserve, the harvesting of totora reeds would become i l l e g a l or that i t would be taxed (Sur 1979b: 13-14). The Decreto Supremo creating Lake T i t ieaca National Reserve "does not affect shore dwellers r ights (of ownership over the bottom of the lake shallows) establ ished by Law No 10842" (D.S. No 185-78-AA, 1978). I t does not affect the i r r ights over the totora either . But since totora reeds are o f f i c i a l l y considered as forestry resources, and since o f f i c i a l texts indicate that aquatic resources (D.L. No 17752, 1969), forestry and w i l d l i f e (D.L. No 21147, 1975) are state owned, shore dwellers have ac tua l l y been l e g a l l y deprived of thei r t rad i t iona l r ights over the totora. Shore dwellers have not l o s t the r ight to use forest and w i l d l i f e resources for subsistence, however. The l a t t e r i s merely defined as "use for home consumption by the extractor's family" (Ibid.) and i s subject to the general regulations promulgated by the MINA for management purposes, though no special l icence or authorization i s required for i t s undertaking. Furthermore, the Forestry and W i l d l i f e Law gives p r io r i t y for extraction permits (for commercial use) to owners of land adjacent to the areas invo lved , p a r t i c u l a r l y when the l a t t e r are involved in tree plant ing. According to th is law, loca l TURF holders should therefore be given preferent ial access to the totora reeds (which they have planted). 103 FIGURE 14: M A P OF THE "L IGA DE DEFENSA DE L A T O T O R A DE C A T U R A PAMPA" (Reduced to 7 5 % of the original and slightly enhanced) N o t e : No sca le ; /„ . ' /'.- "/{ " V . Lj / . I f . on the or ig inal The decision of the regional director of the MINA in Puno "to protect the free use for subsistence purposes of totora reeds in the temporary laguna of Ccota Lacka Pampa by the communities and settlements of Yanaque Catura Pampa" (Resolucion Directoral No 0593-77-DZA-XII, 1977, my translat ion) confirms that p r i o r i t y i s a c t u a l l y given to TURF holders (Figure 14). This decision i s subject to the conditions that "(such a) subsistence oriented extraction w i l l only be carr ied out by members of the communities and settlements indicated (...) according to customs, t radi t ions and internal ru les , providing that they do not c o n f l i c t with the rat ional use of the resource" (Ibid.). Although i t involves only a dozen communities from the Lago Sur, i t does i l l u s t r a t e the discret ional power of loca l administrators in interpret ing ex ist ing laws and in protecting shore dwellers access r ights to loca l aquatic resources. Administrators are careful to stress that preferential access r ights do not enta i l ownership of the resources invo lved , as repeatedly indicated in the documents granting o f f i c i a l recognition to Peasant Communities. The document corresponding to the community of Ramis, for example, s t ipu lates that "the te r r i to ry of the community i s p a r t i a l l y enc i rc led by Lake T i t i caca , but does not include the submerged lands nor the ex ist ing f l o r a . " (Informe f i n a l No 127-75 de Reconocimiento de l a Comunidad de Ramis 1975, my t ranslat ion) . 4.1.2. F isher ies laws Among the numerous Peruvians laws and regulations dealing with f ish ing (Rendon and Rendon 1969), two are of major relevance for the pursuit of small scale f i sh ing . The f i r s t one i s the General Fishing Law (Ley General de Pesqueria, Decreto Ley No. 18810, 1971). I t updates to some extent but mostly confirms the second one (Art iculo 12). The second one corresponds to the by-laws of the Coast Guard, which deal both with navigation and f i sh ing (Reglamento de Capitania y de l a Marina Mercante Nacional, Decreto Supremo No. 105 21, 1951) . A th i rd one dealing with fishermen's gui lds i s worth mentioning because i t indicates, among other things, the conditions under which f ish ing gui lds can be created and how they can operate (Decreto Supremo No 3, 1955; Resolucion M i n i s t e r i a l No 2430, 1955) . a) General Fishing Law. The General Fishing Law (Ley General de Pesqueria, Decreto Ley No. 18810, 1971) asserts that a l l Peruvians have the r ight to f i s h . I t confirms that only the government has the power to d e l i v e r the authorizations and l icenses required to undertake the corresponding a c t i v i t i e s (Art iculo 7). And i t transfers to the Ministry of Fisheries (hereafter MIPE) the respons ib i l i t y of managing a l l Peruvian f i sher ies (Art icu lo 8). According to th is law, the regional o f f i ce of the MIPE in Puno i s l e g a l l y in charge of f i sher ies management on Lake T i t ieaca , although the l a t t e r has done l i t t l e in th is respect. Since i t s creation in the la te 1960s, i t has trained a number of fishermen and of f i sh ing extension agents, but i t has not developed special functions for the l a t t e r , such as the c o l l e c t i o n of catch s t a t i s t i c s . As for the f i sh ing regulations i t promotes, they invo lve minimal sizes of capture for four f i s h species (13 cm for suche and mauri, 14 cm for boga, and 25 cm for trout) , and closed seasons from Apr i l to Ju l y for the trout and August to November for boga, suche and mauri). b) By-laws and regulations of the Coast Guard The by-Taws and regulations of the Coast Guard (Reglamento de Capitania y de l a Marina Mercante Nacional, Decreto Supremo No. 21, 1951) control the conditions under which small scale fishermen can operate thei r c ra f t . They re i terate that f i sh ing i s an a c t i v i t y in which a l l Peruvian c i t i zens can f ree ly part ic ipate (Art iculo 744b). Each fisherman has to be registered at 106 the Coast Guard's headquarters, or Capitania, of the j u r i s d i c t i o n to which he belongs (Art iculo 506). A l l f i sh ing c ra f t have to be registered at the Capitania (Art icu lo 748), and t h e o r e t i c a l l y at l eas t , verbal authorization from the Capitania i s required to undertake a f ish ing t r i p (Art icu lo 757). The Coast Guard regulations a lso include some by-laws designed to minimize interferences between fishermen (Capitulo IV). They give complete j u r i s d i c t i o n to the Coast Guard over a l l navigable waters within the national domain (Art iculo 6), over pr ivate and state-owned landing or docking f a c i l i t i e s (Art icu lo 8c). In addit ion, they give the l a t t e r the respons ib i l i t y of register ing land ownership within the i r j u r i s d i c t i o n (Art icu lo 8q), including the f i f t y meters wide s t r i p of land above the highest l e v e l reached by the sea (Ley No. 17752, 1969: A r t i c u l o 5a). The Coast Guard i s thus the organization that would be in charge of the formal regist rat ion of TURFs, i f the l a t t e r were granted a legal status. In the meantime the Coast Guard supports unrestrained f i sh ing operations by reg istered fishermen. The Coast Guard's internal regulations require each f ish ing settlement to be represented Beach Sargeant or Sargento de PI aya. The Sargento i s often an older fisherman appointed by the Coast Guard, who is expected to perform his duties on a voluntay basis. The most which i s expected from Lake T i t i caca Sargentos i s to keep an up to date l i s t of loca l fishermen, to encourage them to get the i r f i sh ing l icence or to matriculate the i r crafts at the Capitania in Puno, and to inform other fishermen of the f i sher ies regulations promul gated by the MIPE. A br ief analysis of the Peruvian l e g i s l a t i o n dealing with f i sh ing would lead one to be l ieve that both aquatic space and shorel ine areas are under the exc lus ive control ot the state. Although professed by many urban professionals in Puno and elsewhere in Peru, th is view f a i l s to r e f l e c t actual practices. Laws leave ample margin for loca l government o f f i c i a l s to use thei r discretionary powers for the interpretat ion and implementation of these laws and of the corresponding regulations. In addition the perceptions by loca l communities of the r ights they hold over the resources within the i r t e r r i t o r i a l boundaries are more e f fec t i ve in governing the behaviour of thei r members than the laws and regulations weakly and e r r a t i c a l l y enforced by a distant and often careless administration. F i n a l l y , shore dwellers do hold a number of legal and customary r ights over parts of the shore space which al lows them to control access to the shoreline and thus to f i sh ing . 4.2. TURF-RELATED FORMS OF OWNERSHIP Various customary forms of control or ownership over aquatic and shore space, or over f i sh ing intangibles such as f i sh ing techniques and f i sh ing r i t u a l s , can and do af fect the enforcement of Lake Ti t ieaca TURFs. I consider each of them successively here to determine the i r impacts on the enforcement of TURFs and on f ish ing e f fo r t l e v e l . 4.2.1. Ownership of i n t a n g i b l e s . Pr ivate or semi-private ownership of intangib les , such as f i sh ing songs, f i sh ing r i t u a l s and f i sh ing techniques, has never been reported for Lake T i t ieaca , contrary to other regions of the world such as the South P a c i f i c is lands (Johannes 1979; 1981; Pet i t -Skinner 1983; Carr ier and Carr ier 1983). Various ethnographic accounts describe the r i t u a l s associated with f i sh ing around Lake T i t ieaca (Bandelier 1910; LaBarre 1948; Tschopik 1946; Nunez 1982; Chirapo Cantuta 1982), but none indicates that the l a t t e r are p r i va te l y owned. Andean f ish ing r i t u a l s were and s t i l l are often carr ied by specia l i zed 108 div iners , who may either come from neighbouring communities or t rave l from distant regions of B o l i v i a (Tschopik 1946; Bastien 1978). Informants from Laguna Arapa, for example, indicated that up unt i l the ear ly 1950s, the major f ish ing r i t u a l of the year was performed during the f i r s t three days of August by transient Qollahuaya div iners from the distant region of Charazani in northern B o l i v i a . The trade of these div iners might have begun in the ear ly Tiahuanaku times, some 2000 years ago (Browman 1981). Since the same div iners could d i rect or perform the f ish ing r i t u a l of any f i sh ing community, I infer that ownership of r i t u a l s i s more l i k e l y to have been a trade secret of the d iv iners , than of the fishermen themselves. The amulets, charms or u tens i l s required for the performance of these f ish ing r i t u a l s were (and probably s t i l l are) i n d i v i d u a l l y owned. These amulets were found while f i sh ing in p r i v i l eged spots, or obtained from luck ie r kin fishermen or from the Qol1ahuayas div iners themselves (Tschopik 1946). The supply of r i t u a l objects and amulets was (and s t i l l is) unlimited and loca l fishermen had no control over access to these amulets. Today a l l the amulets, u tens i l s and ingredients necessary for the successful completion of a f ish ing r i t e can readi ly be purchased from commercial sources (Nunez 1982) thus e l i m i n i t i n g most of the secrecy which surrounded th is paraphernalia and minimizing whatever control of access to f i sh ing i t gave to loca l fishermen. S i m i l a r l y , ownership of such intangibles as technical knowledge does not occur around Lake T i t ieaca , although fishermen are reluctant to share information about f i sh ing techniques. Ownership of intangibles i s thus u n l i k e l y to have any impact on TURFs enforcement . 4.2.2. Ownership of l i v e f i s h containers. Some fishermen from a couple of communities in the Lago Norte and 109 Northern area of Bahia de Puno keep l i v e f i s h in small pens or bol sas of less than one cubic meter capacity. Four out of seven interviewed fishermen from Esca l lan i in the Lago Norte, for example, use c y l i n d r i c a l pens made of a cheap wooden frame wrapped with small-mesh nylon netting bought second hand. In other communities of the same area, fishermen use modified f i sh ing baskets or col 1anchas, normally used for i sp i f i sh ing . Native f i sh species are kept in these pens, un t i l enough of them has been co l lec ted to j u s t i f y a t r i p to the market, or unt i l the market day comes (Farfan 1974; Bustamante and Trevino 1976). Fishermen report that native species survive in these pens, from a couple of days for the carachis, to more than a week for the loca l ca t f i sh . The pens are sunk by the i r owners in discrete locat ions, though c lose enough to the i r home compound to keep an eye on them while taking care of domestic and agr icu l tu ra l chores. Although a pen i s r e l a t i v l y vulnerable to tampering in the absence of i t s owner, none of those interviewed complained of th is sort of problem. In a couple of instances, both in Laguna Umayo, some fishermen have b u i l t small f i sh holding tanks by ra is ing a wal l around a submerged p lo t of land. In both cases the f i s h held captive had been captured previously , and the submerged p lo t of land was c l e a r l y owned by the fishermen invo lved. Since in a l l cases observed the ownership of l i v e f i s h containers does not r e s t r i c t access to f i s h resources before they have been harvested, i t i s u n l i k e l y to affect f i sh ing pract ices. Although, i t occurs infrequently and only invo lves small scale operations, i t i s worth mentioning because i t may invo lve the private ownership of aquatic space. 4.2.3. Ownership of f i sh ing s i tes A common form of aquatic space ownership i s that of indiv idual f ish ing s i tes . I t may be derived from the h i s t o r i c a l practice of fence f i s h i n g , once 110 common among the Uros (Labarre 1946; Palavecino 1948), which enta i led indiv idual or household property r ights over f ixed f ish ing traps and the neighboring totora beds (Ve l la rd 1949: 100). In the ear ly 1940s fence f ish ing was "perhaps the most common technique used at night", with "fences owned i n d i v i d u a l l y and r e b u i l t each year" (Tschopik 1946: 525). Individual ownership of weirs has been reported for the B o l i v i a n part of the Lago Pequeno as la te as the ear ly 1960s (Sole 1969: 51). Today, i t i s hardly practiced anymore (Bustamante and Trevino 1976). Yet the ind iv idual r ights over the corresponding f i sh ing s i tes may have persisted in many places. The indiv idual ownership of p lots of land submerged by the r i s i n g l e v e l of the lake i s a lso responsible for f i sh ing s i t e e x c l u s i v i t y . Totora reeds transplanted by the owners reinforce the i r claims of exclusive use over the resources within a p lo t . However, the major purpose of these claims i s to protect the totora reeds themselves, rather than the f i sh which hide between thei r stems. Individual ownership of patches of totora reeds i s most commonly observed in type I and type II TURF areas. There i t i s frequent enough to invo lve a majority of the shore dwellers (more than 70% of 53 ind i v idua ls from 10 shore communities). Given that these f ish ing s i tes are included within communal f i sh ing zones, the i r pr ivate ownership i s l i k e l y to reinforce e x i s t i n g TURFs. 4.2.4. Ownership of landing s i tes and navigation channels Landing s i tes and navigation channels commanding access to deeper waters are customarily owned p r i va te l y or semi -pr ivately . They are usual ly found in associat ion, and are p a r t i c u l a r l y common in type I TURF areas (Pacori 1976). There, because of a mi ld bottom slope, boats and balsas loaded with f i s h , harvested macrophytes or passengers would not be able to come close enough to 111 the shore without small channels. People would then have to carry the i r goods, wading in cold water and mud for up to a few hundred meters. Landing s i tes and navigation channels a lso ex is t in type II TURF areas, although the ownership of the landing s i te becomes predominant then, since the navigation channel shortens as the bottom slope becomes steeper. In Yapura and Capano on the western side of the Capachica peninsula, for example, fishermen turn the i r pr ivate landing s i tes into t iny ind iv idual harbours for thei r balsas, using small f l o a t i n g is lands of totora reeds or kyles which have dr i f ted to the shore. To understand the impact of landing s i tes and navigation channels on f ish ing a c t i v i t i e s , one must conceive of the shorel ine in the areas with a . weak bottom slope as a zone up to a few hundred meters wide. The edge of the water moves up and down across th is zone, according to seasonal and long term annual f luctuat ions of the lake l e v e l . This fr inge area i s divided into p lots of land over which ind i v idua ls and the i r fami l ies have l e g a l l y recognized usufructory or ownership r ights . When plots of land are submerged because of a r i s i n g lake l e v e l , small channels have to be dug between contiguous p lots for c ra f t access to the deeper waters of the lake. The fami l ies which own or c u l t i v a t e a p lo t along th is channel can use i t , providing they contribute to i t s maintenance. Although the channel may be ending in the p lots of one set of fami l ies th is year, next year i t may end up in those of another set of fami l ies . One thus has to grant use of a channel to people from whom reciprocity can be expected. Since every channel user i s expected to contribute to the channel maintenance, additional users reduce the or ig inal users' contribution to the task. Because i t e n t a i l s control of access to the aquatic space, th is form of ownership i s l i k e l y to play a s ign i f i cant part in determining who can f i sh in 112 type I and II TURF areas. Given the r e l a t i v e l y high degree of endogamy or inbreeding in the communities of the A l t i p l a n o (Bolton and Mayer 1977), any community member should be able to f ind access to a landing s i t e or a navigational channel in those areas, by act ivat ing his or her kinship t i e s . Members of in land communities are more l i k e l y to face d i f f i c u l t i e s in getting access to such landing s i t e s , although they could always try to estab l ish some f i c t i v e kinship t i e s , or offer compensation. Rights of ind i rect economic gain thus ex is t for these channels, although they may not be the source of substantial advantages for the channel owners. 4.2.5. Ownership of docking f a c i l i t i e s The l a s t type of f i sh ing - re la ted ownership refers to the r e s t r i c t i o n of access to docking f a c i l i t i e s and to the small harbours they protect. These f a c i l i t i e s are only found in type III TURF areas where small harbours are necessary to protect boats and balsas from the waves. Access i s usual ly rest r ic ted to the people who have contributed the i r resources and labour to the construction and maintenance of these f a c i l i t i e s . This may include the members of a community, or i t may be res t r i c ted to the members of one sector of a community, as in Huarisani (Huancane), Llachon (Capachica), and on the is lands of Soto, Amantani and T a q u i l i . Though uncommon communal docking f a c i l i t i e s are l i k e l y to play an important ro le in the control of access to f i sh ing , since the f i sh ing craf ts of outsiders unable to use them would be destroyed by the waves. [Footnote: There i s a lso a number of docks in type II TURF areas which do not protect any harbour and are used pr imar i ly for the landing of goods and passengers, as those of Cariquita (Moho), V i lquechico, Chucuito, Parina, Carana, Tacasaya, Luquina, J u l i , Pomata and Yunguyo.] Having demonstrated that along Lake T i t icaca's shores, there ex is t other forms of ownership over aquatic space than TURFs, I can now try to determine 113 how Lake T i t ieaca shore dwellers use these forms of ownership and ex ist ing laws to enforce thei r i l l e g a l TURFs. 4.3. ENFORCEMENT OF TURFS Because legal documents dealing with aquatic resources and o f f i c i a l practices do leave room for interpretat ion, Lake T i t ieaca shore dwellers are able to use a variety of legal means to enforce thei r i l l e g a l TURFs. 4.3.1. Importance of aquatic macrophytes Totora reeds contribute to the enforcement of Lake T i t ieaca TURFs in a number of important ways. F i r s t they are the most v i s i b l e markers of TURF boundaries and one of the most valuable resources within these areas. Second they leg i t imize the special r ights of shore communities over the lake area immediately adjacent to thei r shores. As mentioned e a r l i e r , the r ights of these communities to the totora beds planted in the i r l i t t o r a l area i s o f f i c i a l l y recognized. This al lows communities with totora beds to discourage outside fishermen from operating within the i r areas. Outsiders usual ly stay at least 200 meters away from the outer edge of these totora beds, not to be accused of damaging or s tea l ing totora reeds. And t h i r d , dense totora beds ensure greater freedom for the enforcement of loca l TURFs, by hindering access to the shore for the coast-guard. Llachu weeds a lso contribute to the enforcement of loca l TURFS. Because they are considered as communally owned grazing resources, support from the administration i s expected for the protection of TURF holders r ight to them. The community of Sajo (Pomata), for example, appealed to the MINA in Puno, during the drought of 1983 to repel 1 trespassers from the neighbouring communities of Chatuma and V i l l a Santiago de Ccama (Sajo, Letter to the 114 Director of the MINA-Puno, Ju ly 11, 1983). 4.3.2. Role of fishermen gui lds and associations For the enforcement of the i r TURFs, the members of various shore communities take advantage of the fishermen's associations which they have created, as in Occosuyo (Amantani), Barco (Chucuito), Ccota (P la te r ia ) , Chimu (Puno), Esca l lan i and Llachon (Capachica), Pusi (Pusi), Socca (Acora), Taqui l i i s l a n d , Kajsi and Patas (on Laguna Umayo), and of the Comite Jose Olaya de  Arapa. These associations may promote and enforce some regulations for the orderly conduct of f i sh ing or for loca l resource conservation with the approval of the MIPE, although the l a t t e r has demonstrated l i t t l e interest in supporting or in co l laborat ing with these associations. The performance of the annual f i sh ing r i t e s , usual ly in the f i r s t week of August (Nunez 1982), i s a major concern of fishermen's associations. The enforcement of communal ly -he1d TURFs and the minimization of c o n f l i c t s between fishermen a lso run high on the i r p r io r i t y l i s t . The "Associacion de  Pescadores Menores de Consumo Humano de Escal1 ani ( d i s t r i t o de Capachica)", for example, which was created in the ear ly 1970's, l i s t e d as objectives "the protection of i t s f i sh ing areas, the sett ing of f ish ing schedules approved by the Sargento de Playa to avoid net thef t , and that of f i sh ing turns for groups of four or f i v e fishermen at a time, to f i s h in Sejnajachi area, during the low season" (Bustamante and Trevino 1976: 78, my t rans lat ion) . The "Comite de Pescadores Jose Olaya de Arapa", which started to organize f i sh ing a c t i v i t i e s for the boga in the western h a l f of Laguna Arapa in the ear ly 1970s, deserves par t i cu la r attention because of i t s success in maintaining t rad i t iona l TURF areas. The leaders of th i s association agreed on prohibit ing nocturnal f i sh ing a c t i v i t i e s (between 5pm and 5am) to avoid 115 f i sh ing accidents and nets' s tea l ing under the cover of darkness. Although TURFs enforcement was never an e x p l i c i t objective of th is federative committee, i t s prohibit ion of nocturnal f i sh ing a c t i v i t i e s forced the members of i t s nine loca l sub-committees to stay within the i r own f ish ing areas. This contributed to TURFs prorogation, despite the strong denials of the persistence of t rad i t iona l f ish ing zones by i t s most educated member (Alberto Gonzales Urquiaga 1984: personal communication). Although extreme th is i s not the only example of fishermen's a b i l i t y to take advantage of ex ist ing regulations to protect thei r TURFs. 4.3.3. Rol e of Sargentos de PI ay a To minimize c o n f l i c t s , the Sargentos de PI aya can use the i r o f f i c i a l status to discourage outsiders from operating in the i r area of j u r i s d i c t i o n which usual ly coincides with the TURF area of thei r community. Although a Sargento has no legal power to exclude an outsider from his area of j u r i s d i c t i o n , i t i s quite easy for him to accuse even f u l l y l icensed intruders of not respecting f i sh ing regulations, of s tea l ing nets or of creating disturbances. No outside fisherman would be foolhardy enough to disregard the opposition of a Sargento backed by his f e l l o w community members (or a fishermen association) when sett ing his nets and getting ready to spend the night in his boat, far away from his home and his own fr iends. If verbal abuse did not discourage him, surreptit ious violence would soon fo l low. 4.3.4. Role of control to shore access Pr ivate or semi-private channels and landing s i tes are wel l respected, f i r s t because they are e a s i l y enforced and second, because one has no choice but to use them where they ex ist . They are eas i l y enforced because one's c ra f t i s exposed to easy r e t a l i a t i o n from the owners of the channel and 116 landing s i te whose prerogatives have not been respected. And one has to use them because i t i s quite d i f f i c u l t to navigate across dense beds of totora reeds. In addit ion, communal author i t ies , l o c a l l y known as rescat is tas , v ig i lan tes de campo or v ig i lan tes de chacra, have the power to give penalt ies to people who trespass on others' totora p l o t s , either cutt ing totora reeds or damaging them by pushing the i r boats through. Because i t i s wel l respected, th is form of ownership plays a s ign i f i cant part in determining who can f i s h in a par t i cu la r area. Although i t would be exceptional for any member of a shore community to be unable to f ind access to a landing s i te or a navigational channel, members of inland communities have much more d i f f i c u l t y getting access to such landing s i tes . Some informants reported that channel owners required compensation for the use of the i r channel. There are very few publ ic landing s i tes or docking f a c i l i t i e s along the shores of the lake, with the notable exceptions of the small dock of Barco (Chucuito) and the large je t ty of the port of Puno. As mentioned e a r l i e r , small docks protecting t iny harbours have been b u i l t in type III TURF areas by loca l residents who have retained control over the use of these f a c i l i t i e s . Road access to the shorel ine i s often l imi ted to small foot paths, which reinforces loca l residents' control over the i r small docks. This a lso applies to pr ivate landing s i tes and the small navigation channels to which they are associated. D i r t roads do lead to shore communities but one often has to fo l low a narrow foot path across the land of the loca l resident's to reach the lake. In communites holding TURFs of type I th is may mean a walk of a couple of ki lometers, which invar iab ly leads one close to the dispersed houses of the residents and subjects one to the i r control (or that of the i r dogs!). I t i s thus hard to think of any shore area to which anybody can have free physical access. Even the fishermen's association of Barco (Chucuito) has 117 discouraged outsiders from using the loca l dock of the Ministry of F isher ies . As for the large je t ty of Puno harbour, none of the dozen of fishermen who operate from the shores of the Inner Bay of Puno uses i t to store his c raf t (Av i la et a l . 1985), because of excessive r isks of theft in an area frequented by many transients. In conclusion, the wide d is t r ibut ion of TURFs implies an almost to ta l control of access to the shorel ine by onshore communities. Very few publ ic f a c i l i t i e s (roads and docks) provide access to the lake for in land dwel lers. Even for ex ist ing publ ic docks competition for l imi ted space with present users e f f e c t i v e l y discourages outsider's interference. And where shore access seems phys ica l l y more p r a c t i c a l , such as along the Panamerican road between Puno and Chucuito for example, outsiders' c ra f t and gear are a lso more vulnerable to transients' larceny. 4.3.5. Comparison of enforcement for d i f ferent types of TURFs For each of the three types of TURFs i d e n t i f i e d , the width of the exc lus ive f ish ing area, that i s the distance from shore to which these TURFs apply i s determined not only by a combination of natura l , whether physical or b io log ica l factors (presence of resources and of v isual markers for boundaries), but a lso of human factors (ease of enforcement, type of f i sh ing technique used). Natural factors are usual ly though not always determinant. In type III TURF areas, for example, human factors may become predominant because of the absence of totora beds thus of c lear boundary markers. The respect of TURFs then becomes a matter of act ive enforcement by loca l fishermen. Human factors also become predominant when communities s i t on opposite sides of narrow bodies of water, because they have to d iv ide the corresponding water space between themselves. Enforcement on the outer edge of type I TURF areas poses obvious 118 l o g i s t i c a l problems. It i s d i f f i c u l t to spot intruders hiding among totora reeds which grow up to 3 m above the water surface (Col l o t 1980) from a low shorel ine. To solve th is problem, the inhabitants of the Uros f l o a t i n g is lands in the Bahia de Puno bu i ld small platforms, one or two meters high, from which they can keep an eye over the i r te r r i to ry . S t i l l , detection of trespassers is not su f f i c ien t to protect one's resources, for one a lso needs to repel or at least to discourage them, and the wider TURFs are, the more d i f f i c u l t enforcement on thei r outer edge i s . I t i s thus common for type I TURF holders to to lerate some degree of trespassing, and to request some compensation, as commonly done by the residents of the Uros f l oa t ing is lands of K'api i n the Bahia de Puno. Enforcement i s not quite so d i f f i c u l t for type II TURFs, because the area they encompass i s not as wide. In addition the steepness of the slope of the bottom of the lake i s re lated to a stronger r e l i e f on the shorel ine, which al lows TURF holders to keep a watchful eye over the i r te r r i to ry almost permanently. Outsiders who trespass under the cover of darkness have a chance of s l ipp ing away before being caught, but they are very l i k e l y to be spotted and wel l advised to make an arrangement with loca l TURF holders, i f they want to avoid los ing the i r nets, or being abused. For type III TURFs, the extension of the f i sh ing zone i s very much influenced by the presence of schools of i s p i . Only when the l a t t e r are detected, do TURFs holding fishermen bother to repel 1 outsiders. 4.3.6. Enforcement p r a c t i c e TURFs can be enforced in two major ways. Trespassers may be repel led ve rba l l y by TURF holders who shout "Out! Out!" (Jota! Jota!) to them, and who "pi ay stupid and refuse to make any deal" (_se ponen brutos y JTO quieren saber  nada de compromiso). Stea l ing nets i s the other common t a c t i c to discourage 119 trespassing. Among the 50 co l laborat ing fishermen, for example, 27 (11%) l o s t at least one net during the f i v e year period between 1980 and 1984. Subsequently, the reputation earned by some fishermen from protecting the i r TURF may be advertised to discourage outsiders from showing up, as I was t o l d during f i e l d work about some communities of the northeast shores of the Puno Bay and of the Lago Grande. These two forms of enforcement are by far the most commonly used, as indicated by 40.6% of the 251 fishermen interviewed in 1976. The majority of them (79.4%) acknowledged that verbal or physical interference ("se atajan") and net theft or destruction ("rompen y roban redes") were the most common tact ics to repe l l trespassers. More v io len t forms of r e t a l i a t i o n may also be used occas ional ly , and rumor has i t that a few trespassers were beaten to death. I t i s unclear, however, whether such events ac tua l l y took place, or whether the rumor i s an advertisement strategy to discourage trespassing. In conclusion, despite the i r informal status, the antagonism of o f f i c i a l texts and the opposition of government o f f i c i a l s , TURFs are s t i l l widely enforced around Lake T i t ieaca . Currently, shore dwellers are able to take advantage of various legal and administrative loopholes to extend them. They are successful at i t because they use e f fec t i ve though i l l e g a l threats of r e t a l i a t i o n to protect the i r resources against competing interests . This shows that exc lus ive use of aquatic resources within t e r r i t o r i a l areas does not ar ise from a lack of competition for these resources, nor from a widespread social prejudice against f i s h i n g , but from TURF holders w i l l i n g e s s and a b i l i t y to enforce the i r t e r r i t o r i a l r ights . This demonstrates that Lake T i t ieaca TURFs do ex is t and are not merely a convenient myth among shore dwel lers . Having demonstrated TURFs existence, I can now consider the effectiveness of these TURFs. 120 CHAPTER 5 EVALUATION OF FISHING RENTS Because the net return to f ish ing labour i s greater than i t s opportunity cost, Lake T i t ieaca fishermen receive an addit ional revenue over what they need to cover the i r f ixed and var iable costs, including an i m p l i c i t wage for themselves and the i r helpers. This addit ional revenue represents the rent from loca l f i sher ies . To demonstrate the capture of such rents by Lake T i t ieaca fishermen, I have estimated the opportunity cost of labour with the net return from a c t i v i t i e s a v a i l a b l e to Lake T i t ieaca shore dwel lers , and compared i t with the net return from f i sh ing and f i sh ing - re la ted a c t i v i t i e s . To ca lcu la te the l a t t e r , I have deducted the costs entai led by these a c t i v i t i e s from the gross return of f i s h i n g , and divided the resu l t ing net return by the corresponding amount of labour. 5.1. FISHING COSTS AND FISHING RETURNS 5.1.1. F i s h i n g costs The costs involved in f i sh ing and f i sh ing - re la ted a c t i v i t i e s include the depreciation and maintenance costs of the f i sh ing c ra f t , gear and outboard engine, the operating costs of the l a t t e r , the expenses for r e t a i l i n g the catch, and some miscellaneous expenses such as regist rat ion costs. I have ca lcu lated these material and labour costs (Appendix D) and l i s t e d them in table 7. 121 TABLE 7: MATERIAL AND LABOUR COSTS (Exchange rate for the Peruvian Sol FOR LAKE TITICACA SMALL SCALE in January 1980: S/250 = US$ FISHERIES 1) PURCHASE Material (Soles) VALUE Labour (Days) COST PER Material (Soles/year) YEAR Labour (Days) Balsa reed boat 10,456 5 12,008 10 Wooden boat 78,825 18,980 13.5 G i l 1 net (for n nets) n(l,195) n(14,713) 30-40 Huayunaccana (trawl) 178,100 30 11,873 2 Fish r e t a i l i n g 54,149 47-70 5.1.2. F i s h i n g returns To ca lcu la te the gross return from f ish ing I have m u l t i p l i e d the monthly catch figures by the corresponding r e t a i l prices (Appendix D). For these ca lcu lat ions I have taken into account the sale of trout and s i l v e r s i d e in B o l i v i a by 16% of the fishermen, for prices 2 to 2.5 times the Peruvian ones respect ively (Appendix D). In my ca lcu la t ions of the net return to f i sh ing labour, I have accounted for the labour dedicated to f i sh ing - re la ted a c t i v i t i e s by members of the fishermen's households and that dedicated to f i sh ing by the fishermen themselves. However, I have not accounted for the labour of thei r helpers (which i s equivalent to assuming that the l a t t e r has no opportunity cost) because helpers are learning as much as they are helping, and because they would be unable to part ic ipate in another a c t i v i t y in view of thei r age. Table 8 l i s t s the corresponding hourly returns to labour. The p r o f i t a b i l i t y of f ish ing varies substant ia l l y with the type of FEU considered (Table 8). FEUs of type 8 (boats with outboard engines and g i l l n e t s ) have the highest hourly return to labour, which i s hardly surprising 122 since a l l FEUs of type 8 s e l l the i r catch on the B o l i v i a n market. FEUs of type 10 (boats and huayunaccanas) have the second highest p r o f i t a b i l i t y , and FEUs of type 4 (boats with g i l l n e t s for native species) and of type 6 (boats with g i l l n e t s for both groups of species) have the th i rd and fourth respect ive ly . F i n a l l y , FEUs of type 1 (balsa reed boats with g i l l n e t s for native species) have the lowest hourly return for labour. TABLE 8: RETURNS TO FISHING AND FISHING-RELATED LABOUR PER FEU TYPE IN LAKE TITICACA SMALL-SCALE FISHERIES (IN PERUVIAN SOLES, JANUARY 1980). FEU type 1 4 6 8 10 A l l Gross return 266,909 287,299 424,746 989,176 453,841 421,638 Fishing costs 77,028 92,708 90,959 315,370 91,697 110,554 Net return 189,881 194,590 333,787 673,806 362,144 311,083 Return/hour 68 72 133 206 175 121 5.2. FISHING RENTS The difference between the net p r o f i t a b i l i t y and the opportunity cost of f i sh ing labour determines the size of the f i sh ing rents. The opportunity cost of labour i s measured by the return th is labour could have brought in the best a l te rnat i ve a c t i v i t y a v a i l a b l e . Because th is a v a i l a b i l i t y depends upon each fisherman's personal contacts, education, or community of o r i g i n , the opportunity cost of labour i s not the same for a l l FEUs. I demonstrate here, that for each type of FEU, the return to f ish ing labour i s higher than i t s opportunity cost, thus that f i sh ing y i e l d s substantial rents. 123 5.2.1. Opportunity cost of labour To estimate the opportunity cost of labour for each type of FEU, I had to determine which was the most prof i tab le of the a l te rnat i ves to f i s h i n g , for the FEU considered. Lake T i t ieaca shore dwellers part ic ipate in a variety of cash-generating a c t i v i t i e s such as mining, br ick lay ing in urban areas of the coast, coffee growing in the jungle , art isanry , trade and transport a c t i v i t i e s accross the border with B o l i v i a , seasonal wage labour on the plantations of the coast, c a t t l e ra i s ing , da i l y wage labour on the A l t i p l a n o and music playing. By discussing the a v a i l a b i l i t y of these a c t i v i t i e s for the fishermen of each type of FEU, start ing with the most prof i tab le (Table 9), I determine the corresponding opportunity cost of f i sh ing . a) Coffee growing in the jungle The opportunity cost of f ish ing for most FEUs of type 8 may be derived from the return to coffee growing in the jungle. Because the l a t t e r requires the investment of substantial amounts of labour and c a p i t a l , and a supportive network, i t i s only a v a i l a b l e to shore dwellers from the d i s t r i c t s of Moho and Conima along the eastern side of the Lago Norte (Co l l ins 1982; Painter 1982), where most type 8 FEUs are found (Bustamante and Trevino 1976). b) Mining, brick lay ing , c i v i l service The return from mining or from brick laying provides an estimate of the opportunity cost of f ish ing for most FEUs of type 10. These trades require long term migrations to distant urban areas of the coast and s k i l l s not usual ly mastered by a g r i c u l t u r a l i s t s . They also require good connections because such rewarding and prestigious jobs are in high demand (Brown 1978). 124 TABLE 9: PROFITABILITY AND AVAILABILITY OF CASH-GENERATING ACTIVITIES FOR THE SHORE DWELLERS OF LAKE TITICACA (JANUARY 1980) A c t i v i t y Return to labour Conditions of a v a i l a b i l i t y (Soles/hour) Coffee production in jungle 138 A v a i l a b l e only to those with access to land, f i v e years after planting Mining, brick laying and c i v i l service 120 - 110 A v a i l a b l e year round, but only to a minority of shore dwellers Trade and transport across the border 107 A v a i l a b l e mostly to those with capi ta l and protections Basic 1egal salary and art isanry 88 Only for urban employment and for s k i l l e d workers (e.g. boat bui lder) Cat t le ra is ing on reeds and weeds 70 A v a i l a b l e year round to those who can purchase c a t t l e Seasonal wage labour on coast 45 - 40 A v a i l a b l e from December to January and May to J u l y , but with migration Dai ly wage 1abour on Al t ip lano 25 A v a i l a b l e from February to Apr i l and August to November (Sources: V e r l i a t 1978; Painter 1986; INE-Puno) 125 Since fishermen belonging to FEUs of type 10 are the only ones with such connections, the corresponding return provides an estimate of the opportunity cost of the i r labour. However, for those type 10 FEUs which do not have these connections, the return from mining, br ick lay ing and other urban trades overestimates thei r opportunity cost of labour. c) Trade and transport across the border Trade and transport of goods across the border between Peru and B o l i v i a i s usual ly at the margin of l e g a l i t y (Painter 1986). I t i s a v a i l a b l e pr imar i ly to shore dwellers who can afford the investment i t requires, and who can count on a number of sympathies or on patronage. Furthermore, i t i s not the most prof i tab le a c t i v i t y a v a i l a b l e to those who can practice i t , because of the s ign i f i cant r isks i t s involves which imply a lower expected value for i t s returns than suggested by a v a i l a b l e estimates (Ibid.). F i n a l l y , smuggling and f ish ing can be eas i l y combined, thus suggesting that the two a c t i v i t i e s are not a l te rnat i ve but complementary. The corresponding returns should thus not be used to estimate the opportunity cost of f i sh ing . d) Artisanry The return from artisanry could be used to estimate the opportunity cost of f i sh ing for some FEUs, because handicrafts such as hat making, weaving, balsa or boat bui ld ing are common a c t i v i t i e s around Lake Ti t ieaca (Galdo 1981). Along the shores of the Bahia de Puno, for example, 40% of the households re ly upon handicrafts for cash (CENFOR-Puno 1982-83, unpubl. data). And art isanry may contribute as much as 37% of the cash income of a-household (Figueroa 1984: 45). However, many of these handicrafts a c t i v i t i e s are not a v a i l a b l e to the fishermen themselves, because they are considered to be female a c t i v i t i e s (e.g. weaving, kni t t ing) . I have approximated the return 126 from art isanry with the basic legal salary because the only rural dwellers who are able to ask for such a salary are boat bui lders and carpenters. However, th is approximation can only be used to estimate the opportunity cost of f i sh ing labour for those fishermen who are a c t u a l l y able to bu i ld a boat, most of whom operate type 6 FEUs. e) Cat t le ra is ing The income from c a t t l e ra is ing might have beeen used as an estimate of the opportunity cost of f i sh ing labour because i t i s a v a i l a b l e to most fishermen, and because i t i s the most prof i tab le a c t i v i t y a v a i l a b l e to those who can neither bu i ld a boat, get a job on the coast, or grow coffee in the jungle. The practice of purchasing scrawny c a t t l e to r e s e l l for slaughter af ter 8 to 10 months of ra is ing on totora reeds, l lachu weeds and stubble i s widely a v a i l a b l e among shore dwellers (Lewellen 1977: 48; Figueroa 1984: 49). More than 90% of them own at least one head of c a t t l e and more than 80% own at least two (Lev ie i l and Goyzueta 1984). This practice i s a lso more prof i tab le than a l te rnat i ve cash generating a c t i v i t i e s such as seasonal wage labour, because i t y i e l d s a return which I have estimated at 70 soles/hour, using data from previous studies (Lewellen 1977: 62; L e v i e i l and Goyzueta 1984). In addition to th is return, cows contribute the i r offspring and milk to the household economy (Lewellen 1977) and b u l l s contribute the i r work when bought in pair and trained for team ploughing (Hickman 1963; Brown 1978: 39). In spite of the above, the return from c a t t l e ra is ing should not be used to estimate the opportunity cost of f i sh ing because these two a c t i v i t i e s are complementary rather than a l te rnat i ve . They do not interfere but contribute to each other. This complementarity may a c t u a l l y explain why f ish ing households have more heads of c a t t l e (3.35) than shore households as a whole (2.85). On one hand, a f i sh ing c raf t can be used to c o l l e c t totora (3.3 127 times/month) and l lachu fodder (5.2 times/month), after re t r iev ing f i sh ing nets (Lev ie i l et a l . 1986). On the other hand, c a t t l e ra is ing contributes to the undertaking of f i sh ing , because c a t t l e i s a form of saving (Montoya 1979: 785) which fishermen often use to finance the purchase of the i r boat. F i n a l l y , both a c t i v i t i e s can be scheduled f a i r l y e a s i l y within the normal working day of a shore household. f) Seasonal wage labour on the coast The income from seasonal wage labour in the plantations of the coast provides an estimate of the opportunity cost of f i sh ing labour for type 1, 4 and most of type 6 FEUs, because of the widespread a v a i l a b i l i t y of th is a c t i v i t y and because of i t s incompat ib i l i ty with f i sh ing . High rates of part ic ipat ion demonstrate that i t i s a widely a v a i l a b l e option for Lake T i t ieaca shore dwel lers . In the mid 1970s, the proportion of household heads involved ranged from 36 to 74% depending on the community (Lewellen 1977: 48; Brown 1978: 151). In the ear ly 1980s, th i s proportion was estimated at 60% for a sample of 8 communities around the Bahia de Puno (CENFOR-Puno 1982-83, unpublished data). Furthermore, th i s a c t i v i t y i s incompatible with f i sh ing because to hire themselves as seasonal labourers on the plantations of the coast, shore dwellers have to become involved in a pendular or c i r c u l a r migration (Hickman 1963; Brown 1978). g) Dai ly wage labour and music playing F i n a l l y , the least prof i tab le option for shore dwellers to earn some cash i s to hire themselves for da i l y wage labour. The corresponding return to labour cannot be considered as the opportunity cost of f i sh ing because almost a l l fishermen have the better option of working for seasonal wage labour in the plantations of the coast. Furthermore, i t i s not a major a c t i v i t y on the 128 A l t i p l a n o where households have t r a d i t i o n a l l y r e l i e d upon the reciprocal exchange of labour for the peak periods of the agr icu l tu ra l cycle (Figueroa 1984: 46). Even nowadays, Only 20% of the households in the southern Andes contract farmhands (Brown 1978: 179; Aramburu and Ponce 1983: 103). The h i r ing out of labour for da i l y wage salary uses less than 8% of the tota l labour force a v a i l a b l e per household (Gonzales 1984: 182), and less than 1% of the labour of Lake T i t ieaca fishermen. Music playing in one of the bands which hire themselves out for loca l celebrat ions, or f i e s t a s , i s f a i r l y common in some shore communities (Hickman 1963; Lewellen 1977). However, the corresponding income does not y i e l d an estimate of the opportunity cost of f i sh ing e i ther , because i t i s only an occasional source of cash for members of a few specia l ized shore communities (Ibid.). Furthermore, music playing and f i sh ing can e a s i l y be combined, since the former i s only an occasional undertaking. 5.2.2. Capture of f i sh ing rents The opportunity cost of f ish ing i s thus determined by the return to labour from seasonal labour on the coast for FEUs of type 1, 4 and most of type 6, by the return from artisanry for some FEUs of type 6, by the return from coffee growing for most FEUs of type 8, and by the return from mining and construction for most FEUs of type 10. As table 10 shows, for each type of FEU the r e l a t i v e difference between the hourly return from f i sh ing and the opportunity cost of labour i s pos i t i ve and ranges from 50 to more than 100%. For the fishermen of type 6, 8 and 10 FEUs who do not meet the requirements for part ic ipat ing in art isanry , mining, br ick lay ing or other urban trade, or coffee growing, th is r e l a t i v e difference i s even greater (from 200 to 500%), since the opportunity cost of thei r labour i s determined by the return from 129 seasonal wage labour on the coast. A l l th is demonstrates that Lake T i t ieaca fishermen are able to capture substantial f ish ing rents. TABLE 10: RELATIVE DIFFERENCE {%) BETWEEN HOURLY RETURNS TO LABOUR FOR FISHING ON LAKE TITICACA AND FOR ALTERNATIVE ACTIVITIES AVAILABLE TO SHORE DWELLERS, ACCORDING TO FEU TYPE (JANUARY 1980). FEU type 1 4 6 8 10 A l l Fishing return (S/hr) 68 72 133 206 175 121 Coffee growing (*: 138 S/hr) 78 Min ing , misc. (*: 120-110 S/hr) 55-67 Basic legal salary (*: 88 S/hr) 51 134 99 38 Seasonal labour (*: 45-35 S/hr) 51-94 60-106 195-232 358-415 290-340 170-203 Dai ly labour (*: 25 S/hr) 172 188 432 724 600 384 (*: Sources: Ve r l ia t 1978; MINA-Puno 1983; Painter 1984; 1986) 5.3. DISCUSSION 5.3.1. Evaluation of non-market s ide-benefi ts Comparing the hourly returns from f i sh ing with those from a l te rnat i ve a c t i v i t i e s assumes that fishermen do not consider non-quantif iable s ide-benefits ( i .e. intangible) . This reduces the i r decision-making process to a u t i l i t a r i a n choice between purely monetary benefits, when some other c r i t e r i a are d e f i n i t e l y involved. F ish ing, for example, implies a l o t of discomfort during the cold nights and ear ly mornings of the dry season when frosts are 130 common, and some inescapable r isk for those who sleep in the i r open boat d r i f t i n g with thei r nets in the middle of the lake. I t a lso e n t a i l s an unpredictable income, contrary to seasonal wage labour, as one fisherman indicated "work on the coast i s preferrable, because wages are secured by contract" (Me gusta mas el trabajo en l_a Costa, es muy seguro en e l contrato). Migration offers some non-monetary benefits r e l a t i v e to f i sh ing . I t brings prestige and opportunities to climb s o c i a l l y (Brown 1978), when there may be some social prejudice against f i sh ing (Lev ie i l 1986), and no opportunity for a rapid social ascension. However, migration requires the head of a household to spend long periods of time away from his family in a foreign and eventual ly h o s t i l e environment. Work in the jungle in par t i cu la r involves numerous health r isks (Painter 1982). When migration l a s t s for more than a couple of months at a time, i t af fects the agr icu l tu ra l cyc le . In the absence of thei r husbands, women must hire farmhands, as demonstrated by the high (64%) proportion of households which contract da i l y wage labour in communities from which numerous household heads migrate to the coast (Brown 1978: 179). I have not included side-benefits in my analys is , f i r s t because they are intangib le , and secondly because I have assumed that they are l i k e l y to cancel one another in the shore dwellers decision-making process. Yet, to f u l l y understand the rat ionale of Lake T i t i caca shore dwellers in choosing between f ish ing and a l te rnat i ve a c t i v i t i e s , such as seasonal migration, one has to consider that they make choices between various types of benefits. S i m i l a r l y , one has to consider that fishermen make choices between combinations of a c t i v i t i e s , which invo lve not only the i r own labour but that of thei r whole household and eventual ly , that of the households with whom they exchange labour. During a fisherman's absence, for example, his household 131 s t i l l uses his f i sh ing c ra f t for totora and l lachu c o l l e c t i o n to feed his c a t t l e , or for waterfowl hunting or egg c o l l e c t i o n (between August and December). A r e l a t i v e may a lso set and ret r ieve the nets of the absent fisherman with his own, and bring them back to the la t te r 1 s wife and chi ldren to pick up the f i s h tangled in the nets. In return, the l a t t e r i s expected to set the nets of the helpful r e l a t i v e when he himself goes away. F i n a l l y , the boat and gear of the absent fisherman may be rented out to a r e l a t i v e . Taking into account shore dwellers' a b i l i t y to combine f ish ing with agr icu l ture , for example, implies accounting for the heterogeneity of household labour, and granting a d i f ferent opportunity cost to the labour spent on f ish ing and that spent on f i sh ing - re la ted a c t i v i t i e s . 5.3.2. Accounting for heterogeneity of household labour The a v a i l a b i l i t y of a l te rnat i ve a c t i v i t i e s i s heterogeneously distr ibuted both among f ish ing households, according to education, connections or community of o r i g i n , and within households according to age and sex. F ish ing -related a c t i v i t i e s , for example, are often carr ied out by women and chi ldren while tending sheep, by elders who cannot undertake heavy work in the f i e l d s any longer, or by fishermen themselves when there i s l i t t l e e lse to do. A d i f ferent opportunity cost of labour could thus be attr ibuted to the labour spent on such a c t i v i t i e s as c ra f t or gear bu i ld ing , cleaning and maintenance, on the presumption that f i sh ing - re la ted a c t i v i t i e s are undertaken in the slack periods of the agr i cu l tu ra l cyc le . Since the only a c t i v i t y competing with f i sh ing - re la ted a c t i v i t i e s is da i l y wage labour, i t i s l o g i c a l to estimate the opportunity cost of labour spent on these a c t i v i t i e s with the hourly return from da i l y wage salary. 132 TABLE 11: RELATIVE DIFFERENCE (%) BETWEEN THE ITS OPPORTUNITY COST, WITH OR WITHOUT GRANTING FISHING-RELATED ACTIVITIES THAN FOR FISHING. RETURN A LOWER TO FISHING LABOUR AND OPPORTUNITY COST FOR FEU type 1 4 6 8 10 Without 51-94 60-106 51 78 55-67 With (25 S/hr) 89-143 116-177 128 135 104-133 Because i t i s the least prof i tab le of a l l the a l ternat ives to f i s h i n g , th is leads to a larger r e l a t i v e difference between f i sh ing returns and the opportunity cost of f i sh ing labour, as i l l u s t r a t e d in Table 11. This analys is thus shows that accounting for the compat ib i l i ty of f i sh ing with agr icu l tu ra l a c t i v i t i e s leads to an increase of the r e l a t i v e difference between the return to labour from f i sh ing and i t s opportunity cost , and to an increase of the rents captured by Lake T i t icaca fishermen. However, i t raises the question of whether fishermen would s t i l l capture f i sh ing rents, i f the opportunity cost of thei r labour was estimated by the return to labour from the subsistence agr icul ture practiced on the A l t i p l a n o . 5.3.3. Comparison of f i sh ing and subsistence agr icul ture To estimate the return to labour from subsistence agr icu l ture , one can use the r e t a i l prices per hour of labour for agr icu l tu ra l products, as done by Painter (1984; 1986) and indicated below after correction for i n f l at ion. Broad Quinoa Bar!ey Dried broad Potatoes Freeze-dried Household beans beans tubers weighted mean 47 48 83 94 147 182 320 133 The r e t a i l price of the f i s h catch per hour of labour i s as below. FEU type 1 4 6 8 10 A l l Retai l price per hour of labour 113 118 206 251 246 179 Comparing r e t a i l prices per hour of labour from f ish ing and subsistence agr icul ture reveals that f i sh ing returns are s l i g h t l y lower than subsistence agr icul ture returns in general, but higher than those from such crops as broad beans, quinoa, barley and dried broad beans. If the l a t t e r were not included in the crop rotat ion, the land would have to be fa l lowed, or f e r t i l i z e r , expensive by l oca l standards, would have to be purchased. Given that Painter (1982) did not include a l l production costs in his ca lcu lat ions of the returns from subsistence agr icul ture (Appendix D), and that his sample of a g r i c u l t u r a l i s t s came from one of the most productive agr icu l tu ra l areas of the lakeshore, i t i s l i k e l y that the difference between agr icul ture and f ish ing returns may be smal ler than apparent. This demonstrates that i f one assumes that the r e t a i l price per hour of labour provides an index of the value of subsistence agr icu l ture , then i t does not make sense economically for shore dwellers to abandon subsistence agr icul ture and to become specia l ized fishermen. However, the return per hour from subsistence agr icul ture does not provide an estimate of the opportunity cost of f ish ing because these two a c t i v i t i e s are complementary rather than a l te rnat i ve . The fishermen who f i s h most often are those who have leas t land, such as those from Soto i s l a n d , from Barco-Chucuito, and those from the flooded areas of the Puno bay. S i m i l a r l y , young s ing le men who have not inherited land from thei r parents yet may spend more time f ish ing than most 134 married men from the same communities. 5.3.4. Persistence of rents over the long term Demonstrating that Lake T i t icaca fishermen captured f i sh ing rents in 1979-80 i s not quite su f f i c ien t to argue that loca l f i sher ies have not reached the i r bioeconomic equi l ibr ium. This equi l ibr ium i s defined over the long term, with the good years cance l l ing the bad ones. The capture of f ish ing rents in 1979-80 could have resulted from a reduction in f i sh ing e f fo r t fo l lowing some bad losses in the previous years. Unfortunately, no h i s t o r i c a l records of Lake T i t i caca f i s h y i e l d s are a v a i l a b l e . However, circumstancial evidence suggests that f i sh ing rents have been captured in the previous years, and that Lake T i t i caca f i sher ies have not reached the i r bioeconomic equi l ibr ium. A comparison between a v a i l a b l e indices of f i sh abundance in 1978 and 1979-80, for example, indicates no apparent change. The captures per unit of e f fo r t for the months of October and November 1978 (1.34 and .905 kg/net/trip respect ively) in the B o l i v i a n part of the Lago Pequeno (Franc et a l . 1986: 18) were s i m i l a r to those encountered by the co l laborat ing fishermen using s i m i l a r f i sh ing techniques over a one year period (.87 to 1.31 kg/net/trip, depending on the type of FEU). They were also s i m i l a r to those measured from two random samples of g i l l netting fishermen in May (1.58 kg/net/trip) and October of 1980 (1.10 kg/net/trip). S i m i l a r l y , a comparison between the catch and the f ish ing e f fo r t of indiv idual fishermen (measured by the number of nets per fisherman) suggests that the catch per unit of e f fo r t (catch per net per night) has remained approximately constant between 1972 and 1979-80 (unless the frequency of f ish ing t r i p s has increased substant ia l l y ) . I have estimated that during th is period, the annual catch and number of nets per fisherman have approximately doubled (from 1 mt to 2 mt, and from 5.7 to 11.1 nets). 135 An analysis of f i s h prices and gear costs over the 1975 to 1980 period shows that gear costs have increased faster than f i sh prices (Figure 15). Carachi and s i l v e r s i d e pr ices , for example, have increased at a slower rate than the o f f i c i a l national consumer price index. Conversely, g i l l n e t s and wooden boat prices have increased at a faster rate than the consumer price index and f i sh pr ices. If the catch per fisherman (i .e. FEU) has not changed during the 1975 to 1980 period, th is implies that f i sh ing costs have increased faster than f ish ing revenues. Therefore, fishermen must have been able to capture even larger rents in 1975 than in 1980, when the i r catch rate was the same but thei r f i sh ing costs were lower. A l l th is suggests that the capture of f i sh ing rents observed in 1979-80 i s l i k e l y to r e f l e c t a long term phenomenon, rather than an exceptional occurrence. Therefore i t suggests that Lake T i t ieaca f i sher ies are u n l i k e l y to have reached the i r bioeconomic equi l ibr ium. This demonstration leads in turn to an invest igat ion of the contribution of TURFs to the capture of f i sh ing rents. 136 FIGURE 15: CHANGES IN ANNUAL PRICE INDICES OF FISH AND FISHING GEAR ON THE PERUVIAN ALTI-PLANO BETWEEN 1 9 7 5 AND 1 9 8 2 (Sources.- MIPE-Puno, IMARPE-Puno, IN E a Becker 1 9 8 3 ) . 3 5 0 0 1 1 1 1 1 r 9 7 5 1 9 7 6 1 9 7 7 1 9 7 8 1 9 7 9 1 9 8 0 1981 1 3 7 CHAPTER 6 TURFS AND THE CAPTURE OF FISHING RENTS So far I have demonstrated that the entire shoreline of Lake T i t icaca i s divided into t e r r i t o r i a l f i sh ing zones and that most fishermen are able to capture substantial economic rents. However, i t has been argued that most indigenous f i sher ies management pract ices, among which TURFs are included, a l l o c a t e access to f ish ing space instead of a l l o c a t i n g resources (Andersen and S t i l e s 1973) and that they are u n l i k e l y to control f i sh ing e f fo r t (McCay 1981: 6). This raises the question of the actual contribution of loca l TURFs to the control of f i sh ing e f fo r t and thus to the capture of ex ist ing rents. The objective of the regulation of f ish ing e f fo r t i s to control f i sh ing costs and the tota l catch from a f ishery in order to maintain a high sustainable y i e l d from th is f ishery. Its log ic stems from the assumption that to ta l catch i s a function of f i sh ing e f f o r t , of i t s effectiveness and of the size of the exploited f i s h population (Ricker 1975). Fishing e f fo r t can be . cont ro l led by l i m i t i n g the number of FEUs par t ic ipat ing in a f ishery , the e f fo r t each FEU develops within th is f ishery , or both. I demonstrate here that membership within a TURF holding community promotes the capture of f i sher ies rents, thus promoting economic e f f i c iency , by c o n t r o l l i n g the number of fishermen and by constraining the i r ind iv idual f i sh ing e f fo r ts . 6.1. EXCLUSION OF INLAND DWELLERS FROM LAKE TITICACA FISHERIES The enforcement of Lake T i t icaca TURFs resu l ts in control of f i sh ing access which can be compared to a l imi ted entry program. It implies the exclusion of in land dwellers from Lake T i t i caca f i s h e r i e s , as demonstrated by 138 numerous accounts (chapter 3) and by the absence of in land dwellers either from an exhaustive census of Lake T i t ieaca fishermen (Bustamante and Trevino 1976) or from the l i s t of fishermen (1,078) o f f i c i a l l y registered by the Coast Guard in Puno. Evidence that by excluding inland dwel lers , TURFs have cont ro l led the number of fishermen on Lake T i t ieaca comes from the apparent s t a b i l i t y of th is number. There has been various estimates of the l a t t e r for the l a s t f i f t y years (F iedler 1940; PDS 1959; V e l l a r d 1963; Everett 1967; MIPE-DIE 1970; Terrazas in Laba 1979). The most r e l i a b l e of these indicate that there were about 3,040 fishermen in 1976 (Bustamante and Trevino 1976), and that th is number has not changed between 1976 and 1984 (Appendix B). I t i s ac tua l l y the same as the one estimated for 1575 (Ibid.). The s t a b i l i t y of th is number in the l a s t decade begs the questions of why there are not more fishermen among shore dwel lers , and why the l a t t e r do not dedicate more resources to f i sh ing . The answers to both must be traced to fishermen's membership in shore communities. 6.2. COMMUNAL CONSTRAINTS ON INDIVIDUAL FISHING EFFORT Membership in a TURF holding community resu l ts in control of ind iv idual f ish ing e f fo r t because to be a member of such a community, one has to hold land in th is community and to keep i t productive. Indiv iduals who wished to f i sh without farming would face the p o s s i b i l i t y of los ing the i r land, the i r status within thei r community and eventual ly thei r membership in th is community. The constraints resu l t ing from the undertaking of agr icu l tu ra l a c t i v i t i e s thus determine how much labour i s a v a i l a b l e for f i sh ing or other a c t i v i t i e s . Membership in a TURF holding community also resu l ts in control 139 of f i sh ing investments. 6.2.1. Constraints on labour a v a i l a b i l i t y Most shore dwellers do not have enough c u l t i v a b l e land to cover the subsistence needs of their household. They h.ave to combine complementary a c t i v i t i e s with agr icul ture which they cannot neglect for fear of los ing the i r land or their status within the i r community. But, because agr icul ture takes precedence over f i sh ing and because i t i s seasonal, i t constrains the amount of labour which they can dedicate to f ish ing or to other a c t i v i t i e s . a) Agr icu l tu ra l requirements A l l fishermen own at least some land, the c u l t i v a b l e part of which i s general ly i n s u f f i c i e n t to support the i r household. The best estimate of the minimum of c u l t i v a b l e land for a family of 5 i s 1.3 has, to which another 1.2 has of unavoidable waste ( fa l low land, home compound) must be added, thus y i e l d i n g a total of 2.5 has (Ferroni 1980: 154) a f igure wel l within the 2 to 3.5 has range of other estimates (Lewel1 en 1977: 44; Poe 1979: 212; Ccama 1981: 122). Few shore dwellers own th is minimum (Ccama 1981: 81; Montoya and Burgos 1981: 32; Painter 1986: 227). This land shortage generates a permanent competition for land among shore dwel lers. Given th is hunger for land, i t s legal ownership by the peasant community as a whole, and the t rad i t ion of i t s redist r ibut ion (Fuenzalida 1970: 78; Brush and G u i l l e t 1985: 26), shore dwellers cannot afford to neglect whatever land they cont ro l , for they might lose i t . The community may repossess unused land (Painter 1982: 123). Other s ib l ings may force a redist r ibut ion of the inheritance. A neighbour may push his own furrows further every year, and l a t e r c laim possession of the land, an easy practice given that many p lots are no more than a few furrows wide by a few meters long ( V e r l i a t 1978). 140 b) Labour supply per household Communal obl igat ions such as celebrat ions, part ic ipat ion to communal projects, administrative r e s p o n s i b i l i t i e s determine the size of a household labour supply. An A l t i p l a n o household comprises s l i g h t l y less than 4.1 members (INE 1981). I t can generate about 3 mandays of labour per work day (Figueroa 1984; Gonzales 1984: 246-7). However, there are only about 250 work days per calendar year because of numerous celebrat ions, or f i e s t a s , within Andean communities (Brush 1977: 132; Gonzales 1984: 174). The celebrat ion of f ies tas i s a consumptive and fes t i ve requirement for community members and a counterpart of thei r access r ights to communal resources (Sherbondy 1982). If the Andean community as a corporate structure were to disappear, many of these celebrations would simply vanish. A standard A l t i p l a n o household thus generates approximately 750 mandays of labour per year, which i s s i m i l a r to estimates for the Cuzco area (Gonzales 1984: 174). Some of the household labour i s dedicated to the communal projects and administrative r e s p o n s i b i l i t i e s entai led by community membership. Community members have to part ic ipate in communal projects, or faenas, for the construction and maintenance of c o l l e c t i v e infrastructures such as roads, fences, i r r i g a t i o n ditches, docks, or communal bui ldings such as schools. Faenas require about 1% of household labour (Brush 1977: 132; Gonzales 1984: 174). Administrative and symbolic r e s p o n s i b i l i t i e s , or cargos, are rotated on an annual basis among community members. I have estimated that faenas and cargos require about 50 mandays of the household labour supply, which leaves about 700 mandays per household for food and income producing a c t i v i t i e s . c) Agr icu l tu ra l labour Agr icu l tura l a c t i v i t i e s are a counterpart of community membership, which determine the size of the household labour surplus a v a i l a b l e for complementary 141 a c t i v i t i e s . The seasonal a v a i l a b i l i t y of th is surplus imposes further contraints on the labour and resources shore dwellers can ac tua l l y dedicate to f i sh ing . To determine the size of th is surplus, I have deducted the tota l amount of labour required for agr icu l tu ra l production from the tota l household 1 abour supply. To determine how much time a shore dwel l ing household requires for agr icu l tu ra l production, I have m u l t i p l i e d for each crop type the cu l t i va ted area per household by the corresponding labour input per hectare, and summed up the labour inputs accross crop types (Table 12). For th i s purpose I have used data on labour inputs per hectare from a sample of 23 shore communities (Brown 1978; V e r l i a t 1978; Painter 1982; 1986), and data on the d is t r ibut ion of cu l t i va ted areas per crop type from a sample of 45 A l t i p l a n o communities (Brown 1978; Ccama 1981; Lescano et a l . 1982; Painter 1982; 1986). For comparison purposes Table 12 indicates separately the time spent on agr icu l tu ra l a c t i v i t i e s in three communities (Luquina, Camacani and Sarata) and in a sample of 20 communities (Ccama 1981: 70-1). The estimate of 1,251 hours (Table 12) for the time spent per household on agr icu l tu ra l production seems the most appropriate, because i t i s based upon a sample of 20 A l t i p l a n o communities (Ccama 1981), and because i t f a l l s within the range ca lcu lated for those communities for which data were a v a i l a b l e . However, i t does not include harvest processing a c t i v i t i e s such as the drying of beans and the freeze-drying of potatoes. The l a t t e r have to be taken into account because they are necessary for harvest conservation for future household consumption, thus an integral part of the agr icu l tu ra l production process. Using Painter's data from Sarata (1986: Table 4 and 5), I have estimated that a total of approximately 72 hours per household was used each year for harvest processing. Given that only 6.5 hours are ac tua l l y a v a i l a b l e per working day for heavy agr icu l tu ra l duties (Painter 1982: 205), a household spends about 203 mandays on agr icu l tu ra l a c t i v i t i e s : 192 mandays for f i e l d labour and 11 mandays for harvest processing. This compares favorably with estimates from various Andean communities (Brush 1977: 130; Gonzales 1984: 174). After these 6.5 hours of heavy work, a g r i c u l t u r a l i s t s can s t i l l spend 2 more hours on miscellaneous time consuming but l i g h t a c t i v i t i e s such as tool-making, the construction and maintenance of containers for crop storage, or the t ra in ing of chi ldren (Ruddle and Chesterf ie ld 1977: 122; Painter 1982: 205). As for the remaining 2 hours of labour a v a i l a b l e per day, shore dwellers normally dedicate them to the i r l i vestock (Painter 1982: 205). This labour must be accounted for as agr icu l tu ra l labour because of the ro le c a t t l e play in the agr icu l tu ra l production process. Teams of b u l l s are used for ploughing and c a t t l e dung i s an important source of f e r t i l i z e r and of f u e l . I have ca lcu lated that i t takes shore dwellers about 50 mandays per year to feed and tend thei r l i ves tock , to t ra in the i r team of b u l l s (Hickman 1963), or to reciprocate the loan of a team i f one i s not owned. Summing up the labour used in a l l the agr icu l tu ra l a c t i v i t i e s undertaken by a standard shore household y i e l d s an estimate of about 253 mandays per year. Once deducted from a supply of 700 mandays of labour, i t leaves a surplus of 447 mandays a v a i l a b l e for complementary a c t i v i t i e s such as f i sh ing . However, th is surplus i s not uniformally d istr ibuted during the year because of agr icu l ture seasonality. 143 TABLE 12: LABOUR REQUIREMENTS FOR AGRICULTURAL PRODUCTION IN VARIOUS SHORE COMMUNITIES OF LAKE TITICACA Crop Labour/ha Luquina Camacani Ccama1 s Sarata (hrs/ha) (hrs) (hrs) (hrs) (hrs) (1) (2a) (2b) (3) (4) Potatoes & Tubers 1,660 460 481 697 714 Beans & Peas 1,290 108 103 103 478 Wheat & Barley 445 77 102 254 76 Quinoa & Canihua 808 61 97 97 32 Other Crops 1,000 1 — 100 7 Total (hours) 707 784 1,251 1,306 (mandays) 112 116 190 205 Area c u l t i v a t e d (has) .61 .71 1.29 1.02 Area owned (has) 1.69 2.25 2.22 -Sources: (1) : Brown 1978; V e r l i a t 1978; Painter 1982; 1986 (2a) : Lescano et a l . 1982 (2b) : Lescano et a l . 1982 (3) : Ccama 1981 (4) : Painter 1982; 1986 144 FIGURE 16: A G R I C U L T U R A L WORK R E Q U I R E M E N T S IN A SHORE COMMUNITY OF L A K E T IT ICACA, PERU (Luquina Grande, Chucui to ; Source: Lescano 1982) Ploughing B H P I a n t i n g immCleaning B U I Harvesting TUBERS P E A S a B E A N S C E R E A L S Q U I N O A , E T C . L A B O U R 9 0 8 0 7 0 6 0 5 0 4 0 30 2 0 10 0 4 1 5 1 6 1 7 M O N T H 8 1 9 ' 10* II i 12 145 d) Seasonality of labour surplus The 447 mandays of labour surplus are a v a i l a b l e during the two slack periods of the agr icu l tu ra l cycle (Figueroa 1984; Gonzales 1984), as demonstrated by the demand curves (Figure 16) for agr icu l tu ra l labour in the community of Luquina (Plater ia) (Lescano et a l . 1982: 144) which i s very s i m i l a r to the one from the community of Sarata (Painter 1982: 243). The f i r s t one corresponds to the peak of the dry season (June to August), the second one to the ear ly part of the rainy season (December to March). To take advantage of th is labour surplus, Lake T i t ieaca fishermen may either f i sh exc lus i ve l y during the slack periods of the agr icu l tu ra l cycle (i .e. seasonal ly) , or they may f i s h year-round. In the mid 1970s, for example, 68% of a sample of 251 fishermen, f ished year-round though not d a i l y , and 32% fished seasonally for an average of 3.3 months a year (IMARPE-Puno 1976: unpublished data). Most (75%) of the seasonal fishermen fished during the rainy season (January to March) which i s often considered the most productive of the year and sometimes c a l l e d the " f i sh season", or Ch'awl 1 a  Phajsi in Aymara (Cordero 1966: 32). i - Year-round fishermen: Year-round fishermen use a large proportion of the i r household's labour surplus in f i sh ing . Using data from 1979-80, I have calculated that they spend a tota l of 396 mandays on f ish ing and on related a c t i v i t i e s : 296 mandays on f i sh ing and 100 mandays on f i s h r e t a i l i n g , c ra f t and gear maintenance (assuming that one can spend 8.5 hours a day on such a c t i v i t i e s ) . This begs the question of why year-round fishermen do not f i sh for the remaining 51 mandays, but use th is remaining surplus labour for other complementary a c t i v i t i e s . The explanation for the use of the remaining 51 mandays on non f ish ing a c t i v i t i e s comes f i r s t from the seasonality of the agr icu l tu ra l cyc le . These 146 51 mandays are a c t u a l l y a v a i l a b l e during the few weeks of the agr icu l tura l slack periods. They do not come in d r i b l e t s of a few hours at a time every so many days, but in series of days during which there i s no agr icu l tu ra l task to carry out with the exception of c a t t l e feeding, for which no more than 2 hours of labour by one adult are necessary. Even i f th i s remaining labour was a v a i l a b l e in short da i l y or weekly lapses uniformally distr ibuted along the year, I demonstrate here that i t would s t i l l not be possible for year-round fishermen to use i t to intensi fy the i r f i sh ing a c t i v i t i e s ei ther by mul t ip ly ing the i r f i sh ing t r i p s or by lengthening them. M u l t i p l y i n g the i r f i sh ing t r ips i s not possible for year-round fishermen because with 296 mandays spent f i sh ing they already operate on almost a l l the working days of the year (cf. section b above). The oldest son of the household could go f i sh ing in a second c r a f t , i f he was o ld enough, but th is would require major investments which would not be worthwhile for 51 mandays only. Lengthening f i sh ing t r i p s i s d i f f i c u l t , although i t could be done either by t r a v e l l i n g further or by mul t ip ly ing the number of nets used. T r a v e l l i n g further i s not worthwhile i f i t increases f i sh ing costs without increasing f ish ing returns. Moreover, i t i s not possible for fishermen normally operating in the l i t t o r a l zone (FEUs of type 1, 4, 6 and 10), because i t would mean trespassing into neighbouring TURF areas and exposing oneself to r e t a l i a t i o n , unless some deal was struck. If the l a t t e r t r a v e l l e d further offshore, they would have to become involved in a pelagic f ishery which requires a completely d i f ferent set of investments: a sea worthy c r a f t , g i l l n e t s of meshsizes larger than 63 mm, eventual ly an outboard engine. As for huayunaccana fishermen (type 10 FEU), towing i s too strenuous and dependent on wind and wave conditions to be carr ied out for much more than 6.6 147 hours a day. F i n a l l y , fishermen operating offshore on type 8 FEUs already have longer f i sh ing t r ips because they can stay offshore for 11.8 hrs/tr ip and spend the night in their c ra f t . Using more nets does not necessari ly enta i l an increase in f i sh ing labour, because sett ing one's nets represents a much smal ler proportion of a whole f ish ing t r i p than the rowing time. I t i s the l a t t e r which determines the length of the t r i p rather than the number of nets set, as suggested by a comparison between number of nets and time spent f ish ing for d i f ferent FEU types. Type 4 FEUs with a mean of 9.7 nets and type 6 with one of 10.3 nets have shorter t r ips (5.9 and 6.5 hrs/tr ip respect ively) than type 1 FEUs with a mean of on ly 6.9 nets (7.3 h r s / t r i p ) . i i - Seasonal fishermen: Seasonal fishermen only use a small proportion of the i r labour surplus for f i sh ing . I have ca lcu lated that they use about 120 mandays during the slack of the rainy season: 70 for f ish ing and 50 for f i sh ing - re la ted a c t i v i t i e s . This leaves them with about 327 mandays a v a i l a b l e for other complementary a c t i v i t i e s . Intensifying thei r f i sh ing e f fo r t would not a l low them to absorb th is large labour surplus, unless they became year-round fishermen. Four hypotheses could explain why seasonal fishermen do not become year-round fishermen to use the surplus labour of the i r households. Either they ac tua l l y have more land than the year-round fishermen thus a smaller surplus labour, or they have much less land thus a much greater need for cash which prevents them from saving enough to intensi fy thei r f ish ing a c t i v i t i e s , or they come from smaller households with a much smaller labour supply. F i n a l l y , the opportunity cost of thei r labour may be higher than thei r f i sh ing return because they have the necessary connections and t ra in ing to work in a 148 l u c r a t i v e and status-enhancing job in an urban or mining area. None of the f i r s t three hypotheses explains the behaviour of the seasonal fishermen. Even with twice as much land as the average year-round fisherman, seasonal fishermen would need at most 506 mandays of labour which would leave them with a surplus of 194 mandays of which f ish ing would only use 120 days. Land-poor seasonal fishermen would s t i l l be able to earn more by f i sh ing from a balsa (type 1 FEU), than by h i r ing themselves for wage labour. F i n a l l y , a household generating only 2 mandays of labour per workday would s t i l l have a supply of 450 mandays and a surplus of 197 mandays a year. Only i f i t had 1.3 times more land than the average shore household, would i t use up i t s 450 mandays of labour: 330 for agr icul ture and 120 mandays for f i sh ing . The fourth hypothesis i s the most convincing of a l l , although i t assumes that the household of the seasonal fisherman can send one of i t s members away for a whole year and s t i l l be able to carry normal agr icu l tu ra l tasks ei ther by h i r ing farmhands for a da i l y wage, or through the reciprocal exchange of labour with other households. Painter (1982) and C o l l i n s (1982) show how in s i m i l a r circumstances, the households of Sarata (Moho) are able to a l l o c a t e the i r labour supply between agr icul ture on the shores of Lake T i t ieaca , coffee or c i t rus growing in the Se lva , and trade (i .e. smuggling) a c t i v i t i e s on the A l t i p l a n o . Although carr ied out mostly by women, the l a t t e r i s for them the equivalent of seasonal f i sh ing . i i i - Synthesis: The seasonal a v a i l a b i l i t y of a labour surplus i s responsible for the involvement of year - long fishermen in other complementary a c t i v i t i e s . To use the remaining 51 mandays of surplus labour of his household, which he cannot use in f i s h i n g , a year-round fisherman may decide to migrate for a month or two to the Costa and hire themselves for wage labour on some p lantat ion , even though he could t h e o r e t i c a l l y earn more f i sh ing than 149 migrating. During his absence, his household may s t i l l use his f i sh ing c ra f t to c o l l e c t totora and l lachu to feed his c a t t l e . A r e l a t i v e may also set and ret r ieve the nets of the absent fisherman with his own, and bring them back to the l a t t e r ' s wife and chi ldren for them to pick up the f i sh s t i l l tangled in the nets. In return, the absent fisherman i s expected to set the nets of the helpful r e l a t i v e when he himself goes away. As for seasonal fishermen, i t i s l i k e l y that the opportunity cost of their household labour i s higher than the i r f i sh ing return and that i t i s economically rat ional for them to send one member of the i r household away for most of the year. Even then, they s t i l l have a surplus of labour during the slack periods of the agr i cu l tu ra l cyc le . I t i s then more prof i tab le for them to engage in seasonal f i s h i n g , than to send another household member to the Costa to hire himself for wage labour. Their f i sh ing investment i s small since they only need a few nets, which they can store or lend for the rest of the year, and a f i sh ing balsa which after 3 or 4 months of f i sh ing can s t i l l be used to c o l l e c t l lachu and totora c lose to shore. The seasonal a v a i l a b i l i t y of a labour surplus a lso determines the f ishery a shore dweller may part ic ipate i n . Fishermen with the least land are more l i k e l y to part ic ipate in the most time consuming f i sher ies . Pelagic g i l l nett ing, for example, i s quite time consuming because i t invo lves long f i sh ing t r i p s offshore, with each type 8 FEU spending 11.8 hours per f i sh ing t r i p compared with a mean of less than 7 hours for other FEU types. Most type 8 FEUs are thus found in communities where agr icul ture is less demanding because of a lack of land, or because of a lack of water as on Soto i s land . S i m i l a r l y , f i sh ing with a huayunaccana requires hours of rowing for at least a couple of adults during dayl ight hours. I t i s thus practiced more in tens ive ly during the slack of the dry 150 season, by fishermen from communities with a high population density such as V i lu rcuni (Yunguyo). The constraints considered so far may not be the only ones to af fect the practices of loca l fishermen. Membership in a TURF holding community may also constrain shore dwellers' f i sh ing investment. 6.2.2. Constraints on f i sh ing investments A v a i l a b l e evidence indicates that community membership resu l ts in few di rect or ind i rect constraints on ind iv idual f i sh ing investments. There i s l i t t l e evidence of d i rect constraints. For example, fishermen are not normally prevented from introducing a new f i sh ing technology in the i r own communities. Even i f the f i r s t boat brought to the community of Cachi Pucara (I lave) was destroyed by angry balsa users ( C a s t i l l o 1978), technical innovation has general ly been wel l accepted around Lake T i t ieaca , once i t s pa r t i cu la r merits had been demonstrated (Appendix B). There i s no evidence either that net theft i s a form of communal control of indiv idual f i sh ing investments. Although Lake T i t ieaca fishermen lose some of the i r nets r e l a t i v e l y frequently (11% of them lose at leas t one net a year), i t i s unclear what proportion of these i s a c t u a l l y s to len. Fishermen with more nets may be more l i k e l y to set some of them in f ish ing areas in dispute and thus more l i k e l y to have them sto len. Yet I could f ind no indicat ion that fishermen steal one another's nets to keep numbers down. More ind i rect forms of constraint on f i sh ing investments may r e s u l t from shore dwellers' hunger for land. Those short of the minimum requirement of 1.3 has of c u l t i v a b l e land, are more l i k e l y to invest into land even of marginal productiv ity when i t becomes a v a i l a b l e , than in f ish ing c ra f t or gear. However, th is constraint i s u n l i k e l y to be a major one simply because only a small amount of land i s ever a v a i l a b l e for sa le . About 90% of the 151 to to ra les , for example, were obtained through inheritance rather than purchase. The sponsorship of f iestas by which many comuneros gain prestige and status within thei r own community and which requires very large expenditures by loca l standards (Equipo J u l i 1974), are also l i k e l y to c u r t a i l investments in f i sh ing gear, even i f mechanisms have been devised to cope with such large expenses (Brown 1978). As for technological knowledge, although i t i s not readi ly a v a i l a b l e , there i s no control on th is form of investment. Members of TURF holding communities can eas i l y learn to f i s h from a r e l a t i v e . Most fishermen (72.3%) have learned to f i s h from a male member of the i r community or from one of the community of the i r wives, either from a male kin (63.2%), or from an in - law or af f ine (9.2%), few (17%) have learned by themselves. This control on crew formation and on t ra in ing does reinforce ex ist ing TURFs, however. The Uros fishermen from Lago Poopo in B o l i v i a , for example, purposely r e s t r i c t the dissemination of technical information to one's kins as a conscious strategy against potential competition from the i r Aymara neighbours (Horn 1981). 6.3. DISCUSSION OF ALTERNATIVE CONSTRAINTS I t has been argued in the l i t e r a t u r e that various factors not re lated to fishermen's membership in shore communities are responsible for the underfishing of Lake T i t i caca f i s h resources. I show that these factors do not prevent loca l f i sher ies from reaching thei r bioeconomic equi l ibr ium, either because they do not occur in the present s i tuat ion , or because, even i f they d id , they would not prevent the bioeconomic equi l ibr ium of loca l f i sher ies from being reached. 152 6.3.1. Cul tural constraints The secrecy over f ish ing locations and techniques by which fishermen may try to protect thei r trade and resources may a lso be reinforced by a prevalent social prejudice against f ish ing which may ref ra in successful fishermen from reinvest ing the i r p ro f i t s into f i sh ing . Some may instead purchase more c a t t l e , or more land i f any i s a v a i l a b l e , or even a t r i c y c l e which can be used for the transport of goods and passengers in J u l i a c a or in Puno. F i n a l l y , some may use thei r f i sh ing income for consumption rather than production. However, i t appears that the prejudice against fishermen and f ish ing which was prevalent on the A l t i p l a n o a few centuries ago and up to the la te 1950's (Ve l la rd 1957: 58-9; Matos 1964: 137) has now withered away. Fishermen are often considered to be better off than other shore-dwel 1 ers (Pacori 1976), and thei r cash-earning a b i l i t i e s are sometimes looked upon with envy. I t has a lso been hypothesized that other cu l tu ra l factors, such as a prohibi t ion of f i sh ing for women, and certain dietary preferences may constrain f i sh ing e f fo r t . However, neither occurs on the A l t i p l a n o . Women do part ic ipate in f i sh ing a c t i v i t i e s , whether wading in the flooded areas with a push net, or rowing a boat while thei r husband sets his nets. A strong enough woman may even row one of the two boats t rawl ing a huayunaccana. I could not f ind any evidence of f i s h avoidance either . Even the aquatic toads (Telmatobius) found in the lake are occasional ly consumed for medicinal purposes. Moreover, dietary preferences would only lower f i s h prices and the explo i tat ion l e v e l at which the bioeconomic equi l ibr ium of the corresponding f ishery is reached. 6.3.2. Economic constraints Orlove (1979) has argued that economic d is incent ives , such as inef fect i ve f i sh ing techniques, the lack of wel l developed and accessible urban markets, 153 r e l a t i v e l y low f i s h prices compared to other food items (which could r e s u l t from loca l dietary preferences), or the u n a v a i l a b i l i t y of cheap cred i t , may be responsible for underfishing on Lake T i t i caca . However, these factors would only be responsible for higher f ish ing costs or for lower f ish ing returns. The bioeconomic equi l ibr ium of the f ishery considered would s t i l l be reached, although at a lower l e v e l of f ish ing e f fo r t (Anderson 1978). In any case, most of these factors do not apply to the s i tuat ion encountered on the Peruvian A l t i p l a n o . Fishermen have access by truck or boat to a large number of weekly l oca l markets (Orlove 1986). Their catches are also exported to most of the large urban markets of the southern Andes (Av i la et a l . 1985). During the drought of 1983, some f i s h vendors even t r ied to export i t as far as Lima. Credit i s r e l a t i v e l y inexpensive on the A l t i p l a n o , although some fishermen complain that i t may be hard to obtain at times. Most (77%) of those who use credi t re ly upon thei r b io log ica l and f i c t i v e kins for i t , while some (5%) use credi t from the Agrarian Bank (Banco Agrario del Peru) normally granted for the purchase of c a t t l e . Many fishermen re ly upon the i r c a t t l e as a source of cash (40%) while others use the i r income from wage labour on the Costa. Furthermore, a young man may be able to purchase his own c ra f t and gear he has been given by his parents and godparents, as he grew up. In exchange for th is form of c red i t , he i s expected to perform menial tasks when s t i l l a c h i l d (Michaud 1973; Lambert 1977) and to help his parents in thei r o ld age (Co l l ins 1982). Therefore, Lake T i t icaca shore dwellers have access to a large enough market to s e l l the i r catch, and to the technical and f inanc ia l a b i l i t y to increase thei r f i sh ing e f fo r t 154 6.3.3. B io log ica l constraints Loubens and Sarmiento (1986) have argued that, because 0^ agassii reaches sexual maturity at smaller sizes than those at which i t i s usual ly captured, i t i s u n l i k e l y to become overfished with the exist ing f i sh ing technology and in par t i cu la r with the meshsizes loca l fishermen use at present. However, thei r f indings only explain why a steady increase of f i sh ing e f fo r t may not lead to a drast ic reduction in the numbers of 0. agassii caught. Such an increase in f i sh ing e f fo r t may ac tua l l y lead to a reduction of the average size and value of the 0. agassii caught which fishermen would probably try to compensate for by reducing the meshsize of the i r nets. These authors only indicate that b io log ica l overf ishing of 0. agassii has not occured yet , but they do not indicate whether th is has lead to the capture of economic rents. The capture of undersized f i s h has been prohibited in a handful of communities. If i t were widely enforced, i t could prevent b io log ica l overf ishing by guaranteeing that most f i s h are able to spawn at least once before being caught. I t would maintain a high b io log ica l product iv i ty for the f ishery considered, thus conserve i t s resources. But i t would not prevent the bioeconomic equi l ibr ium of th is f ishery from being reached. To sum up, I have demonstrated here that Lake T i t ieaca fishermen are able to capture substantial economic rents. I have suggested that these rents are l i k e l y to have persisted over the l a s t few years. And I have traced the or ig in of these rents to shore dwellers' enforcement of communal TURFs, and to the obl igat ions enta i led by fishermen's membership in shore communities (Figure 17). Chief among the constraints on the labour they can dedicate to f i sh ing i s that of keeping agr icu l tu ra l land in production. 155 CHAPTER 7 CONCLUSIONS AND RECOMMENDATIONS 7.1 CONCLUSIONS In try ing to determine whether TURFs could be used for the management of s m a l l - s c a l e f i s h e r i e s , I have demonstrated f i r s t , that TURFs ex is t along the shores of Lake T i t i caca ; second, that they are a c t i v e l y enforced by shore dwel lers ; t h i r d , that substantial economic rents are captured by loca l fishermen; and fourth, that the capture of these rents i s a d i rect consequence of fishermen's membership in TURF holding communities and of TURF holders' control of f ish ing access. 7.1.1. Ex is tence of TURFs I have demonstrated the existence of Lake T i t icaca TURFs with h i s t o r i c a l evidence and the acknowledgment by at least 90% of the fishermen interviewed that they r e s t r i c t the i r operations to the f ish ing area of the i r own community. I have a lso shown that fishermen agreed on the dimensions and customary rules governing these TURFs, and on the conditions under which outsiders could gain access to the corresponding resources. F i n a l l y , I have found evidence of the existence of such TURFs for v i r t u a l l y a l l units within a sample of 80% of the shore communities, thus confirming the widespread d is t r ibut ion of TURFs along the shorel ine. Since TURFs extend the t e r r e s t r i a l t e r r i t o r i e s of the shore communities to the aquatic environment as far as several hundred meters off the outer edge of the communal totora beds, they include v i r t u a l l y the whole shore area of the lake, most of i t s l i t t o r a l area, and even part of i t s pelagic area. 156 7.1.2. Enforcement of TURFs I have documented how Lake T i t ieaca shore dwellers take advantage of a number of legal arguments to enforce the i r i l l e g a l TURFs, and how they use threats and overt or covert v io lence, including net theft , to repel outsiders. Chief among these arguments i s the control they hold over access to the shoreline through semi-private ownership of landing s i tes and navigation channels, communal ownership of docking f a c i l i t i e s , and physical control of the roads and t r a i l s leading to the shore. Communal r ights over nearshore totora and l lachu beds are a lso a legal argument TURF holders frequently use to repel outsiders. To keep the l a t t e r at bay, TURF holders frequently accuse them of damaging or s tea l ing the reeds. F i n a l l y , TURF holders re l y upon the i r Beach Sargents and f ish ing associations, both o f f i c i a l l y recognized, to expel ! outsiders from thei r f i sh ing grounds under the guise of minimizing c o n f l i c t s and net theft by outsiders. 7.1.3. Capture of f i sh ing rents I have demonstrated that Lake T i t ieaca fishermen are able to capture substantial f i sh ing rents. These rents are proportional to the difference between the return to labour from f ishing and the opportunity cost of th is labour, which i s measured by the return to labour from a l te rnat i ve a c t i v i t i e s a v a i l a b l e to shore dwel lers. Because the a v a i l a b i l i t y of such a c t i v i t i e s depends on shore dwellers' education, connections and community of o r i g i n , the difference between the returns to f ish ing labour and the opportunity cost of th is labour ranges from 51 to 106%. Assuming a d i f ferent opportunity cost of labour for f ish ing and f i sh ing - re la ted a c t i v i t i e s on the grounds that the l a t t e r are undertaken when there i s l i t t l e e lse to do, increases th is range to 90-180%. Limited evidence on the s t a b i l i t y of f i sh ing y i e l d s and on a slower rate of increase of f i s h prices r e l a t i v e to input costs suggests rents' 157 persistence over the long term, thus confirming that loca l f i sher ies have not reached thei r bioeconomic equi l ibr ium. 7.1.4. Community membership and the capture of f i sh ing rents F i n a l l y , I have demonstrated that the rents captured by Lake T i t icaca fishermen resu l t from the obl igat ions enta i led by the i r membership in TURF holding communities, and from the exclusion of inland dwellers from loca l f i sher ies (Figure 17). Communal obl igat ions are counterparts to one's membership in a community. Their undertaking i s necessary to guarantee one's access to communal resources, among which the aquatic resources protected by loca l TURFs. They include such obl igat ions as part ic ipat ion in communal projects and fes t i ve celebrat ions, and the undertaking of administrative and ceremonial r e s p o n s i b i l i t i e s which reduce the annual supply of household labour from 940 to 700 mandays approximately. But chief among these obl igat ions i s the necessity for shore dwellers to keep thei r agr i cu l tu ra l land in production. This requires about 353 mandays per year of the i r household labour thus l i m i t i n g to 447 mandays per year the amount of household labour a v a i l a b l e for f i sh ing and f i sh ing - re la ted a c t i v i t i e s . I have a lso shown that a l te rnat i ve explanations proposed for underfishing on Lake T i t icaca are either not relevant to present circumstances, or that i f they were, they would not prevent loca l f i sher ies from reaching the i r bioeconomic equi l ibr ium anyway. This demonstrates the fundamental contribution of the communal structure to the effectiveness of TURFs, and the need for an interpretat ion of th is effectiveness to determine how TURFs could be used for the management of loca l f i s h e r i e s . 158 7.2. INTERPRETATION OF THE EFFECTIVENESS OF TURFS To be e f fec t i ve in c o n t r o l l i n g f ish ing e f fo r t TURFs should incorporate four basic r ights : the r ight of exclusion, the r ight to extract benefits from the use of the resources within the te r r i to r y , the r ight to determine the amount and kind of use of these resources, and the r ight to future returns from these resources (Christy 1982). The effectiveness of Lake T i t ieaca TURFs can be traced to the i r sat is fact ion of these requirements. Lake T i t ieaca TURF holders consider that they have the r ight to exclude outsiders, as demonstrated by the i r enforcement a c t i v i t i e s . Only shore dwellers are able to f i sh in Lake T i t ieaca , because they belong to the TURF holding communities. They a lso have the r ight to extract benefits from the use of the resources within the te r r i to r y , thanks to thei r r ights of d i rect use and of ind i rect economic gain. The former e n t i t l e s them to go f i sh ing at w i l l , as long as they comply with the obl igat ions entai led by thei r membership within a TURF holding community. The l a t t e r al lows them to receive a compensation from outsiders who are allowed to operate within the TURF area, or to keep the i r f i sh ing c ra f t on the community's shorel ine. TURF holders a lso have the r ight to future returns from the resources within the i r t e r r i t o r y , because the TURFs they hold are permanent, although threatened by o f f i c i a l opposition. F i n a l l y , thei r control r ight al lows them to determine the amount and kind of use of the resources within the TURF ter r i to ry . Lake T i t ieaca shore communities, l i k e most Andean communities (Mayer 1985), re ly heavi ly on the i r r ight of control over the explo i tat ion of t e r r e s t r i a l resources (such as communal pasture and f a l l o w lands) and lakeplants beds. However, I found that very few a c t u a l l y control the exp lo i tat ion of thei r f i s h resources, even to a minimal extent. 159 Chief among the addit ional conditions which a TURF system should sat i s fy to be e f fec t i ve i s that harvest outside a TURF area should not af fect the value of the resources within i t (Christy 1982). This means that the f i s h y i e l d within the TURF-control1 ed zone should not be reduced to such a l e v e l that TURF holders would be unable to capture any rents, because an excessively high rate of exp lo i tat ion would be applied either to the portion of a f i s h stock outside th is zone, or to that within the TURF area of a small number of communities. Excessive harvest of adult trout migrating to spawn up the six major t r ibutar ies of Lake T i t i c a c a , for example, an ih i la ted the pos i t i ve effects of TURF-associated constraints on f ish ing e f fo r t , thus contributing la rge ly to the co l lapse of the trout canning industry in the la te 1960s (Appendix B). S i m i l a r l y , heavy f ish ing of the umanto, the suche and the boga outside the TURF-control 1 ed l i t t o r a l zone i s l i k e l y to have contributed to the i r depletion. Fishermen overexploited these valuable species with an increase in f i sh ing pressure resu l t ing from the introduction of new f i sh ing technology (Appendix B), at the same time as these resources had to suffer heavy predation from the rainbow trout. A s i m i l a r s i tuat ion could a lso ar ise in the future for the i sp i i f a large urban-based commercial f ishery were to develop on the lake. Because the i s p i spends most of i t s l i f e cyc le outside of the TURF-protected l i t t o r a l area, ex ist ing TURFs and the constraints resu l t ing from fishermen's membership in TURF holding communities would not af fect f ish ing e f fo r t within such a f ishery. The effectiveness of Lake T i t icaca TURFs in c o n t r o l l i n g f ish ing e f fo r t i s l imi ted , as demonstrated by the i r i n a b i l i t y to protect some valuable species from depletion. This effectiveness i s determined to a large extent by fishermen's membership in TURF holding communities. Formally recognized TURFs 160 would thus not be very e f fec t i ve i f fishermen did not have to respect the constraints resu l t ing from thei r membership in shore communities. Granting TURFs a formal status would not address these l imi ta t ions and would thus be insu f f i c ien t by i t s e l f to take advantage of TURFs l imi ted effect iveness. Instead, ex ist ing TURFs should become a component of a f i sher ies management strategy. Such a strategy would c a p i t a l i z e upon the advantages of t rad i t iona l f i sher ies management practices (Alexander 1980) and contribute to the emergence of a new model of resource management which various authors have advocated (Berkes 1985; Lamson and Cohen 1984; Smith and Panayotou 1984; Pinkerton 1986). The recent evolut ion of planning and development theories suggests a number of p r inc ip les for the formulation of such a strategy. 7.3. TURFS AND THE MANAGEMENT OF LAKE TITICACA FISHERIES A TURF-based management strategy for Lake T i t ieaca s m a l l - s c a l e f i sher ies would c a p i t a l i z e on the advantages of loca l TURFs. These advantages include (cf. section 1.3.4.) the effectiveness of TURFs in c o n t r o l l i n g f i sh ing e f fo r t , their potential for the promotion of equity in the d is t r ibut ion of f i sher ies ' benef i ts , of administrative ef f ic iency and accountabi l i ty in the resource management process, and of a higher f ish ing y i e l d . By reviewing the advantages a TURF-based management strategy would e n t a i l , I indicate-here how congruent with the recommendations from planning and development theories the implementation of such a strategy would be. 7.3.1. TURFs and planning theory The management of natural resources i s a form of planning since both can be defined by the i r central concern with the linkage between knowledge and action in the publ ic domain (Friedmann and Hudson 1974). A TURF-based 161 f i sher ies management strategy should thus fo l low the pr inc ip les of planning theory which has evolved toward decentral izat ion and self-management in the l a s t two decades (Arnstein 1969; Grabow and Heskin 1973; Friedmann 1980). Although self-management has received some scathing c r i t i c i sms for i t s p o l i t i c a l naivety and pract ica l i n f e a s i b i l i t y (Etzioni 1973; F lyvbjerg and Petersen 1982; Hebbert 1982), i t has been argued that planners should incorporate i t s p r inc ip les into the i r professional practice (Bolan 1980; Forester 1982; Kemp 1982) and use i t whenever appropriate (Hudson 1979). The implementation of a TURF-based management strategy would contribute to the adoption of the p r inc ip les of a decentralized approach to f i sher ies management in three major ways. F i r s t , i t would a l low the interests of s m a l l -scale and subsistence fishermen to be better represented in a system which has lent i t s e l f to accusations of i n s e n s i t i v i t y to the needs of the l a t t e r (Alexander 1975). By contributing to the incorporation of the plea of the less favoured resource users into the management process, i t would make th is process less bureaucratic (Maziotti 1974; Davidoff 1978). And providing no small group of fishermen i s allowed to monopolize TURF benefits and to behave as sealords (Christy 1982), i t would promote social j us t i ce . Second, i t would give fishermen long term r ights over the resources in the TURF area of thei r community, which would entice them in part ic ipat ing to the gathering and sharing of valuable information without which no management i s possible. In addit ion, the control r ights of TURF holders over f i sh ing a c t i v i t i e s within the i r TURF area would a l low them to become act ive part ic ipants in the resource management process. A TURF-based management strategy would thus foster a greater involvement of resource users in the information-generating and decision-making processes by al lowing them to become e f fec t i ve and autonomous part ic ipants in the planning process (Arnstein 162 1969), thus contributing to the decentral izat ion and democratization of resource management and to the descaling of social l i f e (Friedmann 1974). Furthermore, TURFs would lead to the combination of fishermen's personal knowledge (i .e. knowledge resu l t ing from subjective experience) with sc ient i s ts ' processed knowledge (i .e. experts' knowledge resu l t ing from factual observation and objective analys is ) , a key objective of s e l f -management ( Ib id . ) . And t h i r d , i t would promote the deregulation and s e l f - c o n t r o l of s m a l l -scale f i she r ies , which would s h i f t the burden of the enforcement of TURFs and of the implementation of f ish ing regulations from administrative agencies to resource users themselves. This would not only reduce bureaucratic ine f f i c ienc ies and administrative costs, but i t would also enhance autonomy and indiv idual freedom (Kemp 1982; Forester 1982) by g iv ing TURF holders control over thei r means of production (Rosenvallon 1976; Beauregard 1978), and by transforming the regulatory ro le of administrative agencies into a new ro le much more supportive of TURF holding communities. In addition to implementing the p r inc ip les of decentra l i zat ion , the adoption of a TURF-based management strategy would a lso contribute to the implementation of the pr inc ip les of adaptive resource management (Walters 1986). By del ineating TURF areas as discrete management uni ts , i t would al low for the undertaking of experimentation within these areas, which would generate useful s c i e n t i f i c information. Different e f fo r t and catch l e v e l s may be purposely adopted by d i f ferent TURF holding communities, for example, to generate more information on the population dynamics of the i r f i s h resources. The resu l t ing s c i e n t i f i c information would thus lead to improved estimates of the b io log ica l parameters of the resources harvested, to a better adaptation of harvest rates to the production potential of these resources and eventual ly 163 to higher f i s h y i e l d s . 7.3.2. TURFs and development theory Because the ultimate purpose of f i sher ies management is to contribute to socio-economic development by increasing the u t i l i t y that society derives from i t s resources, the pr inc ip les of development theory should a lso guide the formulation of a f i sher ies management strategy. In the l a s t decade, development theory has evolved away from a broad and funct iona l l y oriented framework, toward a l o c a l i z e d and t e r r i t o r i a l l y oriented one (Friedmann and Weaver 1979; Stohr and Taylor 1981). Although the t e r r i t o r i a l model of development has been strongly c r i t i c i z e d for i t s economic and p o l i t i c a l i n f e a s i b i l i t y (Flyvbjerg and Petersen 1982; Hebbert 1982) and for i t s adoption of inappropriate assumptions (Gore 1984), i t s recognition of the importance of a s e l f - r e l i a n t approach to development i s gaining wider currency (Stohr and Taylor 1981) and recommendations have been made to modify widely used development strategies accordingly (Cremer et a l . 1984). A TURF-based f i sher ies management strategy would contribute to t e r r i t o r i a l development in three major ways. F i r s t , by reducing or e l iminat ing overf ishing and by improving data c o l l e c t i o n and e f fo r t cont ro l , i t would increase the contribution of f i sher ies resources to the loca l economy. This would help in meeting the needs of the loca l population and promoting the d i v e r s i f i c a t i o n of the loca l economy through the use of loca l resources, both of which are key objectives for t e r r i t o r i a l development (Weaver 1984). In an area l i k e the Peruvian A l t i p l a n o , where few a l te rnat i ve resources are a v a i l a b l e l o c a l l y , these objectives are of pa r t i cu la r relevance. Second, by i n s t i t u t i o n a l i z i n g a cooperative, l o c a l l y based approach to f i sher ies management (Christy 1982; Panayotou 1982), i t would foster communal decision-making not only at the implementation stage, but also at the 164 conception stage where s ign i f i cant orientations for community development are selected (Rondinel l i 1979). I t would thus promote w i l l f u l community action within TURF-holding communities instead of exogenous motivation, and cooperative instead of competitive behaviour, both of which are also key objectives for t e r r i t o r i a l development (Stohr 1981). And t h i r d , by conferring the o f f i c i a l status of TURF holding group to exist ing communities of shore dwel lers , i t would contribute to thei r strengthening as social ins t i tu t ions on which t e r r i t o r i a l development could be based over the long term. Given that these communities are loca l membership organizations according to the def in i t ion provided by Esman and Uphoff (1984) and given that the l a t t e r are economically r a t i o n a l , administrat ively e f f i c i e n t and p o l i t i c a l l y wise for development purposes (Ibid.) thei r strengthening would be a p a r t i c u l a r l y worthwhile object ive. 7.3.3. Recommendations for a TURF-based f isher ies management strategy Presently , f i sher ies management in Peru involves the IMARPE, the MIPE and the Coast Guard (Figure 17-A). Theoret ical ly the IMARPE gathers and processes the information required to formulate s c i e n t i f i c recommendations to the MIPE. The MIPE is responsible for the formulation of such regulations as f ish ing seasons, minimum sizes or harvest rates, and the c o l l e c t i o n of catch s t a t i s t i c s at landing s i tes and markets. F i n a l l y , the Coast Guard i s responsible for the enforcement of the regulations promulgated by the MIPE and for the c o l l e c t i o n of s t a t i s t i c s on the f i sh ing f l e e t . In pract ice, however, neither of the three agencies has the expertise or budgetary means to carry out i t s reponsibi 1 i t i e s . Data c o l l e c t i o n i s poor (Appendix C), the formulation of regulations is tentat ive and thei r enforcement i s arbi t rary . To c a p i t a l i z e on the effectiveness of TURFs, a f ishery management 165 strategy should strengthen the i r most important dimensions. For th is purpose o f f i c i a l recognition of customary TURFs could be granted to shore communities in such a way as to reinforce the necessary r ights of exclusive d i rect use, or ind i rect economic gain and of cont ro l . By entrusting only those r ights which are necessary for the promotion of economic e f f i c iency , the state could retain ownership over the aquatic te r r i to ry of a TURF, as in the case of the t e r r e s t r i a l te r r i to ry of the o f f i c i a l l y recognized Peasant Communities. The formal recognition of ex ist ing TURFs would imply the recognition of shore dwellers' r ight of exclusion, e i ther through the confirmation of TURFs physical and human dimensions, or thei r re -de f in i t i on in ways acceptable to TURF holding communities. I t would automatically reinforce the i r r ight of direct use, and that of ind i rect economic gain. However, the l a t t e r may have to be reconfirmed as a c o l l e c t i v e r ight , so that the benefits of loca l f i sher ies accrue to a shore community as a whole, rather than to a few of i t s members. As for the communal r ight of c o n t r o l , i t would have to be reinforced so that TURF holding communities could implement loca l regulations for the conservation of the i r aquatic resources. F i n a l l y , these TURFs would have to be granted to the i r holders for a long enough period so that they would be guaranteed that by r e s t r i c t i n g the i r own f i sh ing e f fo r t , they would be invest ing in the resources of thei r TURFs, for the i r own future benefit , or that of the i r heirs. Moreover, TURFs should be made non-transferrable, so that a shore community may not r isk los ing i t s f ish ing te r r i to ry to outside pr ivate interests or to a small group of sealords. The creation of federations of TURF holding communities would be required to deal with the problem of use outside of TURF areas, and of the a l l o c a t i o n of catch quotas for each TURF holding community for the more heavi ly exploited species. The ro le of such federative organizations would be to coordinate 166 FIGURE 17: ORGANIZAT ION OF F I S H E R I E S M A N A G E M E N T ON L A K E T IT ICACA. A: EXISTING S Y S T E M 1MARPE-Puno MIPE-Puno Reco mm en-da tions C O A S T -GUARD, Puno "—I R egistration ta x e s\ I I Data collection -4 » Fi shing regula tion s I I i I st En force ment I of r e gul a Hons i I I I k - - » F I S H E R M E N Control of\ fishing I activities^. S H O R E C O M M U N I T I E S B-. T U R F - B A S E D A L T E R N A T I V E IMARPE-Puno i MIPE-Puno "* *-l I L J Catch statistics I COAST-GUARD, Puno an~ Registration taxes Recommendations I Catch I I statistics} I I I I l L Catch st atistics R E G I O N A L F E D E R A T I O N S , Recommen-ida tions S H O R E C O M M U N I T I E S A. | 1 * T U R F s Catch statistics I i Control of fishing i activities FI S H E R M E N management practices between shore communities sharing transboundary stocks. I t would have to be organized in a way sui table for the management of each transboundary stock, with a d i v i s ion for each major ecological zone of the Lake. I t would a lso have to be sui table to foster intercommunity negotiation and to keep transaction costs low, with subdivisions fo l lowing the model of the multicommunal associations already functionning for development purposes on the A l t i p l a n o (Caceda 1984). Under a TURF-based management strategy (Figure 17-B), fishermen would play a much more prominent part in the c o l l e c t i o n of data, the formulation of regulations and thei r enforcement. Given the impact of environmental factors p a r t i c u l a r l y changes in the lake l e v e l , on loca l f i s h resources, harvest l e v e l s w i l l have to be adjusted. This, and the need to negotiate a share of the harvest from transboundary stocks within the federation would require fishermen to gather information on the i r own catches, for which the technical assistance of loca l sc ient is ts would be usefu l . In a TURF-based management strategy, the IMARPE would s t i l l have a major ro le to play but instead of formulating recommendations on harvest rates to the MIPE, i t would make them d i rec t l y to the fishermen themselves, while the MIPE would have to provide the l a t t e r with the needed assessment for the se lect ion of appropriate forms of e f for t control and harvest regulat ion. F i n a l l y , the Coast Guard would remain responsible for boat reg is t rat ion , for the mit igation of c o n f l i c t s between communities sharing transboundary stocks and for the adjudication of t e r r i t o r i a l disputes. 168 BIBLIOGAPHY. ACHESON, J.M. 1975. "The lobster f i e f s : Economic and ecological ef fects of t e r r i t o r i a l i t y in the Maine lobster industry." Human Ecology 3(3):183-207. 1979. " V a r i a t i o n s in t r a d i t i o n a l inshore f i s h i n g r i g h t s in Maine lobstering communities," in North A t l a n t i c Maritime Cultures, edited by R. Andersen, pp. 253-276. 1981. "Anthropology of f i s h i n g . " Ann. Rev. Anthro. 10:275-316. ALEXANDER, P. 1975. "Do f i sher ies experts aid f i sher ies development? The case of Sr i Lanka." Mar i t . Stud. Manag. 3 : 5 - 1 1 . 1977. "Sea tenure in southern S r i Lanka." Ethnology 16(3):231-53. 1980. "Customary law and the e v o l u t i o n of coas ta l zone management." ICLARM Newslet te r . 3 (2) :8 -9 . ALFARO, R.; E. BUSTAMANTE; J . TORRES; H. TREVINO; W. WURTSBAUGH. 1982. "La pesca en el Lago T i t ieaca : Presente y fu tu re" Informe Final Proyecto FAO-PER/76/022. ALLAN, W.; M. GLUCKMAN; D.V. PETERS; C.G. TRAPNELL. 1948. Land hoi ding and  1 and usage among the Plateau Tonga of Mozabuka D i s t r i c t . London: Rhodes Livingstone Inst i tu te , Oxford University Press. AMAT Y LEON, C. 1981. La desigual dad in te r io r en el_ Peru. Lima: Centro de Investigacion de l a Universidad del Pac i f i co . ANDERSEN, R. 1979. "Publ ic and pr ivate access management in Newfoundland f ish ings ," in North A t l a n t i c Maritime Cultures, edited by R. Andersen, pp. 299-336. j 1984. "Bermudian handl ine f i s h i n g i n the s loop e r a : A fisherman's account," in The f i sh ing culture of the world, edited by B. Gunda, pp 777-802. Budapest: Akacfemial Kiadol , e d i t o r . 1979. North A t l a n t i c Mar i t ime C u l t u r e s : Anthropological Essays on Changing Adaptations. The Hague: Mouton. ; G. STILES. 1973. "Resources management and s p a t i a l compet i t ion in Newfoundland f i sh ing , " in Seafarer and Community, edited by P.H. Fr icke, pp. 44-56. New Jersey: Rowan and L i t t l e f i e l d . ANDERSON, L.G. 1977. The economics of f i sher ies management. John Hopkins University Press. ANTUNEZ DE MAYOLO, S.E. 1981. La Nutricion en el Antiguo Peru. Lima: Banco Central de Reserva del Peru, Of ic ina Numismatica. 169 APPLEBY, G. 1978. "Exploi tat ion and i t s aftermath: The spatioeconomic evolut ion of the regional marketing system in Highland Puno, Peru." Unpublished Ph.D Dissertat ion, Anthropology, Stanford Universi ty . ARAMBURU, C.E.; A. PONCE A. 1983. F a m i l i a y t raba jo en e l Peru r u r a l . Lima: Inst i tuto Andino de Estudios en Poblacion y DesarroTTo. ARNSTEIN, S. 1969. "A ladder of c i t i zen part ic ipat ion." J . Amer. I ns t i t . P lan . 35:216-224. ARZEL, P. 1984. "Traditional management of sea weeds in the d i s t r i c t of Leon." FAO F i sh . Tech. Paper No 249. ASADA, Y.; Y. HIRASAWA; F. NAGASAKI. 1983. "F ishery management i n Japan." FAO F i s h . Tech. Paper. No 238. AVALOS D. M., R. 1951. "Changements c u l t u r e l s dans les i l e s du Lac T i t ieaca : L'organisation soc ia le dans T i l e de Taquile." Travaux de VlFEA, 3:40-50 and 74-87, L ima -Par i s . AVILA, L.; M. CHARAJA; D. LEVIEIL; P. PAZ. 1985. "Importancia socio-economica de l a pesca en l a Bahia de Puno," in Polucion en l a Bahia Interna de Puno, edited by P. Morales, D. Levy and T. Northcote. Vancouver: UBC, Westwater. BAILEY, C. 1982. "Natural resource management:. A basis for the organization of s m a l l - s c a l e fishermen." Rural Dev. Par t i c ipa t ion , Winter issue:19-22. BANDELIER, A.F. 1910. The Is!ands of T i t ieaca and Koat i . New York: The Hispanic Society of America. BASTIEN, J.W. 1978. Mountain of the condor: Metaphor and r i t u a l in an Andean  ayl 1 u. St P a u l , Minnesota: West Publ ishing Co. 1985. "Quol 1 ahuaya-Andean body concepts: A topograph ica l -hydraulic model of physiology." American Anthropologist 87(3):595-611 BAZIGOS, G. 1974a. "Design of f i sher ies s t a t i s t i c a l surveys: Inland waters." FAO F i s h . Tech. Paper No 133. 1974j). "App l ied f i s h e r i e s s t a t i s t i c s . " FAO F i s h . Tech. Pap. Nol35. BEAUREGARD, R.A. 1980. "Teaching planning theory: Dilemmas beyond the rat ional model." B u l l . Assoc. Col leg . Sch. P lan . 18(1). BECKER, D.G. 1983. The new bourgeoisie and the l i m i t s of dependency: Mining,  c lass and power in "revolutionary" Peru. Princeton University Press. BERKES, F. 1977. "Fishery resource use in a subarctic indian community." Human Ecol ogy 5(4):289-307. 1985. "The common property resource problem and the c r e a t i o n of l imi ted property r ights." Human Ecology 13(?):187-208. 170 1985. "Fishermen and the tragedy of the commons." Environmental Conservation 12:199-206. BOLAN, R.S. 1980. "The pract i t ioner as theor ist : The phenomenology of the professional episode." Amer. PI an. Assoc. (July):261-273. BOLTON, R. 1977. "The Qo l la marriage process," in Andean kinship and  marriage, edited by R. Bolton and E. Mayer, pp 217-Z39. BOLTON, R.; E. MAYER. Editors. 1977. Andean Kinship and Marriage. American Anthrop. Association. Special Publ icat ion No. 7. BOULANGE, B.; J.E. AQUIZE. 1981. "Morphologie, hydrographie et c l imatologie du Lac T i t icaca et de son bassin versant." Rev. Hydrobiol. Trop. 14:269-287. BROWMAN, D. 1981. "New l i g h t on andean Tiwanaku." American Sc ient i s t 69(4):408-419. BROWN, P.F. 1978. "Fuerza por fuerza: Ecology and culture change among the Aymara of Southern Peru." Unpublished Ph.D d isser tat ion , Anthropology, University of Colorado. BRUSH, S. 1977. "Mountain, f i e l d and family : The human ecology of an Andean va l ley . " Ph i lade lph ia : Univ. of Pennsylvania Press. ; D. GUILLET. 1985. " S m a l l - s c a l e agro -pastora l product ion i n the central Andes." Mountain Research and Development 5( l ) :19-30. BUECHLER, H.C.; J.M. BUECHLER. 1971. The B o l i v i a n Aymara. New York: Hol t , Rinehart and Winston. BUSE DE LA GUERRA, H. 1981. Act iv idad Pesquera. Lima: Documenta, Min is ter io de Pesqueria. BUSTAMANTE, E.; H. TREVINO. 1976. Descripcion de las Pesquerias en el Lago  T i t i caca . Unpublished manuscript. IMARPE-Puno. CALANCHO M., E. 1984. "Economia de los Uros del T i t icaca." Tesis para obtar el t i t u l o de Ingeniero Economista. Puno: Univ. Nacional del A l t i p l a n o . CAMPBELL, H.J. 1981. "Investigations of hydrobiological resources for d i rect human consumption in continental and marine waters: Summary of project a c t i v i t i e s , achievements and recommendations." Interim report. Lima: Project FAO-PER/76/022. CAMPBELL, D. 1984. "Individual transferable catch quotas: Their r o l e , use and appl icat ion." Fishery Report No 11. Department of Primary Production, Austral ian Government Publ icat ion Services. CARLANDER, K.D. 1969. "An operat ional - funct ional c l a s s i f i c a t i o n of f ishery managment techniques." Verh. Internat. Verein. Limnol. 17:636-40. CARRIER, J.G.; A.M. CARRIER. 1983. " P r o f i t l e s s property: Marine ownership and access to wealth on Ponam Is land, Manus Province." Ethnology. 22(2):133-52. 171 CARTER, W.E.; M. MAMANI P. 1982. Irpa Chico: Individuo y comunidad en 1 a cu l tura Aymara. La Paz: Ed i to r ia l Juventud. CASAVERDE, J.R. 1978. "Comunidad andina y dependencia." America Indigena 38(1) :15-41. CASTILLO E., U. 1978. J u l i , su h i s t o r i a , paisaje y personajes anonimos. Puno: Los Andes. CATTARINUSSI, B. 1973. "A soc io log ica l study of an I t a l i a n community of fishermen," in Seafarer and community, edited by Fricke P.H., pp 30-43. London: Groom Helm. CCAMA, F. 1981. "Rural development and small farmers: The case of peasants in the Puno, Peru area." Unpublished Ph.D d isser tat ion , Economics, Iowa State University . CENFOR-Puno. 1979. "Plan maestro de l a reserva nacional del T i t icaca." Lima: Centro de Estudios y proyectos de Inversion y Desar ro l lo , Universidad Nacional Agraria de 1 a Mol ina. CHAPMAN, D. 1979. "Catch evaluation in Lake T i t icaca and the Peruvian Amazon." Lima: Project FAO-PER/76/022. Reprinted in 1981 as IMARPE Informe No 81. CHIRAPO C , E. 1982. "Origen y cambios en l a produccion y consumo en l a nacionalidad Aymara: Los Yanaques." IIDSA, Publ icacion No 2, UNTA-Puno. CHRISTY, F.T. 1982. "Ter r i to r ia l use r ights in marine f i sher ies : Def in i t ions and c o n d i t i o n s . " FAQ F i s h . Tech. Rep., No. 227. CIRIACY-WANTRUP, S.V.; R. BISHOP. 1975. "Common property as a concept in natural resources pol icy ." Natural Resources J ^ 15(3):713-727. CLARK, C.W. 1980. "Toward a predict ive model for the economic regulation of commercial f i sher ies . " Canad. J . f i s h . Aquat. Sc. 37:1111-1129. CLAVERIAS, R.; J . MANRIQUE. Editors.,. 1983. La sequia en Puno: A l ternat ivas  i ns t i tuc iona les , techno!ogicas y populares.Tuno: IIDSA" Convenio UNTA-NUFFIC. COLLINS, J.L. 1982. "Kinship and seasonal migration among the Aymaras of southern Peru: Human adaptation to energy scarcity ." Unpublished Ph.D. d isser tat ion , Anthropology, University of F l o r i d a , G a i n e s v i l l e . COLLOT, D. 1980. "Les macrophytes de quelques lacs Andins." La Paz - Par i s : Convenio UMSA-ORSTOM. COON, T. 1981. " G i l l n e t s e l e c t i v i t y in the endemic cyprinodont (Orestias spp.) f lock of Lake T i t i caca . UC-Davis: Unpublished notes and manuscript. CORDELL, J.C. 1974. "The lunar - t ide f i sh ing cycle in northeastern B r a z i l . " Ethnology 13:379-92. 1978. "Carry ing capac i ty a n a l y s i s of f i x e d t e r r i t o i a l f i s h i n g . " 172 Ethnology 1 7 ( l ) : l - 2 4 . CORDERO E., E. 1966. "Los Uros de Qaama (Pomata, Puno)." Tesis para optar e l t i t u l o de Antropologo. Universidad San Antonio Abad, Cusco. COUTTS, R.R.; M. ROJAS P. 1981. "Estadist icas de l a comercializacion de pescado en La Paz, B o l i v i a : Periodo Febrero 1980 - Enero 1981, Rangos de Precios." La Paz: Min is ter io de Asuntos Campesinos y Agropecuarios. CREMER, M.A.; G. LEAL; R. LEBLANC; B. THEBAUD. 1984. Integrated r u r a l  development: State of the art review. H u l l , Quebec: Canadian International Development Agency. CR0C0MBE, R. 1974. "An approach to the analysis of land tenure systems>" In Land tenure in Oceania, edited by H.P. Lundsgaarde, pp 1-17, ASAO Monograph No 2. HonoluTu: The University Press of Hawaii. CRUTCHFIELD, J . 1979. "Economic and social impl icat ions of the main pol icy a l ternat ives for c o n t r o l l i n g f ish ing effort . " J . F ish. Res. Bd. Can. 36(7):742-52. ; G . P0NTEC0RV0. 1969. The P a c i f i c salmon f i s h e r i e s : A study of i r r a t i o n a l conservation. Resources for the Future. John Hopkins"Press. ; A. ZELLNER. 1962. Economic aspects of P a c i f i c H a l i b u t f ishery. Bureau of commercial F isher ies , US department of the Inter ior . DIREPE. 1981. "Estudio bio-ecologico de las especies nativas boga y mauri de l a Laguna Arapa." Unpublished manuscript. Puno: Min is ter io de Pesqueria. DAVIDOFF, P. 1978. "The red is t r ibut i ve function in planning: Creating greater equity among c i t i zens of communities," in PIanning theory in the 1980s: k search for future d i rect ions , edited by R. Burchel 1 and G.~Tternlieb, pp 69-12. Rutgers Univers i ty : Centre for Po l i c y Research. DAVIES, T.M. 1970. Indian integration in Peru: A hal f century of experience,  1900-1948. L inco ln : Univ. of Nebraska T ress . DAVIS, A. 1984. "Property r ights and access management in the small boat f ishery : A case study from South West Nova Scot ia ," in A t l a n t i c f i sher ies and  coastal communities, edited by C. Lamson and A.J. HansorH Hal i fax : Dalhousie Ocean Studies Program. DEMESTRE M.; A. ROIG; A. DE SOSTOA; F.J . DE SOSTOA. 1977. "Cont r ibuc io a l 'estudi de l a ic t i fauna continetal del de l ta de VEbre." Treb. Hist . Cat. H i s t . Nat. 8:145-226. DEMSETZ, H. 1967. "Toward a theory of property r ights." Amer. Econ. Assoc.  Papers and Proceedings, (May):253-7. DIEZ DE SAN MIGUEL, G. 1964. " V i s i t a hecha a l a Provincia de Chucuito por Garci Diez de San Miguel en el ano 1567." Version paleografica de W. Espinoza Soriano. Lima: Ediciones de l a Casa de l a Cultura. 173 DYSON-HUDSON, R.; E.A. SMITH. 1978. "Human t e r r i t o r i a l i t y : An ecological reassessment." American anthropologist 80:21-41. ECKERT, . The enclosure of ocean resources: Economics and the 1 aw of the sea. Stanford: Hoover Inst i tut ion Press. EMMERSON, D.K. 1980. "Rethinking art isanal f i sher ies development: Western concepts, Asian experiences." World Bank Staff Working Paper No 423. ENCA 1972. Encuesta Nacional de Consumo de Al imentos en el Peru. Lima: Inst i tuto Nacional de Estadist icas . EQUIPO JULI. 1974. "Diagnostico de mercados del eje J u l i a c a - Desaguadero." Puno: Plan Copesco - Orams VIII ESMAN, M.J.; N.T. UPHOFF. 1984. Local organizations: Intermediaries in rural  development. Ithaca: Cornell University Press. ETZIONI, A. 1973. "Commentary on Grabow and Heskin's paper: Foundation for a radical concept of planning." J . Amer. P lan . Ass. 31 (Marchh):114. 1978. "Planning: An h i s t o r i c a l and i n t e l l e c t u a l prospective," in Planning theory in the 1980s, edited by R. Burchel 1 and G. Stern l ieb : 209-211. EVERETT, G.V. 1971a. "The rainbow trout and the f isher ies of Lake Ti t ieaca." Report to the Government of the Republic of Peru. Lima: Unpubl. manuscript. — 1971b. "Estudios pesqueros en e l Lago T i t i e a c a . " R e v i s t a de 1 a UNTA, 3(4):374-?98. 1973. "The rainbow t rou t Salmo g a i r d n e r i (Rich.) f i s h e r y of lake T i t i e a c a . " J . F i s h . B i o l . (5):429^4"4U: ' FA0-RAS/77/P10. 1979. "Population and development in Asian marine f i sh ing communities: A comparative p r o f i l e . Unpublished report. Roma: FAO. FARFAN V., L.E. 1974. "Bases para el estudio estadist ico de l a pesca en el Lago T i t ieaca." Unpublished manuscript. Puno: IMARPE. FERNANDO, S. 1985^ "Costs and p r o f i t a b i l i t y of s m a l l - s c a l e f ish ing orperations in Sr i Lanka," in Smal l - sca le f i sher ies in As ia , edited by T. Panayotou, pp 137-150. 1985J) "The marketing system i n the smal 1 - s c a l e f i s h e r y of S r i Lanka: Does the middleman e x p l o i t the fisherman?" in Smal l - sca le f i sher ies in  A s i a , ed i ted by T. Panayotou, pp 185-196. ; L. DEVASENA; R.M. RANWEERA; H.K.M. SOMAWANTHA. "Impact of Budhism on the smal l - sca le f ishery of Sr i lanka," in Smal l - sca le f i sher ies in  A s i a , pp 205-210. FERRONI, M. 1980. "The urban bias of Peruvian food po l icy : Consequences and a l ternat ives . " Unpubl ished Ph.D d isser tat ion , Cornel l University . 174 FIEDLER, R.G. 1944. "The Peruvian Fisheries." Geogr. Rev. 34:96-119. FIGUEROA, A. 1984. C a p i t a l i s t development and the peasant economy in Peru. Cambridge University Press. FIORAVANTI-MOLINIE, A. 1978. "La communaute aujourd'hui." Annales Econ. Soc. C i v i l . 35(5-6):1182-96. FLORES-CUBA, L.G. 1983. "La caza de subsistencia de los Aymaras." Resumenes de las comunicaciones c i e n t i f i c a s del IX congreso Latino Americano de zoologia, Arequipa, Peru, 9-15 de Octubre 1983. FLORES-GALINDO, A. 1981. "La pesca y los Pescadores en l a costa central ( S i g l o XVII I ) ." H is to r ia 5(2):159-165. FLORES 0 . , J . 1977. Los pastores de P a r a t i a . Lima: Ins t . Estud. Peruanos. FLYVBJERG, B.; V.C. PETYERSEN. 1982. "Planning in the 33 years af ter 1984," in Planning Theory: Prospects for the 1980s, edited by P. Healey, G. McDougall and M. Thomas. London: Pergamon Press. FORBES, D. 1870. "On the Aymara Indians of B o l i v i a and Peru." J . Ethnol.  Soc. 2:193-305. FORESTER, J . 1982. "Know your organizations: Planning and the reproduction of social and p o l i t i c a l re lat ions ." Plan Canada 22(1):3-13. FRANC, J . ; L. LAUZANNE , F. ZUNA. 1985. "Algunos datos sobre las pesquerias de l a parte or iental del lago T i t ieaca Menor. La Paz: Ecologia en B o l i v i a , Revista del Inst i tuto de Ecologia No 7. FRIEDMANN, J . 1974. Retracking America: A theory of transactive planning. Garden Ci ty , New york: Anchor Press. 1980. "Urban communes, self-management, and the r e c o n s t r u c t i o n of the loca l state." Unpublished manuscript. UCLA: School of Architecture and Urban Planning. 1981. "The act ive community: Toward a p o l i t i c a l - t e r r i t o r i a l framework for regional development in Asia." Econ. Dev. Change 29(2):235-262. ; B. HUDSON. 1974. "Knowledge and a c t i o n : A guide to p lann ing theory." J . Amer. I n s t i t . P l a n . 40(1). ; C. WEAVER. 1979. T e r r i t o r y and f u n c t i o n : The e v o l u t i o n of regional planning. Berkeley: University of Caliform"a Press. FUENZALIDA V., F. 1970. "La estructura de l a comunidad de indigenas t r a d i c i o n a l , " in La hacienda, l a comunidad y el campesino en el Peru, edited by R.G. K e i t h , F .Tuenzal i d a , " J - M a t o s , J . CotTer and G. Al"b~erti, pp 61-104. Lima: Inst i tuto de Estudios Peruanos. GALDO P., R. 1962a. "Economia de las colect iv idades indigenas colindantes con el Lago Titieaca.""" Lima: Min is ter io de Trabajo e Asuntos Indigenas. 175 1962b. "El indigena y e l mest izo de Vi 1 quechico." Lima: Min is ter io de Trabajo e Asuntos Indigenas. 1981. " A r t e s a n i a s y pequenas i n d u s t r i a s en e l area c o l i n d a n t e con e l Lago T i t icaca (Puno, Peru)," in Acerca de l a h is to r ia y del universo  Aymara. Lima: CIED Ediciones. GOLDSCHMIDT, W.R.; T.H. HAAS. 1946. "Possessory r ights of the natives of southeastern Alaska." A report to the Commissioner of Indian A f f a i r s . Juneau. GONZALES U., E. 1979. "La economia de l a f a m i l i a (El caso de Antapampa)." Documento No 39. Lima: CISEPA, PUC. 1984. Economia de l a comunidad campesina. Lima: Inst i tuto de Estudios Peruanos. GORDON, H.S. 1954. "The economic theory of a common property resource: The f i s h e r y . " J . Pol i t . Econ. 62(3):126-42. GORE, C. 1984. Regions in question: Space, development theory and regional  po l icy . London, New YorR: Methuen. GOW, R.C. 1981. "Yawar mayu: Revolution in the Southern Andes, 1860-1980." Unpublished Ph.D d isser tat ion , Anthropology, University of Madison, Wisconsin. GRABOW, S.; A. HESKIN. 1973. "Foundations for a radical concept of planning." J . Amer. I n s t i t . P l a n . 39(March):106-114. GRIESHABER, E.P. 1977. "Survival of indian communities in nineteenth century B o l i v i a . " Unpubl ished Ph.D d isser tat ion , History, University of North Caro l ina , Chapel H i l l . HARDIN, G. 1968. "The tragedy of the commons." Science 162:1243-8. HEBBERT, M. 1982. "The new devcentralism: A c r i t ique of the t e r r i t o r i a l approach," in Planning theory, edited by P. Healey, G. McDonald and M. Thomas, pp 102-119. London: Pergamon Press. HICKMAN, J.M. 1963. "The Aymaras of Chinchera, Peru: Persistence and change in a b i - c u l t u r a l context." Unpublished Ph.D d isser tat ion , Cornel l Universi ty . ; W.T. STUART. 1977. "Descent, a l l i a n c e , and moiety in Chucuito , Peru: An explanatory sketch of Aymara social organization," in Andean kinship  and marri age, ed i ted by R. Bo l ton and E. Mayer, pp 43-59. HOLLING, C.S. Editor. 1978. Adaptive Environmental Assessment and Management. John Wiley. HORN, D. 1979. "Uru material cu l ture : An ethno-archeological study of B o l i v i a n marsh dwellers." Unpublished manuscript. Department of Anthropology, Washington University . HUDSON, B. 1979. "Comparison of current planning theories: Counterparts and 176 contradictions. J . Amer. PI an. Assoc. 45(0ct . ) :387-405. HYSLOP, J . 1976. "An archeological invest igat ion of the Lupaca kingdom and of i t s origins." Unpublished Ph.D. d isser tat ion , Columbia Universi ty . ICLARM. 1981. Report on the Coastal Zone Management Seminar-Workshop, November 5 -6 , 1981, ManTTa, Phi 11ipines. INE 1981. Censo Nacional de Poblac iony Vivienda. Inst i tuto Nacional de Estad is t icas , Lima, Peru. ISBELL, B.J. 1977. To defend ourselves: Ecology and r i t u a l in an Andean  v i 1 lage. Austin: Inst i tute of Lat in American Studies. JAMES, M.G. 1936. "Informe sobre l a v i g i l a n c i a para el desarro l lo de pesquerias en el Lago T i t ieaca." La Paz: Unpublished manuscript. JOHANNES, R.E. 1977. "Traditional law of the sea in Micronesia." Micronesia 13(2):121-127. 1978. " T r a d i t i o n a l marine conservat ion i n Oceania and t h e i r demise." Ann. Rev. E c o l . Syst . 9:349-364. 1981. Words of the 1 agoon. Berkel ey: Uni v. Cal i form'a Press . 1982. "Tradit ional conservation methods and protected marine areas in Oceania." Ambios 11(5):258-261. JOHANNESSON, K.; R. VILCHEZ; D. BERTONE. 1981. "Acustic estimation of icthyomass and i t s d is t r ibut ion in lake T i t ieaca." Unpublished report. Lima: FAO, Project GCP/RLA/025/N0R. JULIEN, C.J. 1983. "External control and ethnic i ty : The Uru t r ibute category." Unpublished manuscript . KAPETSKY, J.M. 1981. "Some considerations for the management of coastal lagoons and estuarine f isher ies . " FAO F ish. Tech. Rep. No. 218. KEMP, R. 1982. " C r i t i c a l planning theory: Review and c r i t ique , " in Planning  theory: Prospects for the 1980s, edited by P. Healey, G. McDougal 1 and M. Thomas. London: Pergamon Press. KENT, J.D. 1983. "Periodic a r id i t y and prehispanic settlement in the T i t ieaca basin." Paper presented at the Xlth Congress of Anthropologicl and Ethnological Sciences, Vancouver B.C. KIRBY, E.S.; E.F. SZCEPANIK. 1957. "Special problems of f i sher ies in poor countries," in The economics of f i she r ies , edited by R. Turvey and J . Wiseman. Rome: FAO. KOLATA, A.L. 1978. "The south Andes," in Ancients South Americans, edited by J.D. Jennings. San F r a n c i s c o : W.H. Freeman. (2nd e d i t i o n : 1983) LABA, R. 1979. "Fish, peasants, and state bureaucracies: The development of 177 Lake T i t icaca." Comparative P o l i t i c a l Studies 12(3):335-361. LA BARRE, W. 1948a. "The Aymara indians of Lake T i t icaca plateau, B o l i v i a . " Amer. Anthrop. Assoc. , Memoir 68, part 2. 1948b. "The Uru-Chipaya," in Handbook of South American Indians, edited by J . Steward, pp. 575-585. New York: Cooper -Square. LABBY, D. 1976. The demystification of Yap: D ia lec t i cs of Culture in a Micronesian Island. Chicago: University of Chicago Press. LAMBERT, B. 1977. "Bi 1 a t e r a l i t y in the Andes," in Andean Kinship and  Mar r iage , ed i ted by R. Bo l ton and E. Mayer, pp. 1-27. LAMSON, C ; F.G. COHEN. Editors. 1984. "Proceedings of a workshop on community-based resource management." Ha l i fax : Dalhousie Universi ty . LAUZANNE, L. 1981. "Description de t r o i s Orestias nouveaux du Lac T i t i caca , 0. i s p i n. sp. , 0. f o r g e t i n.sp. et 0. t c h e r n a v i n i n.sp (P isees , Cypri nodlSntTclaeT."" Cybium 5(3):71-"9"2. 1982. " Les Ores t ias du p e t i t l a c T i t i c a c a . " Rev. H y d r o b i o l . Trop. 15:39-70. LAWSON, R. 1984. Economics of Fisheries Development. London: F. Pinter . LESCANO, J . L . ; E. DEL VILLAR; J . L . RAMOS; F. HUAPAYA; D. LOAYZA; M. SALAS; B. LOPEZ. 1982. "Diagnostico tecnico agropecuario y socio-economico de las comunidades de Camacani y Luquina Grande - Chucuito - P l a t e r i a - Puno." UNA-Puno: Centro de Investigaciones en Cul t i vos Andinos. LEVIEIL, D. 1981. "Preliminary estimates of the socio-economic value of Lake T i t icaca s m a l l - s c a l e f isher ies . " Internal report. Lima: Project FAO-PER/76/022 1986. " T e r r i t o r i a l use r i g h t s i n f i s h i n g on lake T i t i c a c a (Peru)." Proceedings of the IIFET-GERMA Conference, Rimouski, Quebec. - ; G. GOYZUETA. 1984. " E v a l u a c i o n de l a ex t racc ion de to to ra para f ines de subsistencia en l a Reserva Nacional del T i t i caca , sector Puno." Informe prel iminar, CENFOR-Puno, Min is ter io de Agr icu l tura . ; P. PAZ. 1984. "Eva luac ion de l a con t r ibuc ion po tenc ia l de 1 os recursos acuaticos al desar ro l lo de los d i s t r i t o s de Arapa, Chupa y Saman del departamaento de Puno, Peru." Consultant report. . ; c. CUTIPA; G. GOYZUETA; P. PAZ. 1985. " Importancia s o c i o -economica de l a extraccion de macrofitos en l a Bahia de Puno," in Polucion en l a Bahia Interna de Puno, Lago T i t i caca , edited by P. Morales, D. Levy andTT Horthcote. westwater Research Center. ; B.S.ORLOVE. 1987. "Ecology of aquatic t e r r i t o r i a l i t y : The del ineat ion of TURF boundaries along the shores of Lake T i t icaca." Manuscript in preparation for submission to Human Ecology. 178 LEVINE, H.B. 1984. "Contro l l ing access: Forms of t e r r i t o r i a l i t y in three New Zealand f ish ing v i l l a g e s . " Ethnology 23(2):89-99. LEWELLEN, T. 1977. "The Aymara in t rans i t ion : Economy and r e l i g i o n in a Peruvian peasant community." Unpublished Ph.D d isser tat ion , Anthropology, University of Colorado, Boulder. LOUBENS, G.; F. 0S0RI0; J . SARMIENTO. 1984. "Observations sur les poissons de l a part ie bol iv ienne du lac T i t i caca . Mi l ieux et peuplements." Rev.  H y d r o b i o l . Trop. 17(2):153-161. ; J . SARMIENTO. 1985. "Observat ions sur Ores t ias a g a s s i i , Valenciennes 1846." Rev. Hydrobiol. Trop. 18(2):159-171. : ~~ MANELIS DE KLEIN, H. 1973. "Los Uros: El extrano pueblo del A l t ip lano . " Estudios Andinos 3(7):129-150. MARTIN, K.O. 1979. "Play by the rules or don't play at a l l : Space d i v i s i o n and resource a l l o c a t i o n in a rural Newfoundland community," in North A t l a n t i c  Maritime Cultures, edited by R. Andersen, pp. 277-298. The Hague: Mouton. MARTINEZ, A. H. 1962. "El indigena y e l mestizo de Taraco (Huancane, Puno)." Revista del Museo Nacional, 31:172-239. Lima. MARTINEZ Z., G. 1981. "Espacio Lupaqa: Algunas hipoteses de trabajo. In Ethnohistoria y antropologia andina, edited by A. C a s t e ! l i , M. Koth and M. Mould, pp Z63-780. Lima: Museo Nacional de H is tor ia . MASUDA, S.; I. SHIMADA; C. MORRIS. Editors. 1985. Andean ecology and c i v i 1 i za t ion : An i n t e r d i s c i p i inary perspecti ve on Andean ecologicaT  complementarity. University of Tokyo press. MATOS M., J . 1964. "La propiedad en l a I s l a de Taquile (Lago T i t icaca) , " in Estudios sobre l a cu l tura actual del Peru, pp 65-142, Universidad Nacional Mayor de San Marcos. MAYER, E. 1985. "Production zones," in Andean Ecology and C i v i l i z a t i o n , edited by S. Masuda et a l . , pp. 45-84. University of Tokyo Press. MAZZIOTI, D.F. 1974. "The underlying assumtions of advocacy planning: P lu ra l i sm and reform." Amer. Ins t i t . P lan . 40(Jan.):38-47. MICHAUD, A. 1973. "Choix de parents r i t u e l s sur l ' A l t i p l a n o peruvien: Strategies a l ternat ives . " J . Soc. Americ. 62:169-83. MINO, G.M. 1984. "Descripcion de l a Provincia de Chucuito comprehendida en l a ju r i sd icc ion de l a Audiencia de las Charcas en el Obispado de La Paz." Revista Andina, No. 2. MALONEY, D.G.; P.H. PEARSE. 1979. "Quantit iat ive r ights as an investment for regul a t i n g commercial f i s h e r i e s . " J . F i s h . Res. Bd. Can. 36(7): 859-66. MIPE 1980. Bolet in estadist ico pesquero No 1. Puno: Direccion Regional de Pesqueria, Unidad de P Ian i f icac ion . 179 MIPE-DIE 1978. "Pesca artesanal en l a s ie r ra sur del Peru con especial referencial al Lago T i t i e a c a . " Mimeo. MONTOYA T. 1979. "Los mini fundios en el A l t i p l a n o Peruano." America  Indigena. 39(4):773-94. ; H. BURGOS 1981. Puno en 1 ucha por 1 a t i e r r a . Lima: CIED. MORLON, P. 1978. "Elementos de descripcion frecuencial no especif ica de las heladas." Puno: Min is ter io de Agr icul tura y Al imentacion, Zona XII. MUJICA, E. 1985. "A l t ip lano-coast re lat ionships in the southeast central Andes: From indi rect to d i rect complementarity," in Andean Ecology and  C i v i l i za t ion , edited by S. Masuda et a l . , pp: 103-14TJ7~^ MUNASINGHE, H. 1985. "Socioeconomic conditions of sma l l - sca le fishermen in Sr i Lanka," in Smal l - sca le f i sher ies in As ia , edited by T. Panayotou, pp 73-83. MUNRO G. 1982. "Fisher ies, extended j u r i s d i c t i o n and the economics of common property resources." Canad. J . Econ. l5(3) :405-425. ; C. KIM LOY. 1978. The economics of f i s h i n g and the deve lop ing world: A Maiaysian case study. LPenangJ,Tul au Pinang: Penerbit Un ivers i t i Sains Malaysia. School of Comparative Social Sciences. MUNRO G.; A.D. SCOTT. 1984. "The economics of f i s h e r y management." U.B.C. Department of Economics, Discussion Paper 84-09. MURRA, J.V. 1984. "Andean societ ies before 153," in The Cambridge history of Lat in Americ. Vol I: Colonial Lat in America, edited by L.I. Bethel 1, pp 59-~9u. Cambridge University Press. 1985a. "El archipelago ve r t i ca l rev i s i ted , " in Andean ecology and c i v i l i z a t i o n , edTted by S. Masuda et a l . , pp 3-14. 1985b. "The l i m i t s and l imi ta t ions of the ve r t i ca l archipelago," in Andean ecology and c i v i l i z a t i o n , edited by S. Masuda et a l . , pp 15-20. McCAY, B.J. 1980. "A fishermen's cooperative l i m i t e d : Indigenous resource management in a complex society." Anthrop. Quart. 53(l) :29-38. 1981. "Development issues in f i s h e r i e s as agrar ian systems." Cul ture and agr icul ture ( l l ) : l - 8 . McGOODWIN, J.R. 1984. "Some examples of se l f - regu latory mechanisms in unmanaged f i s h e r i e s . " FAO F i s h . Report , No. 289, Supp l . 2 and 3. NEAL, R. 1982. "Dilemma of the s m a l l - s c a l e fishermen." ICLARM Newsletter 5 ( Ju ly ) :7 -9 . NEVEU-LEMAIRE, M. 1906. Les lacs des hauts plateaux de TAmerique du Sud. Par is : Imprimerie Nationale. 180 NUNEZ, M. 1982. "La pesca en l a comunidad de Llachon, Capachica." Unpublished report to Prof. B.S. Orlove. University of Ca l i fo rn ia -Dav is . ORLOVE, B.S. 1979. "Lake T i t ieaca fishermen: A preliminary report on anthropological and ecological research." Paper presented at the 78th Annual Meeting of the American Anthropological Associat ion, C inc innat i , Ohio. 1986. "Barter and cash s a l e on Lake T i t i e a c a : A t e s t of competing approaches." Current Anthropology 27(2):85-106. PACORI Q.,H. 1976. "Aspectos socio-economicos de l a pesca y del consumo de pescado del Lago T i t ieaca , en l a loca l idad de Pusi (Huancane, Puno)." Tesis para optar el t i t u l o profesional de Trabajadora S o c i a l , UNTA, Puno. PAINTER, M. 1982. "The p o l i t i c a l economy of food production: An example from an Aymara-speaking region of Peru." Unpublished Ph.D. Dissertat ion. Anthropology. University of F l o r i d a , G a i n e s v i l l e . —i 1986. "The va lue of peasant labour power in a prolonged t rans i t ion of capital ism." J . Peasant Studies 13(4)221-239. PALAVECINO, E. 1948. "Les indiens Uru de I ruito." Runa, Archivo para las Ciencias del hombre 2:59-88. Universidad de Buenos Aires. PALLSON, G. 1982. " T e r r i t o r i a l i t y among Icelandic fishermen." Acta  Sociologica 25(Supplement):5-13. PANAYOTOU, T. 1982. "Management concept for s m a l l - s c a l e f i sher ies : Economic and s o c i a l aspects ." FAO F i s h . Tech. Rep., No. 228. 1984. " T e r r i t o r i a l Use Rights i n F i s h e r i e s . " FAO F i s h . Report, No. 289, Supp l . 2 and 3. Ed i tor .1985. S m a l l - s c a l e F i s h e r i e s i n A s i a : Socioeconomic analysis and po l icy . Ottawa: International Development Research Center. ; j . JITSANGUAN; K. ADULAVIDHAYA. 1985. "Cost s t ruc tu re and p r o f i t a b i l i t y of the Thai coastal f ishery," in Smal l - sca le f i sher ies in As ia , ed i ted by T. Panayotou, pp 163-175. PARENTI, L.R. 1984. "A taxonomic rev is ion of the Andean K i l l i f s h genus Orestias (Cyprinodontiformes, Cyprinodontidae)." Bui 1. Amer. Mus. Nat. Hist . 1/8(2). PEARSE, P.H. Editor. 1979. "Symposium on p o l i c i e s for economic ra t iona l i za t ion of commercial f i sher ies" . J . F i sh . Res. Bd. Can., 36. PEARSE, P.H. 1980. "Property r ights and the regulation of commercial f i s h e r i e s . " J . Business Admin. (1-2): 185-209. PEASE, F. 1985. "Cases and var iat ions of v e r t i c a l i t y in the southern Andes," in Andean Ecology and C i v i 1 i z a t i o n , edited by S.Masuda et a l . , pp. 141-160. 181 1986. "La nocion de propiedad entre l o s Incas: Una approximacion," i n Ethnografia £ h i s to r ia del mundo andino: Continuidad y cambio, edited by S. Masuda, pp:3-33. Universidad de Tokio. PETIT-SKINNER, S. 1983. "Traditional ownership of the sea in Oceania," in Ocean Yearbook No. 4, edited by E. Mann-Borgese. Chicago Univ. Press. PINKERTON, E. Editor. 1986. Cooperative management of loca l f i sher ies : Are  equity, e f f ic iency and superior management achievabTe? University of B r i t i s h Columbia Press. POE, K.M. "Land and labour in the T i t ieaca basin: An ethnohistory of a highland Aymara community." Unpublished Ph.D dissertat ion. Anthropology. University of V i r g i n i a . POLLNAC. R.B. 1984. "Investigating t e r r i t o r i a l use r ights among fishermen," in Maritime Inst i tut ions in the Western P a c i f i c , edited by K. Ruddle and T. Akimichi , Senri Ethnological Studies 17. Osaka: National Museum of Ethnology. 1986. " T r a d i t i o n a l use r i g h t s and f i s h e r y management i n LDCs." Paper presented at the Annual Meeting of the American Fisheries Society, Providence, R.I. POLLNAC, R.; S.J. LITTLEFIELD. 1983. "Sociocultural aspects of f i sher ies management." Ocean Devel op. & Internat. Law 12(3-4):209-246. PRDSP 1959. "Plan Regional para e l Dessarro l lo del Sur del Peru." Informe PS/G/59. Manual de estadist ica Regional. Lima. RAMIREZ, S. 1985. "Social f ront iers and the t e r r i t o r i a l base of curacazgo," in Andean Ecology and C i v i l i z a t i o n , edited by Masuda et a l . , pp. 423-442. RAMIREZ-HORTON, S.E. 1981. "La organizacion economica de l a costa norte: Un analys is prel iminar del periodo prehispanico tardio ," in Ayl 11u y comunidad, edited by A. C a s t e l l i et a l . , pp. 281-297. Lima: Centro de Proyeccion Andina. RENDON. J ; R. RENDON. 1969. Derechos socia les del pescador. Lima: Tarpuy. RETTIG, B.; J . GINTER. 1978. Limited entry as a f ishery management t o o l . Washington Sea Grant publication^ Seat t le : U n i v e r s i t y of Washinton Press. RICHERSON, P .J . ; C.J. WIDMER; T. KITTEL. 1977. "The 1 imnol ogy of Lake T i t ieaca (Peru -Bol iv ia ) , a large, high a l t i t u d e t rop ica l lake." Inst i tute of  Ecology, Publ icat ion No 14, UC-Davis, Davis C a l i f o r n i a . RICKER, W.E. 1975. "Computation and interpretat ion of b io log ica l s t a t i s t i c s of f i s h popu la t ions . " Bui 1. F i s h . Res. Bd. Can. No 191. ROJAS P., M. 1984. "Estadist ica de l a comercial izacion de pescado en La Paz, B o l i v i a : Periode Enero - Diciembre 1983." La Paz: Min is ter io de Asuntos Campesinos y Agropecuarios ROMERO, E. 1926. Monografia del Departamento de Puno. Lima: Torres Aguirre. 182 RONDINELLI, D.A. 1979. "Administration of integrated rural development po l i cy : The pol icy of agrarian reform in developing countries." World P o l i t i c s 31(3):389-416. ROSENVALLON, P. 1976. L'age d'or de Vautogestion. Par is : Le S e u i l . ROSTWOROWSKI, M. 1977. Pescadores, Artesanos y Mercaderes Costenos en el Peru Prehispanico: Etnia y Sociedad. Lima: Inst i tuto de Estudios Peruanos. _- 1981. Recursos Naturales Renovables y Pesca, S ig los XVI y XVII. Lima: Inst i tuto de Estudios Peruanos. ROTHSCHILD, B J . ; J.L. FORNEY. 1979. "Introduction," in Predator-prey systems in f i sher ies management, edited by M. Clepper, Washington: Sport F isheries I n s t i t u t e . RUDDLE, D.; T. AKAMICHI. Editors. 1985. Maritime ins t i tu t ions in the Western  P a c i f i c . Osaka: National Museum of Ethnology. ; R. CHESTERFIELD. 1977. "Education fo r t r a d i t i o n a l food procurement in the Orinoco del ta ." Ibero Americana:53. Berkeley: Univ. C a l i f o r n i a Press. ; R.E.JOHANNES. E d i t o r s . 1985. The t r a d i t i o n a l knowledge and management of coastal systems in Asia and th"e~"Pacific. Jakarta: UNESCO RUNGE, C.F. 1986. "Common property and c o l l e c t i v e action in economic development." World Development 14(5):623-635. SABRI, J . 1976. "Smal l - sca le f i sher ies development in Peninsular Malaysia," in Smal l - sca le f i sher ies development, edited by B. Lockwood and K. Ruddle, pp. 63-/4. Honolulu: East-West Center. SAIGNES T. 1978. "De l a f i l i a t i o n a l a residence: Les ethnies dans les v a l l e e s de Larecaje." Annales Econ. Soc. Civ . 33(5-6):1160-81. SANCHEZ E., R. 1983. "La promocion campesina en una micro-region del A l t ip lano . " Puno: Report to the German technical cooperation serv ices. SCOTT, A.D. 1979. "Development of economic theory of f i sher ies regulation." J . F i s h . Res. Bd. Can. 36(7) :725-41. ; P.A. NEHER. Ed i to rs .1981. The pub! i c regul a t i o n of commercial f i sher ies in Canada. Ottawa: Economic Council of Canada. SCHWEIGGER, E.H. 1930. "Possibi1idades pesqueras del Lago T i t icaca." Bolet in  de 1 a Compania Administradora del Guano. (Agosto 1930):467-473. SCUDDER, T.; T. CONELLY. 1985. "Management systems for r i ver ine f isher ies . " FAQ F i s h . Tech. Paper No 263. SHERBONDY, J.E. 1982. "The canal system of Hanan Cuzco." Unpublished Ph.D d isser tat ion , Anthropology, University of I l l i n o i s , Urbana-Champaign. SMITH, C.L. 1983. "Evaluating human factors." In Methods in f i sher ies 183 research, edited by L.A. Nielsen and D.L. Johnson. Bethesda, Maryland: American Fisheries Society. SMITH, I.R. 1979. "A research framework for t rad i t iona l f isher ies . " ICLARM Studies and Reviews No 2. 1981. "Improving f i s h i n g incomes when resources are o v e r f i s h e d . " Mar. P o l . 5 ( l ) :17 -22 . ; T. PANAYOTOU. " T e r r i t o r i a l use r i g h t s and economic e f f i c i e n c y : The case of the P h i l i p i n e f i sh ing concessions." FAO Fish. Tech. Pap. No.245. SOLC, V. 1969. Los Aymaras de las Is !as del T i t i caca . Inst i tuto Indigenista Interamericano, Serie Anthropologia Social 12, Mexico. SOLDI, A.M. 1978. "Nuevos datos sobre l a antigua Provincia de Chucuito," in Ethnohistoria y Anthropologia Andina, edited by M. Koth; A. C a s t e l l i , pp. 123-138. Lima: Centro de Proyeccion Andina. STOHR, W.B. 1981. "Development from below: The bottom-up and periphery-inward development paradigm," in Development from above or below, edited by W. Stohr and D. T a y l o r , pp 39-69. ; D. TAYLOR. E d i t o r s . 1981. Development from above or be! ow? A radical reappraisal of spat ia l planning in developing countries. N.Y.: wTley. SONQUIST, J.A.; W.C. DUNKELBERG. 1977. Survey and opinion research:  Procedures for processing and analys is . Prentice Hal 1. SUTTLES, W. 1960. "Af f ina l t i e s , subsistence and prestige among the coast S a l i s h . " Amer. Anthrop. 62(2):296-305. TAFT, A.C. 1954. "Estado de l a pesca en B o l i v i a . " La Paz: Unpublished report. TCHERNAVIN V.V. 1944a. "A rev is ion of the subfamily Orestinae." Proc. Zool .  Soc. London. 114:140^33. 1944b. "A rev is ion of some Trichomycterinae based on material preserved in the B r f t i sh Museum. Proc. Zool . Soc. London. 114:234-275. TREVINO, H.; J . TORRES; J . TALAVERA; D. LEVIEIL. 1981. "La Huayunaccana, un arte par l a extraccion de los peces del genero Orestias, en e l Lago T i t i caca (zona del Lago Pequeno)." Unpublished Report. IMARPE-Puno. ; D.A. LEVY; T.G. NORTHCOTE. 1984. "Pesca experimental en aguas negras y l impias del l i t o r a l de l a Bahia de Puno, Lago T i t i caca , Peru." IMARPE Bolet in 8(6) . TROADEC, J.P. "Introduction to f i sher ies management: Advantages, d i f f i c u l t i e s and mechanisms." FAO F i s h . Tech. Pap. No 224. TSCHOPIK, H. 1946. "The Aymara," in Handbook of South American Indians, ed i ted by J . Steward, pp. 501-573. New York: Cooper Square. 184 1951. "The Aymara of Chucui to , Peru. Par t 1: Magic." Anthrop. Pap. AMer. Mus. Nat. H i s t . 44(2). TUTIN, T.G. 1940. "The macrophyte vegetation," in Transactions of the Linnean  Society. The 1937 Percy Sladen Trust Expedition to Iake T i t ieaca. VALLEE.L. 1972. "Cycle ecologique et cycle r i t u e l : Le cas d'un v i l l a g e andin." Canadian Rev. Soc. Anth. 9(3):238-54. VAUX, P.; W. WURTSBAUGH; H.TREVINO; L. MARINO; E. BUSTAMANTE; J . TORRES; R. ALFARO; P. RICHERSON. 1986. "Ecology of the pelagic f ishes of Lake T i t ieaca (Peru-Bol iv ia) . " UC-Davis: Unpublished manuscript. VELASCO M., 0. 1978. Desde 1 a apacheta, sombrero en mano, a_ ti_ Ju l i . Arequipa: Imprenta El S o l . VELLARD J . 1949. "Contribution a Vetude des indiens Uru or Kot'suns." Travaux de l'IFEA. 1:145-209. 1957. "Populations indigenes des hauts plateaux Andins." Travaux de l ' IFEA 6:51-94 1963. C i v i l i s a t i o n s des Andes: Evolut ion des Populations du Haut-Plateau B o l i v i e n . Par is : Gal 1imard NRF. VERGARA, N.E. 1980. "Aspectos pr inc ipales de l a comercializacion y el consumo de productos pesqueros en B o l i v i a . " Rome: FAO, TCP/BOL/8904. VERLIAT, S. 1978. "Etude d'une communaute rurale andine: Chujucuyo Marcayoca, Puno, Peru." Unpublished thes is , Universite de Limoges, France. VILLWOCK, W. 1972. "Gefahren dur die endemische fishfauna durch einbhrgeruungsversuche und akkl imatisat ion vanfremdtischen an beispiel des Tit icaca-Seea (Peru/Bol iv ien) und des Lanao-Sees (Mindanao/Phil 1ipinen)." Verh. Internat. Ver in. Limnol. 18:1227-1234. WACHTEL, N. 1978. "Hommes d'eau: Le probleme Uru (XVI-XV11 s iec les) . " Annales Economies Societes C i v i l i s a t i o n s 33(5-6):1127-1159. WALTERS, C.J. 1986. Adaptive Management of Renewable Resources. New York: Col 1 ierMcMi l lan . ; R. HILBORN. 1975. "Adaptive c o n t r o l of f i s h i n g systems." J . F i s h . Res. Bd. Can. 33( 1): 145-159. ~ 1978. "Ecological optimization and adaptive management." Ann. Rev. E c o l . Syst . 9:157-88. WARD M.D. 1978. The p o l i t i c a l economy of d i s t r i b u t i v e equal i ty versus  inequal i t y . New York: E l sev ie r . WEAVER, C. 1984. Regional development and the loca l community, planning  p o l i t i c s and social context! New York: Wi1ey. 185 WILEN, J.E. 1979. "Fishermen behaviour and the design of e f f i c i e n t r e g u l a t i o n s programs." J . F i s h . Res. Bd. Can. 36(7) WILSON, J.A. 1977. "A test of the tragedy of the commons," in Managing the  commons, ed i ted by G. Hardin and J . Baden, pp 96-111. San F r a n c i s c o : W.H. Freeman. WURTSBAUGH, W.A. 1974. "Biologia y pesqueria del pejerrey (Basi l ichthys  bonariensis) en el lago T i t ieaca." Internal report. Lima: IMARPE. 186 APPENDIX A: ORESTIAS SPECIES ENCOUNTERED IN LAKE TITICACA WATERSHED (TCHERNAVIN 1944; LAUZANNE 1982; PARENTI 1984) COMPLEX SPECIES VERNACULAR NAME FEEDING TYPE Cuvier i 0 . cuv ier i umanto midwater predator 0. pentlandi i boga, b e l l a, ccesi . midwater planktivore 0 . i sp i i spi midwater plankt ivore 0 . forgeti i sp i midwater plankt ivore Gi1soni 0. gi1soni i sp i midwater plankt ivore 0 . taqui r i i sp i midwater plankt ivore 0 . mooni i sp i midwater plankt ivore 0 . uruni 0 . minimus 0 . minutus 0 . tchernavini 0 . tomcooni 0 . imarpe 0, robustus Mul 1 er i 0 . mul1eri carachi gringo benthic feeder 0 . g r a c i l i s benthic feeder 0. crawfordi benthic feeder 0 . t u t i n i benthic feeder 0. incae benthi c feeder Agassii 0 . agassii c. bianco, ch'awlla benthic feeder 0. frontosus 0 . j uss ie i c. amari11 o benthic feeder 0 . puni 0. tschudii 0. ctenolepis 0. richersoni 0. mult ipor is 0. s i l u s t a n i 0 . luteus c. amari11 o, pongo. benthic feeder 0. rotondipinnis c. amari11o benthic feeder 0 . farfani 0 . al bus c. amari11 o, khano. benthic feeder 0. ol i vaceus c. enano, punkhu benthic feeder 187 APPENDIX B: HISTORY OF LAKE TITICACA FISHERIES Formative and Tiahuanuco periods Recent archeological evidence indicates that a succession of sophisticated regional cultures developed in the v i c i n i t y of Lake T i t ieaca between 2000 and 100 BC, despite i t s harsh natural conditions (Browman 1981). During th is formative period, loca l populations made the t rans i t ion from hunting and gathering, to a combination of camel id pastoral ism, tuber agr icu l ture , and lacustr ine resources harvesting, which was su f f i c ien t to support sedentary v i l l a g e s (Ibid.; Kolata 1978). Because f i s h , waterfowl and lake-p lants played such a major ro le in the subsistence of these ear ly cultures (Kent 1983), i t has been hypothesized that Lake T i t ieaca may have acted as a "catalyst for the complex set of economic interactions that resulted in the evolut ion of c i v i l i z a t i o n (in the region)" (Kolata 1978: 248). This evolut ion culminated with the Tiahuanuco (or Tiwanaku) commercial and re l ig ious empire which f lour ished between 100 BC and 1250 AD. The l a t t e r ' s sphere of influence spread over the entire A l t i p l a n o , and even included economic colonies along the warmer v a l l e y s of the P a c i f i c coast, or Costa, and of the Amazonian slopes of the Andes, or Sel va, to which A l t i p l a n o residents migrated p e r i o d i c a l l y (Mujica 1985). Exchange of goods transported by l lama caravans between the A l t i p l a n o and these co lonies , and massive land reclamation programs in the nearshore area contributed to the development of the r ich Tiahuanuco economy (Kolata 1978). L i t t l e i s known of the importance of lacustr ine resources during the Tiahuanuco period, although the locat ion on the periphery of Lake T i t ieaca of most Tiahuanuco s i tes seems to indicate the persistent use of these resources (Mujica 1985). Moreover, a substantial increase in the number and average size of these s i tes suggests a considerable increase of population (Ibid.), and quite possibly of lacustr ine resources use. F i n a l l y , evidence of aquatic transportation of large blocks of stone on totora-reed boats across the lago Pequeno indicates that 2000 years ago, people already knew how to make sea-worthy crafts (Browman 1981) that could have been used for f i sh ing . Inca period The Tiahuanuco empire disintegrated sometimes between 1200 and 1300 AD when i t was replaced by a number of small Aymara kingdoms. A period of chronic wars ensued, to which the Inca conquest put an end around 1450 (Browman 1981; Murra 1984). The end of continuous warfare meant that the Inca administrators could get loca l populations to abandon their strongholds in the h i l l s where they had sought refuge and to s e t t l e back on the lakeshore (Murra 1985). This suggests that after a t rans i t ion period, lacustr ine resources might have regained thei r importance. The A l t i p l a n o , or Qol lasuyu, was the most populated quarter of the Inca empire, or Tawanti suyu. Because of i t s wealth in camel ids and in mineral deposits, i t remained under the di rect control of the Spanish crown after the conquest (Hyslop 1976). Because of t h i s , some documentation on loca l conditions in the ear ly co lon ia l period i s a v a i l a b l e . The o f f i c a l tax rate set in 1575, for example, provides information on the wealth and sources of 188 income of d i f ferent sectors of the A l t i p l a n o population. I t indicates that during the Inca administration of the A l t i p l a n o , f ish ing on Lake T i t i caca was carr ied exc lus i ve l y by Uros Fishermen (Diez de San Miguel [1567] 1964). Given the Inca administrators common practice of adapting ex ist ing social structures rather than replacing them, i t may be hypothesized that th i s c a s t e - l i k e organization might have predated the Inca conquest of the A l t i p l a n o . I t i s commonly bel ieved that the Uros were ear ly s e t t l e r s of the A l t i p l a n o , l a te r pushed into the vast marshes of Lake T i t icaca and of the Desaguadero r i v e r by the Aymaras (La Barre 1948; Manelis 1973). Uros fishermen were poor and landless : they belonged to a caste of in fe r io r social status (Ve l la rd 1963; Manelis 1973). However, not a l l Uros belonged to th i s in fe r io r caste of landless fishermen, many were at least as r ich as the i r Aymara neighbours in land and in camel ids (Wachtel 1978). In the ear ly co lon ia l period, for example, only 40% of the Uros were expected to pay the i r taxes in f i s h , the others had to pay them in precious metal, agr icu l tu ra l or weaving products (Wachtel 1978). Colonial period Lake T i t icaca f i sher ies may have been an important part of the loca l economy, during the co lon ia l period, even though Uros fishermen's contribution in f i s h represented only 3.5% of the tota l t r ibute paid by the region. I t i s l i k e l y to have been low simply because only a small part of fishermen's taxes could be paid d i r e c t l y in f i s h , since the co lon ia l administration preferred precious metal to f i sh (Jul ien 1983). A hundred years l a t e r , the contribution of Lake T i t icaca f i sher ies to the regional economy became more obvious as increasingly large quantit ies of dried f i s h had to be shipped to the growing contingent of laborers forced to work in the s i l v e r mines of Potosi (Soldi 1978). For l a t e r periods, however, a v a i l a b l e documents offer only sketchy information on the exp lo i tat ion of lacustr ine resources in the 18th (Mino 1984) and 19th century (Forbes 1870). Evolut ion of the number of fishermen Although there i s no h i s t o r i c a l record of the number of fishermen on Lake T i t i ca ca , there i s some evidence that neither the number of f i sh ing settlements, nor the number of fishermen have changed since the Spanish co lonizat ion . Under the Inca administration, l o c a l i z e d kinship groups, or ayl 1 us, were granted formal recognition, and exc lus ive r ights over the resources within the i r t e r r i t o r i e s (Murua, in Ramirez 1985: 430). During the co lon ia l period these t e r r i t o r i e s have been reorganized, but the number of human settlements along the shores of the Lake has not been modified since then, as demonstrated by the correspondance between t rad i t iona l ayl 1 us, and modern administrative units (Martinez 1981). Using information drawn from the l i s t s of t r ibu ta r ies , a l l Uros, who had to pay the i r taxes to the Spanish Crown in f i sh (Wachtel 1978; J u l i e n 1983), I have estimated the total number of fishermen on the whole A l t i p l a n o at 6,394 in 1575. This indicates that the number of fishermen was the same in 1575 and in 1976 when the total number of fishermen on Lake T i t icaca was estimated at 6,000: 3,040 for Peru (Bustamante and Trevino 1976) and about as many for B o l i v i a (B. Orlove 1984: pers. com.). 189 This apparent s t a b i l i t y cannot be credited to the enforcement of TURFs alone. The decimation of indigenous populations fo l lowing the Spanish colonizat ion might wel l be responsible for most of i t . Successive epidemics of European diseases, wars, rebe l l i ons , forced labor in dangerous mines and excessive taxing reduced the native populations of Peru and B o l i v i a from 3 m i l l i o n s in 1532 to 608,000 in 1796 (Grieshaber 1977: 49). In the dioceses of Cuzco and La Paz, corresponding to the A l t i p l a n o i t s e l f , the number of t r ibutar ies was reduced by 52% between 1591 and 1754 (Ibid: 51). This decimation continued wel1 into the 19th century, as demonstrated by a decrease of 23% in the indian population of the B o l i v i a n A l t i p l a n o between 1838 and 1877 ( Ib id : 294). In the case of the Peruvian A l t i p l a n o , th is demographic trend seems to have been reversed at the beginning of the 20th century, when the population of the department of Puno increased by 62% between 1896 and 1980 (INE data). However, because of a general urbanization tendency, the lakeshore population (excluding the c i t y of Puno) increased only by 18% between 1940 and 1981, compared to 35% for the rest of the department. S t i l l , an analysis of the change in the number of fishermen during the l a t t e r period, suggests that i t has changed 1 i t t l e . Estimates of fishermen numbers since the beginning of th is century are rare and widely f luc tuat ing . An estimate of 300 nomadic fishermen l i v i n g in totora beds in 1940, for example, seems l imi ted to the Uros fishermen of the Bahia de Puno (Fiedler 1944) and i s thus unre l iab le . Various estimates for the la te 1950's range from 2,800 (PRDSP 1959) to 10,000 fishermen (Terrazas in Laba 1979), although a range of 2,000 to 3,000 fishermen seems more r e a l i s t i c (Ve l la rd 1963: 65). Estimates for the mid and la te 1960s, range from 1,000 (Everett 1971a) to 1,800 fishermen (MIPE-DIE 1978), but the l a t t e r are probably underestimates, because they seem to take into account only those fishermen par t ic ipat ing in the trout f ishery. For the mid 1970s, a r e l i a b l e estimate indicates that 3,040 fishermen operated on Lake T i t ieaca in 1976 (Bustamante and Trevino 1976). F i n a l l y , a more recent estimate of 3,122 fishermen for 1981 i s based on the assumption that the proportion of fishermen in the tota l shore population has not changed since 1976 (Av i la et a l . 1985), and i s of l i t t l e help here. My analysis of the resu l ts of two par t ia l censuses of the f i sh ing population conducted in 34 and 38 communities, in 1980 and 1984 respect ive ly , suggests that the tota l number of Peruvian fishermen i s u n l i k e l y to have changed recently. Summing up the tota l number of fishermen for each sample and comparing these sums with those for the corresponding communities in 1976, indicate that the tota l number of fishermen for these communities might have declined by 17% between 1976 and 1980, but increased by 40% between 1976 and 1984. However, a comparison of fishermen numbers on a community basis reveals that there i s no s t a t i s t i c a l l y s ign i f i cant difference between average numbers of fishermen per community, either for the 1976-1980, or the 1976-1984 period. This suggests that the rates of change in the number of fishermen ca lcu lated above, are u n l i k e l y to apply to the whole fishermen population, and that the total number of fishermen i s u n l i k e l y to have changed noticeably ei ther between 1976 and 1980, or between 1976 and 1984. To sum up, a l l the evidence presented here suggests that the tota l number of fishermen on Lake Ti t ieaca has remained f a i r l y stable during the l a s t forty 190 years, and that even i f i t has increased, i t has done so from the growth of the shore dwel l ing population rather than from an i n f l u x of inland dwellers. This indicates that loca l TURFs may l i m i t entry to loca l f i sher ies . Recent b io log ica l and technical changes Since the middle of th is century some major changes have affected Lake T i t i caca f i sher ies . Exotic f i s h species were introduced to the lake in the ear ly 1940s for the trout (Salmo gairdneri) and la te 1950s for the s i l v e r s i d e or pejerrey (Basil ichthys bonanensis). Subsequently, a successful trout f ishery developed in the ear ly 19faU"s. This f ishery supported up to f i v e art isanal canneries and reached a peak production of 400 mt in 1965, however, overf ishing brought i t s co l lapse and forced the canneries to close down in 1969 (Everett 1971a). After i t s introduction against experts' advice (Taft 1954) the s i l v e r s i d e became the object of a l imi ted f ishery in the ear ly 1970s (Wurtsbaugh 1974) which has kept increasing ever since (Alfaro et a l . 1982; A v i l a et a l . 1985). Fish marketing has also changed considerably in the l a s t 30 years. Urbanization, transportation improvements and the m u l t i p l i c a t i o n of markets on the A l t i p l a n o have opened new opportunities for f i s h traders, with a consequent increase in thei r number. Up unt i l the ear ly 1960s f i s h vendors used to be rare on loca l markets. Fishermen's wives would only s e l l a l i t t l e b i t of f i s h on the beaches ( V e l l a r d 1963). But by 1980, there was at least one f i sh vendor in 76% of the A l t i p l a n o markets, and at least 7 in 54% of them (Orlove pers. comm.). Furthermore, the number of markets had almost doubled from 73 in 1961, to 118 in 1974 (Appleby 1978), so that by 1984 more than 670 f i sh vendors were o f f i c i a l l y registered at the Ministry of F isher ies in Puno. Fish traders have also found new markets in neighbouring departments. They now ship Lake T i t i caca f i s h to the mining centers (eg Toquepala, Cuajone) and urban areas of the Southern Andes (eg Cuzco, Arequipa) and even of the P a c i f i c coast (eg Tacna, Moquegua) ( A v i l a et a l . 1985). Technical innovation has general ly been wel l accepted among Lake T i t icaca fishermen, once i t s par t i cu la r merits had been demonstrated (Ve l la rd 1963). Balsa reed boats, for example, had been the only c raf t used on Lake T i t i caca , for more than a thousand years, yet thei r number has s teadi ly declined since the introduction of wooden boats. The l a t t e r became more widely used in the 1940's, f i r s t for transport, because of a dramatic drought during which the supply of totora reeds dwindled (Aval os-1951; V e l l a r d 1963; Matos 1964), and l a t e r for f i s h i n g , with the development of the trout canning industry. Because boats are l i g h t e r and faster than t rad i t iona l balsas, thei r adoption has meant a greater range of operation for loca l fishermen, now able to t rave l further in the same amount of time. The proportion of balsa among the f ish ing craf ts used on Lake T i t i caca went from 100% in the ear ly 1940's (Schweigger 1930; F ied le r 1944), to about 50% in 1976 (Bustamante and Trevino 1976) and 28% of a sample of 96 fishermen in 1980. Local fishermen have adopted nylon g i l l n e t s even more rapid ly . The l a t t e r were introduced on the lake in 1952 (Laba 1979), and twenty years l a t e r had replaced almost a l l cotton nets (Farfan 1974; Bustamante and Trevino 1976). Their adoption i s l i k e l y to have implied a large surge in f i sh ing e f fo r t , because i t has allowed loca l fishermen to increase the size and number of the i r nets, or the dimensions of the i r t rad i t iona l t rawls . Moreover, i t 191 has allowed them to f i s h more often, since fewer operators were needed, and since drying was no longer required in-between f ish ing expeditions. In the case of the huayuna, for example, the replacement of cotton with nylon has meant a large increase in the size the mouth (from 15 to more than 500 square meters in some cases). Moreover, the nylon huayuna can be used everyday, even during the rainy season i f so desired, and s t i l 1 l a s t longer than a cotton one. F i n a l l y , because i t i s now made by assemblage of ready-made nylon mesh panels, i t s construction has become easier and less time consuming. Conversely, huayunas are now easy to dismantle and to turn into g i l l n e t s . Status of f i sh resources Environmental factors, the introduction of exotic species, the expansion of urban f i sh market, and a large increase of f i sh ing e f fo r t fo l lowing changes in f ish ing technologies are a l l l i k e l y to have affected the status of loca l f i sh resources. For example, i t has been claimed trout were responsible for the demise of the native species, through predation, competition, and the simultaneous introduction of parasites because the introduction of the trout in Lake Ti t ieaca approximately coincided with a decl ine in Orestias catches, (Vi l lwock 1972; Laba 1979). The disappearance of the umanto (0. cuv ier i ) and a strong decl ine in boga (0. pentlandi i ) catches in pa r t i cu la r , were blamed on the trout. After the col 1 apse ot the trout f ishery and the closure of the trout canneries in the la te 1960s, i t has been claimed that the trout had a l l but vanished from the lake (Everett 1977; Laba 1979), and that the more recently introduced s i l v e r s i d e (B. bonariensis) was at least p a r t i a l l y to blame for i t , as wel l as for the continuously low catches of native species (Wurtsbaugh 1974; A l fa ro et a l . 1982; Orlove 1986). Limited a v a i l a b l e evidence, however, shows that a substantial increase in total f ish ing e f f o r t , fo l lowing the introduction of new f ish ing technologies and the expansion of loca l markets, did contribute to the decl ine of the native species. In the Lago Pequeno for example, th is decl ine was at least i n i t i a t e d before the introduction of the trout. I ts occurence has been attr ibuted to the development of truck transportation and of a large urban market for f i s h in the neighbouring c i t y of La Paz (James 1936). As for trout respons ib i l i t y in the infestat ion of native species with various types of parasites, i t had been recorded much e a r l i e r in the century (Neveu-Lemaire 1906). F i n a l l y , the simultaneous introduction of wooden boats and of nylon g i l l n e t s , which allowed the modification of the huayunaccana t rad i t iona l f ish ing trawls are l i k e l y to have lead to a surge in f i sh ing e f fo r t . F i n a l l y , environmental factors such as the dramatic drought of the ear ly 1940s are l i k e l y to have had a major impact on Lake Ti t ieaca f i sher ies , for at least three d i f ferent reasons. F i r s t , because of i t s lowering of the lake l e v e l , a drought concentrates f i sh in a smal ler f i sh ing area where i t i s easier to catch. Second, i t reduces spawning and feeding areas for perimacrophytic species such as the carachis. During the drought of 1983, for example, a decrease by one meter of the l e v e l of Laguna Arapa reduced the spawning area of the boga to less than hal f i t s or ig inal size (DIREPE-Puno 1984). And t h i r d , a drought pushes shore dwellers into f i sh ing , because there is l i t t l e e lse for them to do. Again during the drough of 1983, f ish ing increased so much that the amount of lake f i s h marketed in the c i t y of La Paz increased by 300% (Coutts and Rojas 1984). 192 Catch estimates Data co l lec ted for th is study indicates a tota l catch in the Peruvian part of Lake T i t ieaca of 8,160 mt in 1980 (95% confidence i n t e r v a l : 6,490 to 9,830 mt ). The bulk of i t came from the Orestias genus (69%), fol lowed by the s i l v e r s i d e (15%), and the trout (14%), and the Trichomycterus, mostly mauri, only contributed 2%. This confirms the scarcity of the suche, and that of the boga (0. pentlandi i ) while supporting the c laim that the umanto (0^ cuv ier i ) i s now ext inct (Bustamante and Trevino 1976; Lauzanne 1981; Parenti 1984). This estimate can be compared with an e a r l i e r one based on a very extensive household survey, for which food consumption per household in the Southern region of Peru was recorded over a one year period (ENCA 1972). This survey demonstrated that f i s h accounted for only 2% of the per capita consumption of 66 grams of protein per day in the Puno department (Ferroni 1980). Taking into account the contribution of marine f i s h to th is consumption, and the exportation of lake f i s h to urban markets outside the A l t i p l a n o , or in B o l i v i a , the ENCA data y i e l d s an estimate of 3,301 mt for the year 1972, thus indicat ing an average catch of about 1 mt/year for each of the 3,040 fishermen censused (Bustamante and Trevino 1976). [Footnote: This may be underestimated by as much as 1,000 mt, because the shore dwel l ing population with the highest rate of f i s h consumption was not sampled separately from the rest of the rural population of the department.] Although higher than o f f i c i a l catch f igures of about 1,000 mt for Peru (MIPE 1980), and of 500 mt for B o l i v i a (Vergara 1980), the 1979-1980 f igure i s s t i l l very low in comparison to estimates of 120,000 mt for the standing biomass (Johannesson et a l . 1981) and of 20,000 to 160,000 mt for the annual productiv i ty of the lake (Richerson et a l . 1977). This leads to the hypothesis that loca l f i sh resources are g l o b a l l y underfished, or u t i l i z e d below the i r maximum b io log ica l potential (Orlove 1979). Even i f more valuable species may be rare or ext inct , i t i s l i k e l y that only a small proportion of the standing stocks of pelagic species found offshore (eg i s p i , s i l v e r s i d e ) , i s subject to f ish ing when they come to shore for feeding, or for spawning. 193 APPENDIX C: RESEARCH METHODS 1) F i e l d research For th is study, I have gathered information on f ish ing t e r r i t o r i a l i t y , f i sh ing pract ices, and a number of f i sh ing re lated a c t i v i t i e s not previously described, through di rect observation and in-depth interviewing. Although a number of ethnographic studies did mention the importance of f i sh ing in the loca l economy, few documented the practice of f ish ing i t s e l f , and most only a l luded to f i sh ing t e r r i t o r i a l i t y (Lev ie i l 1986). Even the description of Lake T i t icaca f i sher ies by IMARPE b io log is ts which provided a wealth of information on the various types of c ra f t and gear used by loca l fishermen had l i t t l e to say about TURFs and f ish ing y i e l d (Bustamante and Trevino 1976). I have used di rect observation to gather information on the operation of spec i f ic techniques, including boat bu i ld ing , balsa bu i ld ing , reed and weed c o l l e c t i o n (Lev ie i l et a l . 1985), trout aquaculture, trawl f i sh ing (Trevino et a l . 1981), g i l l n e t f i sh ing , and f i sh r e t a i l i n g (Av i la et a l . 1985). Unobtrusive observation proved impossible without f i r s t securing a verbal autorization from some l o c a l l y elected representative, and e n l i s t i n g the co l laborat ion of at least a few fishermen. Because I rare ly attempted observation in communities where I had not FIRST obtained such guarantees, I encountered few problems, s t i l l I was t o l d to leave two di f ferent communities. In-depth interviewing of a number of shore dwel lers , most of them fishermen, a lso proved a very f r u i t f u l endeavour. Although reserved i n i t i a l l y espec ia l l y with a white foreign observer, casual acquaintances were w i l l i n g to share some of the i r knowledge, and to voice the i r concerns. F i n a l l y , I interviewed shore dwellers not involved in f i sh ing for comparison purposes and members of the loca l bureaucratic e l i t e to ver i fy shore dwellers' accounts. Given TURFs' i l l e g a l i t y , questions re lated to them proved f a i r l y sens i t i ve and I had to use considerable probing and ind i rect quest ion ing to obtain non evasive and r e l i a b l e answers. I asked shore dwellers to report trespassing instead of enforcement in order to avoid se l f - i nc r iminat ion . For cross-checking purposes, I subsequently v i s i t e d the communities from which trespassors came from and asked loca l fishermen to indicate the areas from which they had been repe l led . 2. S t a t i s t i c a l data from o f f i c i a l sources In addition to the information gathered during f i e l d work, o f f i c i a l s t a t i s t i c s were co l lec ted whenever a v a i l a b l e . However, they proved of poor v a l i d i t y and r e l i a b i l i t y , an unfortunately common problem in Third World f i sher ies (eg: Munro and Loy 1978). a) O f f i c i a l regist rat ion l i s t s The fishermen 's regist rat ion l i s t held by the Coast Guard (Capitania) includes about 1,000 fishermen, or about a th i rd of the tota l (Bustamante and Trevino 1976), because fishermen avoid register ing themselves, since i t costs them time and money. Since th is l i s t lacks comprehensiveness, i t could not be 194 used as a sampling frame. The boat reg ist rat ion l i s t a lso suffers from th is same problem, although i t appears that a large proportion of boats i s registered because boats may be confiscated by the Coast Guard. Even, the o f f i c i a l f i sh vendors held by the Puno o f f i ce of the the Ministry of Fisheries (MIPE-Puno) suffers from a lack of comprehensiveness. Moreover, i t i s biased because i t records a higher proportion of urban vendors from the c i t i e s of Puno and J u l i a c a . b) O f f i c i a l catch s t a t i s t i c s Lake T i t icaca o f f i c i a l catch s t a t i s t i c s are i n v a l i d because they indicate the amount of f i s h traded in the c i t i e s of Puno and J u l i a c a instead of the total amount of f i sh harvested from the lake. They are also unre l iab le because of the data c o l l e c t i o n technique used. An untrained MIPE employee v i s i t s a couple of markets, once a day at best, and estimates the tota l amount of f i s h being traded by counting the number of r e t a i l e r s and v i s u a l l y estimating the amount of f i sh for sale in front of a non-random sample of vendors. This method overlooks f i s h r e t a i l i n g on a l l the other markets of the A l t i p l a n o , s e l f consumption by f ish ing households and f i sh export to distant markets. I t even f a i l s to accurately estimate f i s h trade in Puno and J u l i a c a c i t i e s , because i t samples a var iable proportion of vendors on a non random basis, and because with one v i s i t per market per day the MIPE misses the f i s h sold pr ior to the v i s i t as wel l as the f i s h of vendors s t i l l to come. c) O f f i c i a l price l i s t The MIPE establ ishes and enforces a l i s t of o f f i c i a l prices for each f i s h species and even some waterfowl. This l i s t provides a conservative estimate of f i s h prices because, except for the larger f i s h , f i sh vendors dodge price speci f icat ions by s e l l i n g f i s h by the heap or by units rather than by weight as the l i s t s t ipu lates . 3. F isher ies survey Because of the u n r e l i a b i l i t y and i r re levance of o f f i c i a l s t a t i s t i c s , I have used a survey of loca l f i sher ies to estimate fishermen's income from f ish ing . With th is survey I have co l lec ted data on f ish ing y i e l d s and on the time and resources f i s h i n household invest in f i sh ing and f i sh ing - re la ted a c t i v i t i e s . This survey included three components, each represented by a d i f ferent type of questionnaire: a Catch Assessment Survey (or CAS) to evaluate the annual catch of a sample of fishermen, a Coverage Check Survey (or CCS) to check for biases in the CAS, and a socio-economic survey. The CAS was designed to provide information on the da i l y catches and on the f i sh ing e f fo r t ( i .e. duration of f i sh ing t r i p s , number of nets used, number of crew members involved) of a sample of f i f t y co l laborat ing fishermen, over a twelve month period (August 1979 to Ju l y 1980). Two CCS were conducted approximately at six month i n t e r v a l s , to determine whether the f i sh ing y i e l d s and ef for ts of the co l laborat ing fishermen were representative. The socio-economic survey was used to gather data on fishermen's investments in f ish ing c ra f t and gear. The same sample was used for the CAS and the socio-economic survey, and a special sample was drawn for the CCS fo l lowing the recommendations of a s t a t i s t i c a l consultant (Chapman 1979). 195 a) Population, sampling frame and sampling units For th is study, the primary unit of analysis was defined as an indiv idual fisherman (a c ra f t operator owning at least a few nets, or a pushnet in the case of a foot fisherman) rather than either a Fishing Economic Unit (FEU) or a household, although most fishermen operate within the framework of a household economy, and al though most f ishery surveys focus upon f ish ing firms or FEUs. This par t i cu la r unit was selected for three reasons: F i r s t , because focusing upon the indiv idual fisherman s t i l l allowed for the ca lcu la t ion of f i sh ing costs and returns at the household and FEU l e v e l ; Second, because I had to compare the return to f i sh ing labour with the opportunity cost of indiv idual fishermen's labour; F i n a l l y , because a comprehensive sampling frame was already a v a i l a b l e (Bustamante and Trevino 1976), thus f a c i l i t a t i n g the adoption of a frame survey (Bazigos 1974a; 1974b). Although a l l boat operators had a chance of being selected, foot fishermen did not. Women and chi ldren in p a r t i c u l a r , who commonly engage in pushnet f ish ing in flooded areas, were excluded from the population studied. Male fishermen were asked about the involvement of the members of the i r household in such a pract ice, but th is practice may be more common in households which do not include an adult male fisherman, thus leading to i t s underestimation. F i n a l l y , the fear that fishermen might refuse to answer questions because of the i r involvement in i l l e g a l a c t i v i t i e s proved unfounded, since more than one co l laborat ing fishermen acknowledged his involvemnt in such a c t i v i t i e s . b) Sampling procedure For the CAS and socio-economic surveys, a two-stage c lus te r sample of f u l l time fishermen was drawn after s t r a t i f i c a t i o n . For four of these surveys f i f t y col 1 aborating fishermen (units of analysis) from 13 f ish ing communities (clusters) d ist r ibuted in four st rata corresponding to the four zones of the lake were selected on a lo t tery basis (Campbell 1981). For the CCS, a s t r a t i f i e d sample of 16 f ish ing communities, 4 by zone (strata) was selected. However, ind iv idual fishermen (units of analysis) could not be selected on a t ru l y random basis, because i t was not possible to predict how many fishermen would be coming back from a f ish ing t r i p in any given 24 hour period. F i e l d workers were thus recommended to sample fishermen exhaustively, or should that prove impossible, to sample them systematical ly . In either case, the f i e l d workers missed the foot fishermen because the l a t t e r did not operate from the beach where the former were standing. The rat ionale for s t r a t i f y i n g the f i sh ing population in four strata stemmed from geographical differences between the four zones of the Lake. I t was thought that a f a i r l y standard-sized sample of 50 fishermen (Smith 1983) would be appropriate given that only 5 basic types of FEUs could be ident i f i ed . Cluster ing was adopted for budgetary reasons, with only 3 c lus ters or f i sh ing communities per zone being selected for pract ical reasons. The 50 fishermen selected out of 3,040 (a proportion of 1.8%) were thus distr ibuted between 13 communities, with an average of 4 fishermen per community. For each s t rata , they were selected in proportion to the number of fishermen censused in 1976 and assumed to be s t i l l present. 196 SAMPLING FRAME FOR THE SELECTION OF 50 COLLABORATING FISHERMEN (1979-80) Zones Clusters Units Fishermen Sampling rat ios North Lake South Lake Small Lake Puno Bay 3 3 4 3 11 15 13 11 680 936 768 656 16.2 16.0 16.9 16.8 Total 13 50 3,040 16.4 Both c lusters and fishermen were selected within each zone in such a proportion that each censused fisherman had the same probabi l i ty of being selected, regardless of the size of the f i sh ing community he belonged to. In the second stage, fishermen from each selected community (cluster) were selected by lo t tery during meetings attended by a quorum of at least 60% of the fishermen from the communities selected. To compensate for the smal ler chance of ind iv idual fishermen from large communities for being selected in the second stage, each community was given a chance of select ion proportionate to i t s s ize in the f i r s t stage. And since a l l fishermen from the same strata had the same probabi l i ty of being selected, the scheme adopted i s equivalent to s t r a t i f i e d random sampling. d) Data c o l l e c t i o n methods Three d i f ferent methods of data c o l l e c t i o n were used: questionnaires administered by an interviewer for the socio-economic survey, s e l f administered questionnaires for the CAS, and questionnaires with actual weighing and sorting of the f i s h by the f i e l d worker for the CCS. Interviewing for the socio-economic survey involved the use of detai led questionnaires, usual ly within the home compound of the informant, by one the f i e l d assistants or by myself. Verbal pledges of conf ident ia l i t y proved much less convincing for one-time respondents than a pledge of mutual t rust and reciprocity through the r i t u a l g i f t of a handful of coca leaves. The 50 fishermen who had agreed to part ic ipate in the CAS were trained for a couple of days and given spring scales , notebooks, penci ls and recording forms. Although not a l l of them were l i t e r a t e or able to speak good Spanish, most had at least a member of thei r household who was able to f i l l his forms. A f inanc ia l compensation was given to them as an incentive for accurate recording, and as a compensation for the i r time. In addition the t i t l e of "Col laborat ing Fishermen of the IMARPE" was granted to them, with o f f i c i a l l e t t e r s of endorsement from the IMARPE director and small i dent i f i ca t ion cards. As an addit ional incent ive , fishermen were promised that upon completion of the study, o f f i c i a l diplomas would be granted to them and that a number of prizes representing a substantial amount of money would be sorted by lo t te ry during an o f f i c i a l ceremony to which those who would report thei r f i sh ing a c t i v i t i e s accurately would be inv i ted . 197 For the CCS, the cooperation of the loca l Beach Sargeant or, in h is absence, that of the President of the Community or that of a loca l school teacher was sought. Credentials and o f f i c i a l authorizations were presented to them and to any suspicious fisherman, and the objectives of the survey were explained. A small monetary compensation was paid to each informant to demonstrate that the f i e l d workers were not tax c o l l e c t o r s . For the CAS and socio-economic surveys, the response rates were very high. Of the 50 fishermen selected only three refused to col laborate and had to be replaced before the beginning of the CAS. Of those who started co l laborat ing , about ha l f a dozen got a member of thei r household to replace them while they migrated to the coast or to the jungle for seasonal labour, but they resumed co l laborat ion upon thei r return. Although none withdrew before the end of the CAS, a t t r i t i o n reduced the i n i t i a l sample by 10 between 1981 and 1984 when the l a s t survey was conducted (one died, two were j a i l e d , two moved to nearby towns, three moved to Lima, and two refused to answer questions). The ind iv idua ls who had inherited the former col laborator 's c ra f t and gear were taken as replacement whenever possible. e) Interviewers Most of the interviewing for the socio-economic surveys was carr ied by a research assistant , or an interpreter and myself. The presence of an interpreter did f a c i l i t a t e many an interact ion , because foreign looking outsiders are suspicious on the A l t i p l a n o , a hardly surpris ing fact given the tormented history of the region. The f i r s t assistant , a ret i red Adventist school teacher who had previously col laborated on a number of socio-economic studies, spoke both Aymara and Quechua, the two native languages. His past experience with survey work and his knowledge of the area proved invaluable in locat ing the often e lus i ve fishermen. His Adventist connections proved a mixed b less ing , however, since i t was not looked at p o s i t i v e l y by a l l fishermen. The second research assistant , a trained Quechua speaking anthropologist, a l so with considerable experience of f i e l d work in the area, contributed invaluable knowledge of loca l kinship systems. He a lso proved extremely tac t fu l and e f fec t i ve in the invest igat ion of sens i t i ve questions about TURFs' enforcement. f) Questionnaire design procedure For the questionnaires' design, I r e l i e d upon the advice of a consultant (Chapman 1979), upon the experience of IMARPE b io log i s t s with the census of Lake T i t ieaca f isher ies (Bustamante and Trevino 1976), and upon my own experience with the socio-economic evaluation of smal l - sca le f i sher ies in the Peruvian Amazon. I designed the socio-economic surveys fo l lowing Bazigos recommendations (1974a), and I conducted pretesting with a few fishermen from Ichu-Raya and Barco (Chucuito), and with former fishermen now working for the IMARPE. Although the information sought dealt e s s e n t i a l l y with pract ical aspects of f ish ing and f i sh ing - re la ted a c t i v i t i e s , wording was s t i l l a source of d i f f i c u l t i e s , and the help of Dr. B.S. Orlove, an anthropologist spec ia l i zed in Andean studies with the formulation of questions avoided many a confusion. Units of measurements a lso proved a constant source of d i f f i c u l t i e s because fishermen refer to heterogeneous measurements. 198 g) data processing and methods of analys is The forms and note books of the co l laborat ing fishermen were c o l l e c t e d and checked every month by IMARPE sc ient i s ts fo l lowing the instruct ions contained in a small manual. A l l the data from the CAS, CCS and socio-economic surveys was processed and analyzed fo l lowing standard procedures (Sonquist and Dunkelberg 1977) at the Div is ion of Environmental Studies (DES) of the University of Ca l i fo rn ia -Dav is . F i r s t i t was coded on standard 80 columns computer forms by assistants s p e c i a l l y trained for th is task who fol lowed detai led instruct ions recorded in codebooks prepared for th is purpose. The data was then keypunched on standard IBM cards, entered on the computer, and after standard v e r i f i c a t i o n procedures, stored on magnetic tapes. Because of the size of the data set involved (about 13,000 records for catch and ef for t data only) considerable data manipulation was then performed by the s p e c i a l i s t s of DES computing center (Peter Hunter and Rod Thompson) to prepare easy to manipulate data f i l e s . 199 APPENDIX D: CALCULATIONS OF FISHING COSTS AND REVENUES The various costs involved in f i sh ing and f i sh ing - re la ted a c t i v i t i e s include the depreciation and maintenance costs of the f i sh ing c r a f t , gear and outboard engine, the operating costs of the engine, catch r e t a i l i n g expenses and some miscellaneous expenses such as boat and fisherman regist rat ion costs. For each cost or frequency item I have ca lcu lated the arithmetic means, and when appropriate I have indicated i t s 95% confidence interva l and the size of the sample on which these estimates were based. The evaluation of f ish ing costs required a number of assumptions. I have had to assume that each fisherman uses a standard balsa or boat, and that the annual cost of each item i s equal to i t s average value in ear ly 1980, divided by the average number of years that i t could l a s t (straight l i n e method of depreciation with no residual value). I have considered the material expenses for g i l l net maintenance but not the occasional loss and theft of nets. For labour, I have ca lcu lated the number of man-days necessary for each operation involved in f i sh ing or in re lated a c t i v i t i e s . I have neglected the opportunity cost of c a p i t a l , for three reasons: f i r s t l y because there i s very l i t t l e i f any money-lending a c t i v i t i e s for p r o f i t on the Peruvian A l t i p l a n o , and secondly because in 1980, few banks were accessible for shore-dwellers to invest the i r savings. Even in the case of those few banks which had of f ices in the c i t i e s of Puno and J u l i a c a , both about a day away from most f ish ing settlements, the rates of interest offered were general l y lower than the i n f l a t i o n rate. F i n a l l y , i t was not j u s t i f i e d to take into account the opportunity cost of capi ta l in ca lcu la t ions of the hourly return to labour for f i s h i n g , when the l a t t e r had not been considered in ca lcu la t ions of the hourly return to labour for a l te rnat i ve a c t i v i t i e s . Costs have to be borne at a par t i cu la r point in time but returns are obtained year long. Since the Peruvian i n f l a t i o n rate was 67,7% in 1979 and 59.2% in 1980 (Becker 1983), i t had to be taken into account for ca lcu la t ions of f i sh ing p r o f i t a b i l i t y . I have thus chosen to estimate f i sh ing costs as i f they had been incurred at the middle of the catch reporting period ( i .e. January 1980) because i t dtermined two ha l f periods over which the effects of i n f l a t i o n could be expected to cancel one another, which greatly s imp l i f i ed c a l c u l a t i o n s , since i t el iminated the need to correct for monthly i n f l a t i o n on f ish ing costs. I t did assume, however, that no s ign i f i cant changes in the i n f l a t i o n or in the catch rates occurred between the two hal f -per iods considered, an assumption which reasonableness i s demonstrated by a v a i l a b l e catch data and consumer price indices. 1. Fishing costs In my ca lcu la t ions of f ish ing costs I have incorporated the costs of f i sh ing - re la ted a c t i v i t i e s carr ied by the members of the household which owns the FEU. This is equivalent to considering such a household as a v e r t i c a l l y integrated f ish ing enterprise invo lved in c ra f t bu i ld ing , gear making and maintenance, f ish ing and f i sh r e t a i l i n g . Computing the costs of f i sh ing and f i sh ing - re la ted a c t i v i t i e s j o i n t l y implies that the net returns to labour are assumed to be the same for both types of a c t i v i t i e s . 200 a) Balsa reed boats Only 28% of the surveyed fishermen use a 3.8 m long (+ 1.8 m, n=64) balsa for f i sh ing . Few (7.8%) purchase a ready-made one for S/4,850 (+ S/5,276, n=5). Most (90.6%) bu i ld thei r own, c o l l e c t i n g reeds from totora beds to which they gain access for a fee or sometimes from the i r own beds. At leas t 63% of those who bu i ld thei r own balsa have to purchase the 33 bundles of to to ra (32.3 + 1.7 bundles , n=59) at S/100 a bundle (+ S/42.2, n=20) f o r S/3,075 which includes the cost of transportation. To bu i ld thei r ba l sa , ha l f of the informants use a rope made out of chi 11ihua (Festuca do!icophyla) , a grass found on the nearby plateau of the Puna. RcTst (93.3%) of those who use i t have to buy two bundles of i t , or wawas, each worth S/900. The stems have to be soaked and braided (in 46% of the cases) or twisted (in 17% of the cases) to make about 100 meters of grass rope (107 + 6 m, n=57). The whole process takes s l i g h t l y longer for braiding than for twist ing but makes a stronger rope which makes the h u l l of the balsa smoother, thus s l i g h t l y faster and less tiresome to propel 1 . Braiding or twist ing the chi 11 ihua i s usual ly performed by women or elders while tending c a t t l e in three or four days, but f u l l time i t takes about one day. The other ha l f of the informants use a nylon l i n e which has to be purchased in r o l l s , or conos, each worth S/1,800. Two r o l l s are necessary for the construction of a bal sa. The same nylon l i n e can be used repeatedly for three successive balsas which reduces the l i n e cost to S/1,200 per ba lsa , or to S/900 per balsa i f used for four balsas. The increasing popularity of nylon can thus be explained by the economy of time and money i t represents. Furthermore, the nylon l i n e can be tightened very strongly which al lows for the construction of water-t ight and long - las t ing balsas. F i n a l l y , because i t offers less resistance to the water, i t makes the balsa faster and easier to handle. After the totora has dried for 3 to 4 weeks depending on the season, the construction of the balsa takes 3 days of f u l l time work during one of which the bui lder needs the help of another adult man. The l a t t e r i s given food, coca and a l c o h o l , and paid S/200 for his day of labor , or j o r n a l . There i s a discrepancy between the quoted price of a ready-made balsa and that of the materials included in i t s construction which r e f l e c t s the inc lus ion of transportation costs in the price of the reed bundles. Ready-made balsas are purchased in Chimu or Ch'ul luni (Puno) by fishermen l i v i n g nearby, who row or s a i l thei r newly acquired balsa back to thei r home, and who do not include the value of the i r time in that of the purchased balsa. Conversely, bundles of totora are purchased by fishermen from the mouth of the Have r i v e r , who have to pay for the transport of these reeds by truck. Ready-made balsas would be too heavy and bulky to transport by truck and i t would take a long and rather unsafe passage to s a i l them back. Because a balsa becomes waterlogged after a couple of months, thus dangerously heavy and hard to handle, i t has to be replaced p e r i o d i c a l l y , and most fishermen do so about twice a year (1.96 + 0.5 t imes, n=28). Most balsas (94%) are propel led with a 4.2 meters long (+.3m, n=61) wooden pole, the noquena, usual ly made of a branch of eucalyptus (E. globulus) , or of ceo I le (Buddleia l o n g i f o l i a ) for the shorter ones. This l imi ts the navigating range of a balsa to the shallow water areas of the totora beds, or totora!es , and 11achu beds, or 11 achales. The noquena can 201 l a s t an average of three years (2.9 + .5 years, n=56) and i s worth approximately S/2,900 (S/2,896 + 900, n=49) and each balsa comes with a couple of them (1.8 + . 3 , n=64). I sp i f ishermen, fo r t h e i r p a r t , p re fe r a 2.5 meters long s c u l l i n g oar, the pal a, because i t can be used at almost any water depth. I t i s a lso quieter than rowing, since the s c u l l i n g blade stays in the water. Only 22% of the sampled balsas carry a 3.7 m long bipod wooden mast usual ly made of c c o l l e and worth S/655, and 44.4% of them carry a totora s a i l , or ccasana worth S/700. The l a t t e r i s more often used as a small roof at night or when ra in ing, than as a s a i l . P l a s t i c sheets have become increasingly popular for th is purpose because of thei r greater v e r s a t i l i t y and d u r a b i l i t y . Balsas maintenance costs are almost n i l . Al 1 a fisherman needs to do i s to tighten a balsa which has sat in the sun and out of the water for a long time and to repair broken ropes. The ove ra l l expense incurred for the annual operation of a f ish ing balsa thus corresponds to the renewal of the bare balsa and to the depreciation of the accompanying implements. According to the ca lcu la t ions detai led in the table below the expenditure per balsa can be estimated at S/12,008 per year, and 10 days of labour. Item Purchase Value Expected Depreciation Cost (Soles) Use (years) (Soles/year) -Totora reeds 3,075 .5 6,150 C h i l l i h u a rope 1,800 .5 3,600 Labour 5 days .5 10 days Pole (S/2,896 each) 5,126 3 1,704 Mast (22% of S/655) 144 4 36 S a i l (44% of S/700) 311 .6 518 Total (balsa and gear) 10,456 + 5 days 12,008 + 10 days b) Wooden boats The majority of Lake T i t ieaca fishermen (72%) own a wooden boat 4.75 m long (4.73 + .2 m , n=69), 1.5 m wide (1.52 + .1 m, n=69), and 80 cm deep (82 + 4 cm, n=69). Most boats (63%) belong to a s i n g l e owner, some (15%) to a pair of kins (father and son, brothers), and the remaining few (5%) to three owners. However, in some areas l i k e the northwestern shores of Laguna Arapa, co-ownership i s more frequent. Less than 5% of the boats are b u i l t by the i r owners. Instead, spec ia l i zed boat bui lders bu i ld them on order, often on a nearshore piece of land belonging to the prospective owner. I t i s the respons ib i l i t y of the l a t t e r to purchase the material required for the construction of his boat. If the boat bui lder has to move from his community to that of the buyer to bu i ld the boat, the buyer must give h o s p i t a l i t y , in addition of a salary to the bui lder and to provide help whenever necessary. 202 A prospective buyer may a lso order a ready-made boat from one of the specia l ized boat bui lders of Jacantaya (Moho) Huancane, I s l a T a q u i l i , Cachipucara (Pi lcuyo) , J u l i , Yunguyo, I s l a Anapia (Yunguyo), and the B o l i v i a n i s land of Suriqui . The buyer does not have to provide hosp i ta l i t y then. Only in rare opportunities, i nvo l v ing a few boat bui lders from Huancane, do boat bui lders ac tua l l y bu i ld a boat and l a t e r look for a buyer. Second-hand boats can a lso be purchased at a f ract ion of the i r or ig ina l cost, as demonstrated by the proportion of second-hand boats (30.5%) in the regist rat ion l i s t of the Coast Guard in Puno. The fo l lowing table d e t a i l s the costs invo lved in the purchase of a boat. Boat construction expenses Cost (Soles) Wood (50 r i b s , 10 side boards, k e e l , transom, etc) 17,760 Na i l s and screws 4,910 Putty and glue 3,300 Caulking and Painting material 13,410 Labour (4 weeks at S/500 per day for the boat bui lder 19,600 and S/200 per day for his helper) Transport of material 10,720 Total value of 5 meter f i sh ing boat 69,700 Fishermen expect the i r boats to l a s t about 10 years (9.9 + 1.6 years, n=72), which is probably an underestimate, since many ident i fy the remaining years of use of the i r boat with those before a major overhaul, 5 to 8 years after launching, depending of the qual i ty of the wood used and of the construction. In addit ion, informants can rare ly judge how long a boat could l a s t , since few (16%) have owned a boat for more than 7 years . About 40% of the boats use s a i l s for propulsion. Most of them (77%) are made of tocuyo, a cheap fabr ic used to make potato sacks, the others (13%) are made of a cotton material used to make f l o u r bags, and some are made of p l a s t i c (10%). An average-sized s a i l i s worth S/3,635 (+ S/2,315, n=19) and can 1 ast 7.5 years (+ 1.8, n=28). In a d d i t i o n to the sai 1 , a mast and some additional rigging is necessary. Almost a l l masts (97%) are made of eucalyptus wood and the tota l value of the r i g , including the 5 meters long mast (+.6 m, n=28) i s worth about S/1,000 (+ S/300, n=20) and can l a s t about 8 years (8.4 + 2.4 y e a r s , n=27). Most boats (92%) are propel led With a combination of pole and oars. The pole is p a r t i c u l a r l y useful to push the boat to , or away from the landing spots, or along the channels cr isscrossing the totora beds. Most poles (91%) are made of eucalyptus wood. Like those used to propel balsas, they are about 4.5 meters long (+.3 m, n=70) and worth S/1,335 (+ S/145, n=42) each, although more than one can be found in a boat (1.4 + .2 po les per boat, n=27). Boat lloquenas l a s t for almost f i v e years (4.8 + 1 years, n=68), one and a hal f times more than balsa poles, because they are used only when the water i s too shallow, or the macrophyte vegetation too th ick , for the oars to be used. 203 Oars are usual ly (66%) made of eucalyptus, although c c o l l e i s a lso used (22.5%). Fishermen expect thei r 2.5 meters long oars (+.2 m, n=70) worth S/1,620 a p a i r (+ S/330, n=37), to l a s t f o r 4 years (+ 1.1, n=65). In addition to the oars, a pair of iron oarlocks or chumaceras worth S/997 (+ S/186, n=37), which can l a s t f o r 16.5 years (+ 4.7 y e a r s , n=55) i s needed. The annual expenditures incurred by a boat owner are summarized below. Item Purchase Value (Soles/) Expected Use (years) Depreciation cost (Soles/year) Hul l Sa i l Rig Pole Oars (1.4 pole) (1 pair) Oarlocks (1 pair) 69,700 3,635 1,003 1,870 1,620 997 10 7.5 8.5 5 4 16.5 6,970 485 118 374 405 60 Total 78,825 8,412 Wooden boats must be painted twice a year, although fishermen paint them less often (1.4 + .2 times a year, n=40). Expenditures in material for painting include paint, o i l and caulking material for a tota l of almost S/6,984 (+ 2,032, n=20). Each painting and caulking session requires about 9 man-days (9.1 + 2.6 days, n=35). Boats a l so have to go through a major overhaul s ix years af ter launching, during which rotten r ibs and boards are changed. This overhaul i s general ly combined with boat painting. I t requires about S/5,678 (+ S/2,191, n=30) of wood and n a i l s and 3.65 man-days (+ 1.2 days, n=40) of work. The expenses for boat maintenance are l i s t e d below. Expenses for boat maintenance Labour (man-days) Expense (Soles) Painting and caulking - Frequency: 1.4 + .2 t imes/year (n=40) - Mate r ia l : S/6,984 + 2,032 (n=20) - Labour: 9.1 + 2.6 days (n=35) 12.8 9,603 Repair - Frequency: .17 times/year - M a t e r i a l : S/ 5,678 + 2,191 (n= 30) - Labour: 3.65 + 1.2 days (n=40) Total 0.6 13.5 965 10,568 204 c) Outboard engines Few fishermen (14%) own an outboard engine which horsepower ranges from 12 to 40 hp, with a mean of about 25 hp (24.75 + 5.5 hp, n=8). Engine owners expect thei r engines to l a s t only 5 years ( 5 + 2 years, n=7), because few have owned the i r engines for long enough to know that they can l a s t 8 to 9 years on the lake. Their year ly expenses are S/15,216 (+ S/7,542, n=12) for maintenance and repair , and the i r monthly expenses S/8,941 {+ S/5,503, n=8) for gas and S/2,460 (+ S/1,476, n=ll) for lubr icants . In sum they spend S/152,028 per year on the operation and maintenance of the i r engines which depreciation cost vary between S/65,000 and 85,000 depending of the i n i t i a l value of the i r engines. d) Gi 11 nets Lake T i t icaca fishermen use nylon g i l l n e t s made of imported nylon mesh panels. They purchase those of meshsizes under 63 mm for S/17,000 and div ide them in three or four longitudinal sections which they use for native species. And thay purchase those of meshsizes equal or larger than 63 mm for S/19,000 and div ide them in two or three sections which they use for introduced species. They a lso div ide the large f l o a t s which they purchase second hand into smal ler units which they use to make the f l o a t l i n e of the i r nets. F i n a l l y they use small stones and o ld batteries as sinkers. Fishermen expect thei r nets to l a s t for ten years (9.9 + 1 year, n=151). Thus the average value of a net and i t s year ly depreciation cost can be ca lcu lated as detai led in the fo l lowing table . Cost per item (in Soles) Native species net (< 63 mm) Introduced species net (> 63 mm) Longitudinal section Nylon l i n e Floats Labour 4,960 2,700 900 2 days 7,910 2,700 1,350 2 days Cost per net 8,560 + 2 days 11,960 + 2 days Expected years of use 10 10 Depreciation cost per year 856 + .2 days 1,196 + .2 days G i l l n e t s do require maintenance, e i ther cleaning the organic detr i tus which c log the meshes and make the net too v i s i b l e in the water, or repair ing the holes resu l t ing from wear and tear. Interviewed fishermen claim to perform cleaning operations alone (21%), or with the help of an associate fisherman (8.8%), but mostly with the help of the women and chi ldren of thei r household (70.2%). They indicate that they clean thei r nets every two or 205 three weeks, and that they repair them only occas ional ly , though they perform both types of a c t i v i t i e s more frequently during the rainy season. The fo l lowing table indicates the mean material and labor expenditure for the cleaning and repairing of the nets of a standard FEU. These estimates also apply to the maintenance of trawl nets or huayunaccanas. Net Maintenance Operation Labour Material (per year) (Hours) (Soles/) Cleaning (n=50) 133.5 + 5.6 Repairing (n=50) 105.3 + 5.4 2,840 + 796 These are mean estimates for the sample of 50 co l laborat ing fishermen. The time dedicated by each fisherman to these a c t i v i t i e s i s ac tua l l y a function of the number of nets he owns. Given the r e l a t i v e l y large share of net cleaning and maintenance in f i sh ing expenditures, th is has to be taken into account. I have thus used for each fisherman a weighting factor equal to the rat io between the number of nets he owns and the average number of nets owned within the sample of 50 fishermen. e) Huayunaccana Fishermen claim to use thei r huayunaccanas for at least 15 years. Since none i s so ld brand new, and since fishermen were unable to evaluate the value of the i rs , I have had to estimate the value of the material and of the labour invo lved in the construction of a huayunaccana to estimate i t s depreciation cost, as detai led in the table be 1 ow. Expense categories Labour (Days) Cost (/Soles) 8 panels (meshsize <63 mm) 7 1arge f1 oats 8 rol 1s of nylon 1 ine 160 meters of nylon cord 30 136,000 2,100 24,000 16,000 Total 30 178,100 Depreciation (over 15 years) 2 11,873 f) Reta i l ing costs Fish r e t a i l i n g costs include transportation costs for the vendors and for the f i s h they s e l l , a municipal tax and a warm meal, as detai led in the table below. Fish vendors undertake two market t r ips per week, each of which takes 206 them a ha l f day to a whole day each according to market a c c e s s i b i l i t y . I have taken th is into account in my ca lcu la t ions of the time dedicated by each household to f i sh r e t a i l i n g according to market a c c e s s i b i l i t y . I have also taken into account the common practice among fishermen of the Lago Pequeno to s e l l the i r catch to B o l i v i a n f i s h buyers who come in the i r own boats to buy f i sh r ight on the lake, thus saving fishermen r e t a i l i n g expenses. Expense categories Costs per t r i p (Soles/) (Days) Frequency: 93.1 + 6.8 times/year (n=251) 93.1 Fare: S/328.7 + 55.4 (n=104) 30,592 Tax: S/23.37 + 3.2 (n=71) 2,175 Meal: S/229.7 + 44 (n=36) 21,382 Annual tota l 54,149 + 9,550 47 to 70 g) Miscellaneous costs. Fishermen must register themselves and the i r wooden boats at the Coast Guard headquarters in Puno, which may require more than one t r i p (depending on the whims of loca l bureaucrats) and a small annual tax. Some fishermen also use some addit ional implements to f i s h or to keep f i s h a l i v e , which depreciation costs has to be considered. Fishermen from Esca l lan i (Capachica), for example, use a small holding pen to keep the i r catch a l i v e unt i l i t can be marketed. S i m i l a r l y , a few fishermen own and use a small pushnet or sajjana, which they or a member of the i r household uses occasional l y to f i s h for carachis in shallow water areas. Either one of these i s worth about S/500 and can be expected to l a s t for 5 years or more, with an annual depreciation of S/100. 2. Fishing returns To ca lcu la te the gross return from f ish ing I have had to mul t ip ly the monthly catch of each species by the corresponding prices which I gathered from the o f f i c i a l price l i s t (Appendix C). However, I used B o l i v i a n prices for the trout and s i l v e r s i d e so ld by 16% of the co l laborat ing fishermen because prices are 2 to 2.5 times higher in B o l i v i a (Coutts and Rojas 1981). This lead to a 15 to 16% increase of the hourly return to f i sh ing labour as demonstrated by the table next page. Assuming that r e t a i l prices minus r e t a i l i n g costs represent the value of the f i s h used for subsistence or for non-monetary exchange, implies assuming that fishermen are free decision-makers, and not subject to any constraint in deciding how to dispose of the i r catch. Assigning to subistence goods a value based on what s e l l i n g brings or could bring i s , however, more conservative than assigning one based on what would have to be paid to purchase the same or comparable goods on a r e t a i l market. Using the l a t t e r implies imputing the vendor's margin of p ro f i t to the value of the good. 207 Costs and returns to f i sh ing labour (per FEU) Peruvian prices Actual prices Gross return 371,449 421,570 Total costs 110,554 110,554 Net return 260,895 311,016 /FEUHOUR 106 121 Return to /ALLHOUR 74 86 Labour /TRIP&DAY 758 917 3 . Opportunity costs To estimate the opportunity cost of f i sh ing , I have ca lcu lated the returns per hour of labour for a number of a c t i v i t i e s from published sources. I have estimated the return from c a t t l e fattening at 70 S/hr, for example, by actua l i z ing previous studies and correcting for i n f l a t i o n (Lewellen 1977: 62; L e v i e i l and Goyzueta 1984). For a c t i v i t i e s such as trade and transport,and coffee growing, estimates of the returns to labour were already a v a i l a b l e , and a l l I had to do was to correct for i n f l a t i o n . For coffee growing, I have corrected Painter's estimate of the return to labour (Painter 1982; 1984; 1986) for i n f l a t i o n only, although I bel ieved his f igures to be overestimated, because i t allowed me to ca lcu la te a conservative estimate of the r e l a t i v e difference between f ish ing and coffee growing returns. Painter's f igures of the gross value of coffee production are overestimated because he used a mid-season price rather than the much lower price a c t u a l l y obtained by producers for the i r crop. Coffee prices dropped suddenly at the end of the 1980 season, to the dismay of the producers who had been enticed into holding on to the i r crop by r i s i n g prices. Some of them actua l l y had to borrow money to pay thei r helpers (Painter 1984). Painter's f igures are a lso overestimated because he did not include depreciation costs in his ca lcu la t ions . The l a t t e r are substantial because new plantations have to be created pe r iod ica l l y (Painter 1982: 248). A very large amount of labour i s involved in gaining a plantat ion on the jungle and maintaining i t for f i v e years before i t becomes productive. After 20 years of production, th is plantat ion becomes unproductive, because of environmental conditions and of the lack of f e r t i l i z e r s and pest ic ides: the s o i l f e r t i l i t y i s l o s t , the top s o i l i s eroded, the plants are invaded with pests and diseases. Each prospective coffee grower must a lso invest a considerable amount of time to learn the trade, to create and to maintain a pool of labour from which to draw help. 208 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0097449/manifest

Comment

Related Items