UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Dilemmas in decision making : a methodological test case in economic anthropology Prattis, James Ian 1970

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

Item Metadata

Download

Media
831-UBC_1970_A1 P73.pdf [ 11.42MB ]
Metadata
JSON: 831-1.0103965.json
JSON-LD: 831-1.0103965-ld.json
RDF/XML (Pretty): 831-1.0103965-rdf.xml
RDF/JSON: 831-1.0103965-rdf.json
Turtle: 831-1.0103965-turtle.txt
N-Triples: 831-1.0103965-rdf-ntriples.txt
Original Record: 831-1.0103965-source.json
Full Text
831-1.0103965-fulltext.txt
Citation
831-1.0103965.ris

Full Text

DILEMMAS.GF DECISION MAKING:  A METHODOLOGICAL  TEST CASE IN ECONOMIC ANTHROPOLOGY  by JAMES IAN PRATTIS B . L i t t , Oxon, 1967  A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n the Department of ANTHROPOLOGY AND SOCIOLOGY  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA July, 1970  In  presenting  this  an a d v a n c e d  degree  the  shall  I  Library  further  for  scholarly  by h i s of  agree  this  thesis  in p a r t i a l  fulfilment  of  at  University  of  Columbia,  the  make  it  that permission  available  for  It  financial  is  gain  of  The U n i v e r s i t y o f B r i t i s h V a n c o u v e r 8, C a n a d a  by  the  Columbia  shall  not  requirements  reference copying of  I  agree  and  copying or  be a l l o w e d  for  that  study.  this  thesis  Head o f my D e p a r t m e n t  understood that  written permission.  Department  for  for extensive  p u r p o s e s may be g r a n t e d  representatives. thesis  freely  British  the  or  publication  without  my  ABSTRACT  The problem addressed here i s the examination of the dilemmas of decision making i n d i f f e r e n t substantive contexts.  The contexts i n -  clude peasant farmers deciding whether or not to accept an a g r i c u l t u r a l innovation, sophomore students gambling, and fishermen on B r i t i s h Columbia's West coast making up their minds where they intend to f i s h . The major conclusion i s that the structure of decision making i s a constant  c r o s s - c u l t u r a l variable.  This implies that s o c i o - c u l t u r a l  factors can most p r o f i t a b l y be viewed as a framework within which a similar structure of decision making occurs.  From this consideration  i t follows that decision making i s a basic b u i l d i n g block for the study of s o c i a l behavior. I d e n t i f i c a t i o n of the basic structure of decision making i s i n terms of a theory of r i s k taking which relates the type of decision strategy used i n any s i t u a t i o n to considerations of the resources, i n formation and u t i l i t i e s that p a r t i c u l a r individuals possess with regard to the event the decision i s about. From an i n i t i a l substantive concern with Third World farmers deciding to adopt or reject a g r i c u l t u r a l innovations a number of statements about individuals and r i s k .  I generalise to Risk taking refers  to behavior i n situations where there i s a desirable goal and a lack of certainty that i t can be achieved with attendant p o s s i b i l i t i e s of loss.  - ii -  Three main sources are used to test the assertions about r i s k taking —  f i r s t a laboratory experiment, then fieldwork and f i n a l l y  secondary sources.  The argument made to j u s t i f y this procedure i s  that these situations constitute p a r t i c u l a r empirical settings i n which the  propositions about r i s k taking could legitimately be tested.  The  argument rests on the assumption that the same scope conditions are met i n each substantive s e t t i n g . The scope conditions considered here place an i n d i v i d u a l decision maker within parameters of resources and subjective  utility  with regard to some outcome, information and incentive conditions for  any r i s k .  The propositions predict the type of decision strategies  that would be employed for given values of the above parameters.  This  l e v e l of abstraction, which i s not tied to s i t u a t i o n a l boundaries i s , I submit, a necessary prerequisite f o r e f f e c t i v e c r o s s - c u l t u r a l analysis.  Thus my thesis i s not about peasant farmers, or fishermen  or gambling, the work attempted here i s concerned with individuals and r i s k . The tools^used draw upon a t r a d i t i o n of model building extant i n economics with reference to decision theory.  Thus the work  attempted here i s part of the growing formal t r a d i t i o n i n economic anthropology. The model b u i l t was not acperfect f i t with data but adequate enough to give one confidence i n the set of methodological assumptions which were fundamental to i t s construction.  Also the testing procedure  - iii -  employed has implications for the manner i n which anthropologists  may  conduct enquiry, as i t establishes that laboratory contexts are as legitimate a source of v e r i f i c a t i o n as f i e l d  contexts.  It i s from these two considerations that I o f f e r a test case for methodology i n economic anthropology.  - iv -  TABLE OF CONTENTS  Page Abstract  i  Chapter 1:  Introduction Economic Anthropology: Controversies i n Methodology — The Anthropology of Choice — Progression of Thesis.  Chapter 2:  The Problem and Its Conceptualization Introduction — Farmers and Innovations — LWHU State of Nature — Some Propositions — Supplementary Propositions — S o c i a l Mobility — Social Marginality — Cooperative Structure — Whither Next? — Decision Theory — Theoretical Propositions — Testing — Summary.  Chapter 3:  Test Case: The Experiment Introduction — Experimental Design — Decision Task and Incentive Conditions — Wealth Gradient — Playing for Real Stakes — Information — Subjective U t i l i t y — Data C o l l e c t i o n — Testing — Farmers, Risks and Experimental Data — Implications for F i e l d Test.  Chapter 4:  Test Case: Fishermen as Risk Takers Introduction — The Fishing Industry i n B r i t i s h Columbia — G i l l n e t t i n g — Constraints — Fieldwork — Data C o l l e c t i o n — Risk — Wealth — U t i l i t y — Information — Strategies as Risks — Rivers Inlet — Johnston S t r a i t s — Closure of Northern Fisheries — Investments on Technology as Risk — Case No. 1: Ernie M c A l l i s t e r — Case No. 2: N e i l Anderson — Test of Propositions — Strategy Risks — Investment Risks — Congruence Between Investment and Strategy Risks — Relevance of Cultural Factors — Summary.  1  19  56  96  - v -  Page Chapter 5:  Farmers as Risk Takers Introduction — Case No. 1: Malayan Rubber — LW Farmers and Risk — Wealth and Risk — Information and Risk — U t i l i t y Considerations — Moturiki Community Development Scheme — Food Fishing i n A l e r t Bay — Spanish American Farmers and Hybrid Corn — Some Implications.  157  Chapter 6:  Conclusion Introduction — Methodological Implications — On Models and Research Cycles — Editing — Intervening Variables — (1) Education — (2) Social Marginality — Scope Conditions: The U t i l i t y Parameter — New Problems — Theoretical Implications — P r a c t i c a l Implications — Conclusion.  187  Bibliography Appendices:  216 Appendix I:  Duplex Gambles Master Sheet  223  Appendix I I : Instructions to Subjects  224  Appendix I I I : Transformation of Experimental Data, Tables IV-VII.  256  Appendix IV: Chi Square Median Test on Propositions.  260  Appendix V: Scale f o r Wealth S t r a t i f i c a t i o n for A l e r t Bay G i l l n e t t e r s .  276  - vi -  LIST OF ILLUSTRATIONS  Page TABLES Table I:  Summary of Scope Conditions  65  Table I I :  Data from Experiment  67  Table I I I :  Bets Categorised by Riskiness  68  Table IV:  Bets Categorised Per Subject  256  Table V:  Frequency of Bets Played Per C e l l  257  Table VI:  Raw Data Matrices  258  Table VII:  Corrected Data Matrices  259  Table VIII:  Chi Square Median Test:  Proposition 1'  74  Table IX:  Chi Square Median Test:  Proposition 2'  79  Table X:  Chi Square Median Test:  Proposition 3'  83  Table XI:  "Sure Thing" Bets  86  Table XII:  Low Payoff Bets  87  Table XIII:  Relative Weighting of Risk Components  89  Table XIV:  Summary of Experimental Test  90  Table XV:  Risks on Technology  Table XVI:  Information C i r c u i t s  113  Table XVII:  Strategy Risks i n Rivers Inlet  117  and Strategies  106  Table XVIII: Strategy Risks i n Johnston S t r a i t s  118  Table XIX:  Northern Closure and Strategy Risks  '  120  Table XX:  Investment Risks and Range of E x p l o i t a t i o n  124  - vii-  Page  Table XXI:  Risk Taking by Ernie M c A l l i s t e r  134  Table XXII:  Risk Taking by N e i l Anderson  139  Table XXIII:  Data on A l e r t Bay G i l l n e t Fishermen  140  Table XXIV:  Strategy Risks  141  Table XXV:  Proportion of High and Low Risks Taken by A l e r t Bay G i l l n e t Fishermen  142  Investment Risks by Japanese and Yugoslav G i l l n e t Fishermen  150  Table XXVI: Table XXVII:  Adoption of Hybrid Corn by Spanish American Farmers 181 FIGURES  Figure 1:  States of Nature  22  Figure 2:  Propositions 1 - 3  28  Figure 3:  Propositions 4 - 6  31  Figure 4:  General Model  45  Figure 5:  Payoff Matrix f o r Innovations  47  Figure 6:  Duplex Gamble  53  Figure 7:  Gambling Wheel  59  Figure 8:  Data Matrix  66  Figure 9:  Data Matrix  70  Figure 10:  Information C i r c u i t s Model  145  Figure 11:  Revised Information C i r c u i t s Model  202  - viii -  Page GRAPHS Graph 1:  LW Preference for Low Risk Strategies: Proposition 1 (SR Bets)  71  LW Preference for Low Risk Strategies: Proposition 2 (FR and VR Bets)  72  Graph 3:  Risk Taking with Respect to Wealth (SR Bets)  77  Graph 4:  Risk Taking with Respect to Wealth (FR and VR Bets)  78  Graph 5:  Risk Taking witn Respect to Information (SR Bets)  81  Risk Taking with Respect to Information (FR and VR Bets)  82  Graph 2:  Graph 6:  MAPS Map  1:  Coastal Fisheries of B r i t i s h Columbia  99  - 1-  CHAPTER 1 INTRODUCTION The main s t i m u l i f o r this d i s s e r t a t i o n were considerations of underdevelopment i n the Third World and a concern with methodology i n economic anthropology.  In a sense I am combining idealism with a  philosophical basis f o r enquiry, as both led me to focus on dilemmas of decision making. My own personal experiences i n community development projects i n Greece and Sarawak l e f t me with a broad general interest i n development problems.  However, the translation of this i n t e r e s t into  workable models was hampered by the vast number of variables seemingly pertinent to the development process, and the fact that my i n i t i a l attempts to explain the development process merely offered another plausible account rather than a d e f i n i t i v e study. One theme that remained  consistent from my own f i e l d ob-  servations and e a r l i e r work (Prattis:  1967) was that I suspected an  i m p l i c i t r a t i o n a l i t y lay behind the decisions made by the peasant farmers I was working with. The parameters or framework within which decisions were made varied, but I f e l t that the process of weighing up the advantages and costs attached to any alternative, subject to external constraints, was predictable. attention.  I t was to this type of problem that I turned my  - 2-  Rather than deal with the whole range of peasant decision making I decided to confine my attention to decisions made about new agricultural practices.  Thus my i n i t i a l substantive concern i s with  the c r i t e r i a by which Third World farmers decide to accept or reject innovations. This concern involves a movement away from broad general perspectives about the development process and attempts rather to solve a p a r t i c u l a r problem i n addition to simply i d e n t i f y i n g a problem area. To remain with a broad, .general perspective places the anthropologist at a disadvantage with regard to the application of h i s creed.  In the p r a c t i c a l world of development planning and the anth-  ropologist's contribution to i t , Moerman has stated "Anthropological notions of c u l t u r a l systems and of i n s t i t u t i o n a l interconnections are so imprecise that he has no s p e c i a l competence, aside from immersement i n native l i f e , to make the predictions that his job description and s e l f - d e f i n i t i o n often demand" (Moerman 1968: 84). To a l t e r this portrayal of the applied anthropologist requires an increasing concern with p a r t i c u l a r "tool k i t s " that can be successfully applied to s p e c i f i c problems.  And i t i s this concern  with tool k i t s that brings me to the second major stimulus i n my research —  methodology i n economic anthropology.  - 3-  Economic Anthropology:  Controversies i n Methodology  Anthropologists have long argued among themselves and with economists regarding the range of a p p l i c a b i l i t y of economic theory, so much so that there has been a p o l a r i z a t i o n into two d i s t i n c t camps —  the formalists and the s u b s t a n t i v i s t s . Polanyi's d i v i s i o n of the term "economic" into two d i s t i n c t  meanings i s basic to the whole controversy that has divided scholars i n economic anthropology  for over t h i r t y years.  Polanyi discriminates the substantive from the formal meaning of economic.  By the former he refers to man's interchange with  his natural and s o c i a l environment " i n so far as this results i n supplying him with the means of material want s a t i s f a c t i o n " (Polanyi 1957: 243). The l a t t e r derives from the l o g i c a l character of the meansends r e l a t i o n s h i p , " i t refers to a s i t u a t i o n of choice, namely, that between the d i f f e r e n t uses of means induced by an i n s u f f i c i e n c y of those means" (1957: 243). This categorization covers what i s known as formal economics, and r e s t r i c t s one's range of study to choice i n duced by s c a r c i t y s i t u a t i o n s .  The implication i s that the communi-  ties anthropologists study do not always have the same s c a r c i t y problems possessed by i n d u s t r i a l economies. Dalton goes on from Polanyi's d i s t i n c t i o n and argues that the substantive aspect of economics i s the sphere of enquiry for economic anthropologists (Dalton 1961, 1969).  - 4 -  He opposes the use of formal economic theory on the grounds that i t grew out of and was made to account f o r situations of indust r i a l economies and i s thereby inapplicable to any other s i t u a t i o n . "The differences between primitive economic organisations ( i . e . where market transactions and produce are absent or present only i n petty amounts) and our own are so great that a s p e c i a l set of concepts, leading ideas and terms are necessary to analyze these subsistence economies... the concepts of conventional economics r e l a t i n g to economic organisations are not f r u i t f u l l y applicable outside of market systems" (1969: 65). This i s the key to Dalton's whole argument; yet i t remains unclear as to what he means by market transactions.  I t i s also not  clear whether the d i s t i n c t i o n he i s making i s between the form of market transactions ( i . e . physical market place) or the structure of exchange. If he refers to the former, he has missed the point that i t i s not necessary to have a physical structure to indulge i n market transactions.  I f he i s r e f e r r i n g to the l a t t e r , then he i s c l e a r l y  wrong as there i s no evidence to indicate that exchange i n market and non-market economies d i f f e r s i n terms of structure.  In both setups the  actors involved i n exchange are balancing payoffs against costs i n terms of the p a r t i c u l a r u t i l i t y dimensions paramount at the time of the exchange.  Dalton seems to be confusing substantive situations  with a n a l y t i c a l p r i n c i p l e s . Robbins devastates  the m a t e r i a l i s t d e f i n i t i o n of economics  which was l a t e r employed by Polanyi and Dalton and argues that whether or not a good or service i s material has nothing to do with whether i t i s economic (1935).  - 5 -  He defines economics as the science which studies human behavior as a relationship between ends and scarce means which have a l ternative uses.  This d e f i n i t i o n has frequently confused  gists of the substantivist persuasion only to western i n d u s t r i a l society.  anthropolo-  as they f e e l i t i s applicable However Robbins i s quite clear  about the range of a p p l i c a b i l i t y of h i s d e f i n i t i o n : "We do not say that the production of potatoes i s economic a c t i v i t y and the production of philosophy i s not. We say rather, that insofar as either kind of a c t i v i t y involves the relinquishment of other desired a l t e r n a t i v e s , i t has i t s economic aspect." (Robbins 1935: 15) This consideration and weighing up of alternatives further implies that economic anthropology i s about choice.  In fact F i r t h  makes the problem of choice a major point of emphasis i n the study of s o c i a l organisation.  He states:  "The working arrangements by which a society i s kept i n being, the ways i n which relations between groups are made operative and become e f f e c t i v e rest upon i n d i v i d u a l choice and decisions." (Firth 1964a: 46) Implicit i n the writings of the early economic  anthropolo-  gists such as F i r t h (1939), G-oodfellow (1939) and Herskovits 1952)  (1940,  i s the notion of r a t i o n a l choice b u i l t around a p r i n c i p l e known  as the "calculus of maximisation" (LeClair and Schneider 1968:  6).  This assumes that people make decisions and choices i n a r a t i o n a l manner, between known a l t e r n a t i v e s .  I t also assumes that  the choice i s made according to determinable p r i n c i p l e s . From this one can i n f e r that any human a c t i v i t y directed to choosing between alternatives has  the component of economizing.  Economizing i s the  - 6 -  a l l o c a t i o n of scarce resources among alternative ends, which means that an individual can economise with respect to any scarce resource. There are a number of conclusions which can be drawn from the discussion so f a r . F i r s t of a l l the logic of the premises of formal economics i s that economic theory i s not so much concerned  with  a n a l y t i c a l p r i n c i p l e s for western, i n d u s t r i a l i z e d society but with a way of looking at behavior. Secondly the assumptions used by an economist, about resources, wants and choices, i n his formal analysis are s u f f i c i e n t l y abstract so as to be applicable to any human society. For instance the theory of maximization, which i s at the heart of neoclassical economic theory says nothing about what i s maximized. I t i s generally assumed that p r o f i t i s being maximized, but this represents one application of the theory and not maximization  theory  itself. One could say that i t i s u t i l i t y which i s being maximized -the value which an individual places on certain ends.  The end can  vary -- maximize p r o f i t , r i t u a l s o l i d a r i t y or status, and decisions taken with respect to that end are constrained by the u t i l i t i e s attached to competing ends, but maximization  theory i s held to be universally  applicable (Robbins 1935; Burling R. 1962). The third important inference to make i s that Polanyi's implication that some primitive societies do not have the same scarcity problems as possessed by i n d u s t r i a l economies i s c l e a r l y wrong.  I  - 7 -  stated e a r l i e r that one can economize with respect to any scarce resource, and the process of economizing  i s the same whether i t i s ap-  p l i e d to material commodities or intangible goods such as prestige. This implies that the structural implications of scarcity are a major focus for enquiry i n economic anthropology. Systematic rebuttals of the substantivist'.s position have been provided by Burling (1962), L e C l a i r (1962) and Cook (1966). , They show the l o g i c a l and empirical d e f i c i e n c i e s i n the s u b s t a n t i v i s t s argument and Cook points out very succinctly a gross  1  misunderstanding  of the nature of formal economics by both Polanyi and Dalton. Dalton's argument that formal economics i s based on 19th century -industrialization i n B r i t a i n merely alludes to a particular system of p o l i t i c a l economy and the 19th century European market system, and t o t a l l y ignores the subsequent refinements, modifications and new additions to economic science. Dalton seems not to acknowledge the fact that though one  may  use concepts from other d i s c i p l i n e s this does not necessarily e n t a i l the use of the same operational d e f i n i t i o n s for such concepts. The substantive argument tends to confuse analytic p r i n c i p l e s with the s p e c i f i c contexts to which they are applied.  Dalton's con-  cern with distinguishing separate economic contexts with separate anal y t i c p r i n c i p l e s misses the point of s c i e n t i f i c enquiry.  I would submit  that our concern should primarily be with p r i n c i p l e s , and that we use d i f f e r e n t substantive contexts to test the range of a p p l i c a b i l i t y of the p r i n c i p l e s .  - 8 -  Within the wider problem area of development I have elected to concentrate on the choice processes that underlie the acceptance and r e j e c t i o n of innovations by peasant producers.  This i s an i n i t i a l  substantive context from which I intend to generate a set of more general statements. The need for wider generality can be met by borrowing concepts from economics, as considerations of choice behavior are fundamental to modern economic theory. I r e j e c t the substantivist orientation as i t s concern with unique contexts prevents g e n e r a l i s a b i l i t y .  The concern with choice  behavior i s not new i n anthropology (see below), though i t lacks explanatory power on a more general cross c u l t u r a l l e v e l .  The synthesis  of the two stimuli to my research i s i n terms of development interests bringing me to the subject area of choice behavior, while methodology suggests that I borrow analytic tools from economics i n order to study it.  The Anthropology of Choice  I f economic anthropology i s concerned with problems of scarc i t y and choice then i t would seem necessary to construct models of choice behavior which can predict the type of decisions made as cons t r a i n t s of scarcity vary.  - 9 -  Raymond F i r t h emphasised as long ago as 1939 that the study of decision making should be a major focus for enquiry, but i t i s only i n recent years that systematic attempts have been made to explicate the process of decision making and i t s structural implications. The concept of choice i s fundamental to Barth's notion of a transactional model (Barth 1966).  He maintains that i n any s o c i a l re-  lationship we are involved i n a flow and counterflow of prestations, of appropriate and valued goods and services.  He uses the notion of  flow and counterflow of prestations to determine the extent to which p a r t i c u l a r statuses can interact.  The term 'transaction' denotes  those sequences of i n t e r a c t i o n which are systematically governed by reciprocal behavior. Barth i s concerned with the incentives and constraints on the choice behavior of an individual actor i n terms of strategies opted f o r and the calculations of gains and losses. can be along both s o c i a l and material dimensions.  This c a l c u l a t i o n  By considering d i f -  ferent dimensions of value Barth points out how a l l parties can gain from a transaction. To i l l u s t r a t e his arguments Barth analysed the s o c i a l  organ-  i z a t i o n of a Norwegian seine boat to show how the t r a d i t i o n a l and expected relationship between statuses -- captain, netboss and crew -was changed as a function of the transactional nature of s o c i a l r e l a tions on board ship (Barth 1966).  - 10 -  On the seiner he noted that the t r a d i t i o n a l subordination crew to captain was  relaxed.  The crew spent a great deal of time on  the bridge and also monitored the equipment that was f i s h and navigate.  of  used to locate  This went against maritime convention which speci-  f i e d the bridge as the exclusive domain of the captain, unless a crewman  was  on watch. In his analysis Barth focussed  and how goal was new  on the goals of the captain  these were effected by the use of his authority. to maximize f i s h intake and  type of seine net.  This new  His primary  to this end he had invested i n a  technology, which enabled the captain  to better achieve his goal, required a committed and well trained crew as the co-ordination of a c t i v i t y was the old types of d r i f t n e t .  greater than that required with  Mistakes or slowness on the part of the  crew at c r u c i a l periods of the workcycle could lead to loss of catch or damage to gear. As a r e s u l t of these new and crew entered  expectations  of crewmen, the captain  into a kind of bargaining relationship.  As part of  the transaction, the crew were allowed priveleges which impinged upon the captain's t r a d i t i o n a l authority i n return for the extra performance required with the new  technology.  The captain could maintain his t r a d i t i o n a l authority, but only at the cost of his primary goal -- maximization of catch. change i n technology and the d i f f e r e n t expectations crewmen which resulted, permitted  The  of labour from the  a set of transactions which called  - 11 -  for a p a r t i a l relinquishment of authority by the captain and increased performance on the part of the crew. Barth maintains that the necessity of choosing between a l t e r natives ( i n this case authority and size of catch) can operate to change values as new decisions become systematised. this new  He also emphasises that  consensus of values has to adjust to further experiences and  events. Blau takes as his starting point the notion of exchange as a means of characterising social interactions.  He defines s o c i a l exchange  as "voluntary actions of individuals that are motivated by the returns they are expected to bring, and t y p i c a l l y do i n fact bring, from others" (Blau 1964:  91).  He specifies that exchange takes place when s p e c i f i c  resources are d i f f e r e n t i a l l y d i s t r i b u t e d , and interaction permits the flow of these resources between actors.  He i l l u s t r a t e s this with an  analysis of s o c i a l processes i n a work group (1964). The exchange commodities considered i n his analysis are advice and s o c i a l status which he relates i n the following manner: asking for advice raises the status of the advisor, giving advice l o wers the status of the advisee.  Blau extends this assumption i n terms  of a consideration that i f advice on a work problem i s highly valued then other favours are i n s u f f i c i e n t to discharge one's obligations for it.  I t then becomes necessary to reciprocate for advice with respect  and compliance, visor.  i n other words accept a lower status v i s - a - v i s the ad-  The resources d i f f e r e n t i a l l y d i s t r i b u t e d i n this example are  status, as a function of compliance,  and advice.  - 12 -  Recalling Barth's remarks about a l l parties p r o f i t i n g from a transaction, the exchange i n the work group results i n a mutual prof i t for the participants. I f one labels Person A as the advisee and Person B as the advisor i n the workshop situation then Person A p r o f i t s from the advice that enables him to perform his job better, and Person B p r o f i t s from the superior status he i s accorded i n exchange for his expert advice. Other authors have looked at choice behavior as fundamental to a general notion of exchange and have viewed the process of s o c i a l interaction i n cost-benefit terms. Homans i n his essay "Social Behavior as Exchange" makes his major point with respect to considering aspects of s o c i a l i n t e r a c t i o n from the point of view of c a l c u l a t i o n of "costs", "rewards" and "prof i t s " , and views p a r t i c u l a r decisions as the outcome of balances or imbalances i n the calculations (Homans 1958). The g e n e r a l i s a b i l i t y of the notion of exchange i s documented by Gouldner (1960).  Drawing upon Malinowski and Mauss, Gouldner hypo-  thesizes that a norm of r e c i p r o c i t y i s universal -- a dimension to be found i n a l l value systems. "We owe others certain things because of what they have previously done for us, because of the history of previous i n teraction we have had with them. I t i s this kind of o b l i gation which i s entailed by the generalised norm of reciproc i t y ".(I960: 171).  - 13 -  Gouldner emphasizes the important d i s t i n c t i o n between complementarity and r e c i p r o c i t y .  He notes that the former connotes that one's  rights are another's obligations, and vice versa.  On the other hand,  r e c i p r o c i t y denotes that each party has rights and duties. Gouldner argues that i f there were only rights on the one side and duties on the other, then there would be no exchange as there can only be stable patterns of r e c i p r o c i t y qua exchange insofar as each party has both rights and duties.  The significance of r e c i p r o c i t y as  exchange for r o l e systems i s that i t tends to structure each r o l e so as to include both rights and duties. This all-pervasive aspect of exchange i s attested to by Levi-Strauss: "Goods are not only economic commodities, but vehicles and instruments for r e a l i t i e s of another order, such as power, influence, sympathy, status and emotion and the s k i l l f u l game of exchange consists i n a complex t o t a l i t y of conscious or unconscious manoevres i n order to gain security and to guard oneself against r i s k s brought about by a l l i a n c e s and by r i v a l r i e s . " ( 1 9 6 7 : 54) A l l the authors I have mentioned i m p l i c i t l y view s o c i a l behavior as an "economy" with calculations of rewards, costs and optimum points as functional to choice and decision making. Blau (1964) and Barth (1966) go further i n that their view of exchange suggests that one can view s o c i a l structure not as a given, but as the outcome of choice behavior. seem imperative  From this viewpoint  i t would  to construct models of how people make decisions under  varying circumstances,  as this endeavour would then explicate what may  be a basic building block for the study of s o c i a l  behavior.  - 14 -  Though the discussion has moved from choice to exchange i t rests on the same assumption of calculus of maximization and r a t i o n a l i t y i n choice.  It also assumes that the tools of economics -- decision  theory and maximizing models -- can also be the tools of the anthropologist. These are the " i n p r i n c i p l e " p o s s i b i l i t i e s , and annumber of anthropologists have endorsed the idea of using formal decision theory to solve anthropological problems (Buchler and Nutini 1969:  Davenport 1960).  However a balance between anthropological data and formal models has to be kept clear i n that more s t r i c t l y anthropological considerations serve to i d e n t i f y the constraints and parameters within which decision models can be applied. This means that ethnographic description of p a r t i c u l a r value systems and cultures permits one to designate the c u l t u r a l l y perceived alternatives that are appropriate to p a r t i c u l a r decision s i t u a t i o n s , and to demarcate the p r i n c i p l e s which are determinate (or seem to be) for choosing between the a l t e r n a t i v e s . And  i t i s i n this way  that the anthropologist, with his t r a -  d i t i o n a l tool k i t of analysing value systems and c u l t u r a l l y  appropriate  ways of doing things, can make an enormous contribution to the crossc u l t u r a l study of decision making.  - 15 -  The basic assumption  that links formal models with  anthropo-  logy i s that men act i n order to get things they value, through the r a t i o n a l use of the rules and resources of their physical and c u l t u r a l environments.  The rules, rewards and man's perception of both can  change and vary through time and space, but I am submitting that the process of r a t i o n a l calculation remains constant. However,methodological pronouncements and statements of what should be done can get one only so f a r . While I regard methodology as a necessary philosophical hurdle, i t i s of l i t t l e use when not applied. I t i s obvious where my methodological preferences l i e and i t should be evident that I am taking part i n a growing formal t r a d i t i o n i n economic anthropology.  I t i s now necessary to get on with the job.  The methodological controversy i n economic anthropology has many " i n p r i n c i p l e " advocates on both sides, but very few p r a c t i t i o n e r s . Perhaps the debate can be put into a perspective by viewing economic theory as being made up of axioms which have standard interpretations in a market context, but consists of empirically uninterpreted variables i n the non-market context. With this view, I agree with Orans (1968) that the r e a l question i s whether or not empirical interpretations can be given to the axioms of economic theory so that they cover a domain wider than that of the market.  -  1 6 -  Th-is means that instead of talking about domains of relevance for  one approach as against another, one must get on with the job, con-  struct models and test them to see i f they work. set  Then one can see which  of methodological assumptions one can have confidence i n .  Progression of Thesis In development  Chapter 2, i n keeping with my i n i t i a l concern with underi n the Third World, I pose a problem of farmers making de-  cisions about innovations and offer a solution i n terms of a model of r i s k taking that predicts the sort of decisions p a r t i c u l a r types of farmers w i l l make under specified conditions. From this s t r i c t l y substantive orientation I then move to a higher level of generality i n that I generalise from my i n i t i a l propos i t i o n s to a series of statements about individuals taking r i s k s , i . e . making decisions under conditions of uncertainty and prospective loss. With this more theoretical l e v e l of operation, the testing procedure I use to assess the v a l i d i t y of my statements includes both laboratory and f i e l d settings, with supplementary evidence from secondary sources used by way of corroboration. In the  the laboratory I ran an experiment whereby I manipulated  conditions under which subjects took r i s k s on gambles (Chapter 3);  and i n the f i e l d I observed the constraints upon decision strategies taken by g i l l n e t fishermen on the west coast of B r i t i s h Columbia ter  4) .  (Chap-  - 17 -  I used evidence on the way  peasant farmers took decisions  with respect to innovations as a supplement to these tests (Chapter 5). In each case " t e s t i n g " meant the observation  of indicators  of the concepts i n the propositions and the assessment of the degree to which their observed occurrence corresponds to the t h e o r e t i c a l predictions. However, for the tests to be at a l l v a l i d , I have to show that the e x p l i c i t scope conditions specified by my propositions are met  i n each test s i t u a t i o n . That this i s possible i s a function of the l e v e l of abstrac-  tion I have chosen to use.  My  set of t h e o r e t i c a l propositions do not  refer to farmers and a g r i c u l t u r a l practices or students and gambling decisions, but individuals taking r i s k s under constraints of uncertainty, state of their resources  and the incentive conditions (both p o s i t i v e  and negative) that apply to the r i s k . This means that data on farmers and innovations, students and gambling, and fishermen deciding where to f i s h , can be used as test cases and are comparable, once i t has been shown that the scope conditions s p e c i f i c to the abstract formulation have been  met.  Thus i n a rather clumsy manner I w i l l be attempting to adopt a procedure suggested by Homans when he states "The  current job of  theory i n small group research i s to make the connection  between experi-  mental and r e a l - l i f e studies, to consolidate the propositions that empi-  - 18 -  r i c a l l y hold good i n the two f i e l d s , and to show how these propositions might be derived from a s t i l l more general set." (1958: 606) The conclusion, as i s the function of a l l conclusions, i s intended  to p u l l the disparate threads of discourse together and to  s p e l l out the implications of the thesis f o r economic anthropology.  - 19 -  CHAPTER 2 THE PROBLEM AND  ITS CONCEPTUALIZATION  Introduction From the l i t e r a t u r e on Third World farmers one can establish a number of propositions which link the decisions made by a subsistence farmer about a g r i c u l t u r a l innovations to his state of resources, incentive conditions attached to an innovation, and information and u t i l i t i e s attached to the outcome "increase i n p r o d u c t i v i t y . " From tbese statements s p e c i f i c to peasant farmers, I generalize to three statements about individuals and r i s k taking.  In the interests  of parsimony i t would perhaps be better to s t a r t with the general statements and proceed immediately to testing procedures.  However,  given the lack of precedents f o r the type of methodology implemented here and the explicational nature of my research, I think i t i s necessary to show the step-by-step progression from one level of generality to another.  It i s also important to i l l u s t r a t e the source  from which my statements originate. Farmers and Innovations The i n i t i a l substantive problem I am dealing with i s the explication of the factors that influence farmers t u r a l innovations.  1  decisions on a g r i c u l -  The conceptual approach I intend to use i s that of  decision theory, so both my substantive and conceptual concerns are within limited boundaries.  - 20 -  But even with such limited boundaries, i t i s possible to have a m u l t i p l i c i t y of variables that seem to be relevant.  There are  many things going on i n the decision making process with many possible combinations of factors.  At the outset some are regarded as absurd,  others as p o s s i b i l i t i e s which indicates that I have a number of preconceived  notions of the sorts of things that are relevant. The variables I am assuming to be relevant to the problem  can be categorised under three headings: (1)  the state of nature that the i n d i v i d u a l decision maker i s i n ,  (2)  the incentive conditions attached  (3)  access to information.  to the innovation, and  My choice of variables could be considered  a r b i t r a r y , but i n  actual fact i s not, as they are a l l i m p l i c i t to decision theory considerations. relevance  My major j u s t i f i c a t i o n i s that I w i l l be testing for the  of each variable and am prepared to r e j e c t any proposition  that cannot be v e r i f i e d i n empirical data.  In my i n i t i a l  formulations  I do not use e x p l i c i t s o c i o - c u l t u r a l variables (though these are implic i t i n the state of nature consideration) c h i e f l y on methodological grounds.  This i s because I am attempting to ascertain just how much  of decision-making can be accounted for i n p a r t i c u l a r situations by using a limited number of variables. Before I can proffer s p e c i f i c propositions I have to define what I mean by a state of nature.  F i r s t of a l l i t i s a c l a s s i f i c a t i o n ,  a set of categories into which I can place decision-making i n d i v i d u a l s .  - 21 -  But this taxonomic exercise i s not merely to garner d i f f e r e n t coloured b u t t e r f l i e s to my bosom -- an exercise that Leach parodies  i n 'Rethink-  ing Anthropology' (1961) -- but i s a preliminary step i n ordering  the  complex phenomena that a f f e c t decision-making into states of nature, the c h a r a c t e r i s t i c s of which I can then make statements about. The variables I use to define states of nature are (1) wealth and (2) subjective u t i l i t y attached  to an outcome.  The u t i l i t y part of  the state of nature i s a numerical measure of the strength of a farmer's preference  for a p a r t i c u l a r outcome.  cular substantive context remains constant  The outcome i n this p a r t i -- "increase i n productivity."  The implications of the u t i l i t y consideration are that the i n d i v i d u a l farmer w i l l react ( i n terms of decision strategies used) according the way  to  he sees a p a r t i c u l a r s i t u a t i o n . I t i s reasonable to assume that as subjective u t i l i t i e s for  the outcome 'increase i n productivity' vary, so w i l l decision strategies on events d i r e c t l y associated with that outcome, (events  such as the  consideration of an innovation). I am interested i n two implications of the wealth consideration: f i r s t l y , individual farmers have varying buying power and d i f f e r e n t i a l l y able to absorb losses according  are  to that wealth; second-  l y , d i f f e r e n t wealth levels w i l l have d i f f e r e n t i a l access to information about innovations. the resources outcome.  Thus, the wealth part of the state of nature defines  that any i n d i v i d u a l can mobilise to e f f e c t a p a r t i c u l a r  - 22 -  I f we posit three levels to each one of these variables, (wealth and subjective u t i l i t y f o r increasing productivity) -- low, medium and high -- there are nine possible states of nature, as below (Figure 1). Figure 1:  States of Nature Wealth L  M  H  L  1  4  7  M  2  5  8  H  3  6  9  Subjective Utility  1  The reason for defining a state of nature i n these terms i s partly to get round the problem of ethnocentrism, and p a r t l y to reduce the phenomena I am interested i n to manageable proportions.  I t i s an  analytic device useful f o r c r o s s - c u l t u r a l comparison, because any part i c u l a r state of nature either does or does not occur.  Where i t does  my statements should hold, where i t does not my statements are empirically irrelevant. I t would be natural to expect man as a culture-bearer to make decisions i n l i g h t of h i s c u l t u r a l values.  I am assuming that  the subjective u t i l i t i e s that any i n d i v i d u a l i n any culture w i l l have -- with respect to increasing a g r i c u l t u r a l productivity -- w i l l be a d i r e c t function of a number of factors, of which c u l t u r a l values i s only one, though a major one.  - 23 -  I think i t well nigh impossible to estimate the nature of this function, but one does not have to 'observe' values i n order to take them into account. subjective  In other words an individual's estimation of  u t i l i t i e s i s his perception (as an individual culture bearer)  of a p a r t i c u l a r outcome.  That this perception w i l l vary with d i f f e r -  e n t i a l wealth s t r a t i f i c a t i o n accounts for the wealth variable i n the state of nature. The  use of a state of nature prevents one  from making gross  statements about the significance of socio-cultural and  economic fac-  tors with reference to the decision-making process, as i t forces to be s p e c i f i c about the relationship between parameters and  one  situa-  tions . In other words, this form of conceptualisation  sets boundaries  to the relevance of p a r t i c u l a r variables, because each state of nature has  certain c h a r a c t e r i s t i c s and  isations applicable cable to any  to one  conditions which implies  state of nature are not necessarily  to an innovation.  but also considers the l i k e l i h o o d of the innova-  tion either l i v i n g up to i t s expectations or f a i l i n g . sideration involves  costs attached  This i s calculated partly i n terms of expected y i e l d s  c a p i t a l outlays,  cost.  appli-  other state.  Incentive conditions refer to the payoffs and  and  that general-  The  l a t t e r con-  the calculation of a further cost -- a l i v e l i h o o d  - 24 -  Access to information tion of information discussed  refers to the d i f f e r e n t i a l d i s t r i b u -  to p a r t i c u l a r categorised  individuals.  (This i s  i n d e t a i l later i n the chapter.)  LWHU State of Nature The  following remarks are pertinent  to state of nature No. 3.  They refer to a set of conditions whereby a farmer has a low wealth ranking, yet attaches a high u t i l i t y to increasing his productivity and i s generally analogous to subsistence agriculture. A major assumption I would make following Schultz  (1964) i s  that peasant production operates at or near an optimum i n terms of existing production factors and state of knowledge, i n this state of nature. That a farmer has a high preference f o r increasing  producti-  v i t y implies that he w i l l have taken up a l l or most of the 'slack' i n terms of existing production p o s s i b i l i t i e s . The  low l e v e l of productivity that i s c h a r a c t e r i s t i c of much  peasant agriculture i s often attributed to peasant i n e f f i c i e n c y , and their r e f u s a l to adopt new technologies  and practices to c u l t u r a l re-  sistance to change; whereas I would submit that production i s highly e f f i c i e n t i n terms of the given techniques and state of knowledge and the r e f u s a l to change a r a t i o n a l economic c a l c u l a t i o n . Generally,  farmers have a given set of production factors  which have been unchanged over generations, with the r e s u l t that a form  - 25 -  of equilibrium has been reached with reference to maximum production from given areas of land with given factors of production. The  implications of this assumption can be categorised i n  terms of two related properties: (1)  one cannot assume that better a l l o c a t i o n of the e x i s t i n g production factors would increase a g r i c u l t u r a l production;  (2)  farmers bound by t r a d i t i o n a l agriculture have usually exhausted a l l p r o f i t a b l e opportunities to invest i n the a g r i c u l t u r a l factors at their disposal. The farmer i n this state of nature i s i n a s i t u a t i o n where  the r i s k s from his occupation  are high, but they are f a m i l i a r and well  known through past experience and can thus be expected.  There i s un-  certainty with respect to these f a m i l i a r r i s k s i n that the farmer can never be sure that he w i l l be able to produce enough for his needs. The penalties for not doing so include hunger and debt. Innovations involve new and unfamiliar r i s k s , and i t i s highly unlikely that a farmer w i l l decide to a l l o c a t e from his e x i s t i n g stock of resources  the necessary wherewithall to obtain an innovation  the conditions attached  are highly favourable  unless;  to him.  Some Propositions  The propositions that follow apply only to state of nature No.  3, where the farmer cannot afford to lose.  - 26 -  Proposition 1:  Innovations with a high payoff factor which c a l l for a minimal diversion of stocks from existing resources w i l l tend to be adopted.  The whole idea of high investment  returns at low costs i s an  obvious one, but i t i s rare that innovations presented to LWHU farmers have these conditions attached.  In those instances where these condi-  tions are present, innovations w i l l be accepted. Proposition 2:  Innovations that offer a low return tend not to be adopted.  A c q u i s i t i o n of an innovation w i l l require that a farmer  takes  from his existing stock of resources ( c a p i t a l , land, labour, time) that which i s necessary for him to obtain the use of a new production factor.  As h i s e x i s t i n g resources are barely s u f f i c i e n t for his present  needs, i t i s obvious that special incentives are required i n order to persuade him to r i s k part of his existing stock of resources ( i . e . deplete his survival k i t ) f o r some new production factor. I t i s necessary that inputs and factors of production have to change i n order to get increased productivity, but they must be of such a nature that somehow overcomes the high-risk element involved i n a farmer i n this state of nature investing i n new production factors. Farmers are not going to invest i n new unproven factors unless they see that the return r e l a t i v e to the cost i s so much higher than their existing methods of production. r i s k s for marginal benefits.  They are unwilling to take  - 27 -  Proposition 3:  Individual farmers seek to minimise the cost factor of an innovation, hence they are more concerned with cost reduction than high returns.  The process of a decision has to balance between the possible gains and the possible losses.  I f a decision i s made to adopt an i n -  novation and the innovation f a i l s , the attendant  costs may  be high.  Not only has the farmer l o s t the resources invested i n the innovation, but also he had to deplete his 'survival k i t ' with the r e s u l t that the odds are now  against his surviving.  The costs have to be assessed  r e l a t i v e to the goals and  values  of the decision-making i n d i v i d u a l , and i n this p a r t i c u l a r state of nature I submit that the deterrent value of the costs for making a wrong decision exceeds the a t t r a c t i o n value of the possible gains.  We  are  dealing with a s i t u a t i o n where the i n d i v i d u a l cannot afford to lose. This means that though the 'High investment returns' thesis is an obvious one, proposition 3 sets a l i m i t i n g condition to i t . may  It  prove to be a more f r u i t f u l l i n e of enquiry ( i n terms of p o l i c y im-  plementations) to find the conditions whereby the loss factor can be reduced. These propositions can be represented low (Figure 2):  schematically as be-  - 28  Figure 2:  Propositions  1-3  Payoff  Adopt Innovation  Information Flow The concern with information flow and access to information c i r c u i t s indicates that the decision-making  process i s one of a per-  son struggling to integrate several sources of information into a single choice.  One could go into the whole area of information-pro-  cessing (with respect to the selection and weighting  of d i f f e r e n t  types  of information), but as this i s highly complex I prefer.';to regard i t as a 'black box' operation and trust that the state of nature and variables employed adequately  r e f l e c t the internal process.  I f we v i s u a l i s e information as flowing i n d i s t i n c t c i r c u i t s , i t i s readily apparent that some people w i l l gain access to more c i r cuits than others and thus possess more information.  - 29 -  But to make information a useful v a r i a b l e for analysis there must be some c r i t e r i a by which individuals can be d i f f e r e n t i a t e d i n terms of the amount' of information they have access to. I think most useful are wealth and education.  The  criteria  I discussed wealth ear-  l i e r i n terms of a varying a b i l i t y to absorb costs.  With reference to  information c i r c u i t s I w i l l be hypothesising that people at d i f f e r e n t wealth levels have d i f f e r e n t i a l access to information. This assertion breaks down into two considerations: (1)  The absolute cost of obtaining information about innovations places the poor farmer at a r e l a t i v e disadvantage with respect to the r i c h farmer.  (2)  The number of information c i r c u i t s available and the amount of information-dispensed w i l l increase with wealth. I am assuming that wealthy farmers w i l l tend to associate  more with other wealthy farmers and mutually  reinforce one another with  respect to information that i s exchanged between them, but not exchanged with poor farmers. Also the information media associated with s e l l i n g farm technology and innovations w i l l concentrate i t s agents and advertising effects on those individuals who  can most readily afford them.  The relationship between information and innovation i s a d i r e c t l y proportional one.  As r i c h  farmers have greater access to  information c i r c u i t s they tend to innovate more because an obstacle to r i s k taking has been removed, as increased information reduces the uncertainty aspects attached to any innovation.  - 30 -  Education could be tied i n i n a variety of ways to the decision-making problem I am interested i n ; however, my concern with the education variable i s the way  i t operates i n terms of allowing i n d i v i -  duals d i f f e r e n t access to information c i r c u i t s .  I am hypothesising  that as education increases so does the a b i l i t y to gain access to more information. Between wealth and education there are c e r t a i n l y some feedback effects, as i t i s reasonable to expect that higher forms of education are most l i k e l y to r e s u l t from a higher wealth base.  I am also  assuming that a higher l e v e l of education w i l l r e s u l t i n a greater range of exposure to information about innovations and a greater awareness of changing market conditions and productive techniques. I t could be argued that an increase i n education would drive individuals away from the farms.  This may be so, but the point of the  model i s that of those remaining (or i f those leaving were to remain), the individuals with higher education w i l l have access to more i n f o r mation c i r c u i t s than those with less education. But there are some problems here.  In addition to consider-  ations of formal education, one has to consider the q u a l i t a t i v e weighting that can be given to relevant experience of the situation i n hand. The propositions dealing with information are: Proposition 4:  A b i l i t y to gain access to information increases with wealth.  - 31 -  Proposition 5: Access to information increases with  education.  Proposition 6: Innovative behavior increases with information. These propositions are concerned with comparisons between states of nature.  Assuming that subjective u t i l i t y remains constant  we are dealing with the LWHU, MWHU, and HWHU states of nature. These three propositions can be represented  schematically  as below (Figure 3): Figure 3: Propositions 4 - 6 No. of Information circuits H -  a b—V*  /  H  S  Wealth—^M  Education^ -A  _ - -  *  c  H - - - - - - - d  f >Education-)M - - - - - - -  \ /  8  X  High Information ^HIGH  N E N H 0 A  x  e  ,s  Medium ^ Information w  :  v  \  Education-^M - - - - - - \^  MB  h  i 1  i  ^ Low LOW-""E ^ ^ ^ j f l n f ormation<^  The diagram has 9 d i f f e r e n t c i r c u i t s ( a - i ) , which indicates a p o s s i b i l i t y of 9 d i f f e r e n t gradations of information.  In terms of  the propositions that relate wealth and education to information I can say that a > b > c ; d ^ e ^ f ;  and g )> h > i .  - 32 -  There are a number of l o g i c a l l y possible ways to obtain an ordering f o r the whole sequence, depending on the r e l a t i v e weightings given to wealth and education. By assuming that education i s s l i g h t l y more important than wealth i n terms of access to information, one of the orderings I could get would be to have d / c and g > f and for the whole sequence a  >  b  >  d  _ > c 7 e 7 £ 7 f >  h> i .  This ordering enables me to break the nine categories down into gross categories of high, medium and low that I am using f o r the other variables.  The ordering I have i s an assumed one and the sig-  nificance of testing procedures w i l l be to see whether the assumptions for ordering are v a l i d or not. Supplementary Propositions Although the propositions I have stated do not contain exp l i c i t reference to s o c i o - c u l t u r a l variables, this does not mean that I t o t a l l y disregard these l a t t e r factors. The testing of the model w i l l function to e s t a b l i s h the r e l e vance or irrelevance of factors such as s o c i a l mobility, s o c i a l margin a l i t y and co-operative behavior.  structure i n terms of i n d i v i d u a l r i s k - t a k i n g  Where my propositions are inadequate to explain the behavior  they are "supposed" to I w i l l i n s e r t additional propositions to see i f a better f i t between model and data r e s u l t s .  - 33 Social M o b i l i t y As an index of what I s h a l l c a l l 'social mobility' I intend to use a measure of the degree to which "role s l o t s " are present i n the society which are recruited by achievement c r i t e r i a , rather than by a s c r i p t i v e c r i t e r i a .  This rests on a conceptualization  of roles  as; s l o t s or positions i n a s o c i a l structure for which individuals compete.  may  My concern with t h i s index i s with the rules of competition  which define the flow or c i r c u l a t i o n of persons between positions. The  index measures the degree to which 'achievement r o l e s l o t s '  are present i n a p a r t i c u l a r society.  Thus a society with a high index  of s o c i a l mobility indicates that a greater number of people can compete for roles than i n a society with a low s o c i a l mobility index. The follows.  implications  of s o c i a l mobility for innovation  are  as  If .we v i s u a l i z e the competition for a p a r t i c u l a r r o l e as  requiring the attainment of c h a r a c t e r i s t i c s (a  n); i t follows  that  persons aspiring to that r o l e have to acquire those c h a r a c t e r i s t i c s . Societies low i n s o c i a l mobility are characterized  by exclusion  rules  which prevent everyone i n the population from acquiring the necessary prerequisites f o r p a r t i c u l a r r o l e s .  Societies high i n s o c i a l mobility  do not have these b a r r i e r s . The  c h a r a c t e r i s t i c s (a  expected behavior patterns and,  h) may  consist of prestige symbols,  for want of a better word, 'entry fees';  and can be acquired i n a variety of ways.  If prestige symbols have to  be attained, t h i s means that an individual has  to involve himself i n  - 34 -  assymetrical  exchange relations with others, with reference to ideas  and goods, i n order that others accord him prestige.  C l a s s i c a l l y this  can be done by exchanging goods for enhanced prestige, and usually requires  increased  c h a r a c t e r i s t i c s may  productivity on the material  require c a p i t a l outlays  pected behavior patterns may  level.  Other  for 'entry fees', and  ex-  involve a considerable amount of expense.  P a r t i c u l a r role c h a r a c t e r i s t i c s may l e v e l or standard of  the process  be d i r e c t l y related to an income  prosperity.  In a society where roles are open to competition ( i . e . where there i s high s o c i a l mobility) people without this optimal l e v e l of wealth have:  (a) the incentive and  (b) the opportunity through i n -  creased productivity to a t t a i n these r o l e c h a r a c t e r i s t i c s . I f one  i s a farmer one way  adopt a g r i c u l t u r a l innovations.  of increasing productivity i s to  I would predict that i n a society  where roles are more open to achievement, the incentives  to  increase  productivity are present i n a far greater degree than i n a society allocates roles on a more a s c r i p t i v e basis. the incentives  In the l a t t e r s o c i e t i e s ,  to aspire to certain roles, and  sion to increase productivity, are constrained certain categories Proposition  7:  whi  hence by l o g i c a l extenby the fact that only  of individuals are e l i g i b l e for p a r t i c u l a r r o l e s . The proportion  of innovative  society increases s o c i a l mobility.  individuals i n a  with the society's index of  - 35 -  Social Marginality S o c i a l marginality implies that individuals are at the margin of a given culture or are i n a s o c i a l p o s i t i o n i n which they straddle more than one culture. When this i s accompanied by some form of sectarian i d e n t i t y , the marginal i n d i v i d u a l —  not bound by wider group norms yet being  to draw upon the resources  of his own  immediate community —  able  has a  very wide range of a c t i v i t y . The lack of constraining ties and the p o s s i b i l i t y of m o b i l i sing more economic resources  should place the marginal i n d i v i d u a l i n  a s i t u a t i o n where the opportunities and resources  for innovation are  much better than for the non-marginal i n d i v i d u a l . The role of marginal individuals i n diverse economic pursuits i n many underdeveloped countries gives support to the notion that there i s a relationship between innovation and s o c i a l marginality (Hoselitz and Moore 1966)  —  the Chinese i n South East Asia and Indians i n East  A f r i c a are perhaps the most obvious examples. But marginality i n i t s e l f i s not a s u f f i c i e n t c r i t e r i o n , as i t could be argued that marginal individuals may others to succumb to anomie.  be more prone than  I t i s when a cohesive and marginal group  operates within a wider c u l t u r a l context can we then expect high i n novative behavior.  - 36 -  S o c i a l l y marginal trader groups are innovative because they are f;reer to try novel economic behavior and because they restructure the culture of deference and the rewards system rather than accomodate themselves to i t . I stated e a r l i e r that decisions were the outcome of balancing up the perceived costs and payoffs of p a r t i c u l a r outcomes (p. 27). S o c i a l marginality implies that there are certain costs to being marginal i n terms of relationships with the wider society, which may  mani-  f e s t i t s e l f through s o c i a l and l e g a l discrimination. Thus marginality v i s - a - v i s the wider society has  inherent  costs with the r e s u l t that day to day existence fraught with costs i s a reality.  Because of this factor I would submit that the cost e l e -  ment attached  to any alternative i s not as big a deterrent to innovating  as i t would be for a non-marginal i n d i v i d u a l . Thus reinforced by his group and with a d i f f e r e n t i a l appraisal of costs, the s o c i a l l y marginal i n d i v i d u a l i s more l i k e l y to innovate than the non-marginal i n d i v i dual. These considerations suggest the following proposition: Proposition 8:  The proportion of innovating persons i n soc i a l l y marginal business groups i s higher than i n non-marginal business groups.  Co-operative Structure  A supplement to proposition 3 (Individual farmers seek to minimise the cost factor of an innovation)is the notion that forms of  - 37 -  productive organisation which spread the r i s k and cost factors over a c o l l e c t i v i t y greater than a single farmer w i l l r e s u l t i n more innovations being adopted.  This i s by merit of p r e v a i l i n g conditions of c o l -  l e c t i v e insurance. Recent work carried out i n experimental psychology indicates a d e f i n i t e tendency for individuals to take more risky decisions when they are i n a group context as against their acting alone. phenomenon has been l a b e l l e d as the "risky s h i f t " concept  This (Kogan and  Wallich 1967) and suggests the following proposition. Proposition 9:  Individuals making decisions i n the context of a co-operating group of individuals w i l l adopt more innovations than an i n d i v i d u a l decisionmaker acting alone.  Whither Next? It would appear that that would be as f a r as I could take my statements  about farmers and the sorts of risks they w i l l tolerate  with reference to new production factors. However, though I could regard this as a substantive l i m i t a tion on my endeavours, conceptually i t i s possible to go much further. Desire for more generality leads one from the immediate substantive concern for peasant producers deciding on innovations to a higher l e v e l of generality i n terms of a theory of risk-taking, whereby data about p a r t i c u l a r sets of individuals constitute separate empirical tests of the theory.  - 38 -  Cancian i n a somewhat confusing yet stimulating a r t i c l e makes similar types of suggestions  (Cancian 1967).  He develops a theory of s t r a t i f i c a t i o n and r i s k taking and tests i t oh a g r i c u l t u r a l innovation, though he states that i t could be applied to any s i t u a t i o n involving s t r a t i f i c a t i o n and r i s k taking.  His  i n i t i a l theory which predicts that a ranking based on the possession of a valued resource has a negative relationship to r i s k taking, i s modified to develop a predicted c u r v i l i n e a r relationship between resource and r i s k . In the application of the theory to agriculture he takes wealth as the valued resource and the early adoption of a new t i c e as a r i s k .  a g r i c u l t u r a l prac-  His i n i t i a l assumption i s that "the better your finan-  c i a l position, the more you have to lose and the less you have to gain from taking chances; therefore, insofar as adopting an innovation i s risky, the richer you are, the less l i k e l y you are to adopt." (Cancian 1967:  913). He defines r i s k as "a c h a r a c t e r i s t i c of situations of ex-  change i n which the rate of return on investment of resources i s uncertain; the greater the uncertainty, the greater the r i s k . " (1967: 913) We do not know what s p e c i f i c " c h a r a c t e r i s t i c " Cancian refers to, but his use of the term i s synonymous with the presence or absence of uncertainty. He does not consider as important the incentive conditions attached to the innovation (yield v i s - a - v i s cost) which means he does not d i f f e r e n t i a t e between expensive innovations with long term y i e l d s and cheap innovations with immediate r e s u l t s .  - 39 -  It i s evident that considerations of r e l a t i v e cost and y i e l d would influence a farmer's decision whether to adopt an innovation, which implies that the indicators of innovation that Cancian uses i n his comparative study are not i n fact very good ones. From Marsh and Coleman's study (1955) he takes the rate at which Kentucky farmers adopted bluestone lime, and uses Gross's (1942) study of Iowa farmers adopting hybrid corn seed for comparative purposes. The lime was an expensive f e r t i l i z e r with results expected a f t e r a three-year period of use, while hybrid corn seed was an innovation with immediate results promoted by the Department of Agriculture as part of the post depression a g r i c u l t u r a l e f f o r t .  The adoption rates  of these are not s t r i c t l y comparable unless some method of standardization or control f o r v a r i a t i o n of r e l a t i v e costs and y i e l d s i s used. Cancian does not use any controls either for these two instances or for the f i v e other case studies that he uses. So Cancian does not have an accurate assessment of r i s k taking behaviour i n the d i f f e r e n t farming communities he was  concerned  with, and i n fact h i s theory of r i s k taking i s a misnomer, as his theory i s about information d i f f u s i o n and i s not d i r e c t l y about r i s k taking. His i n i t i a l theory predicts an inverse relationship between wealth and early adoption of innovations and he labels this the " i n h i b i t i n g e f f e c t " of wealth. Where the conditions of h i s theory are not applicable (perfect d i v i s i b i l i t y of r i s k s ; knowledge equally spread over a l l wealth ranks) then we have what he c a l l s the " f a c i l i t a t i n g " e f f e c t of wealth;  - 40 -  a p o s i t i v e relationship between wealth and early adoption. develops a " c u r v i l i n e a r e f f e c t " i n order to explain  conservatism on the  part of the middle wealth rank i n terms of early adoption. a number of rationales  He then  He offers  for this c u r v i l i n e a r e f f e c t but by way of under-  cutting his own e f f o r t s states, "None of the arguments offered cons t i t u t e s a satisfactory•theory  explaining  the c u r v i l i n e a r e f f e c t " (1967:  916) and " i t s (curvilinear effect) theoretical status remains f a r from satisfactory because two d i s t i n c t and inconsistent  arguments for the  effect were made during the development of the modified theory." (1967: 925) The point to emphasize i n Cancian's concern f o r d i f f e r e n t types of " e f f e c t s " i s that some of the conditions of h i s o r i g i n a l theory do not apply when he refers to f a c i l i t a t i n g and c u r v i l i n e a r e f f e c t s , which means that he i s dealing with discrete sets of scope  conditions.  (Scope refers to the set of phenomena a p a r t i c u l a r statement has r e f erence to.  To test the v a l i d i t y of a statement i n a scope other than  that i n i t i a l l y ;  specified for the statement i s not to o f f e r a test at  all.) This further implies that h i s statements with respect to one  set of scope conditions are not applicable  discrete sets.  to any of the other  Cancian seems to ignore this aspect of comparative  methodology and as a result one may doubt the ..value of h i s comparisons .  - 41 -  In operational terms Cancian v i s u a l i z e s the adoption process as having at least two stages:  "Stage 1 i n which i n c l i n a t i o n to r i s k  i s important, and stage 2 i n which i n c l i n a t i o n to r i s k i s substantially less important" (1967: 917). Suddenly, we are introduced to " i n c l i n a t i o n to r i s k " as defi n i t i v e of adoption stages.  Cancian does not o f f e r a measure f o r this  or even a proposition r e l a t i n g i t to either wealth or adoption.  He  states that r i s k i s high i n stage 1 and low i n stage 2, on the basis of there being more information available i n stage 2.  But this cate-  gorization of r i s k into high and low, as a function of information, has very l i t t l e to do with a farmer's actual decision to adopt.  An  expensive innovation i s s t i l l a f i n a n c i a l r i s k i n stage 2, whereas subsidized a g r i c u l t u r a l practices with high, immediate y i e l d s would be low risks i n stage 1. Cancian does not consider investment factors as important i n his c a l c u l a t i o n of adoption rates and this i s a major shortcoming i n his conceptualisation of r i s k taking. His  comparative analysis supports the c u r v i l i n e a r effect of  middle class conservatism, and he argues that there are instances where the i n h i b i t i n g and f a c i l i t a t i n g e f f e c t are operative.  However, due to  confusion of scope conditions and lack of standardization of measures, too much importance should not be attached to the r e s u l t s . However, I found his suggestions for further research extremely i n t e r e s t i n g , mainly because they endorsed the conceptual approach attempted here.  - 42 -  It should be pointed out that his i s presented as an introductory exercise, a working paper.  But, be this as i t may,  I gained a great  deal of encouragement from his suggestions that as part of further research " d i r e c t tests may  be possible i n laboratory s i t u a t i o n s " (1967:  and that "theories of r i s k taking may  926)  be more important than theories  of information d i f f u s i o n for the study of the process of adoption of new  a g r i c u l t u r a l p r a c t i c e s . " (1967:  927)  I found his conceptualization of r i s k unsatisfactory and his lack of concern for incentive conditions an oversight.  His concern  with the implications of information d i f f u s i o n had already been incorporated into my model, while his consideration that future lay with r i s k taking theory was of nature and  fundamental to my  research  concern with states  utilities.  To i l l u s t r a t e the r e l a t i v e merits of my model i t i s necessary to further explicate i t s basic b u i l d i n g blocks, and the f i r s t i s the conceptualization of r i s k . If we were to think about r i s k i t would l i k e l y conjure  up  images of uncertainty of achieving desirable goals and of the penalties that may  ensue from the f a i l u r e to a t t a i n the desired goals.  I use these two aspects —  In fact  lack of certainty and prospect of loss  —  as d e f i n i t i v e of r i s k - t a k i n g . If we look at the decisions common to everyday l i f e , the pervasiveness  of the risk-taking concept becomes apparent.  a career or of a marriage partner i s fraught with r i s k : to buy  The choice of the decision  a house, to invest i n bonds, to drop out of society —  as long  - 43 -  as they c o n t a i n a t t r i b u t e s of u n c e r t a i n t y and p o t e n t i a l l o s s — c o n s i d e r e d as To  can  risks. t a l k about r i s k - t a k i n g ,  then, i s to r e f e r to b e h a v i o r i n  s i t u a t i o n s where there i s a d e s i r a b l e g o a l and a l a c k of  certainty  t h a t i t can be a t t a i n e d , w i t h a t t e n d a n t p o s s i b i l i t i e s of l o s s . t a k i n g i s thus a s p e c i a l category of d e c i s i o n - m a k i n g , the f a c t o r s i n v o l v e d i n the proeess developed by d e c i s i o n t h e o r i s t s Phillips  be  1965;  and  Risk  to e x p l i c a t e  I i n t e n d to use some of the  (Edwards 1954;  Von Neumann and Morgenstern  tools  Edwards, Lindman and  1953).  D e c i s i o n Theory  The  c l a s s of models which most d e c i s i o n t h e o r i s t s have been  concerned w i t h have been t o t a l l y  devoted  ence to what i s deemed to be r a t i o n a l  to c h o i c e outcomes w i t h  refer-  criteria.  An i n d i v i d u a l i n a d e c i s i o n s i t u a t i o n e v a l u a t e s the p r o b a b i lities  and p a y o f f s a t t a c h e d to a p a r t i c u l a r outcome and  then  s e l e c t s the a l t e r n a t i v e t h a t maximises something of v a l u e to  rationally him.  The v a l u e a t t a c h e d to any d e c i s i o n i s a r r i v e d a t by t a k i n g the v a l u e s of the p o s s i b l e outcomes, m u l t i p l y i n g them by t i v e p r o b a b i l i t i e s of o c c u r r e n c e and  then summing up  their  respec-  these p r o d u c t s .  For example i n the case of an i n d i v i d u a l h a v i n g  to^decide  whether to i n v e s t c a p i t a l i n a r i s k y e n t e r p r i s e , c o n s i d e r the v a l u e s a t t a c h e d to two outcomes — the r i s k pays o f f , and does n o t .  the v a l u e a t t r i b u t e d to the e n t e r p r i s e i f  the v a l u e a t t r i b u t e d to the e n t e r p r i s e i f i t  ( T h i s l a t t e r v a l u e w i l l be  negative.)  - 44 -  If we denote the f i r s t value by U  g  and the l a t t e r value by  and l e t P represent the l i k e l i h o o d of the r i s k paying o f f and (1-P) be the p r o b a b i l i t y of the r i s k not paying o f f , then the value of the enterprise  (V) w i l l be:  V = U .P + U . ( l - P ) . f  V can be d i f f e r e n t i a l l y calculated by a series of models according to whether the values of the outcome ( i . e . the u t i l i t i e s ) and the p r o b a b i l i t i e s attached to the outcome are subjective or objective. (An objective view states that there i s only one unique r e a l number that can represent a p a r t i c u l a r outcome or p r o b a b i l i t y irrespective of the variety of decision-making i n d i v i d u a l s .  The subjective view  represents a degree of b e l i e f on the part of the i n d i v i d u a l decisionmaker and i s not uniquely determined.) A l l the models used by decision theorists have i n common the notion that the u t i l i t i e s and p r o b a b i l i t i e s w i l l be considered by the decision-maker i n adherence to s p e c i f i c l o g i c a l rules that  define  rationality. This i s why descriptive theories of decision-making have a l ways been closely linked with normative models.  When l o g i c a l  consider-  ations indicated that people should behave according to certain prescribed p r i n c i p l e s , those p r i n c i p l e s were incorporated  into descriptive models  against which performance could be compared. While I think i t i s useful to employ normative models as s t a r t i n g points for descriptive theory, preoccupation with normative aspects, seems i n the case of decision theory, to have led to the neg-  - 45 -  l e c t of possible psychological aspects of the decision making process and has often'completely ignored the situational*context. • The r e s u l t has-'been that when these models have been applied to gambling situations, only a moderate l e v e l of success i n predicting choices between bets has been achieved — (Kogan and Wallich 1967: 118).  between 55% and 70% accuracy.  This i s s l i g h t l y better than predictions  that would emanate from a randonu generator. The model builders i n attempting to erect a comprehensive mathematical theory of decision-making have been unable to account f o r the limited set of decisions which constitute the domain of human gambl i n g behavior.  This indicates that further e x p l i c a t i o n of the nature  of decision-making i s required before mathematical  formalisation be-  comes meaningful i n an applied sense. The reason for this may be found i n the consideration that extraneous influences that bear upon the decision-maker, while not changing the calculated value of any alternative w i l l d i r e c t l y  influ-  ence any decision with respect to that p a r t i c u l a r a l t e r n a t i v e . It i s important to think of decision-making as multidimens i o n a l , and that the four general components considered i n the models I have discussed are only part of the process. I would go as f a r as to say that any general decision-making model, of the.sort described, w i l l meet with no more than limited success i f i t ignores the variety of motivational and s i t u a t i o n a l factors that enter into the decision making process.  - 46 -  My concern with states of nature i s an attempt to remedy this defect.  Similarly my interest i n variations of information i s a  selected s i t u a t i o n a l factor.  These broad considerations can be repre-  sented diagramatically as. below, where D i s the decision-making  indivi-  dual (Figure 4):  Figure 4:  General Model  INCENTIVE CONDITIONS /S.  A.  U  U„  RISK  P  The linkages between the d i f f e r e n t parts of the model w i l l be provided by propositions that relate a p a r t i c u l a r decision to variations i n one or more of the other categories. To show the relevance of the decision theory format just described, consider the farmer deciding whether to take a r i s k on an  - 47 -  innovation.  Recall that the value*of a decision i s defined i n terms  of the value of the outcome " r i s k pays o f f " times the p r o b a b i l i t y that i t w i l l pay o f f , plus the value of the outcome " r i s k does not pay o f f " times the l i k e l i h o o d that i t w i l l not. i . e i V = U .P + n . s f  (l-P)  If we consider an innovation as having incentives analogous to this equation, we get U  = expected y i e l d or payoff i f innovation works  P  = chance that innovation w i l l come up to expectations = calculated losses i f the innovation f a i l s  l-P = chance that the innovation w i l l  fail.  Thus i f we consider an innovation as either succeeding or f a i l i n g , and a farmer as deciding to either accept or reject i t , we can .construct a 2x2 payoff matrix as below to account f o r every poss i b l e s i t u a t i o n (Figure 5): Figure 5:  Payoff Matrix for Innovations  INNOVATION  Succeeds Adopts  Rejects  Fails  Y-I  -I-L  0  0  - 48 -  In the 1,1  c e l l , i f the farmer adopts an innovation and i t  succeeds he w i l l have a payoff of (Y-I), where Y i s the y i e l d from the innovation and I i s the i n i t i a l c a p i t a l investment i n the innovation. The 2,1 c e l l where the farmer rejects an innovation that would have worked, we have a statement of opportunity cost. off  The pay-  i n this c e l l would be zero. In the 1,2  c e l l where a farmer adopts an innovation which  f a i l s , the payoff i s negative i n that the farmer has l o s t h i s investment c a p i t a l (I) and i f unable to absorb this loss may his  livelihood  suffer cost to  (-L).  The 2,2  c e l l has a payoff value of zero.  In decision-theory terms the value attached to the outcome "innovation succeeds" i s the sum of payoffs  i n the f i r s t column, which  i s (Y-I). i.e. U  = Y-I. s  The value attached to the outcome "Innovation f a i l s " i s the sum of payoffs i n the second column, i.e. = -I-L. We can consider U and  as losses he may  s  as gains that would accrue to the farmer  have to absorb.  On a t h e o r e t i c a l l e v e l what I intend to show i s that there are certain combinations of expected gains and calculated losses which determine which decision strategies individuals w i l l employ i n d i f f e r ent s i t u a t i o n s .  -  49  -  Theoretical Propositions D e c i s i o n s t r a t e g i e s can be c a t e g o r i s e d i n t o low and r i s k s according  high  to whether the d e c i s i o n maker chooses t h a t outcome  which maximises g a i n , m i n i m i s e s l o s s o r a f f o r d s a l o n g High r i s k s t r a t e g i e s i n c l u d e maximisation  shot.  of g a i n —  choosing  the a l t e r n a t i v e t h a t would g a i n the most i f the outcome i s p o s i t i v e (i.e.  i f the r i s k pays o f f ) and l o n g  w i t h the l o w e r p r o b a b i l i t y o f s u c c e s s .  shots — Low  choosing  the a l t e r n a t i v e  r i s k s t r a t e g i e s seek to  m i n i m i s e l o s s and i n v o l v e c h o i c e of the a l t e r n a t i v e t h a t would l o s e the l e a s t i f the outcome i s n e g a t i v e , i . e . the r i s k does not pay o f f . With these c o n s i d e r a t i o n s I have g e n e r a l i s e d from my mediate p r o p o s i t i o n s about farmers and i n n o v a t i o n s  inter-  to t h r e e p r o p o s i t i o n s  r e l a t i n g i n d i v i d u a l s to r i s k t a k i n g s t r a t e g i e s as a f u n c t i o n of v a r i a t i o n i n wealth, u t i l i t i e s  and  information.  (The supplementary p r p p o s i t i o n s a r e not p a r t of t h i s they remain as t h e i r name s u g g e s t s — The  suplementary.)  f i r s t t h r e e p r o p o s i t i o n s a l l r e f e r to a p r e f e r e n c e  by LWHU farmers f o r low r i s k s t r a t e g i e s ; " m i n i m a l d i v e r s i o n of from e x i s t i n g r e s o u r c e s " marginal  process,  ( p r o p o s i t i o n 1); "no  r e t u r n s " ( p r o p o s i t i o n 2); and  shown stocks  r i s k of r e s o u r c e s  "farmers seek to m i n i m i s e  for the  c o s t f a c t o r of an i n n o v a t i o n " ( p r o p o s i t i o n 3). These t h r e e s t a t e m e n t s can be g e n e r a l i s e d t o one tion:  proposi-  - 50 -  1.  Antecedent Conditions 1.  Every decision maker can be c l a s s i f i e d on a two way matrix using wealth and subjective u t i l i t y f o r some outcome as parameters .  2.  Consider the LWHU state of nature. Proposition 1':  Individuals w i l l show a preference f o r low r i s k strategies with respect to decisions involving the outcome to which HU i s attached.  The statements about wealth, information and innovative behavior can be generalised to the following two propositions: 2.  Antecedent Conditions 1.  Same as 1 above.  2.  Consider states of nature where U remains constantly high. Proposition 2':  Individuals w i l l show an increased tolerance for high r i s k s t r a t e g i e s , with respect to the outcome to which HU i s attached, as wealth increases.  (Wealth i n this formulation i s defined with reference to the s p e c i f i c outcome i n question and could be considered as the resources possessed by an i n d i v i d u a l to a t t a i n a p a r t i c u l a r end.) 3.  Antecedent Conditions  1.  Same as 1 above.  2.  Same as 2 above.  - 51  Proposition 3':  -  Individuals w i l l show an increased  tolerance  for high r i s k strategies, with respect to the outcome to which HU i s attached, formation  as i n -  increases.  (An increase i n information reduces uncertainty and removes a deterrent to high r i s k strategies.) Testing I have constructed a very simple model of r i s k taking and information processing.  The f i r s t proposition refers to one p a r t i c u l a r  state of nature, the second and third refer to changes across states of nature.  Given a state of u t i l i t y and a wealth ranking, and control-  l i n g for information, predictable.  the types of risks an i n d i v i d u a l w i l l take are  But to be at a l l useful the model has to be tested i n  empirical data. It i s possible to use three main sources to test my about risk-taking —  assertions  f i r s t i n the laboratory, then i n the f i e l d and f i -  n a l l y from secondary sources.  In following this through I anticipate  objections to the p o s s i b i l i t y of generalising from one s i t u a t i o n to any other.  But this objection can be rendered i r r e l e v a n t by  consider-  ing precisely the methodological steps that should be taken. I should emphasise that my research i s concerned with theoretical  considerations.  - 52 -  Theories consist of systems of assumptions and conditions which specify the scope or range of appropriate cular statements.  application of p a r t i -  The assumptions assert relations among abstract  concepts, and scope conditions specify the circumstances under which the relationship s p e c i f i e d i n the assumptions i s expected to hold  true.  Thus the laboratory, f i e l d and l i b r a r y are p a r t i c u l a r research settings i n which predictions from the theory may legitimately be subjected  to empirical test.  In each of these three cases theory testing means the observation of indicators of the concepts i n the propositions, and the measurement of the degree to which their relationships correspond to the t h e o r e t i c a l predictions. In each s e t t i n g the consequences of testing are the same — so long as the scope conditions have been met, either f a i l u r e or success of the theory's predictions has d i r e c t implications f o r the assumptions of the theory.  The scope conditions that I am concerned with place  an i n d i v i d u a l decision maker within parameters of wealth and subjective u t i l i t y for some outcome, information and incentive conditions for any risk.  The model i s attempting to predict the type of decision s t r a t e -  gies employed f o r given values of the above parameters. It i s important to note that i f there i s no theory, and i f there are therefore no e x p l i c i t scope conditions, then no generalisations of the results of a laboratory study, or of any other study, i s permissable.  - 53 -  The reason for this i s that without theory, there are no grounds for asserting that the results of any given empirical study were produced by a small set of s p e c i f i a b l e factors.  The results may  as well have been the consequences of the i n t e r a c t i o n of an i n f i n i t e number of non-recurring be  circumstances and therefore would themselves  non-recurring. If any p a r a l l e l s are to be drawn between a number of empiri-  cal studies then there must be a theory which shows that the studies have i n common the factors s p e c i f i e d by the scope conditions. So long as the e x p l i c i t scope conditions are met, whether inside the laboratory or i n the natural environment, the assumptions of the theory w i l l enable prediction of empirically testable consequences.  My testing procedure begins i n the laboratory, moving thence  into the f i e l d and f i n a l l y uses secondary source material for corroborative evidence of the f i r s t two testing procedures.  There i s a  good s t r a t e g i c reason for s t a r t i n g one's set of tests from a laboratory base. In any natural s e t t i n g there are a large number of variables that one cannot control for, whereas i n the laboratory one can s e l e c t a small number of variables and systematically control their v a r i a t i o n to see what happens.  Thus i t i s more l i k e l y that any theory w i l l be  able to make more precise predictions for the laboratory s e t t i n g . And this i s where the value of the laboratory l i e s —  i t can provide the  'purest' possible test of the predictions of the theory, dealing with  - 54 -  a small set of variables and conditions at a time.  Confirmatory re-  sults from such a rigorous test can then increase one's confidence  in  applying the theory to a natural s e t t i n g . If the laboratory test i s confirmatory  of the theory's  pre-  dictions, and then the theory i s applied to a natural s e t t i n g (where the same scope conditions are met) l a t t e r test may  any difference that occurs i n the  then be attributed to variables other than the ones  controlled for i n the experiment. I was  (and s t i l l am)  interested i n more s t r i c t l y  socio-  c u l t u r a l variables with respect to decision-making, and formulated  a  series of propositions r e l a t i n g s t r u c t u r a l aspects of group membership such as s o c i a l mobility, s o c i a l marginality, and co-operative  structure  to risk-taking behavior. My  reasons for excluding these a d d i t i o n a l variables from  the immediate model included the fact that they made model b u i l d i n g very d i f f i c u l t , and also that I was vance.  not too convinced of their r e l e -  However, the primary reason was  I intend to use may,  because the testing procedure  i n f a c t , show their  relevance.  By f i r s t testing my statements i n the laboratory and then i n the f i e l d , a lack of congruence between the two  tests could indicate  that there are other constraints operating which were not controlled i n the laboratory.  It i s then imperative  to bring i n a d d i t i o n a l  propositions r e l a t i n g these additional constraints to r i s k taking behavior i n order to see i f there i s a better f i t between data and model.  - 55 -  However, i f the results from the d i f f e r e n t tests agree, i t means that the constraints outlined i n the supplementary propositions are not overly relevant to risk-taking i n the situations considered.  Summary In a somewhat pedantic manner, I have arrived at a set of general propositions v i a the intermediary considerations of farmers making decisions about a g r i c u l t u r a l innovations.  This was necessary  i n order to show that my second order propositions did not appear out of thin a i r but were based i n i t i a l l y on s p e c i f i c substantive considerations . The advantage of operating at a higher l e v e l of generality i s that the scope of my statements i s not r e s t r i c t e d to the domain of farmers and a g r i c u l t u r a l innovation,  but i s bounded by considerations  of u t i l i t i e s , information, and resource gradients with reference to any  risk. V e r i f i c a t i o n of these general statements requires that they  be operationally defined i n terms of a number of substantive and tested.  contexts  Provided that the scope conditions s p e c i f i e d by the second  order of propositions can be shown to be the same i n each substantive context, then the laboratory and the f i e l d are legitimate sources to test the same propositions.  - 56 -  CHAPTER 3 TEST CASE:  THE EXPERIMENT  Introduction Recall that i n Chapter 2, I stated an intention to test my propositions, with respect to individuals and risk-taking, i n laboratory and f i e l d settings. The argument used was that once the scope conditions were adequately met then this testing procedure could be regarded as v a l i d . An additional argument was made for a precedence of laboratory work i n terms of affording the experimenter greater control over the variables. The point to be made i s that each substantive  s i t u a t i o n (stu-  dents playing bets; fishermen selecting strategies; farmers deciding whether to accept an innovation) provides a test for the t h e o r e t i c a l propositions.  These l a t t e r statements r e l a t e u t i l i t y preferences (for  some outcome), to the calculated incentive conditions attached  to an  event (associated with that outcome) i n terms of the decision strategies p a r t i c u l a r categorised  individuals are predicted to take with res-  pect to the event. In other words the scope conditions place an i n d i v i d u a l dec i s i o n maker within parameters of wealth and subjective u t i l i t y for some outcome; present him with a decision task associated with the outcome; and subject him to variations i n incentive conditions and i n formation attached  to the event.  - 57 -  Thus, i f I can reproduce the conditions of my concern i n a laboratory setting, and of the n variables, control (n-1) while manipulating one variable i n a systematic way, then I w i l l be able to observe what effects d i f f e r e n t combinations of the variables have on taking  risks. In this way, the laboratory becomes one test s i t u a t i o n for  general statements I am making about risk-taking behavior. Experimental Design  In the following experimental design I am c o n t r o l l i n g for wealth, u t i l i t i e s , information and incentive conditions.  The experi-  mental design has to reproduce: (1)  a decision task whereby the incentive conditions vary with respect to a given outcome;  (2)  a wealth gradient;  (3)  an information  (4)  another consideration i s that a l l the propositions refer to states  gradient;  of nature where the u t i l i t y for the outcome remains constantly high (see Theoretical Propositions, Chapter 2). This means that I have to screen a l l subjects who p a r t i c i p a t e i n the experiment as being analogous to the HU category.  - 58 -  Decision Task and Incentive Conditions  Recall that decision models comprised loss and gains factors i n terms of the equation V = U xP g  s  + U^xP^; and that I conceptua-  l i z e d V as the value attached to the decision, U xP as the u t i l i t y s s function of the r i s k paying o f f and U_xP as the u t i l i t y function of 1 JC s the  r i s k not paying o f f .  These considerations can be reproduced i n a  duplex gambling design developed by Slovic (1968). The gamble has four components, a chance of winning (P ), w an amount to win ($ ), a chance of losing (P,) and an amount to lose w I  ($!>• This can be v i s u a l i s e d i n terms of two discs (Figure 6): Figure 6:  Duplex Gamble  The l e f t hand disc represents the chances of making gains (60%  chance of winning $4), the right hand disc represents the odds on  accruing losses ( i . e . 40% chance of losing $6). These 4 components can be varied systematically, i n terms of the  gross categories of high, medium and low that I have been dealing  with throughout, and this gives me 27 discrete gambles (Appendix 1). The decision task was for subjects (selected on the basis of HU, see below) to decide which of the 27 gambles they would l i k e to  - 59 -  play for r e a l stakes.  They were asked to rate the bids on a -5 to +5  degree of attractiveness scale. scale represented  They were told that the zero on the  an indifference point where the gamble appeared n e i -  ther a t t r a c t i v e nor unattractive.  Then they were asked i f they would  bet on the gamble. Those gambles that they decided  they would l i k e to play, they  i n fact played for r e a l stakes (rather than for hypothetical payoffs and losses) after completing  the decision task for a l l 27 gambles.  The actual playing was conducted on a roulette wheel marked into percentage segments as below (Figure 7): Figure 7:  Gambling Wheel  A pointer which could be spun was attached  to the centre of  the wheel and to determine how a subject would fare on any p a r t i c u l a r bet two sp'ins at the wheel were necessary. For instance consider the bet: chance of losing $6.  60% chance of winning $4, 40%  On the f i r s t spin i f the pointer comes to r e s t  anywhere i n the 10-60 area of the wheel then the subject wins $6, i f i t f a l l s outside this area ( i . e . i n the 70-100 area) then he does not win $6.  (The d i v i s i o n s on the disc represent p r o b a b i l i t y i n spatial  - 60 -  terms.)  On the second spin, i f the pointer finishes up i n the  10-40  area then $6 i s l o s t , i f i t f a l l s outside this area ( i . e . i n the 50-100 area) then no loss i s incurred. So i n actual playing i t i s possible to win and lose, win  and  not lose, not win and lose, and not win and not lose depending on the vagaries of the spinning wheel. I should emphasize that the actual process of playing the wheel for real stakes was not important from the experiment was  concerned.  as far as the data collected  The playing of bets was  to simulate  r e a l i s t i c conditions. The information I wanted lay i n the questionnaire -- the record of actual decision making -- as i t was  here that the subject made  his choices of what bets he would l i k e to play on the wheel for r e a l stakes. The whole questionnaire was played.  completed before any bets were  Whether or not subjects won.^or lost was not important  data  for the purposes of this experiment. While completing  the decision task (before playing), the sub-  jects were instructed to regard each gamble as discrete i n terms of the constraints that were imposed upon them (Appendix I I : Subjects). These constraints consisted of: (a)  v a r i a t i o n i n starting stake -- wealth f a c t o r ;  (b)  v a r i a t i o n i n information about the gamble;  (c)  a $1 ante for every bet they decided to play.  Instructions to  - 61 -  Wealth Gradient  The  subjects were given a salary for p a r t i c i p a t i n g i n the  periment, but they had in salary was The  to use their salary to bet with.  ex-  The  variation  $1 ante for any bet played by the subjects was  irrespec-  $1 (low), $3 (medium) and $5  (high).  tive of the subject's salary. The  implications  that the low wealth ($1) for  any bet played, and  to lose; the medium and and  of this for the actual playing of bets i s  i n d i v i d u a l has  to surrender his whole salary  thus i s in a position where he cannot afford high wealth individuals can absorb the $1 ante  s t i l l have a chance of remaining i n the game. Recall i n my  i n i t i a l discussion  of peasant farmers, a major  c h a r a c t e r i s t i c of a farmer being in the LWHU state was  that he was  in  a p o s i t i o n where he could not afford to lose. In order to calculate wealth or resource gradients for other substantive settings I use this c h a r a c t e r i s t i c as a baseline  i n order  to estimate an appropriate gradient v i s - a - v i s context. In the context of hard up students gambling away their experimental salaries I estimated that a $1 salary f u l f i l l s the requirements of the baseline  category.  - 62 -  Playing For Real Stakes  I t has been a feature of much of the experimental work on decision making that the experimenters have assumed that an i n d i v i d u a l  1  behavior w i l l be l i t t l e d i f f e r e n t under hypothetical conditions than in r e a l l i f e situations where he has to accept the consequences of h i s actions. Slovic showed that when subjects played for hypothetical stakes they maximised gain and ignored p o t e n t i a l , hypothetical losses. However when subjects knew they would actually play some of the gambles they were more cautious, preferring r e l a t i v e l y higher P , lower P^ and lower $•]_.  (Slovic 1969) His study also points to an area that has largely been neg-  lected by t h e o r i s t s , namely the e f f e c t that the i n i t i a l f i n a n c i a l posi t i o n may have upon decision making. In this experiment I attempt to accomodate these two factors. F i r s t of a l l the bets were played f o r real money, and the subjects gambled with their own money -- (their s a l a r i e s ) . Also, subjects were grouped into Low, Medium and High salary groups so that, a l l other things being equal, I could observe the d i f f e r e n t i a l e f f e c t s on decision making caused by a change i n f i n a n c i a l standing.  - 63 -  Information  The 27 gambles were the same for every subject, the only difference being that the information dispensed varied. The L.Inf subjects were given a vague verbal description of each bet i n terms of the number of people who play i t and whether or not  they were good gamblers (see Appendix I I ) . The gradations i n description were a function of the ex-  pected value (E.V.) of the bet as below.  (See Appendix I)  E.V.  Information Given  y 2  Many good gamblers have played this bet.  .5-2  Of the people who have t r i e d t h i s , some have been good gamblers.  -.5 - .5  Some people have t r i e d this bet.  -2.5 - -.5  Of the people who have t r i e d t h i s , some have been poor gamblers.  <-2.5  Only poor gamblers try this bet.  The subjects were told that good gamblers were generally successful at the wheel while poor gamblers were generally unsuccessful. M.Inf subjects were given a verbal description but with more specifics,  e.g.:  Bet-A: Good Chance of High Winnings, Poor Chance of Moderate Losses 0? ) W  .  C$w)  (P^  ($i)  They were informed of the range of p o s s i b i l i t i e s of the verbal description.  - 64 -  A good chance of winning/losing  -~  70% - 1007<>  A f a i r chance of winning/losing  =  40% -  707>  A poor chance of winning/losing  =  07o -  407>  . and High winnings/losses;  =  $5 - $10  Moderate winnings/losses.-  =  $2 - $5  Low winnings/losses-  =  $0 - $2  The H.Inf subjects received the same information as the M.Inf subjects, but they could also have any 2 of the 4 bet components spec i f i e d as to the precise monies and p r o b a b i l i t i e s . So the task for the subject was to make a decision whether to play a gamble or not i n terms of the information he has, his i n i t i a l salary and the knowledge that the bet w i l l cost him $1 to play. I t was emphasised to the subjects that they regard each bet as discrete i n terms of the above constraints, while they completed the questionnaire.  Subjective U t i l i t y  The scope conditions specified that decision-makers have a high subjective u t i l i t y for the outcome associated with the event they are making decisions about. In the experiment the decision situation involves gambles f o r money, therefore the outcome to which HU must be attached i s more money.*  *  Abbreviations are used throughout for experimental  categories.  - 65 -  To screen my subjects (male students at U.B.C.) into an HU category, I obtained a l i s t of male students who were registered at the U n i v e r s i t y Placement Office on the assumption that students so registered would have a high u t i l i t y for increasing money stocks as they were looking for part time jobs.  (I took only those students who  had not found jobs). The second part of the screening was to ask subjects to i n dicate their preference rating for increasing money stocks from the experiment.  Only those who had a high preference rating ( i . e .  subjec-  tive u t i l i t y ) took part i n the experiment.  Table I : Summary of Scope Conditions  Wealth Information  MW $3  LW $1 LOW  ;  vague description :  Incentive Conditions  Investment Factor  Subjective Utility  Med ium description..  P„$ " w  -  gains factor  P-^$^  -  loss factors  -$1  HW $5 High description and data  27 d i f f e r e n t combinations -$1  A l l Subjects screened as HU for increase i n money stocks  -$1  - 66 -  Data C o l l e c t i o n The data from the experiment was put into a simple 2 way matrix (using wealth and information as parameters) f o r easy reference (Figure 8) . Figure 8:  Data Matrix WEALTH  Information  L  M  H  L  1  2  3  M  4  5  6  H  7  8  9  ( I t should be pointed out that subjective u t i l i t i e s are held constant at HU.) The bet components i n c e l l s 4, 5, 6 were converted into numbers by taking an average of the range to which the verbal description applied. For  instance:  "High Chance of Moderate Payoffs" referred to  a 707 - 1007o p r o b a b i l i t y range, moderate payoffs indicate a range of o  $2-$5.  This data was-recorded as 857» chance of $3.50. In c e l l s 7, 8, 9 the same calculations were used, except f o r  the  two components for which information was sought.  As the master  sheet consisted entirely of averages of the range of p o s s i b i l i t i e s , the  bet components for these c e l l s were the same as f o r c e l l s 4, 5, 6.  The type of data collected i s recorded i n Table I I .  - 67 -  Table II".'. Data From Experiment Components of Bets (Averages of Range)  Information Sought by Subjects  Cell No  Bets Played  Rating of Bets  1 2 3 4 5 6 7 8 9  X X X  X  X X X X X  X X  X  X  X X  X  X  X  X  X X X X X  X X  X  In order to categorise the bets into levels of riskiness I obtained a random sample of 20 from my ninety experimental subjects and asked each one to evaluate the gambles on the master sheet i n terms of whether they thought the bet was  s l i g h t l y risky, f a i r l y r i s k y , or very  risky. From these 20 sets of evaluations, I assigned bets to one of these three categories i n terms of 80% agreement among evaluating subj e c t s , i . e . i f 807> of the subjects agreed to assign bet i to category j then bet i was assigned to category j . Had  there been instances of less than 807° agreement on any  bet then I would have assigned i t to one of the three categories i n terms of the E.V.  of the bet.  But this did not arise.  of bets into categories of riskiness was determined  Thus my  scaling  empirically (Table I I I ) .  An even more important implication i s that the objective values given to the variations i n incentive conditions corresponded the subjective evaluations by the subjects.  Had  closely with  there been large diver-  - 68 -  gences, i n subjective evaluation of the bet dimensions, between subjects then this would have shown i t s e l f i n the categorisation of bets into degrees of r i s k i n e s s . This i s s i g n i f i c a n t i n terms of comparative purposes, as the data on the fishermen (who  constituted the f i e l d test) was on their  subjective evaluation of alternatives. Table I I I ;  Bets Categorised by Riskiness S l i g h t l y Risky [SR]  Bet No:  F a i r l y Risky  Very Risky  [ER]  [VR].  7 10 12 14 16 18 23 26  1 3 4 11 15 17 20 21 24  2 5 6 8 9 13 19 22 25 27  8  9  10  Total  The data was  f i r s t of a l l recorded per c e l l i n terms of (a)  bets categorised by subject (Table IV) and (b) frequencies of bets played (Table V).  These raw data were then transferred into a 2 way  matrix for each l e v e l of riskiness (Table VI). The data i n these matrices were corrected for the unequal number of bets i n each r i s k category ( v i z : there were 8 SR bets, 9 FR bets, 10 VR bets).  V I N V  ^  . .  I'IJ T  A  (N  )  . .  rb'ij  til ttl i s the c e l l entry i n i row and j column; B i s the number of bets taken i n that c e l l ;  ( N ) . . x (N . ).. i s the number of possible bets I i] rb i ] that could have been taken i n that c e l l where N^. i s the number of T  individuals i n the c e l l and N , i s the number of risky bets i n that c e l l . rb This means that i n order to get the new c e l l entry, the number of bets taken i n each c e l l i s divided by the number of possible bets that could have been taken.  M u l t i p l i c a t i o n by 100 converts the index  into a percentage (Table VII).  These stages of data  transformation  are represented i n Tables IV-VII (Appendix I I I ) . Testing Recall that my t h e o r e t i c a l propositions were as follows: 1'.  Individuals w i l l show a preference f o r low r i s k strategies with respect to decisions involving the outcome to which HU i s attached.  Conditions: 2'.  LW and HU.  Individuals w i l l show an increased tolerance f o r high r i s k s t r a tegies, with respect to the outcome to which HU i s attached, as wealth increases.  Conditions: 3'.  HU.  Individuals w i l l show an increased tolerance f o r high r i s k s t r a tegies, with respect to the outcome to which HU i s attached, as information increases.  Conditions:  HU.  - 70 -  The data matrix reproduced below i s to f a c i l i t a t e  reference  to c e l l numbers (Figure 9): Figure 9:  Data Matrix WEALTH  Information  L  M  H  L  1  2  3  M  4  5  6  H  7  8  9  Proposition 1' Given low wealth, individuals w i l l show a preference  for low  r i s k strategies with respect to decisions involving the outcome to which HU i s attached. In operational terms proposition 1' means that individuals i n c e l l s 1, 4, 7 w i l l take fewer risky bets than individuals i n c e l l s  2,  5 and 8 respectively. The accompanying graphs p l o t SR bets, then a combination of FR and VR bets, over a wealth increment from low to medium; holding i n formation  constant.  The graphs indicate that the ordinal relationships  required to support proposition 1' do i n fact hold —  but they also show  much more. Graph 1 shows that there i s a s i g n i f i c a n t difference between L.Inf and M.Inf categories, but not such a large difference between M.Inf and H.Inf categories.  From Graph 2 i t i s reasonable to i n f e r  - 71 -  GRAPH 1 L.W P r e f e r e n c e for  Low Wealth %  of  in S.R.  S  . 1  Bets  31.3  Low  Risk  Strategies:  Med.  Info. 3  1 81.3  51.3  Info.  Proposition 1  High  3  1  91.3  85  Info. 3 93.8  100'  H.INF. M. INF 80-  60  % of SR. BETS TAKEN  L INF 4  0  20  1  WEALTH IN S  3  GRAPH 2 LW  Preference  for  Low Wealth in S %  of  & VR.  Low  Risk  Info.  1  3  0  3.3  Strategies:  nfo.  Med.  F.R.. Bets  H i g h Info.  3  1  25.6  178  Proposition 1  1  3  25.6  38.4  100  80-  60-^ % o f and  V R.  ' BETS TAKEN 40'  ..-H. INF ^  M. INF  20-  L.INF \1  i  i  ~  3  WEALTH IN %  - 73 -  that the three information  levels show differences, but that within  the L.Inf category the ordinal relationship may not indicate a s i g n i f i c a n t difference. On the basis of these r e s u l t s one could accept the data as providing support for proposition 1.  However, there i s a need for  further testing i n that one could argue that the differences between c e l l s may not be s i g n i f i c a n t . Recall that I v i s u a l i s e d low r i s k strategies as attempting to minimize the loss i n a decision s i t u a t i o n . I can get an additional test to proposition 1 by showing whe ther or not individuals i n d i f f e r e n t c e l l s made their decisions as a function of minimizing $-^. The method i s to calculate the maximum pos s i b l e loss for each subject i n the c e l l s I am interested i n ( i . e . the bets played by each subject are summed as i f each one had a negative outcome). The next step i s to calculate the range of maximum possible losses for each c e l l , and to do a chi square median test on the c e l l s I want to compare, to ascertain whether the differences between the c e l l s are s t a t i s t i c a l l y s i g n i f i c a n t . The n u l l hypothesis to be tested i n each case i s that the median amount of loss w i l l be the same for each c e l l .  (Appendix IV)  - 74 -  Table VIII:  Chi Square Median Test:  Chi  Squ.  Proposition 1'  Significance level  Null Hypothesis  C e l l 1 cf C e l l 2  7.46  p =  .01  Reject  C e l l 4 cf C e l l 5  5.00  p =  .05  Reject  C e l l 7 cf C e l l 8  0.20  Not Significant  The c h i square of 7.46 are.very  Not Reject  for c e l l s 1 and 2 implies that there  s i g n i f i c a n t differences between the medians of both c e l l s ,  and that there i s only a 1% p o s s i b i l i t y that these r e s u l t s are due chance.  The c h i square for c e l l s 4 and 5 s i m i l a r l y enables one  to  to  r e j e c t the n u l l hypothesis at the 5°L l e v e l of significance. But the c h i square of 0.20  for c e l l s 7 and 8 means that the  n u l l hypothesis cannot be rejected, and implies that at high levels of information there i s no s i g n i f i c a n t s t a t i s t i c a l difference between the way LW and MW  individuals seek to minimise their losses.  The graphs provide data to support proposition 1' as they indicate that the ordinal relationships required of the data do i n fact occur.  However the c h i square median test shows a facet concealed  by the graphs, v i z :  that LW  do indicate a preference tion i s high.  individuals (as d i s t i n c t from MW  individuals)  for low r i s k strategies except when informa-  From both tests we can infer that proposition l holds at 1  low and medium levels of information but not under conditions of high information.  (In terms of this l a s t finding i t i s s i g n i f i c a n t that i n  - 75 -  the f i e l d and from secondary sources I did not discover one empirical example of a LW individual having access to high information.  This  seems to indicate that while LW/HINF i s a l o g i c a l p o s s i b i l i t y and  can  be set up i n an experiment i t i s not to be found i n the empirical world. This lack of empirical correspondence suggests that there are a number of intervening variables which makes LW  incompatible with high  informa-  tion. However, what this r e s u l t states i s that i f LW individuals do have HINF (with respect to a p a r t i c u l a r outcome) then they w i l l similar types of r i s k s as do MW/HINF individuals.  The important  take  task  is to delineate the intervening variables that prevent the LW/HINF category from occurring.)  Proposition 2' "Individuals w i l l show an increased tolerance for high r i s k strategies, with respect to the outcome to which HU i s attached, as wealth increases." In the accompanying graphs I have plotted the proportion of bets taken as wealth increases (holding information constant) for each l e v e l of information.  The f i r s t graph deals with SR bets, the second  with a combination of FR and VR  bets.  In operational terms proposition 2' means that for each l e v e l of information, increased wealth should r e s u l t i n an increase i n the number of bets taken for each r i s k category.  - 76 -  The f i r s t graph shows some interesting r e s u l t s .  MW/L.Inf  subjects take 20% more SR bets than LW/L.Inf subjects, but there i s only a 6% difference between MW/L.Inf and HW/L.Inf.  To a lesser extent  this relationship holds for M.Inf and H.Inf subjects; i . e . that there i s a bigger increase i n r i s k s taken as wealth moves from low to medium, then when wealth moves from medium to high. From the second graph at L.Inf, HW and MW subjects took the same proportion of risky bets, (which was higher than LW i n d i v i d u a l s ) , though the M.Inf and H.Inf levels showed a more nearly linear increase with respect to wealth. These results seem to indicate that a sort of "step function" i s operative i n that there i s a much greater increment i n risky bets taken between LW and MW individuals than there i s between MW and HW individuals.  The data i s s u f f i c i e n t to provide overall support for  the ordinal relationship between wealth and r i s k taking, but supplementary tests may indicate more about the precise nature of this r e l a t i o n ship . By doing the obverse of the procedure used to test proposition 1', a chi square median test can be established.  I f we consider high  r i s k strategies as those that seek to maximize gain ( i f the outcome i s p o s i t i v e ) ; then by summing the maximum possible payoff f o r each subj e c t one has an index of h i s tolerance f o r high r i s k strategies. The n u l l hypothesis being tested i s that the median amount of gains w i l l be the same f o r each c e l l .  GRAPH RISK  TAKING  Low  w.r.t.  Info.  1  3  5  % of S.R. Bets Taken  313  51.3  57.5  1  WEALTH  Med.  W e a l t h in 8  1  3  1  1  81.3  Info. 3  5  91.3  91.3  1  3 WEALTH IN $  High  Info.  1  3  5  93.8  96.3  85  !  1  5  - 78 -  GRAPH RISK  Wealth in S %  TAKING  w.r.t.  WEALTH  Info.  High  Info.  Low  Info.  1  3  5  1  3  5  1  3  5  0  3.3  3.3  17.8  25.6  30  25.6  38.4  46  of F.R.  & VR. Bets  4  Med.  100i 90807060% of FR.&VR. 5 0 i BETS TAKEN "  "MM*-.  H.INF  A 0  M.INF  30' 2010-  UNF  WEALTH IN S  1 - 79 -  Table IX: Chi Square Median Test;  Proposition 2'  Chi Squ.  Significance level  1 cf 2 cf 3  15.2  Reject  1 cf 2  3.4  p = .01 close enough to P = .05  2 cf 3  1.8  not s i g n i f i c a n t  Not Reject  4 cf 5 cf 6  6.8  P = .05  Reject  5.0  P = .05  Rej ect  0.1  not s i g n i f i c a n t  Not Rej ect  7 cf 8 cf 9  3.2  close to p = ..05  Rej ect  7 cf 8  0.1  not s i g n i f i c a n t  Not Reject  8 cf 9  1.8  not s i g n i f i c a n t  Not Reject  4 cf 5 5 cf 6  '  N u l l Hypothesis  Rej ect  The f i r s t six results confirm the idea of a step function. Overall, as wealth increases, individuals i n the M.Inf and H.Inf categories do show a tolerance for higher r i s k strategies. But when this i s broken down one finds for L.Inf a s i g n i f i 2 cant difference between c e l l s 1 and 2 (X r e s u l t between 2 and 3.  = 3.4), and a non-significant  S i m i l a r l y for M.Inf, the difference between 4  and 5 i s s i g n i f i c a n t , while the difference between 5 and 6 i s not.  o The test for H.Inf l e v e l , as wealth increases, gives an X of 3.2 which i s close enough to the .05 l e v e l of significance (3.8) to make the inference that, o v e r a l l , subjects with H.Inf do tend to indicate d i f f e r e n t i a l tolerance levels for r i s k as wealth changes.  - 80 -  When this i s broken down, however, one finds that there are not s t a t i s t i c a l l y s i g n i f i c a n t differences between c e l l s 7 and 8, 8 and 9, v i z :  and  H.Inf conditions produce non-significant differences i n  tolerance for r i s k , as wealth increases. For each information gradient the graphs and test support  chi-square  the notion of tolerance for r i s k increasing with wealth.  More s p e c i f i c a l l y there i s a cutting off point at MW where though the graphs show a difference, the chi square test indicates that i t i s not statistically significant.  One  mental d i s t i n c t i o n between MW  reason for this may  ($3) and HW  be that the experi-  ($5) i s inadequate.  However the important inference to make i s that there i s a big jump i n tolerance for high-risk strategies as wealth moves from low to medium,except under conditions of high  information.  Proposition 3'  Individuals w i l l show an increased tolerance for higher strategies, with respect to the outcome to which HU i s attached,  risk as  information increases. In operational terms this means that as information increases (holding wealth constant) more risky bets w i l l be taken for each wealth level.  Graph number 6 indicates that these ordinal relationships hold  for FR and VR bets. indicate two things:  With respect to SR bets graph number 5 seems to (a) a threshold for r i s k at the MW  very l i t t l e difference between the MW  and HW  level,  general categories.  (b)  - 81 -  GRAPH 5 RISK  TAKING  Low Info. %  of  w.r.t.  Wealth  L  M  31.3  81.3  INFORMATION  Med H  High  Wealth  L  M  51.3  91.3  H  Wealth  L  M  H  57.5  91.3  96.3  S.R.  Bets Taken  85  93.8  - 82 -  GRAPH 6 RISK  TAKING  w.r.t.  L o w Wealth  Info. % of  INFORMATION  Med. W e a l t h  L  M  H  L  M  0  178  25.6  3.3  25.6  High  H  Wealth  L  M  H  3.3  30  46  FR.  & VR. Bets  38.4  100-1 908070-  INFORMATION (N. B.:  Ordinal  Not  Metric  Intervals)  - 83 -  The important implication from the f i r s t graph i s that L.Inf subjects took 50% fewer SR bets than M.Inf subjects for each wealth level.  The supplementary test to assess the significance of these d i f -  ferences i s exactly the same as for proposition 2', the only difference being that d i f f e r e n t c e l l s are compared.  The n u l l hypothesis being  tested i s that the median amount of gains w i l l be the same for each cell.  Table X:  Chi Square Median Test:  Proposition -3'  Chi Squ.  Significance level  1 cf 4 cf 7  13.5  .01  Reject  1 cf 4  16.1  .01  Reject  ___  Not Reject  4 cf 7  1.8  Null  Hypothesis  2 cf 5 cf 8  16.8  .01  Rej ect  2 cf 5  16.1  .01  Reject  ___  Not Reject  .01  Reject Reject  5 cf 8 3 cf 6 cf 9  0.2 16.8  3 cf 6  9.7  .01  6 cf 9  1.8  _  Not Reject  These results lend support to the threshold notion, as at each W l e v e l there i s not a s t a t i s t i c a l l y s i g n i f i c a n t difference between M.Inf and H.Inf.  - 84 -  There are two possible implications of this — experimental  one that the  d i s t i n c t i o n between M.Inf and H.Inf was not i n fact a  good dividing l i n e , or that increments i n information a f t e r the medium l e v e l are not s i g n i f i c a n t i n terms of increased tolerance for high-risk s t r a t e g i e s . Of these a l t e r n a t i v e inferences, i t i s reasonable  to accept  the second one, as the test of the f i r s t proposition indicated that H.Inf has an effect on LW individuals such that they do not take s i g n i f i c a n t l y d i f f e r e n t risks when compared to MW/H.Inf individuals (this was  i n terms of a preference for low r i s k s t r a t e g i e s ) .  Whereas  M.Inf operates on LW individuals to produce a s i g n i f i c a n t l y d i f f e r e n t pattern of r i s k preferences when compared to LW/M.Inf individuals. to these d i f f e r e n t effects i t i s reasonable  Due  to assume that High Infor-  mation i s d i f f e r e n t from and greater than Medium Information. The implication of this r e s u l t i s that information increments after the range of odds, costs and payoffs have been s p e c i f i e d (M.Inf) do not s i g n i f i c a n t l y influence an i n d i v i d u a l ' s tolerance for high r i s k strategies. Risk was loss.  defined as having attributes of uncertainty and  Information acts to lower the uncertainty, but the experiment  shows that there i s an optimal point at which tolerance for high r i s k i s not affected by additional increments i n information.  This cutting  off point seems to be a s p e c i f i c a t i o n of the range of odds and payoffs s p e c i f i c to a p a r t i c u l a r decision.  In order to extract the maximum  - 85 -  information from the controlled s i t u a t i o n extant i n the experiment, I intend to test some of the f i r s t order propositions to see i f the experimental data can provide additional information as to the domain of relevance for my assertions. Farmers, Risks and Experimental Data From Chapter 2 r e c a l l that I made a number of propositions r e l a t i n g to farmers and innovations, e.g.: 1.  Innovations with a high payoff factor which c a l l for a minimal diversion of stocks from existing resources w i l l tend to be adopted.  2.  Innovations that o f f e r a low return tend not to be adopted.  3.  Individual farmers seek to minimise the cost factor of an innovation.  Conditions:  LW/HU.  By translating these propositions into more abstract statements r e l a t i n g to r i s k I should be able to use the experimental data to test the f i r s t series of statements. I conceptualised the decision to accept or reject an innovation i n risk-taking terms, so the propositions above  with regard to  farmers and innovations can be translated into statements  relating  individuals and r i s k i n the following way: A.  Risks with a high payoff factor and a low cost factor w i l l tend to be taken.  B. *  Risks with a low payoff w i l l tend not to be taken.* Proposition 3 has been d i r e c t l y tested.  - 86 -  Conditions:  LW/HU.  A can be operationalised by considering gambles whose u t i l i t y functions f o r payoffs ($ .P ) have an E.V. } 2 and whose u t i l i t y funcw w tion for losses ( ^ ' ^ j ) have an E.V. < 1.  What I am testing i s whether  or not LW individuals take "sure thing" bets. The calculation of r e l a t i v e E.V.s l i m i t s consideration to bets 7, 12, 14, 16, 18 and 23. subjects at M. and H.Inf l e v e l s .  In the table below I have compared LW (The data collected f o r low informa-  tion levels does not permit a c a l c u l a t i o n of E.V.) Table XI:  Bet No. Cell 4 LW/M.Inf Cell 7 LW/H.Inf  "Sure Thing" Bets Subjects Per C e l l  7  12  14  16  18  23  9  10  10  8  10  7  10  7  8  10  7  10  8  10  Of the 12 cases i n the table, there are 5 cases of a l l subjects playing the bets, 1 case with 90%, 3 cases with 80% and 3 cases with 70%. These data are s u f f i c i e n t to provide support for proposition A. Proposition B:  Risks with a low payoff w i l l tend not to be taken.  Operationally this means that subjects i n LW categories w i l l not take bets 6, 8, 11, 13, 15, 17, 19, 21 and 27.  The table below rep-  resents data collected from LW/M.Inf and LW/H.Inf c e l l s .  (Information  - 87 -  supplied to L.Inf subjects  did not specify whether payoffs were low,  medium or high). Table XII:  Low Payoff Bets No. of Subjects Playing Bet  Bet  LW M.Inf (4)  (7) LW H.Inf  6  0  0  8  0  0  11  5  5  13  0  0  15  1  3  17  0  1  19  0  0  21  1  1  _0  _0  10  10  No.  27 No. of Subjects Per C e l l  With the exception of bet 11 f o r c e l l 4 and 7, and bet 15 for c e l l 7, the data i n the table supplies general support f o r the contention of proposition B. Bet 11 referred to a good chance of low payoffs, and a low chance of low losses. Bet 15 referred to a good chance of low payoffs, low chance of moderate losses.  - 88 -  In both c e l l s 4 and 7 50% of the subjects took bet 11 which indicates a tolerance for low payoffs when (a)the loss factor i s n e g l i g i b l e a r e  ^- ) ow  a n  ^ ^ ) when the chance of the low payoff i s good  (P i s high). w In c e l l 7 30% of the subjects took bet 15, i n d i c a t i n g that some LW subjects, under conditions of H.Inf, would take a low payoff bet when (a) the chance of the low payoff i s good (P high) and (b) w the loss factor has a low  and a moderate P^.  These cases indicate that there are instances where low payoffs are a t t r a c t i v e to LW subjects —  and the attractiveness seems to  be a function of the p r o b a b i l i t i e s of winning and l o s i n g . These results led me to test for the o v e r a l l relevance of the chance factor for LW subjects. I d i d this by estimating the r e l a t i v e weightings given by LW subjects to each of the 4 r i s k dimensions ($ , P , $.. , P,). w w l I  This can  be done by c o r r e l a t i n g each subject's evaluation .of bets with the l e vels of each r i s k dimension across the 27 gambles. I am assuming that the correlations between r i s k dimensions and evaluation of bets i s d i r e c t l y proportional to the weights i n a linear regression equation c h a r a c t e r i s t i c of each subject. The model underlying this analysis uses A(G) = u + w,$ + w P + w $, + w.P, (Slovic 1968: 12) lw 2w 31 41 n  where A(G) i s the evaluation of the gamble, u i s the mean attractiveness of a set of gambles, and w_^ are weights r e f l e c t i n g the r e l a t i v e importance of each dimension.  - 89 -  This model assumes a d d i t i v i t y of components and also that the import of p r o b a b i l i t i e s and payoffs i s a l i n e a r function of t h e i r objective values  (Slovic 1968:  13).  The value of this format i s that the model permits independent assessment of the r e l a t i v e influence of a l l 4 incentive dimensions,  Table XIII:  Relative Weighting of Risk Components Subject No.  Correlations Between Risk Dimensions and Ratings of Attractiveness $ w  P  1 2 3 4 5 6 7 8 9 10  0.55 0.38 0.28 0.32 0.40 0.60 0.16 0.42 0.47 0.42  0.66 0.68 0.74 0.77 0.65 0.49 0.47 0.66 0.68 0.76  -0.40 -0.44 -0.32 -0.40 -0.52 -0.28 -0.39 -0.47 -0.41 -0.36  -0.66 -0.68 -0.74 -0.78 -0.65 -0.49 -0.44 -0.66 -0.68 -0.76  1 2 3 4 5 6 7 8 9 10  0.32 0.34 0.58 0.54 0.45 0.28 0.33 0.38 0.28 0.35  0.66 0.76 0.50 0.66 0.72 0.79 0.74 0.75 0.87 0.77  -0.38 -0.39 -0.17 -0.22 -0.39 -0.17 -0.42 -0.12 -0.03 -0.38  -0.64 -0.75 -0.50 -0.65 -0.72 -0.77 -0.74 -0.73 -0.87 -0.77  y  Cell 4 LW M.Inf  Cell 7 LW H.Inf  w  $. l v  P,  1  In both c e l l s 9 out of 10 subjects weighted the chance factors (P ; P T ) higher than the monetary factors, from which one can i n w  1  fer their r e l a t i v e importance to a LW operator. account for the exceptions  This finding helps to  to proposition B, i n that the payoff factor  - 90 -  (low $ ) was w  not a d e t e r r e n t when the P  r e s u l t s are summarised i n T a b l e  T a b l e XIV:  w  was  h i g h and P.. low. i.  XIV.  Summary o f E x p e r i m e n t a l  Test  Accept/Reject  C o n d i t i o n s of A p p l i c a b i l i t y  Exceptions  Prop.  1'  Accept  L . I n f ; M.Inf;  H.Inf  Prop.  2'  Accept  LW  MW  Prop.  3'  Accept  L . I n f - M.Inf; LW;  Prop.  A  Accept  LW  Prop.  B  Accept  P (low - med); ? (Med hxgh); $ (med - h i g h )  - MW;  L . I n f ; M.Inf; MW;  HW  ±  one  - HW;  H.Inf  M.Inf - H.Inf None  1  From T a b l e XIV  These  -  P h i g h ; P-L low; $ j low w  can see t h a t the major e x c e p t i o n s to  my  a s s e r t i o n s about r i s k t a k i n g u s u a l l y i n c l u d e a s i t u a t i o n where c o n d i t i o n s of High I n f o r m a t i o n are p r e s e n t . t h a t i t was  reasonable  I argued  above ( page 84 )  to a c c e p t t h a t High I n f o r m a t i o n was  greater  than Medium I n f o r m a t i o n . T h i s l e a v e s one w i t h the c o n s i d e r a t i o n of a t h r e s h o l d n o t i o n , whereby once an o p t i m a l p o i n t i s reached w i t h r e s p e c t to i n f o r m a t i o n r e c e i v e d , then f u r t h e r increments  i n i n f o r m a t i o n a f t e r t h i s p o i n t do  not make any s i g n i f i c a n t d i f f e r e n c e s experiment  i t i s p o s s i b l e to i d e n t i f y  to d e c i s i o n making. this  t h r e s h o l d as knowledge of  the range of odds and p a y o f f s a t t a c h e d to any The  From the  alternative.  t e s t of p r o p o s i t i o n 2', which r e l a t e d wealth  i n d i c a t e d t h a t HW  i n d i v i d u a l s d i d not take s i g n i f i c a n t l y  to r i s k ,  different  - 91 -  types of risks to MW i n d i v i d u a l s .  One could accept this r e s u l t as i t  stands, though one could also argue that the experimental d i s t i n c t i o n between MW and HW was not i n fact a good one. However, the important thing to note i s that the results are quite conclusive i n terms of supporting  the notion that LWHU individuals prefer low r i s k s .  This  was further endorsed by the test of proposition A which produced no exceptions  and proposition B —  the exceptions  to which i n actual fact  support the above contention. With reference to proposition B I had argued that LWHU i n dividuals would not take risks for marginal benefits.  The exceptions  to this proposition indicated that some LWHU people w i l l take r i s k s for marginal payoffs when the remaining incentive conditions are strongly i n their favour, i . e . when conditions of high P , low P w 1 and low $^ are present. So the testing of propositions 1', A and B gives  strong  support to the notion that low r i s k strategies are preferred by LWHU individuals.  Other findings from the experiment suggest that the  chance factors involved i n a decision are the most important ones f o r LW i n d i v i d u a l s . Implications for F i e l d Test  In the f i e l d context of B.C. g i l l n e t fishermen I w i l l be subjecting the same propositions to t e s t .  The sample of fishermen  used w i l l be characterised by HU for the outcome "increase f i s h i n -  - 92 -  take", and I w i l l be analyzing their r i s k taking behavior subject to constraints of information, wealth and the incentive conditions attached to alternatives. As a result of the experimental test I w i l l be expecting s i g n i f i c a n t differences i n r i s k taking behavior between low wealth and medium wealth fishermen, i n terms of the former showing a d i s t i n c t preference for low r i s k strategies.  I w i l l also expect low  wealth g i l l n e t t e r s to take advantage of high return/low cost investments but not to go f o r marginal payoffs except when the remaining incentive conditions are strongly i n their favour.  In other words  they w i l l only go for marginal payoffs when there i s a good chance that they w i l l be successful, and a low chance that they w i l l lose anything by i t .  tor  It w i l l also be i n t e r e s t i n g to see whether the chance fac(P , P ^ that was so important to LW subjects i n the experiment w J_  i s as important to LW fishermen i n the f i e l d . These considerations discussed above a l l refer to a combination of factors which amount to a preference f o r low r i s k s .  The  main expectation of the f i e l d test (based on the experimental results) i s that LWHU fishermen w i l l be low r i s k takers. The wealth gradient for fishermen i s calculated from the same baseline or property used i n the experimental calculation, i . e . a LW i n d i v i d u a l i s defined as one who  cannot afford to lose.  It could  be argued that the experimental d i s t i n c t i o n between MW and HW was  not  i n fact a good one.  This makes i t imperative to have a better opera-  t i o n a l d e f i n i t i o n f o r the f i e l d test.  To this end, I consulted the  Department of Welfare's minimum wealth standards of e l i g i b i l i t y for f i n a n c i a l assistance i n addition to assessing from fishermen their " f o l k " view of the dividing lines between wealth l e v e l s . From these two sources I constructed a s l i d i n g scale which matched o f f a man's income with the number of dependants he had to support. The experiment showed s i g n i f i c a n t increases i n tolerance for high risks as wealth moved from low to medium.  I t did not show  s i g n i f i c a n t increases as wealth moved from medium to high, (possibly due to the poor d i s t i n c t i o n between MW w i l l expect MW  and HW).  So i n the f i e l d I  fishermen to take s i g n i f i c a n t l y more high risks than  LW fishermen, and w i l l be trying to assess the difference that HW makes to decision strategies chosen. One of the most interesting was  results from the experiment  the establishment of a threshold for the influence of information  upon decision making. i n the f i e l d .  I w i l l be assessing whether or not this operates  I f i t does then my main concern w i l l be with an i n f o r -  mation gradient up to and including the s p e c i f i c a t i o n of the range of odds, payoffs and costs (medium information i n the experiment).  Field  data should show that increments i n information after this l e v e l do not affect decision strategies.  The experiment showed that increments  i n information from low to medium were accompanied by an increased  -  tolerance  f o r high  increasingly  risks.  take h i g h  94  In the f i e l d  risks  I w i l l expect f i s h e r m e n to  as i n f o r m a t i o n  increases  to the medium  level. Another r e s u l t t i o n operates on a LW ence f o r low This  from the experiment was  i n d i v i d u a l i n such a way  r i s k s t r a t e g i e s no  different  that high  informa-  as to make h i s p r e f e r -  from t h a t of a MW  f i n d i n g c o n f l i c t s w i t h the i n f o r m a t i o n  individual.  c i r c u i t s model (with  gard to t e c h n o l o g i c a l investments) i n Chapter 2.  re-  T h i s model r u l e d  the p o s s i b i l i t y of a LW/H.Inf c o n f i g u r a t i o n e x i s t i n g .  The  be  configuration  to see whether there are e m p i r i c a l examples of t h i s  i n the f i e l d .  I f there are,  model w i t h r e g a r d  then t h i s  to investment r i s k s .  t i o n i s t h a t the e x p e r i m e n t a l d e s i g n which has  no  task  out  r e f u t e s the i n f o r m a t i o n I f there are not,  dealt with a l o g i c a l  will  circuits  one  implica-  possibility  empirical referent.  Results  contrary  to those expected as a consequence of  experimental findings require lating s o c i a l marginality,  t h a t the supplementary p r o p o s i t i o n s  s o c i a l m o b i l i t y and  to r i s k t a k i n g ) be brought i n to see These l a t t e r f a c t o r s were not  the  co-operative  (re-  structure  i f they can account f o r them.  c o n t r o l l e d f o r i n the experiment  and  c o n s t i t u t e d a s e t of c o n s t r a i n t s which I thought might bear upon r i s k taking behavior. T h e i r use of p r o p o s i t i o n s taking.  i s predicated  (1', 2',  3')  on the f a i l u r e of the secondary s e t  to account f o r the phenomena of  risk  Generally, as a function of the experimental expect LW fishermen  findings, I  to be low r i s k takers, and that increments i n  wealth to the medium l e v e l w i l l be accompanied by a progressively increased tolerance for high r i s k s t r a t e g i e s .  I also expect  fishermen  to take more high risks as information increases up to the medium level.  Beyond t h i s , increments i n information should not make any  difference to decision making.  - 96 -  CHAPTER 4 TEST CASE:  FISHERMEN AS RISK TAKERS  Introduction The move from the laboratory to the f i e l d setting involved fieldwork i n a f i s h i n g v i l l a g e on Northern Vancouver Island. This provided another substantive s e t t i n g i n which to test my t h e o r e t i c a l statements.  Given that I can demonstrate that I am  dealing with the scope conditions which applied to the laboratory setting, then I can assume that both situations are v a l i d tests f o r the same set of propositions. The scope conditions specify that fishermen can be categorised according  to parameters of wealth and subjective u t i l i t y for an  outcome (the outcome i n this context i s "increase f i s h intake").  I  w i l l be analysing decisions associated with that outcome, i n terms of their v a r i a t i o n subject to constraints of information and the i n centive conditions that apply to the alternatives the fisherman has to decide between (cf my statements i n the Introduction of Chapter 3).  The Fishing Industry i n B r i t i s h Columbia The f i s h i n g industry i n B.C. i s the province's gest industry.  fourth l a r -  In 1968 the wholesale value of f i s h and l i v e r products  produced i n B.C. amounted to $119.3m.  Of this figure salmon products  accounted for $100 m. (Department of Fisheries of Canada 1969).  - 97 -  7,548 boats and 12,133 fishermen exploit the west coast f i s h eries (1968).  These are supported by 1200 workers on packers and other  a u x i l i a r y vessels and approximately 5,000 persons i n fish-processing, packing and handling plants. However the c y c l i c a l nature of the f i s h runs  (particularly  salmon) means that the majority of these 18 1/2 thousand individuals are f u l l y employed i n the f i s h i n g industry for only s i x months of the year.  Despite this fact, the importance of the industry i s that i n  many areas i t i s the only opportunity f o r g a i n f u l employment. Information from the Department of Fisheries indicates that the number of fishermen each year does not vary d i r e c t l y with the catch. Approximately  the same numbers of men f i s h i n poor years and bad years.  As the fishermen know beforehand whether i t i s going to be a poor year for f i s h i n g (from Department of Fisheries catch estimates), one reason for this " i n e l a s t i c i t y " may be the fact that the fishermen cannot f i n d suitable alternative employment. Both primary and secondary sectors of the f i s h i n g industry are seasonal, though the primary sector appears to be b a s i c a l l y more variable than the processing, packing and handling sector. The primary object of this m u l t i - m i l l i o n dollar industry i s the P a c i f i c salmon. sockeye (oncorhynchus  Five p r i n c i p a l species of salmon are caught: nerka), coho (oncorhynchus  corphynchus tschawytscha), pink (oncorhynchus corhynchus keta).  kisutch), spring (on-  garbuscha) and chum (on-  - 98 -  These s p e c i e s d i f f e r c o n s i d e r a b l y i n s i z e , m i g r a t o r y p a t t e r n s and i n commercial v a l u e .  A l l f i v e s p e c i e s come i n from the P a c i f i c to  spawn i n f r e s h water, and a r e most abundant between Southern B r i t i s h Columbia and Western A l a s k a .  W i t h i n t h i s l e n g t h o f c o a s t l i n e the geo-  g r a p h i c range o f each s p e c i e s o v e r l a p s c o n s i d e r a b l y , so t h a t some spec i e s a t the same time occupy n o t only the same g e o g r a p h i c t e r r i t o r y b u t f r e q u e n t l y the same stream. The spawning and m i g r a t o r y h a b i t s have been s t u d i e d over the y e a r s , and a f a i r l y up.  a c c u r a t e p i c t u r e o f the salmon c y c l e has been b u i l t  So much so t h a t the Department o f F i s h e r i e s i s s u e s a s e r i e s o f  p r o j e c t i o n s f o r the t i m i n g , s i z e and p l a c e o f the d i f f e r e n t salmon runs f o r each y e a r , w i t h a s u r p r i s i n g degree o f a c c u r a c y . The salmon a r e caught on t h e i r way to the spawning grounds by t h r e e main methods —  s e i n i n g , g i l l n e t t i n g and t r o l l i n g .  cess I s t u d i e d i n the f i e l d was t h a t o f g i l l n e t t i n g . map i n d i c a t e s the main c o a s t a l f i s h i n g areas  The p r o -  The accompanying  (Department o f F i s h e r i e s  S t a t i s t i c a l Map o f B.C. F i s h e r i e s ) .  Gillnetting  G i l l n e t t i n g b o a t s range i n s i z e from 16 f e e t to 40 f e e t and are  g e n e r a l l y o p e r a t e d by one man.  G i l l n e t s take a l l v a r i e t i e s o f  salmon b u t more p a r t i c u l a r l y s o c k e y e s , cohos and p i n k s .  The p r i n c i p a l  waters f i s h e d w i t h the g i l l nets a r e the e s t u a r i e s o f the l a r g e r  rivers  such as the Nass, Skeena and F r a s e r and those i n l e t s which have salmon streams f l o w i n g i n t o them, e.g. R i v e r s I n l e t , Smith I n l e t , K n i g h t Burke Channel, e t c .  Inlet,  • _ 99 -  M A P 1: C O A S T A L F I S H E R I E S OF BRITISH COLUMBIA  - 100 -  The g i l l n e t snares the salmon by the g i l l s while they are en route to spawning grounds.  The net i s wound around a drum i n the  stern of the boat and the drum i s operated by a power take-off from the main engine and i s controlled by a clutch operated by the f i s h e r man as he stands at the stern. The net i s hung from a cork l i n e at the surface and i s weighted down by a lead l i n e which runs along the bottom of the net. l i k e a curtain i n the water.  It hangs  A buoy i s attached to one end of the  net and a f t e r the net i s set out over the stern, the boat and net d r i f t with the current or tide. The net i s taken i n by being rewound on the drum.  As the  net comes i n over and between sets of r o l l e r s , the fisherman p u l l s out the salmon. Constraints  The Inshore Salmon Fisheries on the West coast i s r i g i d l y policed by the Federal Department of F i s h e r i e s .  Regulations govern  the s i z e and type of nets and the manner i n which they can be used. Certain areas, usually near the mouth of r i v e r s , are closed to commerc i a l f i s h i n g at a l l times. Federal law also c a l l s for a minimum weekly closed period of 72 hours, and conservation frequently demands that the f i s h i n g week be reduced to two days or l e s s .  - 101 -  These regulations are a safeguard to the salmon as a resource, as the contemporary f i s h i n g f l e e t , working with nylon nets, e l e c t r o n i c navigation instruments and capable of ranging a l l over the coast to find f i s h can wipe out a salmon run i n a matter of days or hours.  The  closures thus place time as a prime factor of consideration f o r the fisherman.  I f he has only one or two day's grace a week to exploit  the salmon resource then the way i n which he uses his time during the openings i s very  important.  Fieldwork The f i s h i n g v i l l a g e of A l e r t Bay on Cormorant Island provided the base for the f i e l d test.  Located o f f the north-east t i p of Vancouver  Island i n Johnston S t r a i t s , A l e r t Bay straddles the main waterway between Vancouver and Prince Rupert. Two administrative structures — council —  served the population of 1500,  a v i l l a g e council and a band two-thirds of whom were Indians.  The v i l l a g e was incorporated as part of the Mt. Waddington D i s t r i c t i n 1966 and had domain over 800 people, the majority of whom were white. The Nimpkish band council administers two Indian Reserves with a j o i n t population of 700. A l e r t Bay has two further s i g n i f i c a n t features i n that i t serves as the c u l t u r a l centre for the Kwakiutl nation, and as a service centre for Northern Vancouver Island.  - 102 -  With r e s p e c t to i t s s i g n i f i c a n c e as a c u l t u r a l focus f o r the K w a k i u t l , A l e r t Bay has w i t n e s s e d a l o n g p r o c e s s i o n o f a n t h r o p o l o g i s t s from Boas through Codere to the h o s t o f museum o f f i c i a l s day exponents who t h r o n g r t h e summer p o t l a t c h e s .  and l a t t e r -  As a s e r v i c e c e n t r e ,  the v i l l a g e b o a s t s a 70-bed h o s p i t a l , b o a t - b u i l d i n g wharves, 15 r e t a i l s t o r e s , a l i q u o r s t o r e and e i g h t R.C.M.P. o f f i c e r s . In terms o f b o t h s e r v i c e and o c c u p a t i o n A l e r t Bay's dependance upon the f i s h i n g i n d u s t r y i s e x t a n t . opens i n June and c o n t i n u e s to November.  The i n s h o r e salmon f i s h i n g D u r i n g the i n t e r v e n i n g months,  the p o p u l a t i o n o f A l e r t Bay i s s w o l l e n by as many as 1500 fishermen, and  the r e t a i l and s e r v i c e i n d u s t r i e s a r e l a r g e l y dependant on t h i s  transient  custom. In the w i n t e r most A l e r t Bay f i s h e r m e n take j o b s as l o g g e r s .  There i s no shortage o f w i n t e r employment p o s s i b i l i t i e s , camps a r e always  as the l o g g i n g  short-handed.  Three f i s h i n g companies operate out of A l e r t Bay — P a c k e r s , Canadian  F i s h Co., and Nelson B r o s . L t d .  Packers and Nelson B r o s . —  B.C.  Two of these —  B.C.  a r e owned by the same p a r e n t company, b u t  they o p e r a t e i n d e p e n d a n t l y o f one another.  The companies' o p e r a t i n g  concern was w i t h f l e e t s o f s e i n e b o a t s ; t h e i r main r e l e v a n c e to a g i l l n e t t e r was i n terms o f t h e p r i c e p a i d f o r f i s h and the s e r v i c e s g i v e n to r e t a i n a fisherman's  affiliation.  Minimum p r i c e s f o r landed  catch a r e n e g o t i a t e d between the U.F.A.W.U. and company owners, l a n d i n g p r i c e s vary from p l a c e to p l a c e .  though  \  - 103 -  Each fisherman i s a f f i l i a t e d w i t h one o f the companies, which means t h a t i n r e t u r n f o r c e r t a i n loans, etc. — ticular  services —  delivering  nets,  floating  the f i s h e r m a n i s o b l i g a t e d to s e l l h i s c a t c h to a p a r -  company.  Data-Collection  Prior  to the salmon opening  i n June I spent s e v e r a l months  making c o n t a c t s w i t h both I n d i a n and white w i t h the r e s u l t  t h a t I was  when I began f i e l d w o r k . Fishermen  The Department of F i s h e r i e s ,  c o n s u l t e d b o t h b e f o r e and d u r i n g my  the U n i t e d  the N a t i v e Brotherhood  the f i s h i n g  were  p e r i o d of f i e l d w o r k .  I l i v e d a t the wharf on a boat  (with the  l i k e l y name of " L i n t u " ) which meant t h a t even when I was sea w i t h a p a r t i c u l a r  Bay  known p e r s o n a l l y by about s i x fishermen  and A l l i e d Workers Union and  In A l e r t Bay,  f i s h e r m e n from A l e r t  f i s h e r m a n I was  un-  n o t out a t  i n continuous i n t e r a c t i o n  with  community.  In c o l l e c t i n g i n f o r m a t i o n on g i l l n e t t i n g a c t i v i t i e s I over the whole coast from R i v e r s I n l e t , where I spent weeks a t Wadham's Landing, f i n a l week of f a l l  two and a h a l f  to the F r a s e r R i v e r where I observed  f i s h i n g i n November.  f o l l o w e d the f i s h runs, I had  the A l e r t Bay  to be w i t h them.  c i p a n t o b s e r v a t i o n on each boat gave me s i v e i n t e r a c t i o n w i t h each p a r t i c u l a r by r e g u l a r c o n t a c t throughout  As  ranged  the  fishermen  The p r o c e s s o f p a r t i -  a minimum o f f o u r days e x c l u -  f i s h e r m a n , and t h i s was  the r e s t o f the f i s h i n g  season.  augmented  - 104 -  In any average week I would spend three days ashore and four days at sea.  The time spent ashore, p a r t i c u l a r l y i n the fishcamp  at Wadham's Landing, was used to f i l l up any information gaps I had on any fisherman and to cross check the r e l i a b i l i t y of the data I already had. In A l e r t Bay my sample of independent g i l l n e t t e r s was composed of 12 Indian fishermen and 14 Anglo-Canadian  fishermen.  The composition  of the sample was such that I should be able to observe whether or not c u l t u r a l differences accounted for any difference i n decision-making. Of these 26 boats I went out on 10 and d i r e c t l y observed the decision-making of the i n d i v i d u a l captains. Of the remaining boats I used data gathered from observing t h e i r boats while on the f i s h i n g grounds and supplemented  this by talking to the captains on shore, to  ascertain whether the inferences I had made about t h e i r decision-making behavior were i n fact accurate. This method worked extremely well i n some cases but not i n others.  For instance, a group of boats —  May, Flora M, Miss Lorna, and Sea View — the same way.  Annie L, Beatrice S, Eddie always fished together and i n  Thus, one had to go out on one boat to observe a l l s i x  with respect to f i s h i n g strategies employed and perform follow-up i n terviews on the remaining f i v e to obtain the necessary background material.  I worked on the assumption that another g i l l n e t t e r faced with  the same alternatives as the captain I was with could be c l a s s i f i e d as the same type of r i s k - t a k e r , i f his strategies were the same as on "my" boat.  - 105 -  But down.  t h e r e were two boats  f o r which t h i s assumption  broke  I c o u l d not be s u r e t h a t the a l t e r n a t i v e s were the same as f o r  my  " s k i p p e r " , and as the s h o r t n e s s o f the f i s h i n g season  me  from g o i n g out w i t h these p a r t i c u l a r  from my  prevented  f i s h e r m e n , I dropped  sample on the grounds t h a t I had inadequate  them  data f o r them.  Risk  I am  d e a l i n g w i t h two  them the same a n a l y t i c a l l y . ment r i s k s costs),  taken on new  c a t e g o r i e s o f r i s k , though I  F i r s t o f a l l I am  technology  treat  concerned w i t h  invest-  (I do not i n c l u d e g e n e r a l maintenance  and s e c o n d l y w i t h s t r a t e g y r i s k s w i t h r e s p e c t to  exploiting  f i s h as a r e s o u r c e . The if  reason f o r t h i s d i s t i n c t i o n i s t h a t I want to d i s c o v e r  there i s consistency i n r i s k - t a k i n g ,  or whether c a p i t a l r i s k s  taken on new  p l o i t a t i o n a l r i s k s being  taken.  F o r i n s t a n c e , an investment tainty  a c r o s s the board, as i t were;  technology r e s u l t i n lower  i n r a d a r would reduce  ex-  the uncer-  a s p e c t to f i s h i n g i n f o g ; i n depth sounders would g i v e more  i n f o r m a t i o n about p r o x i m i t y to r e e f s and shore,  whether o r not  this  r e d u c t i o n o f u n c e r t a i n t y l e a d s to the t a k i n g o f h i g h r i s k s i s a  fac-  t o r I wish  to  establish.  R i s k s on both culation factor  technology and s t r a t e g y c h o i c e i n v o l v e a c a l -  on the p a r t o f the f i s h e r m a n o f p a y o f f s , c o s t s and  as i n d i c a t e d below (Table XV):  the chance  - 106 -  T a b l e XV:  R i s k s on Technology  and S t r a t e g y  Technology  Strategy  Payoff  Calculated increase i n f i s h intake A b i l i t y to compete b e t t e r f o r f i s h resource  Calculated intakes from d i f f e r e n t l o c a tions  Costs  C a p i t a l Cost Calculated loss i f payoff does not accrue  P o s s i b l e l o s s of gear P h y s i c a l danger Time Cost o f t r a v e l l i n g  Chance  Required c o n f i d e n c e l e v e l f o r f i s h e r m a n to i n v e s t capital  C a l c u l a t i o n of the chance o f a l t e r n a t i v e l o c a t i o n s g i v i n g expected y i e l d s .  C a l c u l a t i o n of chance of i n c u r r i n g l o s s e s due to h a z a r d s , e t c .  High r i s k i s d e f i n e d as the d e c i s i o n to maximise g a i n or to take a l o n g shot — mize c o s t s .  low r i s k s t r a t e g i e s are those t h a t attempt  The d i s t i n c t i o n between the two  c a t e g o r i e s of r i s k i s  simply a h e u r i s t i c d e v i c e to (a) s o r t out the d a t a and test  to m i n i -  (b) p e r m i t me  to  the i n f o r m a t i o n c i r c u i t s model (Chapter 2) as t h a t model i s spe-  cific  to r i s k s on t e c h n o l o g y .  Wealth  F o r a wjalth c l a s s i f i c a t i o n I used a combination o f Department o f W e l f a r e standards and fishermen's e s t i m a t e s of r e l a t i v e w e a l t h to d e f i n e the low w e a l t h  category.  - 107 -  Department of Welfare standards involved a s l i d i n g scale whereby a minimum income v i s - a - v i s the number of dependants was  cal-  culated to establish whether a person q u a l i f i e d for welfare payments. Using the same p r i n c i p l e I constructed another s l i d i n g scale but gave s l i g h t l y higher figures than the minimum necessary to stay off welfare as a result of the " f o l k " view of what constituted low wealth (Appendix V).  Recall that i n my discussion of farmers taking r i s k s ,  and of the experimental was  subjects taking r i s k s , the low wealth i n d i v i d u a l  i n a p o s i t i o n where he could not afford to lose.  In order to meet  the necessary scope conditions low wealth fishermen must be s i m i l a r l y categorised —  i . e . they are i n a p o s i t i o n where they cannot afford to  lose. Medium wealth ranged from the d i v i d i n g l i n e discussed above to +$3,000, high wealth was  categorised as +$6,000.  Utility  I i n f e r r e d a fisherman's u t i l i t y for increasing f i s h intake from two sources of data: intent and  (a) from verbal statements about f i s h i n g  (b) a subjective evaluation (on the part of the fishermen)  of increasing f i s h intake. The l a t t e r involved a scale score 1-10,  and followed ques-  tions from me as to the importance of increasing f i s h intake.  I asked  the fishermen to rate themselves on the scale and categorised HU having a score between 7-10,  MU 4-7  and LU 1-4.  as  This method e l i c i t e d  a number of verbal statements that often l e f t no doubt as to the p a r t i cular fisherman's preference l e v e l for the outcome "increase i n f i s h intake"  - 108 -  Information  (a)  With respect to f i s h i n g strategies decided upon, there were  a number of information sources that could be used. (1)  Department of Fisheries projections.  (2)  Other fishermen.  (3)  Radio.  (4)  Performance of other boats.  (5)  Depth sounder.  (6)  Fish jumping. The f i r s t few categories give information of a general nature,  the l a t t e r two of a more s p e c i f i c kind. The Department of Fisheries projections give estimates of timing, place and s i z e of d i f f e r e n t salmon runs. as a general pool of information exchange — ised by d i s t o r t i o n .  Other fishermen act  an exchange often character-  Much of the time during closure i s spent  informa-  tion-broking. In the f i s h camps and taverns a gaming-like s i t u a t i o n exists i n that fishermen are seeking as much information as they can while attempting not to divulge any themselves.  Different channels of i n f o r -  mation are regarded as r e l i a b l e , but most queries from one  fisherman  to another are met by evasion and d i s t o r t i o n . It was  as though there were d i f f e r e n t networks of information  within ithe community of fishermen, and entry into one or more networks seemed to be a function of a variety of factors.  Kinship played a role  - 109 -  amongst both Indian and white fishermen i n that information was pensed freely and accurately between k i n . l i m i t to kinship as a useful c r i t e r i o n —  dis-  But there seemed to be a one's peers i n an  informa-  t i o n network had to.be more or less equal i n terms of performance with respect to f i s h , i n t a k e . Implicit contractual ties also defined p a r t i c u l a r networks. This sometimes operated as a "buddy" system whereby two or more f i s h e r men  a c t i v e l y helped and directed one another on the f i s h i n g grounds,  but more generally i t consisted of a r e c i p r o c a l pooling of r e l i a b l e information.  Here again the l i m i t s of the network seemed to be  fined by performance.  Highliners did not dispense accurate  de-  informa-  tion to non-highliners and vice versa.  (The term " h i g h l i n e r " refers  to a top fisherman i n terms of recorded  catch.)  One network that I saw operating at Rivers Inlet consisted of the skippers of the V i r g i n i a M, Miss C h r i s t i e , Aunna, Ruby S, Slave ( a l l A l e r t Bay boats) plus one g i l l n e t t e r from Sointula. The network was  defined by a l l s i x skippers at one time or  another during the closure period eating a meal on each one of the other boats.  They did not take food on anyone else's boat, and nobody  outside the s i x ate on t h e i r boats.  This was  a " h i g h l i n e r " network,  i n that a l l s i x were considered to be top fishermen.  So whether the  network was based on k i n or contract, the l i m i t s to i t seemed to be defined by performance.  - 110 The radio was  also an important  source of information, though  d i f f e r e n t fishermen had d i f f e r e n t evaluations of i t .  Most  operated with the radio on, but the information dispensed was  rarely regarded  fishermen over the radio  as r e l i a b l e .  I f a fisherman was doing well he would hardly broadcast i t , as he would have 100 boats breathing down his neck i n less than an hour.  I f a "buddy" system was  operative statements over the radio  would be coded -- so that only the "buddy" would know what was meant. Some makes of depth sounder (Ekolite, Marconi) were equipped with a f i s h - f i n d e r function which would give readings on a trace-out from which a fisherman  could infer the presence or absence of f i s h .  Perhaps the most highly valued piece of information the sighting of f i s h jumping.  was  As salmon generally run i n schools,  one  or two jumpers usually means there are quite a l o t more there. Fishermen watch one another very closely -- to see how  they  are doing on a comparative basis -- and one can i n f e r from observing other boat's catches more about the general nature of f i s h running i n the area than simply from one's own  haul.  Sources (1) to ( 4 ) , even i f each one i s tapped, only gives the fisherman information of a general nature.  The sounder and jumpers  give more s p e c i f i c information about the presence of f i s h .  So for stra-  tegies the former defines low information, the l a t t e r high information. (N.B.  High information for strategies i s thus analogous to  the medium information l e v e l i n the experiment.  R e c a l l that medium i n -  - Ill -  formation i n the experiment specified the range of odds, payoffs and costs -- high information i n the f i e l d bilities. for  There was no e f f e c t i v e way  the fisherman.  s i m i l a r l y gave a range of possithat information could be increased  The most that he could get was  jumping and sounder readings.  a function of f i s h  Thus for the f i e l d test there was no cor-  responding high information l e v e l with respect to strategies decided upon.) Another important could draw upon was  his own  source of information that the experience and s k i l l .  fisherman  This s t i l l , however,  gave information of a general nature with respect to tides, currents and salmon runs. (B)  The information c i r c u i t s model d i f f e r s from my preceding re-  marks i n that with this model I was c i r c u i t s with respect to new cation (Chapter 2).  stating that access to information  technologies varied with wealth and edu-  The implication was  that individuals with greater  access to information c i r c u i t s would be more innovative, i . e . take high r i s k strategies. I use the same d e f i n i t i o n for wealth as previously.  Educa-  tion, with respect to this model, i s defined i n terms of formal education.  Schultz and others have discussed the general benefits of  education with regard to productivity (Schultz 1963; F i r t h 1964b) -- I am attempting by inshore  to show i t s precise relevance for investment  r i s k s taken  fishermen. Low  education i s defined as below Grade 6, medium education  consists of standard secondary education, high education i s categorised as a function of post secondary training -- university, etc.  - 112 -  Information C i r c u i t s  The following are the main information sources that can be tapped: (1)  Fishing journals.  (2)  Department of F i s h e r i e s .  (3)  Manufacturing companies.  (4)  F i s h i n g companies.  (5)  Union.  (6)  Other fishermen. I constructed a scale for access to information c i r c u i t s  by assigning 1 unit for each information source tapped by p a r t i c u l a r fishermen. There were three f i s h i n g journals current on the West Coast (one unit for each); the Department of Fisheries ran courses every year which gave out information about new exploitative and technical devices and also issued projections for the f i s h runs (two units for this).  One unit was assigned for any i n i t i a t i v e taken by a fisherman  i n approaching manufacturing companies with regard to technology; also the f i s h i n g companies operated as information brokers.  Consideration  of both the union and other fishermen as additional sources of i n f o r mation gives me a scale of 0 — ^ 1 0 .  - 113 -  Table XVI:  Information  Circuits  Fishing Journals subscribed to  3  Department of Fisheries Courses Salmon run projectsion  2  Manufacturing companies (a) electronic (b)  net  2  Fishing Companies  1  Union  1  Other Fishermen  1  I define access to information c i r c u i t s as high when a score i s greater than 7, medium with a score between 4 and 7 and low as having a score less than 4. Every fisherman used (6) —  other fishermen —  as one source  of information, but this was generally regarded as u n r e l i a b l e . Dist o r t i o n of information and evasion to questioning was the general rule. This was not surprising i n terms of the very f i e r c e competition f o r the salmon as a resource. Each source could generally give information that others could not.  For instance the department of f i s h e r i e s offered catch  projections, the journals offered general information on the manufacturing  technology,  companies offered demonstrations of technology, and  the union gave information as to v a r i a t i o n i n landing price f o r f i s h . Even where duplication of information e x i s t s , the point of the scale i s to measure the number of information sources a p a r t i c u l a r  - 114 -  fisherman uses on the assumption that more than 7 sources gives him higher information than 5 sources, and 5 gives him greater information than 3. The point to be made i s that i f a person seeks out a number of  information c i r c u i t s with regard to technology even though there may  be some duplication, he w i l l be more l i k e l y to take high r i s k s on i n vestment.  I w i l l also be trying to assess whether the information  category a p a r t i c u l a r fisherman i s i n i s the one  he i s "supposed"  to be i n as a function of h i s wealth and education. Case Histories of Risk-Taking by A l e r t Bay G i l l n e t t e r s I intend to present i n d e t a i l an analysis of two types of fishermen with respect to the risks they took on technology and s t r a tegies.  The exercise i s to make clear the c r i t e r i a that I use to ar-  rive at categories of risk-taking.  For the remaining 24 fishermen I  present the end result of the analysis i n tabular form —  to write 26  case h i s t o r i e s i n f u l l would make this section longer than the rest of the thesis.  Strategies as Risks In order to i n f e r whether a p a r t i c u l a r strategy i s high or low r i s k one has f i r s t of a l l to be aware of the c h a r a c t e r i s t i c s of salmon runs and secondly of the alternative means of getting at them.  - 115 -  There were three major situations where I was on hand to observe which strategies d i f f e r e n t fishermen elected to choose —  (a)  Rivers Inlet f i s h i n g ; (b) Johnston S t r a i t s f i s h i n g ; and (c) the closing of the Northern F i s h e r i e s . This was supplemented by other instances, but these three situations are the major ones that were used to categorise an i n d i v i d u al's risk-taking behavior. Rivers Inlet  Rivers Inlet stretches 40 miles inland from the Queen Charlotte Straits.  The main channel has a number of breaks provided i n the  form of reefs, shoals and islands, and the mouth of the i n l e t has quite a number of offshore islands and reefs. Runs of sockeye coming i n from the P a c i f i c , turn the t i p of Vancouver Island, and a large proportion head d i r e c t l y across Queen Charl o t t e S t r a i t s into Rivers Inlet, using the offshore islands as a guide. Inside the i n l e t the schools tend to follow the shore, reefs and islands as pointers to fresh water.  Over shoals and reefs the salmon r i s e c l o -  ser to the surface and are less l i k e l y to swim under the net the closer a set i s made to the reef. The schools tend to run more predictably i n bad weather conditions and with the tide.  Rain has the result of making the salmon  bunch together and head faster f o r fresh water — closer to the surface.  i t also brings them  - 116 -  From a knowledge of salmon runs, i t i s perhaps an understatement that a l l the i d e a l spots are hazardous.  To f i s h the sides of  the channel and the reefs meant the p o s s i b i l i t y of fouling on the rocks. Fishing outside meant an 8-10  foot groundswell and the choice spots  involved locations close to the reefs and islands. In addition to these locations, another i d e a l spot to f i s h was  the boundary; i n this case the hazard to be feared was  ermen.  Boundary f i s h i n g was  almost a law unto i t s e l f .  other f i s h -  At the head  of Rivers Inlet a demarcated boundary indicated the extent to which commercial f i s h i n g was  permitted.  This was where the f i s h were heading to the main salmon stream.  Because of this fact 100-150 boats would be packed t i g h t l y  together along the boundary, which meant there were very r e a l dangers of fouling nets and ruining gear.  Also a r i p current would frequently  drag a boat and net across the boundary whereupon a f i s h e r i e s p a t r o l would be ready with a fine and confiscation of catch. With the number of boats, and narrowness of the boundary, the s p e c i a l feature about boundary f i s h i n g was  the p o s s i b i l i t y of a very  large catch, and also the very r e a l p o s s i b i l i t y of losing equipment (fouling on other boats) and catch (crossing the boundary with a net out constituted an offence against federal f i s h i n g regulations and resulted i n a fine as w e l l as confiscation of any f i s h that had been caught). Rivers Inlet had choice spots and conditions where f i s h could be caught, but they were a l l hazardous.  The i n t e r e s t i n g thing, analy-  - 117 -  t i c a l l y , i s what a p a r t i c u l a r fisherman does when he gets p o s i t i v e indicators ( v i z . f i s h jumping) off a hazard.  Does he make a set (high  r i s k or go somewhere safer (low risk)? Fishing inside i n the main channels away from reefs and shoals involves low r i s k inasmuch as the danger of losing equipment (either from fouling or hazards) i s much less though the catch w i l l be smaller.  During the f i s h runs there i s always the prospect of getting  a f a i r catch, but the large catches involve playing the hazards.  Table XVII;  Strategy Risks i n Rivers Inlet  Fishing Locations  High r i s k  Boundary  x  Reefs and Shoals  x  Outside - o f f reefs  x  Low  risk  Centre Channel  x  Protected shores  x  Johnston S t r a i t s The v i l l a g e of A l e r t Bay was boat wishing to f i s h Johnston S t r a i t s .  a central focus point for any The same remarks about salmon  habits hold, though there are two a d d i t i o n a l hazards i n the S t r a i t s . One i s the preponderance of large kelp beds, the other i s the use of the S t r a i t s as a major waterway. The former i s s i g n i f i c a n t i n that salmon seem to "play" around kelp beds at the estuary of certain streams and they o f f e r a  - 118 hazard inasmuch as that, though f i s h are to be found close to kelp, the process of extraction usually brings up a whole l o t of kelp which may take 2-4 hours to pick out. this was  With a f i s h i n g week very often of 1-2  days,  time that could cost a fisherman one or two sets. The considerable amount of t r a f f i c through the S t r a i t s , es-  p e c i a l l y i n the daytime, constituted a hazard to nets. of  The narrowness  the channel and the s i z e and disregard f o r nets of the large tugs  and l i n e r s made the prospect of losing gear a very r e a l one. The S t r a i t s do have a number of r e l a t i v e l y safe spots clear from the t r a f f i c lanes —  Growler's Cove, Robson's Point —  which can  be considered low r i s k areas i n that losses due to fouling can be regarded as n e g l i g i b l e , though only moderate catches can be taken. Shore f i s h i n g i n Johnston S t r a i t s i s generally regarded as a prerogative of seine boats.  Though some g i l l netters may dispute  t h i s , there i s very l i t t l e that they can do when a 200 ton seiner nudges them away from a beach location.  Table XVIII:  Strategy Risks i n Johnston S t r a i t s  Fishing Locations  HR  Shoals and reefs  x  Kelp beds  x  T r a f f i c Lanes  x  Safe Areas Shore f i s h i n g  LR  x x  - 119 -  Closure of Northern Fisheries In August a s i t u a t i o n arose that affected a l l A l e r t Bay netters.  gill-  Due to an inadequate amount of spawning i n the salmon streams  the entire Northern Fisheries was  closed as a conservation measure.  In the f i r s t week of the closure only four areas were l e f t open: (1)  The Rivers Inlet area was open f o r 3 days (Areas 9, 10 and 11).  (2)  Johnston S t r a i t s was open f o r 2 days (Areas 12 and 13).  (3)  Central Inshore Fisheries (Areas 14, 15, 16 arid 17) was open for two days.  (4)  The Fraser River area was open for 2 days. Of these openings, Rivers Inlet was  terms of expected volume of salmon.  the poorest prospect i n  The summer sockeye run was over,  and no more would be expected t i l l the f a l l . The S t r a i t s showed promise of a moderate run, as steady returns had been reported for the previous week.  The central part of  the inshore f i s h e r i e s were p r a c t i c a l l y devoid of salmon. A run of sockeye was expected at the Fraser, though the previous week's catches had been very low on the average, which was not a good indicator, of an impending run. To f i s h Rivers Inlet or central areas with Johnston S t r a i t s and the Fraser open would be a useless exercise.  The decision facing  the A l e r t Bay g i l l n e t t e r s was either to f i s h Johnston S t r a i t s and remain close to A l e r t Bay or go south to the Fraser and f i s h 2 days.  -  120 -  The calculated advantage i n going south would be that there was a low to moderate chance of a good catch.  But against this had to  be offset the cost of t r a v e l l i n g there ( 2 days journey from A l e r t Bay), the hazards from t r a f f i c (which were worse than for Johnston  Straits),  and the warning from the Department of Fisheries that they were not sure whether the run would materialise. Fishing Johnston S t r a i t s offered an advantage i n that though no high catches would be expected, there was a f a i r chance of moderate hauls (the week p r i o r had shown a steady run).  The cost factor was  reduced by lack of t r a v e l l i n g costs. From this one could i n f e r that a high r i s k would be to go south; a low r i s k to stay put i n Johnston  Straits.  Table XIX: Northern Closure and Strategy Risks Johnston S t r a i t s  Fraser  Payoffs:  Moderate  Payoffs:  High  Chance:  Moderate  Chance:  Low  Costs:  Hazards and T r a f f i c  Costs:  Hazards and T r a f f i c and Travel  LOW RISK  HIGH RISK  Investments on Technology as Risk The options on technology open to the g i l l n e t t e r are very wide and cover a variety of functions.  These options can be broken  down into general categories of technology i n terms of either (a) l o cational technology or (b) exploitative  technology.  - 12* >  (a)  Depth sounders with f i s h - f i n d e r functions are a recent  entry to the market and as w e l l as being a safety device, readings on the trace out can give information about the presence of f i s h .  However  these are expensive additions to a fisherman's repertoire as the cheapest ones cost $1300 (Marconi). Ship-to-ship and ship-to-shore radio telephones  can give the  fisherman an extensive range of information from which he can draw i n order to make decisions.  Someone f i s h i n g the Fraser can f i n d out what  i s going on at the Nass River, over 400 miles away.  Again, this tech-  nology i s not cheap, as each type of radio telephone  costs upwards of  $150,  and some are much more  expensive.  Compass and charts are standard navigational devices; though quite a number of boats are beginning  to carry radar.  As w e l l as a  navigational aid, radar functions to increase f i s h i n g time during fog conditions.  Most boats without  radar stayed i n port during fog, while  those with i t had the f i s h i n g grounds v i r t u a l l y to themselves. (b)  The main exploitative device used on the g i l l n e t t e r s i s  the power drum, which can be either h y d r a u l i c l y or gear-operated. hydraulic drum i s much smoother, and a fisherman sets much faster.  The  can make and pick up  The conversion from a gear-driven drum to a hydrau-  l i c system costs i n the region of $300. Most nets are made out of nylon with either a monofilament web (single strand) or a polyfilament net (several strands).  The l a t -  ter are f a i r l y standard throughout the industry whereas the monofilament  - 122 -  nets are quite new introductions.  They cost h a l f as much again as the  standard nylon nets, but their advantage i s i n terms of longer l i f e and a better " l i e " i n the water.  To exploit the inshore f i s h e r i e s  e f f e c t i v e l y a fisherman requires at least s i x d i f f e r e n t nets —  varia-  tions i n colour and web-size are the main considerations. Most marine engines used i n the g i l l n e t t i n g f l e e t were capable of a steady eight knots, but recent years have seen the advent-: of boats with high speed engines and planing h u l l s .  A marine firm (Albion Co.  Ltd.) at Haney, B.C. started putting these on the market i n 1968. advantage of a high-speed  The  engine was that i t increased the area of  exploitation possible during an opening as t r a v e l l i n g time was cut i n half. The slower boats had more or less to s t i c k to one major f i s h ing area as they would lose too much time t r a v e l l i n g during an opening. This was not a handicap to the high-speed  boats, though their operating  costs were very much higher.  Operating  Costs  The average operating costs for a g i l l n e t t e r worked out at around $2,000 per annum.  This included general maintenance and repairs  as well as replacements for f i s h i n g gear. costs on new technology).  (This figure does not include  The high speed g i l l n e t t e r s , because of t h e i r  very much greater f u e l consumption, had operating costs of around $2,700 per annum.  - 123 -  One r e s u l t of the technological explosion i n recent years has been to decrease the hazardous nature of the business (improvement in safety features as well as better e x p l o i t a t i o n a l devices), but there is s t i l l a large measure of uncertainty attached to the process — as the prospect of losing l i f e and c a p i t a l i n pursuit of the salmon i s an everyday fact of l i f e to the fisherman. Perhaps the major impact of technology has been to change the patterns of mobility and exploitation.  In the past g i l l n e t t e r s  tended to f i s h i n areas adjacent to their home base and e f f e c t i v e l y operated as trappers — through.  they attempted to catch the f i s h as they passed  But the larger and technologically more p r o f i c i e n t boats  capable of withstanding rough seas permit the fishermen to range the whole of the inshore f i s h e r i e s hunting the salmon, i . e . going i n purs u i t of the salmon, instead of l y i n g i n wait f o r i t . This increased mobility has been coupled with improved  com-  munication systems and navigation aids, making i t possible f o r the fishermen to keep i n close contact with f i s h i n g conditions i n d i f f e r ent areas and to take advantage of unexpected salmon runs.  The dis-  t i n c t i o n between hunting and trapping goes a long way to explain d i f ferences i n technological investment. If a fisherman i s concerned mainly with trapping and stays close to h i s home base then there i s often no need f o r the l a t e s t technology.  The fishermen would have such an intimate knowledge of the  area and conditions, to an extent that would render things l i k e radar and depth sounders worthless to h i s purposes.  - 124 -  A hunter, on the other hand, depends far more on  technology  as his range of exploitation requires the use of devices other than intimate knowledge of a p a r t i c u l a r area. There i s a very close correlation between range of exploitation and investment i n technology, and i t appears as though two t i n c t investment  cycles are operative.  dis-  Hunting reinforces c a p i t a l i n -  vestment, while trapping discourages i t . The following table indicates this r e l a t i o n s h i p . Table XX:  Investment Risks and Range of E x p l o i t a t i o n No. of Fishermen Taking Investment Risks  Fisherman's Range of Exploitation L M H  H  8  0  2  6  L  18  3  15  0  Despite the vast array of technology available, i t offers to the g i l l n e t fisherman only a marginal advantage i n exploiting the f i s h resource.  There are no high payoffs to be expected from technological  investment, and the cost of new  equipment i s generally high.  Thus the investment of c a p i t a l i n new  f i s h i n g technology i s  always a high r i s k inasmuch as the calculated payoff from i t i s marginal, i n terms of increased f i s h intake, and the c a p i t a l cost i s usually steep. However i n an industry as highly competitive as the inshore salmon f i s h e r i e s , the marginal advantage i s a s u f f i c i e n t incentive for some fishermen to take high c a p i t a l r i s k s .  - 125 -  Case No. 1:  Ernie M c A l l i s t e r  I travelled up to Rivers Inlet from A l e r t Bay with Ernie M c A l l i s t e r on his boat the V i r g i n i a M.  During the journey I had ample  opportunity to find out the types of investment risks he was prepared to take. He had b u i l t the boat himself a year ago; the design and technology  incorporated indicated a very astute appraisal of conditions  i n the f i s h i n g industry. At 40 feet long and with a 12-foot beam the V i r g i n i a M was much bigger than the ordinary g i l l n e t t e r . the boat being used as a multipurpose  This was i n a n t i c i p a t i o n of  vessel.  The large beam meant  that the owner could f i s h outside i n unprotected waters i n large swells.  Also the boat was b i g enough to go t r o l l i n g off the West coast  of Vancouver Island (once t r o l l i n g poles were f i t t e d ) ; and could also be rigged to long l i n e for h a l i b u t .  The boat had been b u i l t with  tougher than normal timbers i n order that i t could also operate as a salvage and beachcombing vessel. Ernie had anticipated a possible change i n f i s h i n g operation. At present the companies had f l e e t s of collectors and packers that c o l lected f i s h from the boats on the f i s h i n g grounds. Ernie saw the day coming when the companies would cut costs by demanding that fishermen i c e and f i l l e t their own f i s h and deliver to a port, thereby excluding the cost of maintaining a packing  fleet.  To this end, a large hold was incorporated i n the design of the boat  - 126 -  to f a c i l i t a t e what was f i l l e t his own  an eventuality (to Ernie) of having to i c e and  catch.  Technologically, Ernie had invested i n a very large power drum —  much bigger than was  necessary for a g i l l net.  This was  so he  could, i f he wanted, put a small seine net on his boat. Thus the V i r g i n i a M was  designed to exploit the West coast  f i s h resource i n terms of g i l l n e t t i n g , t r o l l i n g , seining and  long-  l i n i n g , as well as having supplementary use as a salvage vessel. He had also invested i n a Marconi depth sounder which had a f i s h - f i n d e r function, as well as two radio telephones (ship-ship; shipshore), and an automatic p i l o t . monofilament web  He was  on his outside side.  trying f o r the f i r s t time a He calculated that this t o t a l  investment of $26,000 would give him a marginal advantage over most other fishermen i n exploiting the West coast f i s h resource; terms of my  d e f i n i t i o n of risk-taking he was  so i n  taking high investment  risks.  Range  The s i z e and strength of Ernie's boat, gave him a range of mobility denied to smaller vessels.  He fished from the Fraser to the  Queen Charlotte Islands following the f i s h runs.  He had seen his  exploitative patterns change over the years since he f i r s t fished  own —  he had never ventured very f a r from his home base then, and even with  - 127 -  better technology through the years the tendency was s t i l l to "trap" the f i s h as they passed near to his home base. In the l a s t ten years the technology and information explosion had l e d to him hunting the f i s h rather than trapping them. Information dispersed by the f i s h e r i e s department on the s i z e and place of runs were f a i r l y r e l i a b l e ; also the technology available permitted a g i l l n e t t e r to range the entire length of the inshore f i s h e r i e s . Born into a f i s h i n g family, Ernie at 55 was an acknowledged highliner i n A l e r t Bay. He had fished since he was a boy and remembered the days when he operated from a rowing boat with a net which had to be hauled i n by hand.  Thus his experience and knowledge of the West  coast f i s h e r i e s was extensive, having fished on a l l types of boats. He put himself through university at U.B.C. but on graduation the most that a teacher could earn per annum was $800, so he stuck to fishing.  (However, he was a great believer i n education, and was very  proud of the fact that he had put his daughter through the University of B r i t i s h  Columbia). He had very much of a "protestant e t h i c " attitude towards  work —  "You have to work hard at i t to be a good fisherman" -- and he  had a high incentive to increase his take as he wanted to pay o f f the loan on the boat as soon as he could.  When I asked him to rate himself  on a 1-10 u t i l i t y scale, he laughed and said that I should put him at around 100 (HU).  - 128 -  His taxable income worked out at around $12,000 per annum. The  only dependent he had was  t a i l store i n A l e r t Bay.  his wife who  worked part-time  His wealth category was  at a re-  high.  He seemed very well informed on issues a f f e c t i n g the f i s h i n g industry from both a union and management aspect. secretary of the A l e r t Bay l o c a l of the U.F.A.W.U.  At one time he  was  Thus i t i s not  sur-  p r i s i n g that he tapped just about every possible information source. He subscribed  to two f i s h i n g journals (2), had taken a f i s h e r i e s course  at U.B.C. and used the department's salmon run projections (2).  He made  a point of seeing his net company and electronics company agent at least once a year on what was  new  (2), and also drew upon his f i s h i n g company,  union and other fishermen for information (3). This gave him a score of 9 which indicated that he had access to information Strategy  high  circuits.  Risks  He planned to spend the next two weeks at Rivers I n l e t as the sockeye run was  expected to be at a peak.  took a day and was  The journey from A l e r t  Bay  through Queen Charlotte S t r a i t s which were notoriously  rough -- other areas such as Johnston S t r a i t s or Hardy Bay which did not involve the cost of a long journey or rough waters were rejected as p o s s i b i l i t i e s as the chances were that more f i s h were expected at Rivers I n l e t .  - 129 -  The week before he had planned to f i s h Rivers I n l e t , but  on  a r r i v i n g there and talking around he calculated that the sockeye would peak at Smith's I n l e t this week and Rivers I n l e t the following week. So he pulled out of Rivers I n l e t and went to Smith's, knowing that he would lose at least half a day's f i s h i n g which was factor.  (There was  important as a cost  a two-day opening).  But his c a l c u l a t i o n paid o f f . Rivers I n l e t took 200  That week the top boat i n  sockeye, Ernie at Smith's I n l e t took 250,  so  he was well pleased with his decision. Ernie used a s p e c i f i c set of indicators for f i s h presence. He kept the radio on most of the time but did not evaluate i t too highly as he did not think the fishermen would be t e l l i n g the truth. He never broadcast over the radio himself.  He used i t to pick up  casional t i d b i t s  pf information between c o l l e c t o r s and packers.  He was  very observant of tide and weather conditions.  oc-  He  fig-  ured that f i s h ran more predictably i n rough weather and with a f a i r tide or current and these were the conditions he chose over any  other.  Other boats would often be f u r i o u s l y setting or picking up, but not Ernie.  On the average he would make 5-6  sets a day, but he  would be very careful as to when and where he did so.  He took into ac-  count a number of information factors that added up to a c a l c u l a t i o n of when the f i s h would l i k e l y be running and i n which d i r e c t i o n . As the f i s h ran with the tide, the best time to set was j u s t before the tide changed and  then pick up an hour after slack water --  - 130 -  a 3-hour set. A swell of 8-10 feet and rough weather conditions ( r a i n and 15-20 mile per hour wind) would drive the f i s h d i r e c t l y into the i n l e t when f i s h i n g outside.  Good weather conditions and no tides per-  mitted the f i s h to "play" and not go d i r e c t l y into the I n l e t . He preferred overcast and rough conditions as the f i s h could less e a s i l y see the net, thus his preference for outside f i s h i n g as the inside waters were always more protected. He made his f i r s t set inside the I n l e t , off a reef close to Wadham's Landing.  He could have fished the main channel, but he calcu-  lated that any f i s h coming into the main channel would use the reef as a guide and also come closer to the surface. had to tow the net away from the reef.  On this set he constantly  He opted for this location on  L.Inf -- we had not sighted any jumpers and the readings from the depth sounder were ambiguous. When he pulled i n the set he found he had one f i s h .  Other  boats i n the main channel, nearby had pulled i n an average of twelve. Ernie swore at himself and immediately set off to sea.  He realised  his net was too dark for that part of Rivers I n l e t ; the f i s h would see i t i n the water and swim round i t . At Rivers I n l e t one needed at least three nets to e f f e c t i v e l y f i s h the waters -- a very l i g h t green one for the boundary (known as a head net), as the s i l t brought down by the salmon streams made the water cloudy and milky.  A darker green was necessary for the central parts  of the i n l e t , while a very much darker green was needed for f i s h i n g outside.  (This l a t t e r net had a monofilament  web.)  - 131 -  Ernie had planned to f i s h the inside reefs (which were f a i r l y well down the i n l e t ) , and the outside reefs -- so had kept his outside net on the boat on the c a l c u l a t i o n that he could save time, and also exploit both areas e f f e c t i v e l y .  He had wondered whether his net would  be too dark for the inside reefs, but had decided to take a chance on it. He did have an inside net i n the l o f t at Wadham's Landing F i s h camp, but to put this on would have prevented  him from f i s h i n g  outside.- So he was d e f i n i t e l y out to maximise gains; he appreciated the cost aspect of using his outside net inside but s t i l l went ahead. On his way  to the mouth of the I n l e t , he saw three jumpers,  immediately stopped and made a set -- the cost factor of his net type was  s t i l l operating.  He only caught 18 and as he reckoned he should  have had at least sixty he immediately roared out to sea and fished the reefs off Table and Egg Islands off the mouth of Rivers I n l e t for the remainder of the opening.  Time  By deciding to f i s h outside, i t meant that Ernie had to journey back inside the I n l e t to deliver f i s h , as i t was  too rough for the  c o l l e c t o r s to come outside. This time loss -- costing him at least one set per day  --  could have been used to get his net changed so he could f i s h the inside reefs, but he calculated that weather conditions and runs outside would  - 132 -  give him more f i s h . . Also the inside fishermen were not doing p a r t i c u l a r l y well. Weather conditions were quite rough and there was a large groundswell (8-10 f e e t ) , perfect conditions for outside f i s h i n g (according to E r n i e ) . low information.  He consistently fished the reefs on both high and  No other boats were around him -- but this did not  seem to bother him. He was interested i n the way other boats were doing only i n terms of r a t i o n a l i s i n g whether his own decision to f i s h a particular location was good or not. Off the reefs outside, Ernie averaged 40 f i s h per set which was  above average.  A constant watch had to be kept on the net's d r i f t ,  tide and wind conditions to keep from foundering on the reefs. On one piece of high information Ernie took the plunge again. Over the radio he picked up a c a l l from a t r o l l e r that was operating off the North East coast of Vancouver Island, who was excitedly claiming that he had run into a "fucking great school of sockeye".  Ernie calcu-  lated that the run coming round the t i p of Vancouver Island would head d i r e c t l y to Rivers I n l e t and be further out from the islands he was fishing off. Conditions were very rough, near gale force winds and r a i n with a 12-foot groundswell,  but Ernie pulled further out to sea, made  a set and took 70 and 50 sockeye on two sets, which was exceptional for daylight sets (generally expected to average 25-30). So despite his disastrous start Ernie's f i r s t week at Rivers I n l e t gave him a catch well above average.  Each decision he had made  - 133  both on low ing  and  -  high i n f o r m a t i o n was  to maximize f i s h  h i g h r i s k i n t h a t he was  attempt-  i n t a k e even when the c o s t s c o u l d have been con-  siderable. The  second week a t R i v e r s I n l e t he f i s h e d the boundary,  t i n g on h i s head net. II,  During  the c l o s e d p e r i o d between week I and  he l e a r n e d that.the high boats were those  (though the lowest He  was  M c A l l i s t e r " and  week  t h a t f i s h e d the boundary,  boats a l s o f i s h e d the boundary).  fully  other boats but he  put-  aware of the c o s t f a c t o r of g e t t i n g f o u l e d by  s t a t e d t h a t "where t h e r e ' s  as an a f t e r t h o u g h t  "Anyway my  fish,  you'll find Ernie  boat's b i g g e r  than any-  one e l s e ' s . " I d i d not  see him  f i s h i n g J o h n s t o n S t r a i t s , but  week of the N o r t h e r n c l o s u r e s he headed south c i s i o n was  i n terms of h i g h i n f o r m a t i o n .  to the F r a s e r .  of the c o s t of t r a v e l l i n g and But  had  of F r a s e r junk and  at the F r a s e r .  He  He was traffic  he c a l c u l a t e d t h a t a l l t h i n g s being  chance of making a k i l l i n g  de-  depart-  t h a t p r i o r to the Nor-  been taken.  get a moderate c a t c h i n J o h n s t o n S t r a i t s , while  first  This  He knew the f i s h e r i e s  ment were d o u b t f u l about the s i z e of the run and t h e r n c l o s u r e s v e r y s m a l l catches  i n the  equal  a l s o aware hazards.  he c o u l d  there was  a  only  slight  agreed t h a t i t would  s a f e r and more " r e l i a b l e " to s t a y i n J o h n s t o n S t r a i t s but observed " i f you d i d n ' t got f o r the b i g c a t c h e s , you would never get  them."  be that  - 134 -  Table XXI: Risk Taking by Ernie M c A l l i s t e r Summary:  Ernie M c A l l i s t e r HEd/ H. Inf C i r c . / H. Inv. R./ HW Strategy Risks INF  Strategy  Risk  Peak expected, lose 1/2 day's f i s h i n g by leaving RI.  H  Max Gain  H  1. Peak Expected, a l l f i s h i n g locations chosen hazardous. 2. Using outside net inside, hoped to optimise two d i f f e r e n t locations; cost of net being inappropriate. 3. Fishing inside o f f reef, cost of fouling on rocks. 4. Fishing by jumpers inside; cost factor of inappropriate net. 5. Fishing o f f reefs outside cost i n time of d e l i v e r i n g fish. 6. Fishing further out to sea large school, dangerous conditions.  H  Max Gain  H  L  Max Gain  H  L  Max Gain  H  H  Max Gain  H  Max Gain  H  H  Max Gain  H  Location  Factors  Smith's I n l e t  Rivers I n l e t Week I  of Decision  Rivers I n l e t Week I I  1. Fished boundary - l a s t weeks high boats were here, cost of fouling.  H  Max Gain  H  N. Closures Week I  1. Went to Fraser.  H  Max Gain  H  - 135 -  Case No. 2:  N e i l Anderson  I went out with N e i l on his boat, the Storm Winds, for one day at Rivers Inlet and for two days i n Johnston S t r a i t s . He valued his boat at $4,500 and had bought i t eight years ago.  I t was 34 feet long with an 8 foot beam, with a gear rather than  a hydraulic drum.  In the eight years that he had had the Storm Winds  he had not invested i n any new technology -- had merely maintained equipment and replaced nets. He rejected the idea of investing i n new technology on two grounds -- the prohibitive price and the fact that i t would only marg i n a l l y increase his f i s h intake.  Thus, a hydraulic drum was not con-  sidered as a replacement for h i s gear drum -- though with the former he could make and pick up sets much faster.  I t was cheaper:, to maintain  his gear drum than to invest $400 i n a hydraulic system.  Similarly,  depth sounders, radar, etc., had not been adopted c h i e f l y because of their prohibitive price. This concern with keeping costs down was exhibited i n his choice of nets -- he had a combination inside/outside net for Rivers I n l e t (a d u l l green net) thus cutting down the cost of owning two nets for two f i s h i n g areas, though i t was not as good ( i n terms of optimising catch) as two separate nets.  He exploited a range between Namu and  the Fraser River. The major source of information about technology used by N e i l was other fishermen, though he occasionally dropped i n to see his net  - 136 -  company to shoot the breeze.  He did not use the f i s h e r i e s o f f i c e or  companies as information sources, neither did he subscribe to any journals though he occasionaly read the union newspaper i f one was l e f t lying around i n the taverns.  Thus his score of three puts him i n the  low category with respect to access to information c i r c u i t s . He had been brought up on an Indian Reservation near C h i l l i wack -- his father was white, his mother Salish.  He had turned his  hand to a variety of professions -- logging, clerk, nightwatchman -and had been i n the f i s h i n g industry for f i f t e e n years.  A l l his exper-  ience had been gained on g i l l n e t t e r s . He was hoping for a good season, and rated his desire to i n crease his f i s h intake as high.  Against this had to be weighed his  fear of fouling his nets or boat as he could not afford heavy maintenance costs. He had grade 10 education which puts him i n the MEd category, though his wealth ranking was low. His taxable income worked out at around $6,000, and he was supporting a wife from whom he was separated. He drew unemployment benefit during the winter, and one of his preoccupations was " f i s h i n g for stamps".  In order to qualify for  this benefit, he had to show that he had been employed f u l l time as a commercial fisherman during the f i s h i n g season.  This preoccupation be-  came evident i n the second week of the Northern closures. During this week Rivers I n l e t was l e f t open, everything else was closed (including Johnston S t r a i t s ) except the Fraser.  - 137 -  The run at Rivers I n l e t was over, so one could not expect many f i s h there.  The Fraser River run had not materialised and boats had  done very badly the week before. So N e i l could either stay tied up i n A l e r t Bay, go to Rivers I n l e t where there were few f i s h , or go to Fraser where there were few fish.  I t was cheaper to go to Rivers I n l e t than to the Fraser i n terms  of t r a v e l l i n g costs and there were no t r a f f i c hazards.  To stay tied  up i n A l e r t Bay might have been a good strategy as there were no f i s h anywhere -- but he would have f o r f e i t e d a stamp and run the r i s k of being d i s q u a l i f i e d for winter unemployment benefit.  So i n this s i t u a t i o n  he went to Rivers I n l e t , caught 5 f i s h and collected his stamp. His strategy was aimed e n t i r e l y at minimizing costs.  I t was  cheaper^ and safer to go to Rivers I n l e t as against the Fraser; i t was also a cost-minimizing strategy to go to Rivers I n l e t , though he knew he would not catch much, because staying tied up may endanger his unemployment benefits. At Rivers Inlet he never fished the boundary as he was a f r a i d of getting his net torn and snagged; neither d i d he f i s h the reefs.  He  calculated that he would always be able to catch something by playing i t safe i n the main channels and by f i s h i n g outside under calm conditions. He r e l i e d heavily on what other boats did around him -- when they set, he set, when they picked up he picked up.  He liked other  boats around him, because then he f e l t that not everybody could be wrong about f i s h i n g a p a r t i c u l a r area.  - 138 -  He had the radio on a l l the time and was constantly asking other fishermen how they were doing.  He did not have a depth sounder.  He always chose a low r i s k strategy when information was low. Once when he saw four jumpers off a reef he almost made a set, but a strong current into the reef put him off so he went 300 yards down the i n l e t before setting, and was very nervous about d r i f t i n g  into the reef.  He  only l e f t the set i n for a half-hour before picking i t up. Apart from this one instance, he never took a high pective of information.  risk,irres-  On another occasion, I observed two jumpers  near a shoal -- he ignored them and stayed i n the main channel. In Johnston S t r a i t s N e i l did not f i s h the beach for fear of seiners, nor near kelp beds.  He fished Growler's Cove and Robson's  Point, both of which were safe areas. Bay during the day.  He also travelled back to A l e r t  There were two reasons f o r t h i s :  ston S t r a i t s were always unproductive,  day sets i n John-  averaging 6-10 per set; and also :  there was a tremendous amount of day t r a f f i c .  I t did not seem worth  his while to dodge the t r a f f i c hazards f o r a mere six f i s h per set, though some day sets would take as many as twenty f i s h . The f i r s t week of the Northern Closures, N e i l elected to f i s h the S t r a i t s rather than go to the Fraser on the grounds that his costs ( i n travel and time) and i n terms of hazards, would be less i n the S t r a i t s than at the Fraser.  - 139 -  . Table XXII:  Risk Taking by N e i l Anderson Strategy Risks  Location  Factors of Decision  Rl  1. Fished main channel, knew more f i s h at boundary/ reefs, afraid of costs to fouled gear, etc. 2. Fished safe spots -- main channel. 3. Fished near reef when jumpers sighted. 4. Ignored jumpers when near reef.  J.S.  5. 6. 7. 8. 9.  North Closures  Never fished shore. Never fished kelp. Fished Growler's Cove. Fished Robson's Point. Returned to A l e r t Bay during day.  10. Fished Areas 12 and 13  Inf  Strategy  Risk  H  Min Loss  L  L  Min Loss  L  H  Max gain  H  H  Min Loss  L  H & L H & L L L H  Min Min Min Min Min  Loss Loss Loss Loss Loss  L L L L L  H  Min Loss  L  These two cases represent polar extremes i n terms of risk-taking and could be regarded as "ideal types". fishermen conformed  However, not a l l the other  i n risk-taking to either one of these stereotypes.  Some took high investment r i s k s and low strategy r i s k s , others took high investment r i s k s and a mixture of low and high strategy r i s k s . The results of the analysis of the other 24 boats are presented i n tabular form (Table XXIII) -- the case h i s t o r i e s of Ernie M c A l l i s t e r and N e i l Anderson were merely to indicate the c r i t e r i a by which I categorised individuals and their risk-taking behavior.  4J  Boat *Virginia M *Storm Winds Aunna Franton Gulf M2 *Lumarco Laurelin *Miss C h r i s t i e Miss M e l l May S *Annie L Beatrice S *Broken Islands Eddie May Flora M Miss Lorna North Wind *Harco P o l l y Ann *Ruby Ann Sea View *Sea Kay Seal Spray *Slave Carodic 12 K 14 AC *  crj CO  Skipper E. M c A l l i s t e r N. Anderson G. J o l l i f f e L. Stoffer G. Hardy G. Huson K. J o l l i f e T. Christiansen M. Stoffer S. Huson P. Bruce J . Bruce A. Johnson A. James G. Bruce N. Bruce N. Johnson A. Dick B i l l Scow G. M e l l i s h A. Bruce A. B r i s t o l L. B r i s t o l L. White S. Commano R. Bianco  Boats that I went out on.  Xi  H H H H H H H H H  H H H H H H H H H  c o •r-1 4J  cd O  W  H H L M H M M M L M M M M M M M M L L L L L L L L L L L L L L L L M M L M M M M L L L M L M M M M L M H  Index:  c o  •I-l +J  cn  j-i  •H  D  S-l Mo o •rH  CO  4-1  C  a) B  4-1  co  co  M  CU  CO  >  T-l  Strategy Risks  CO  fl o fl erf LINF H H H L H H L L H H 3 H H 4 L L 5 M L 3 4 H H 4 1 H H 4 M L 4 L L 5 L L 3 L L 4 L L 3 L L 4 L L 3 L L M L 2 3 2 4 M H 4 M L H H L L 4 L L 4 L L 5 M L 1 4 M L 3 1 4 H H  Q.C: A.B.: Rl:  m O  HTNF H L 6 1 5 5 4 3 3 5 6 2 3 5 3 4 3 4 3 4 4 1 3 1 1 3 6 5 5 5 2 3 6 1 1 5  cu  'A  u CO  to  cod Kwtx  55 56 50 30 55 40 28 50 45 50 49 27 50 32 26 19 40 45 50 35 26 29 27 30 50 32  40 15 30 10 15 20 8 20 20 30 30 9 30 10 10 2 20 30 35 10 4 11 9 10 35 2  CJ  <ti  i-l  o fl  c o o c U  CO  Range of E x p l o i t a t i o n A.C. 1/2 caste A.C. A.C. A.C. K A.C. A.C. A.C. K K K K K K K A.C. K K A.C. K A.C. A.C. A.C. K A.C.  Queen Charlotte Islands A l e r t Bay Rivers I n l e t  Q.C.-Fraser Namu-Pender Harbour Q.C.-Fraser Q.C.-Fraser Low; around A.B. Rl-Pender Harbour Q.C.-Fraser Q.C.-Fraser Namu-Fraser Hardy Bay-J.S. Rl-Pender Harbour Rl-Pender Harbour A.B. area Rl-Pender Harbour Rl-Pender Harbour Rl-Pender Harbour Rl-Pender Harbour Namu-Fraser Rl-Pender Harbour Q.C.-Fraser Rl-Pender Harbour RI-Fraser RI-Fraser RI-Fraser RI-Fraser Rl-Pender Harbour J.S.: A.C: K:  H M M H L M H H M L M M L M M M M M M H M M M M M M  Johnston S t r a i t s Anglo-Canadian Kwakiutl  Ul  .4  Test of Propositions My 1'.  t h e o r e t i c a l propositions were:  Individuals w i l l show a preference for low r i s k strategies with respect to decisions involving the outcome to which HU i s attached. Conditions:  2'.  LW.  Individuals w i l l show an increased tolerance for high-frisk strategies, with respect to the outcome to which HU i s attached,  as  wealth increases. 3'.  Individuals w i l l show an increased tolerance for high r i s k strategies, with respect to the outcome to which HU i s attached,  as  information increases.  Tests on Strategy Risks Table XXIV shows the number of high and low r i s k strategies taken as information and wealth varied.  A strategy i s categorised as  high or low r i s k upon analysis of actual decisions of where to f i s h . Given a set of alternatives a strategy was  c l a s s i f i e d as high r i s k when  the fisherman chose that alternative which maximised gain, and as r i s k when an a l t e r n a t i v e was Table XXIV:  Information  chosen which minimised loss.  Strategy Risks  L H  L 2 8 n=12  Wealth M 23 40 n=12  H 6 1] n=Z  Number of high r i s k strategies taken.  Information  L H  L 52 52  Wealth M H 25 0 19 0  Number of low r i s k strategies taken.  low  - 142 -  I n T a b l e XXV t h i s raw data has been converted t i o n s of h i g h and low r i s k s  T a b l e XXV:  into  propor-  taken.  P r o p o r t i o n o f High and Low R i s k s : Taken by A l e r t Bay G i l l n e t Fishermen  Wealth  Medium  Low  Low  Information  Proportion of High R i s k s Taken  Proportion of Low R i s k s Taken  4%  90%  High  High  Low  High  Low  High  13.4%  48%  67.7%  100%  100%  86%  52%  32.3%  0%  P r o p o s i t i o n 1'  From the data i n T a b l e XXV one can see t h a t low w e a l t h fishermen  took low r i s k s t r a t e g i e s 96% o f the time when i n f o r m a t i o n  was low and 86.6% o f the time when i n f o r m a t i o n was h i g h . trasted with  the f a c t t h a t medium wealth  fishermen  When con-  took low r i s k  s t r a t e g i e s o n l y 52% o f the time under c o n d i t i o n s o f low i n f o r m a t i o n and  32.3% o f the time when i n f o r m a t i o n was h i g h , one can take the  r e s u l t s as s t r o n g l y s u p p o r t i v e o f p r o p o s i t i o n 1' which s t a t e s t h a t low  r i s k s t r a t e g i e s a r e p r e f e r r e d by LW i n d i v i d u a l s .  medium wealth category.  fishermen  are d i r e c t l y  The low and  comparable as N=12 f o r each  0%  - 143 -  Proposition 2'  This proposition related an increased preference for high r i s k strategies to increases i n wealth. Under conditions of low information medium wealth f i s h e r men took 44% more high risks than low wealth fishermen.  There was  a 54.3% increase i n high risks taken when information was high. Though these categories of fishermen are d i r e c t l y comparable, comparison i s d i f f i c u l t between medium wealth and high wealth categories as there were only two high wealth fishermen.  However  the fact that these l a t t e r individuals took high risks a l l the time i r r e s p e c t i v e of information lends support to proposition 2', though too much s i g n i f i c a n c e should not be attached to a sample of two.  Proposition 3' This proposition related increased preference for high r i s k strategies to increases i n information. Table XXV shows that low wealth fishermen took high risks 4% of the time when information was low and 13.4% of the time when information was high.  The information increase produced a 9.4%  increase i n the proportion of high risks Medium wealth fishermen  taken.  took high risks 48% of the time  under conditions of low information whereas 67.7% of the strategies  - 144 -  opted for under high information were high r i s k .  This represents  an increase of 19.7% i n the proportion of high risks taken as i n formation increases. For the high wealth category, information did not make any difference to r i s k taking as both skippers consistently took high risks i r r e s p e c t i v e of information, but as I pointed out previously the small sample s i z e i s not a sound basis for comparison. Test on Investment Risks When r i s k - t a k i n g with respect to c a p i t a l investment was tested for i n the A l e r t Bay sample the f i t was less good. Proposition 1' was supported,  as a l l 13 low wealth f i s h e r -  men took low investment r i s k s . However proposition 2' was not confirmed as some of the l o g i c a l states (MW L.Inf; HW L.Inf; HW M.Inf, etc.) were not observed i n the empirical world. The same can be said for proposition 3'.  I t was not  empirically possible to get the o r d i n a l increments i n r i s k - t a k i n g to test the propositions.  This means that one or more intervening  variables, functional to risk-taking, operate i n such a way as to prevent p a r t i c u l a r configurations or states from appearing. The f i r s t possible intervening variable I would l i k e to consider i s education.  R e c a l l the information c i r c u i t s model I discussed  - 145 in Chapter 2 -- i n that I v i s u a l i z e d access to information c i r c u i t s about technology  as a function of wealth and education, and that r i s k -  taking on technology was a function of information category (Figure 10).  Figure 10:  Information C i r c u i t s Model  Of the 26 A l e r t Bay cases, the model was correct for 18.  The  eight cases that, did., not conform to the model were:  No. 4 7 8 9 17 18 20 25  -  MW -MWMW MW LW MW MW MW  MEd HInf MEd HInf MEd HInf LEd MInf MEd MInf LEd MInf MEd HInf LEd MInf  Predicted Inf Category MInf MInf MInf LInf LInf LInf MInf LInf  Of the 7 MW MEd configurations four had medium access to i n formation as predicted, three had high access.  This difference can be  located between the significance of c i r c u i t e with respect to d for part i c u l a r individuals.  For 4 fishermen e'ifj^d, for three  c\~r—P.-  - 146 -  Similarly of the three MW/LEd combinations, a l l had MInf i n stead of LInf as the modal predicted.  Of the f i v e LW/MEd, only one case  did not conform to the p r e d i c t i o n ; f o r this individuals h * ^ g , and h ^ f . Thus each one of these s i x contrary cases involve a d i f f e r e n t i a l s i g n i f i c a n c e attached to only adjacent information c i r c u i t paths. This factor and the fact that almost 70% of the cases ran according to prediction makes i t reasonable  to accept the model as holding for the  majority of cases, though i t i s evident that the r e l a t i v e  weightings  of wealth and education have to be revised for some parts of the model. The model i s less accurate with respect to the next step of predicting r i s k - t a k i n g i n terms of the l e v e l of access to information circuits.  A l l the cases i n the H.Inf category and the L.Inf  category  took investment r i s k s as predicted by the model, but of the eight f i s h ermen i n the M.Inf category two took High Investment Risks, and s i x took Low Investment Risks.  This suggests that i n terms of c a p i t a l i n -  vestment on new technology H.Inf as measured by access to information c i r c u i t s i s the generally necessary  requisite for high r i s k .  Congruence Between Investment and Strategy Risks  For 23 of the A l e r t Bay cases there was a 1:1 congruence between investment r i s k s and strategy r i s k s , but the three exceptions are espec i a l l y noteworthy.  - 147 -  17.  North Wind -- The skipper on this boat had not invested  in any new technology for his boat, yet was observed  to take high stra-  tegy r i s k s f a i r l y consistently. Of f i v e low information situations, he twice elected to take high r i s k s , and on five high information situations he took high r i s k s four times. On my shore interview I learned two things:  (a) that higher  r i s k s than were safe f o r the boat were being taken; (b) the lack of capi t a l investment  i n the boat was mainly a function of the fact that i t  was not worth investing in.' The skipper, N e i l Johnson, was looking ahead i n terms of getting a new boat, but i n order to f l o a t a loan from his f i s h i n g company he would have to turn i n a good performance with regard to the amount of f i s h he landed.  His elected strategy to achieve this end  was to take higher than usual exploitative r i s k s . 25. low investment  Slave -- Sonny Commano, the skipper of the Slave took r i s k s on his boat, but on both high and low information  consistently elected to take high strategy r i s k s . One explanation for this could be attributed to the general consensus amongst A l e r t Bay fishermen that no one knew the currents and tides quite l i k e Sonny, which would mean that some modern technology would be superfluous to him -- radar, depth sounder, etc.  (But there  were others l i k e him, who took high strategy r i s k s yet also invested in  technology.)  - 148 -  Another explanation may sense".  be a t t r i b u t a b l e to Sonny's " f i s h  But, i n general he just seems to be an exception. 26.  Carodic -- Roy Bianco, skipper of the Carodic had inves-  ted $18,000 i n the boat and much of the l a t e s t technology,  but usually  always opted for low r i s k strategies. He informed me  that this was  he had only been a fisherman  c h i e f l y due to the fact that  for two years, and didn't f e e l he knew  enough about tide, current and run conditions to f i s h the hazards. though he had taken high investment risks on technology,  he had  So  decided  to take low strategy r i s k s . These exceptions point out something very important  -- v i z .  that r i s k - t a k i n g as a means to increase f i s h intake can be used i n one or both of two ways. Investment r i s k s on technology provide" a fisherman with a better means to exploit the general f i s h resource.  Better equipped  boats had a greater range and mobility and could f i s h areas that poorly equipped boats were denied access to. High r i s k s on f i s h i n g strategies served to place the f i s h e r men  i n a pas i t i o n where they could exploit the greatest quantity of  salmon; though this often placed the fisherman i n a hazardous p o s i t i o n with respect to his boat's and own physical well-being.  - 149 -  Relevance of Cultural Factors  None of the Kwakiutl fishermen i n the sample had a high range of exploitation, and i t could be argued that this has a c u l t u r a l base v i z . Indians prefer to exploit areas close to their home base while whites prefer to go further a f i e l d . But I think an economic explanation s u f f i c e s .  Of the twelve  Indians i n the sample, 11 took low investment r i s k s on technology, which implied that they were prohibited from exploiting the entire inshore f i s h e r i e s because of d e f i c i e n t technology.  Also, white fishermen who  s i m i l a r l y took low investment r i s k s had a limited range of exploitation for the same reasons. However, some of the data I collected gave support to the notion that c e r t a i n c u l t u r a l factors are very relevant^to differences i n risk-taking.  During the course of fieldwork on the A l e r t Bay boats,  I was constantly struck by the technological ware displayed on two part i c u l a r categories of boats --Japanese  and Yugoslav g i l l n e t t e r s .  Time constraints prevented me from doing an adequate  survey,  but data collected from f i v e Japanese boats and three Yugoslav boats with respect to investment i n technology indicated that the i n d i v i d u a l fishermen concerned took far higher investment r i s k s than one would expect. The table below shows that the wealth and education combinations p r o d u c e 1  a higher than expected information l e v e l (not just ad-  jacent information c i r c u i t paths), resulting i n the expected high investment r i s k s .  Skipper  Boat  Kamachi  Utility  W  Ed.  Inf. Circ.  Inv. Risks  Age  Experience  Reference Group  H  M  L  H  H  50  28  Japanese  H  Range  Hikari  M.  Nugget  T. Asano  H  M  L  H  H  45  20  Japanese  H  Symphony I I .  S. Muramatsu  H  M  L  H  H  30  10  Japanese  H  Libra  M.  H  M  L  H  H  48  25  Japanese  H  Bonnie L o c k  T. Oikawa  H  M  L  H  H  35  15  Japanese  H  A l b i o n Ranger  I . Malavoz  H  M  L  H  H  45  18  Yugoslav  H  A l b i o n Commando  R. S e k u l a  H  M  M  H  H  35  10  Yugoslav  H  Tony  A. P e r a d e n i c  H  M  H  H  55  20  Yugoslav  H  Kamachi  - 151 -  Two of the Japanese boats, H i k a r i and Symphony I I , carried radar.  The other boats did not, largely as a function of the fact  that the f i v e boats always fished together, and information from the boats with radar would be relayed to the other three. The Nugget meant more speed.  and Symphony II had f ibreglass planing h u l l s which  A l l had the very l a t e s t i n e l e c t r o n i c equipment  radio telephones, automatic p i l o t s and depth sounders.  —  This group of  f i v e represented a considerable investment i n f i s h i n g technology. A l l f i v e Japanese fishermen l i v e d i n the Japanese sector of Steveston, within walking distance of one another.  Japanese was spoken  at home, and as f a r as I could gather the majority of their s o c i a l relationships were within the Japanese community. The skippers of the H i k a r i and Libra had been interred dur1  ing World War II and had had t h e i r seine boats confiscated.  They f e l t  they had received a very bad deal over this and a f t e r twenty years there was s t i l l a note of bitterness about i t . The three Yugoslavs possessed what were perhaps the most beaut i f u l g i l l n e t t e r s on the entire West coast.  B u i l t at a cost of $30,000  each, by the Albion Boat Co. of Haney, B.C., the boats were the very l a s t word i n technology —  high-speed engines and planing h u l l s , radar,  automatic p i l o t , hydraulic drum, depth sounders with f i s h finder functions —  absolutely everything! It i s interesting to note that of the seven boats b u i l t by  this firm since 1968 four are owned by Yugoslavs.  -  152  The three Yugoslav skippers l i v e d i n West Ladner where 60% of the population was  Yugoslav.  The influence of ethnic ties were  quite extensive i n that a s p e c i a l Yugoslav school operated at weekends, and s o c i a l relations were more with other Yugoslavs than with English-speaking  Canadians.  Both groups of fishermen are thus categorised by an ethnic i d e n t i t y and an idea of ethnic separation, and both are i n t r u s i v e minority groups —  i . e . they came to Canada.  These considerations  may  make explicable their preponderance for innovation i n terms of a consideration of s o c i a l marginality  (Proposition 8, Chapter 2).  The implication of s o c i a l marginality i s of individuals at the margin of one given culture, yet occupying a p o s i t i o n where they operate i n more than that culture. The marginal i n d i v i d u a l i s not bound by wider group norms and i n addition has the resources  of his own  i d e n t i t y group to draw  upon.When this i s accompanied by a lower evaluation placed upon cost factors than by non-marginal persons, then, i n p r i n c i p l e , i t would appear that the marginal i n d i v i d u a l has a much wider scope for activity. The data I have on the Japanese and Yugoslav fishermen i s only suggestive  for support of the supplementary proposition r e l a t i n g  s o c i a l marginality to r i s k .  This evidence can only be taken as i n d i c a -  tive and not conclusive, due to the very small sample s i z e .  - 153 -  Summary The f i e l d test gave general support to my propositions on r i s k taking with respect to strategies preferred by fishermen, but there was less good a f i t when one came to consider investment risks on technology. Intervening variables of education and of s o c i a l marginality were shown to be relevant i n some cases.  However the most important  ing was that low wealth individuals took low r i s k s —  which was one of the  major contentions I made i n my i n i t i a l discussion (Chapter also the most strongly supported  find-  2), and was  result from the experiment (Chapter 3).  Fishermen exhibited an increased tolerance f o r high r i s k s t r a tegies as wealth increased which supported proposition 2'.  As wealth  moved from low to medium this increase was especially marked.  Increased  high r i s k taking was also noted as wealth moved from medium to high, but as there were only two HW fishermen i n the sample one cannot place too much emphasis on the r e s u l t , though i t does support the o v e r a l l contention of proposition 2'. Except f o r the two high wealth fishermen, everyone else i n the A l e r t Bay g i l l n e t f l e e t indicated a higher tolerance f o r high risks as information increased. information.  The former took high risks i r r e s p e c t i v e of  This gives p a r t i a l support to proposition 3'.  The experimental  data indicated the importance of the proba-  b i l i t y consideration of taking r i s k s f o r LW i n d i v i d u a l s .  I t was less  easy i n the f i e l d to find out exactly what chance factors were operative for a p a r t i c u l a r fisherman, as h i s estimated payoffs and losses were  In-  u t i l i t y functions ($ .P : and $,.P.) and not straight calculations of ww 1 1 J  payoffs and costs. However, I did discover that the chance factor was more a r t i culated by LW fishermen when "hazard" f i s h i n g was  envisaged.  To f o u l  on any hazard meant at least $200 for net repairs, and the chance of fouling had to be pretty low (tide running away from hazard, etc.) before a LW fisherman would consider s e t t i n g o f f a hazard.  In these  instances the (P^) appeared to be the most s i g n i f i c a n t factor. I would have l i k e d to assess whether the threshold notion v i s - a - v i s information and r i s k taking also operated under f i e l d  condi-  tions . However as I pointed out e a r l i e r (page 110 ) high information for strategies was experiment.  analoguous to the medium information l e v e l i n the  There was  no e f f e c t i v e way  that information could be  increased for the fisherman beyond specifying the range of odds, payoffs and costs for any a l t e r n a t i v e . Thus I was  unable to see whether  increments i n information a f t e r the threshold point ( i d e n t i f i e d i n the experiment) made any difference to decision strategies employed. Similarly the information c i r c u i t s model related r i s k taking  to the number of c i r c u i t s of information a person had access to  i n terms of h i s wealth and education.  The argument made was  that i f  a person a c t i v e l y sought out a number of information c i r c u i t s then one would expect him to idee higher r i s k s on  technology.  The scaling of information with regard to technology was dinal, and i t gave no provision for i d e n t i f y i n g cut o f f points such  or-  - 155 -  as the s p e c i f i c a t i o n of the range of odds, payoffs and costs.  This  was because the information c i r c u i t s model was designed for a d i f ferent problem —  that of r e l a t i n g a person's risk-taking behavior to  the number of information c i r c u i t s that he tapped. So the threshold notion seems to occupy the status of a by-product of experimental serendipity as I was unable to adequately assess how i t operated i n the f i e l d , due to adverse s i t u a t i o n a l factors v i s - a - v i s f i s h i n g strategies and the lack of an operational d e f i n i t i o n for the threshold with reference to investment risks on technology. The assumptions of weighting education v i s - a - v i s wealth which were fundamental to the construction of the information  circuits  model were shown to be v a l i d i n the majority of cases, though a p a r t i a l r e v i s i o n of the model i s necessary because of the exceptions. The proposition with respect to high investment returns at low cost was not tested i n the f i e l d , simply because there were not any strategies or technologies  that had these features.  However, i n the next chapter,  I s h a l l be marshalling evidence  to show that low wealth individuals w i l l respond p o s i t i v e l y to risks of this nature. Despite the lack of congruence i n testing procedures between the experiment and the f i e l d ,in that I was unable to test i n the f i e l d a l l the propositions that were tested i n the laboratory, both have shown the preference  contexts  f o r low risks exhibited by LWHU i n d i v i d u a l s .  - 156 -  I now have to show among other things that this i s the case for Third World farmers. I w i l l also be looking for cases that seem to refute my assertions, to see whether i n fact these contrary examples f a l l within the s p e c i f i e d scope conditions which are pertinent to my statements .  - 157 -  CHAPTER 5 FARMERS AS RISK TAKERS  Introduction In this chapter I w i l l be examining evidence from development l i t e r a t u r e to assess whether or not my assertions about r i s k taking behavior are, at a prima f a c i e l e v e l , applicable to the  do-  main of Third World farmers. I should emphasize that the evidence i s not d e f i n i t i v e . Short of doing fieldwork i n a variety of ethnographic contexts  I  am unable to o f f e r an e x p l i c i t test of my assertions, as I have to rely on work done by scholars interested i n d i f f e r e n t sorts of problems. Where possible I i n f e r incentive conditions, resources  information,  and u t i l i t y dimensions from the e x i s t i n g l i t e r a t u r e .  Ex-  cept i n those instances where the data i s readily amenable to this type of treatment, I should acknowledge that the process of inference i s by and large a rule of thumb method. Despite my  i n a b i l i t y to adequately test s p e c i f i c proposi-  tions i n this context, I w i l l be subjecting the general model to quite a severe scrutiny, as I w i l l be looking for cases that appear to refute my  contentions.  This i s i n order to set the precise boun-  daries for the domain of relevance  for my assertions about r i s k - t a k i n g .  - 158 -  Roger's book "Diffusion of Innovations"  (1962) i s considered  to be a c l a s s i c sourcebook i n this f i e l d and inasmuch as i t reviews and attempts to synthesise available research findings on the d i f f u sion of innovations i t certainly i s . But inasmuch as i t continues the Rural Sociology t r a d i t i o n (Beal, Rogers and Bohlen 1957; and Taves 1952;  Copp 1958)  Gross  of studying the c h a r a c t e r i s t i c s of i n d i -  viduals associated with innovativeness, i t does l i t t l e to develop any theoretical considerations.  Rogers does not attempt a theory,  though he suggests further research should be concerned with a more s t r i c t l y theoretical endeavour. He i s primarily concerned with a taxonomy of adopter categories whereby individuals can be categorised i n terms of the speed with which they adopt innovations.  The categories he uses —  inno-  vators, early adopters, early majority, late majority and laggards are not used i n any way  to specify why  —  one i n d i v i d u a l decides to  adopt an innovation and another does not.  His concern i s to esta-  b l i s h , a f t e r the fact, the patterns of adoption i n any given community. His variables are selected on an a p o s t e r i o r i basis and this i s a feature of most of the work conducted on a g r i c u l t u r a l innovation. Consequently no claims can be made for advancing tiveness .  a theory of innova-  I t i s not u n t i l one has an a p r i o r i s e l e c t i o n of variables,  which when accompanied by the s p e c i f i c a t i o n of their i n t e r r e l a t i o n ships can one say there i s a theory.  - 159  -  I am a l s o none too sure t h a t Roger's concern w i t h is particularly  u s e f u l , and wonder whether a more s t r i c t l y  taxonomy theoretical  approach would render h i s c a t e g o r i e s s u p e r f l u o u s . For i n s t a n c e , the d i f f e r e n c e between " I n n o v a t o r s " and Adopters"  c o u l d be viewed i n terms of a s e r i e s of p r o p o s i t i o n s r e -  l a t i n g a s e t of antecedent p r e d i c t a b l y w i l l be  c o n d i t i o n s to the types of r i s k s t h a t  taken.  " I n n o v a t o r s " may mation,  "Late  have a t t r i b u t e s of h i g h access  to i n f o r -  be capable o f e f f e c t i v e l y m o b i l i s i n g r e s o u r c e s i n a d d i t i o n  to b e i n g s o c i a l l y m a r g i n a l .  One would then p r e d i c t t h a t i n d i v i d u a l s  w i t h these c h a r a c t e r i s t i c s would take h i g h The low access  "Late Adopter" may  risks.  be c h a r a c t e r i s e d by low w e a l t h  to i n f o r m a t i o n and p r e d i c t a b l y w i l l take low r i s k s .  when a l a r g e number of farmers  around him have adopted  and Only  an i n n o v a t i o n ,  such as h y b r i d c o r n , can the " l a t e a d o p t e r " assess to h i s s a t i s f a c t i o n the r e l a t i v e advantages of a d o p t i n g .  His lateness i s a f u n c t i o n  of the f a c t t h a t he c o n s i s t e n t l y  r i s k s and wants to see  takes low  everyone e l s e f a r e s b e f o r e a d o p t i n g .  how  I n these terms adopter c a t e g o r i e s  do not make too much sense as they can be accounted  f o r by r e f e r e n c e  to d i f f e r e n t i a l r i s k t a k i n g . S i m i l a r l y Rogers i s concerned w i t h d i s t i n g u i s h i n g i n the a d o p t i o n p r o c e s s — (1962: 81).  stages  awareness-interest-evaluation-trial-adoption:  From the p o i n t of view of d e c i s i o n theory one  only w i t h t r i a l and a d o p t i o n .  is  concerned  The p r e c e d i n g c a t e g o r i e s are merely  various  - 160 -  configurations of "access to information" and " u t i l i t i e s " .  The d i f -  ference between t r i a l and adoption i s simply the difference between decisions at time T and time T ,,. I f the incentive conditions as n n+1 perceived by the i n d i v i d u a l change as a function of increased  informa-  tion or changing u t i l i t i e s , then one can expect some change i n decision at T ,,. I f a d d i t i o n a l information does not r e s u l t i n a change i n n+1 incentive conditions, then the decision at T w i l l be the same as n+1 at T , a l l other things being equal, n What I am trying to point out i n this b r i e f discussion i s that categories and taxonomies without a theoretical referent remain mere categories and taxonomies, and indicate an i n t e r e s t i n b u t t e r f l y c o l l e c t i n g (Leach 1961) rather than systematic knowledge. A more considerable drawback, for my purposes, i s that Rogers i s not e x p l i c i t l y concerned with the actual process of the i n d i v i d u a l .  decision-making  The calculations of a p a r t i c u l a r farmer  with respect to y i e l d s and payoffs v i s - a - v i s costs aia not emphasised. Also other researchers  (Cancian F. 1967; Ryan and Gross 1943) have  tended to regard a l l innovations as equivalent units from the viewpoint of analysis with the r e s u l t that the s i g n i f i c a n c e of the incentive conditions ( i n terms of both material and s o c i a l calculation) i s obscured. I have argued e a r l i e r that i t i s the incentive conditions that define the nature of the r i s k i n adopting an innovation. also maintained  I have  that incentive conditions, both p o s i t i v e and negative  - 161 -  can be calculated i n s o c i a l as well as i n material terms.  In this  way decision making can be viewed as an expanded cost-benefit analysis  (subject to certain constraints) and the preoccupation with  t u r a l obstacles to adoption of innovations  cul-  that characterises the  work of Foster (1962), Spicer (1952) and Hogbin (1958) can be put i n a more r e a l i s t i c  perspective.  Cultural norms become obstacles to adoption of innovations when adoption would involve some perceived c u l t u r a l cost, such as disapproval of peers or loss of status.  An innovation may be rejec-  ted on the grounds that i t c o n f l i c t s with group norms and values.  A  subsistence farmer, dependent on s o c i a l a f f i l i a t i o n for access to land, labour and means of production, cannot readily f l o u t group norms, as this may invoke sanctions that would cost him his access to means of l i v e l i h o o d . In this sense c u l t u r a l norms as "obstacles" are merely an additional dimension to the material incentive conditions to the innovation.  attached  Certain problems e x i s t i n the estimation of c u l -  t u r a l costs and payoffs v i s - a - v i s a decision to adopt an innovation, but this i s a problem of measurement and not of conceptualisation. Thus i t i s that my approach takes l i t t l e d i r e c t l y from the Rogers and Rural Sociology t r a d i t i o n , as my discussion of Rogers' adopter categories indicates my concern i s with d i f f e r e n t kinds of questions.  G r i l i c h e s (1957) anticipates much of my approach with h i s  emphasis on economic variables as the major determinants of the pattern of technological innovation.  He emphasises the predominance of  - 162 -  profitability innovation's an  —  the d i f f e r e n c e between economic r e t u r n s  economic c o s t —  and  the  i n e x p l a i n i n g the r a t e of a d o p t i o n of  innovation. While I endorse h i s n o t i o n of p r o f i t a b i l i t y ,  I feel  i t has  to be broadened to i n c l u d e c a l c u l a t i o n s of p r o f i t a b i l i t y  in social  terms.  factors  Peasant farmers do r e a c t to p r o f i t a b i l i t y  f a c t o r combinations, but  they do so as  the c o n s t r a i n t s of t h e i r own  o f new  they view p r o f i t a b i l i t y  and  through  t e c h n o l o g i c a l (and s o c i a l ) systems.  I am not p r e p a r e d to argue t h a t s o c i o - c u l t u r a l f a c t o r s c a n c e l one  another out  l e a v i n g s t r i c t l y economic v a r i a b l e s as  prime movers o f t e c h n o l o g i c a l change. overlooks dual's  to do  the f a c t t h a t the important t h i n g to a s s e s s i s the  perception  terms but  G r i l i c h e s i s , and  of p r o f i t a b i l i t y ,  i n terms o f s o c i a l p r o f i t .  s t a t e s of n a t u r e , u t i l i t i e s  and  not  only i n s t r i c t l y  T h i s i s why  I am  subjective evaluations  the so  indivi-  material  concerned w i t h of p a r t i c u l a r  outcomes. I t should  be  apparent t h a t I cannot r e a d i l y use  done on a g r i c u l t u r a l i n n o v a t i o n s , d e c i s i o n theory  as  the work  the parameters r e q u i r e d  are r a r e l y f u l l y s p e c i f i e d .  Thus I w i l l o n l y  for draw on  those s t u d i e s where I can i n f e r or e s t i m a t e the parameters n e c e s s a r y to a s s e s s d i f f e r e n t a s p e c t s of my  concern w i t h the a c t u a l p r o c e s s  of  d e c i s i o n making. To s t a r t o f f w i t h a case t h a t g i v e s l i n e o f enquiry  I would l i k e to c o n s i d e r  Rubber S m a l l h o l d i n g s  (Bauer 1946).  a p o s i t i v e colour  to  my  Bauer's examination of Malayan  - 163 -  Case No. 1:  Malayan Rubber  His c r i t i c a l report on the administration of and s t i p u l a t i o n s with respect to smallholdings  includes a detailed documentation of two  instances where opportunities f o r investing i n technical change were passed up by the vast majority of smallholders. These instances concerned the processes new planting respectively.  of replanting and  Replanting involves the digging up of o l d  rubber trees and planting new ones i n their stead, while new planting refers to the process of planting new trees on new acreages.  New plant-  ing was prohibited by the government, although the regulations were relaxed i n 1939-40. As part of h i s survey Bauer distinguished between peasant, medium and estate holdings as a function of scale.  Peasant holdings  were characterised as having a resident owner, tapped by family l a bour and comprising  an area of 10-12 acres.  Medium holdings of 60-100 acres saw the increased use of contract labour for tapping, while the large estates r e l i e d e n t i r e l y on wage-labour f o r tapping.  The great bulk of the Malay-owned acreage  i s i n peasant holdings; some of the smaller Chinese holdings and a few Indian holdings also f a l l i n this category.  The estates were l a r g e l y  controlled by Europeans.  Replanting The process of replanting involved the uprooting of e x i s t i n g trees and their replacement by a new stand planted on the same area.  - 164 -  Replanting enabled the rubber producer to weed out older unproductive trees and to replace them with new e s s e n t i a l l y a long-term investment, new  clonal types.  This was  as i t took several years for the  stand to reach maturity. The cost for labour i n the process of digging out the o l d  trees, and the heavy expenditure required f o r manuring to produce a second stand of mature trees on what i s more often than not unsuitable ground, constituted a prohibiting factor f o r the peasant smallholder. Before World War  II there was hardly any replanting on small-  holdings, and what there was, was  almost e n t i r e l y on the larger  Chinese  and Indian holdings, which are hardly smallholdings i n the accepted sense of the word. (1946:  Replanting was, however, widely adopted on estates  91). While i t i s feasible to replant on large estates i t i s  impractical to do so on peasant smallholdings f o r economic and technical reasons.  The lack of c a p i t a l required to pay not only for the  heavy expenses involved but also to bridge the loss of incomes from the time of the f e l l i n g of old trees to the maturity of the new  stand  constrained peasant holders from replanting. Also i t was  technically impossible to replant part of a  holding of a few acres successfully, as the area replanted would be closely surrounded by mature trees, which would intercept the sunlight and whose roots would compete f o r food with the undeveloped rootlets of the newly planted trees, resulting i n stunted trees.  The combined  - 165 -  result of these various factors has been v i r t u a l l y to confine replanting to estates (1946: 93). When Bauer questioned the smallholder as to the absence of replanting —  the reply was  always the same —  i n a b i l i t y to afford the  loss of income from the f e l l e d stand and the technical d i f f i c u l t i e s of replanting part of a peasant holding (1946: 108). If one assumes that wealth i s coterminous with size of holdings, Bauer's account of replanting strongly supports proposition 1' which states that under conditions of LW and HU,  individuals w i l l  show a preference for low r i s k strategies with respect to decisions involving the outcome to which HU i s attached. This preference for low r i s k i s compounded of two main factors : —  peasant smallholders did not go for marginal returns to inves tment.  —  also they seemed to be more concerned with cost minimization:: than gain maximization.  That I can assume that U for increased productivity i s high i s borne out by Bauer's frequent references to the energy and market orientation of the peasant smallholders.  In fact the whole thesis of  his report i s to change rubber planting r e s t r i c t i o n s i n favour of the smallholders mainly on the grounds that they can do more for the export industry than can the larger, more costly estate operations. Also the progression from peasant - medium - estates i s d i r e c t l y proportional to an increase i n replanting, from none at the peasant  - 166 -  l e v e l , to a number of instances at the medium l e v e l to most estates at the upper l e v e l .  This evidence i s supportive of proposition 2'which  relates tolerance for high r i s k to increasing wealth.  New Planting New planting to the extent of 5% of the 1938 planted acreage was permitted i n 1939-40. The peasant category took v i r t u a l l y no advantage of the small amount of new planting permitted, and what area was replanted was almost e n t i r e l y on the larger Chinese and Indian holdings (medium) (1966: 14). "I did not find a single instance of an owner of less than 20 acres undertaking new planting without purchasing planting r i g h t s . In other words, every instance of an area planted i n 1939-40 I saw i n the course of this extensive tour belonged to a man who either had the c a p i t a l to buy planting rights, or had s u f f i c i e n t land to be classed as the owner of a medium holding rather than as a peasant smallholder." (1946: 116) New planting was organized i n terms of share c e r t i f i c a t e s i n denominations and multiples of 1/20 of an acre which e n t i t l e d the owner to plant rubber to the extent of 5% of h i s 1938 reigstered acreage under rubber.  Thus an owner of 3 acres would be e n t i t l e d to plant 1/7  of an acre, while the owner of a 5 acre holding was e n t i t l e d to plant 1/4 of an acre.  - 167 -  Any l a n d h e l d by peasant s m a l l h o l d e r s rubber) u s u a l l y had o t h e r etc.,  crops  (which was n o t under  growing on i t —  fruit  trees, tapioca,  and new p l a n t i n g on t h e i r own l a n d c o u l d o n l y be undertaken a t  the c o s t o f d e s t r o y i n g another source Owners who had no unplanted  o f income. r e s e r v e l a n d had to apply to  the l a n d o f f i c e s f o r the a l i e n a t i o n o f l a n d on which to c a r r y out new planting.  F o r the peasant s m a l l h o l d e r  the amounts o f l a n d were s m a l l  and u s u a l l y a f a i r d i s t a n c e from h i s e x i s t i n g h o l d i n g . i s o f t e n prepared  to operate  A  smallholder  a h o l d i n g a m i l e o r two away from h i s  house, b u t i t i s not f e a s i b l e f o r him to go f a r a f i e l d  to tap 16, 24  or 32 t r e e s . The to  f r a c t i o n a l acreage o f new p l a n t i n g was thus o f no v a l u e  the i n d i v i d u a l peasant s m a l l h o l d e r ,  u n l e s s he had unused  reserve  l a n d o r i f t h e r e was u n a l i e n a t e d l a n d c l o s e to h i s own h o l d i n g . These l a t t e r  two c o n d i t i o n s were very  Bauer concludes  r a r e l y met.  t h a t "New p l a n t i n g by the g r e a t m a j o r i t y o f  the s m a l l h o l d e r s was thus r u l e d o u t . " (1946: 120) Once a g a i n i t was the e s t a t e s and w e a l t h i e r medium h o l d i n g s planting. tions.  t h a t c a p i t a l i s e d on the new  T h i s i n s t a n c e i s a g a i n s u p p o r t i v e o f my f i r s t  two p r o p o s i -  LW producers a r e observed t o take low r i s k s by m e r i t o f t h e i r  attempts to keep costs to a minimum and not to go f o r m a r g i n a l nomic advantages i n terms o f b e t t e r rubber t r e e s .  The w e a l t h  ecogradient  from peasant-medium-estate h o l d i n g s was a g a i n accompanied by an i n crement i n r i s k t o l e r a t e d w i t h  r e s p e c t to new p l a n t i n g .  - 168 Also Bauer makes frequent references  (1946: 121, 122,  135)  to the fact that the estates and some medium owners were much better informed about the precise nature of the 5% new  planting  permitted,  whereas the peasant holders very often did not know what the whole process was  about.  While I cannot glean from Bauer's report any  information  with respect to education, his references give support to my  asser-  tion that the taking of high risks increases as a function of i n f o r mation (Proposition 3').  Summary One  could view the lack of i n i t i a t i v e i n terms of new  and  replanting as peasant conservatism, but this would be doing them an injustice.  I t i s clear from Bauer's report that the peasant small-  holders made very shrewd economic calculations and rejected the oppor tunities offered i n terms of cost considerations and also because i t was  not worth their while (economically)  to either replant or  new  plant. Bauer states i n paragraph 63 "...the decisions of the small holders seemo quite d e f i n i t e l y influenced by f a i r l y r a t i o n a l comparisons between the advantages of the alternatives i n terms of income, expenditure, e f f o r t and r i s k . " (Bauer  1946)  - 169 -  LW Farmers and Risk From an example of general support I would l i k e to move to the s p e c i f i c and take each proposition i n turn. Proposition 1' referred to LWHU individuals indicating preferences for low r i s k strategies.  This assertion can be examined i n  terms of three substantive aspects with respect to a g r i c u l t u r a l innovations : (A)  Innovations with a high payoff factor which c a l l for a minimal diversion of stocks from existing resources w i l l tend to be adopted;  (B)  Innovations that o f f e r a low return tend not to be adopted;  (C)  Farmers seek to minimise the cost factor of an innovation, hence they are more concerned with cost reduction than high returns.  (A) way  The high investment return at low cost thesis goes a long to explain the present a g r i c u l t u r a l revolution i n A s i a . The basis of the revolution l i e s i n  a technical  breakthrough  i n the choice and recombination of desirable genetic sources of a small number of high-yielding v a r i e t i e s of r i c e , wheat and coarse grains at the International Rice Research I n s t i t u t e (IRRI) near Manila and other research centres i n Mexico and India. These v a r i e t i e s , being v i r t u a l l y hand-tailored p h y s i o l o g i c a l l y and morphologically for f e r t i l i z e r responsiveness, showed f a n t a s t i c results i n widespread t r i a l s and demonstrations  with y i e l d s double  even treble those of the t r a d i t i o n a l v a r i e t i e s grown under the same conditions of good i r r i g a t i o n and heavy  fertilization.  and  - 170 -  This factor accompanied by f e r t i l i z e r subsidies (Punjab) and seed price subsidies (Philippines) produced a low-cost high  investment-  return innovation. Since 1966 when the hybrids were f i r s t d i s t r i b u t e d i n the Punjab,  Indian farmers i n two years planted about 5 m i l l i o n acres each  of the high-yielding v a r i e t i e s of r i c e and wheat and nearly 3 m i l l i o n acres of those of corn, sorghum and m i l l e t s (Hsieh Sam-Chung 1968). Pakistan had planted approximately  2 m i l l i o n acres of Mexican  wheat by 1968, and 1 m i l l i o n acres of IRRI r i c e .  In the Philippines  the area planted to the "miracle r i c e " IR-8 increased to 400,000 acres by 1967. Yet i n 1965 Schultz sounds l i k e a prophet of doom i n his book "Economic C r i s i s i n World Agriculture".  Before the introduction of the  hybrids there was a gloomy, almost desperate concern for the i n a b i l i t y of Asian farmers  to produce enough food for basic n u t r i t i o n a l needs;  while population growth was accelerating. But i n 1967 Dr. D.S. Athwawl of the Punjab A g r i c u l t u r a l Univ e r s i t y announced that by 1970 India would be s e l f - s u f f i c i e n t i n grain and r i c e crops (Economist  1967).  In the' 1969 annual report of the  United Nations Food and Agriculture Organisation the director-general forecast that the major problem facing some underdeveloped countries i s that of handling a surplus of a g r i c u l t u r a l production (F.A.O. 1970). Extrapolating from this report the international science journal "Nature" predicts that even countries such as India are now faced with the p o l i t i c a l l y contentious problems of how to manage a surplus (Nature 1970).  - 171 -  In these few short years I do not think the surplus i s a function of any d r a s t i c change i n the Indian peasant farmer. maintain that he i s s t i l l operating on a low r i s k strategy.  I would The  dif-  ference l i e s i n the economic opportunities offered by the hybrid grains and  rice. The opportunity  to invest i n these new  crops constitutes a  low r i s k strategy for the farmer, as his costs are minimised through subsidies and the incentive conditions of high y i e l d s reassures at the very worst, of a harvest bigger than that which can be  him, obtained  through using t r a d i t i o n a l v a r i e t i e s . (B)  The evidence from Bauer indicates a shrewd assessment of eco-  nomic p o t e n t i a l with regard to new planting and replanting on the part of the Malayan smallholder. marginal the smallholder  Where the returns were seen to be  did not r i s k his investment c a p i t a l .  Schultz' thesis i n "Transforming T r a d i t i o n a l A g r i c u l t u r e " (1964) i s based on the notion that peasant farmers i n t r a d i t i o n a l a g r i c u l t u r a l systems are not going to invest further i n t r a d i t i o n a l production factors because the return to investment i s so low. assertion i s generalisable to any production  I think his  factor, be i t t r a d i t i o n a l  or not, as no farmer i n the LWHU state can afford to r i s k his "surv i v a l k i t " for marginal benefits. His discussion of the Guatemalan community of Panajachel shows that the community i s not an i s o l a t e d subsistence economy, but i s closely integrated into a larger market economy.  - 172 -  "Yet hoes, axes and machetes are not replaced by better tools and equipment.  There i s not even a wheel.  Coffee leaves used as f e r -  t i l i z e r are not replaced or supplemented by chemical f e r t i l i z e r s .  Tra-  d i t i o n a l v a r i e t i e s of corn are not replaced by hybrid seed. T r a d i t i o n a l breeds of chickens are not replaced by better hens for producing  eggs  and b r o i l e r s for producing meat." (1964: 35) This i s not because the Guatemalian Indian i s not looking for ways to improve his l o t .  Tax notes that "He i s on the lookout  for new and better seeds and ways of planting" (Tax 1953). But such improvements as the Indian could use have an exceedingly small e f f e c t upon production.  Tax calculated that the return on  further investment i n available tools, domestic animals and growing crops was too low to provide an investment incentive (1953: 118 f f ) . A s i m i l a r argument i s advanced by Schultz to explain the s i t uation i n Senapur, India.  In Senapur there i s a larger stock of ma-  t e r i a l c a p i t a l than i n Panajachel.  But despite this the two v i l l a g e s  have i n common a low rate of return to investment i n a l l forms of capit a l used i n a g r i c u l t u r a l production. The s i m i l a r i t i e s of both v i l l a g e s could be considered as a function of the fact that both economies perform highly e f f i c i e n t l y i n using the factors at hand, and farmers i n both v i l l a g e s have shown a reluctance to invest i n new production factors mainly as a function of the marginal productive advantage of the new factors.  - 173 -  Before leaving the "no r i s k for marginal return" hypothesis, r e c a l l that the experimental findings indicated that a low payoff was a t t r a c t i v e to some LW individuals when they were reasonably sure that the payoff would accrue, and that their costs were low and prospects of f a i l u r e low. Unfortunately, i n the f i e l d and from the l i t e r a t u r e I was unable to find empirical instances of this s i t u a t i o n so I cannot o f f e r an adequate f i e l d assessment of the experimental findings. However, inasmuch as the experiment has had s i g n i f i c a n c e for other f i e l d results I think i t would be reasonable to suggest as a development strategy that some people w i l l go f o r new production factors that o f f e r marginal productive advantages i f (a) the calculated advantage i s l i k e l y to accrue; (b) the cost of the factor i s low; (c) the prospect of the factor f a i l i n g i s also low. (C)  The implications of C are r e a l l y the obverse of B.  Because  farmers are poor they cannot afford marginal investment returns as the prospect of loss i s too great a r i s k to their " s u r v i v a l " k i t . The cases c i t e d by Bauer and Schultz indicate that the maintenance of the status quo with regard to production factors i n Malaya, Panajachel and Senapur i s a function of i n h i b i t i n g costs for investment.  Wealth and Risk Proposition 2' related tolerance f o r r i s k d i r e c t l y to increases i n wealth.  E.M. Rogers i n his book "Diffusion of Innovations" (1962)  - 174 -  gives general support to this notion though he does not s p e c i f i c a l l y deal with r i s k . Wealth i s often associated with other predictors of adoption of a g r i c u l t u r a l practices l i k e simple a b i l i t y to afford the innovation, s p e c i a l i z a t i o n , mental a b i l i t y and access to the best s c i e n t i f i c i n f o r mation.  Though there may be a complex of intervening variables, wealth  i s a. good gross predictor of a tendency  to adopt innovations.  In a study of 108 Indian v i l l a g e s F l i e g e l , Roy, Sen and K i v l i n (1968) sought to determine the extent to which v i l l a g e s d i f f e r e d i n u t i l i z a t i o n of modern a g r i c u l t u r a l technology.  Taking the v i l l a g e  as their unit of analysis they focussed attention on innovations i n a g r i c u l t u r a l technology and constructed an index which measured the rate of adoption of a g r i c u l t u r a l innovations.  Taking this as a de-  pendent variable the authors correlated i t with a series of other factors as independent variables to assess the significance of these external constraints on adoption rate. As a measure of affluence, the proportions of families paying Rs. 10 or more i n v i l l a g e property taxes was  found to be p o s i t i v e l y  and s i g n i f i c a n t l y related to adoption rate (30 zero order;.24 1st order p a r t i a l ) (1968: 56).  The e f f e c t of this measure was  found to be inde-  pendent of other measures s i g n i f i c a n t l y related to adoption. This measure served as a rough guide not only to family affluence as v i l l a g e resources, but also to the a b i l i t y of v i l l a g e s to tax themselves.  With l i m i t e d state and federal resources, this a b i l i t y  may be c r u c i a l for successful development.  - 175 -  That v i l l a g e affluence i s related to adoption indicates a r e c i p r o c a l and reinforcing e f f e c t . cash i n v e s t m e n t s —  Modern a g r i c u l t u r a l inputs require  v i l l a g e s with r e l a t i v e l y high levels of affluence  have the economic resources either i n the hands of the farmer or a v a i l able from l o c a l people i n the form of loans, to invest i n modern inputs. The authors state on page 60 "Any increase i n levels of l i v ing  t  a s  measured by tax index] w i l l provide more opportunity to adopt  practices which may  require i n i t i a l investment or r i s k taking."  Information and Risk Proposition 3' related tolerance for r i s k to increments i n information. F l i e g e l et a l i a i n their study,while not e x p l i c i t l y  concerned  with information, correlated a number of factors with adoption rate that could be taken as i n d i c a t i v e of d i f f e r e n t levels of information. Their sample had three types of v i l l a g e s i n terms of time of exposure to development programs.  The stage III v i l l a g e s which had  had more than ten years of exposure to the community development program showed more a g r i c u l t u r a l innovation than the v i l l a g e s with less exposure.  The adoption index was  correlated with the stages measuring  period of development exposure and the data indicated an association which was s t a t i s t i c a l l y s i g n i f i c a n t at the 1 per cent l e v e l (0.25 correlation coefficient).  - 176 -  Adoption rate was  also found to be a function of the proximity  to a g r i c u l t u r a l and extension services. s i x i n s t i t u t i o n s -- V.L.W. HQ, Centre, Godown, Block HQ,  —  A proximity scale covering  Co-operative, Panchayat, Veterinary  was  s i g n i f i c a n t l y related to a g r i c u l t u r a l  adoption (correlation c o e f f i c i e n t of 0.28; level).  s i g n i f i c a n t at 1 per cent  I t i s reasonable to assume that proximity i s d i r e c t l y related  to the proportion of information that i s made available to farmers. The amount of contact each v i l l a g e had with the extension agent was highly correlated with adoption.  A combined index of four  measures of contact had a correlation c o e f f i c i e n t of 0.50 with rate.  adoption  Again I am operating on the assumption that increased contact  with an extension agent i s an indicator of increased information being made available to farmers. These three measures (a) length of exposure to development programs, (b) proximity to extension and co-operative and  headquarters  (c) contacts with the a g r i c u l t u r a l agent are a l l p o s i t i v e l y and  s i g n i f i c a n t l y related to the adoption rate of a g r i c u l t u r a l innovations. A l l these measures imply a function of information and d i s semination of information through various channels, and the data supports the notion that r i s k taking on innovations increases with information. In a previous chapter I v i s u a l i s e d access to information c i r c u i t s as a function of wealth and education. i t i s not possible to assess adequately  From the Indian data  the combination  of these fac-  tors with regard to information, but i n a round-about way the significance of education.  one can show  - 177 -  The authors found that v i l l a g e s with a high proportion of l i t e r a t e s tended to score high on the measure of a g r i c u l t u r a l adoption, and show that l i t e r a c y i s part of a complex of "modernization" variables associated with development. The l i n k between l i t e r a c y and information i s a f a i r l y obvious one —  I am assuming that by being l i t e r a t e a farmer has access to more  information channels than i f he was n o n - l i t e r a t e . Utility  Considerations The preceding  assertions; I now  sections indicate empirical support for my  propose to examine several contrary cases to see i f  i n fact my propositions are refuted. Recall that my u t i l i t y parameter.  remarks about risk-taking were bounded by a  I stated that my  concern lay with decision making  i n a s i t u a t i o n where a high u t i l i t y was that outcome may  held for the outcome, whatever  be.  In the empirical world of contrary cases one can expect  two  things: (a)  the high u t i l i t y s t i p u l a t i o n may  not be present i n the decision  s i t u a t i o n or (b)  the high u t i l i t y s t i p u l a t i o n may  be present i n the decision s i t u -  ation, but be i n c o n f l i c t with another high u t i l i t y outcome which takes p r i o r i t y . The following case h i s t o r i e s are examples of the situations described above.  - 178 -  Moturiki Community Development Scheme  Moturiki Island i n the F i j i group was  chosen as the locale  for a community development project i n 1948 by the F i j i a n government (Hayden 1954).  The scheme saw the introduction of a copra marketing  board, training of Moturikans at a l o c a l a g r i c u l t u r a l s t a t i o n , the planting of coconut nurseries and hardwoods, the establishment of an a g r i c u l t u r a l school for boys with piggeries and poultry. New  cash crops such as r i c e and dalo were introduced and  markets for copra, timber and cash crops were assured. development team was  A community  resident on the island for a year and i n i t i a t e d  these projects. Two years after the team's withdrawal from the community the planting of cash crops had ceased, the clearing of coconuts and care of the nursery had been abandoned, the nursery and reforestation schemes no longer functioned and nobody attended the a g r i c u l t u r a l school. The scheme had had the advantages of ready markets f o r cash crops and products and of providing Moturiki with considerable i n creases i n a g r i c u l t u r a l productivity —  yet these opportunities were  not taken advantage of. The major reason can be found i n Hayden's own words " L i f e was pleasant and leisured on Moturiki —  food was not a problem" (1954: 6).  Thus one could not expect the islanders to increase their productivity i n food at the expense of l e i s u r e when they had enough for their needs.  -  E x t r a food was a g r i c u l t u r a l schemes. change, i t was  179  the way  As  -  i n which they regarded  there was  more than they  the  extensive  c o u l d eat and  ex-  o b v i o u s l y not needed, so the p r o j e c t s were abandoned.  The Moturikans  e x h i b i t e d a low u t i l i t y  a g r i c u l t u r a l p r o d u c t i v i t y , and  for increasing their  the whole scheme seemed to be more a  f u n c t i o n of the community development team's i d e a of what was r a t h e r than as a response  F o o d - F i s h i n g by  required,  to the i s l a n d e r s wants o r needs.  Indian G i l l n e t t e r s  In the v i l l a g e of A l e r t Bay,  from A l e r t  Bay  where I based my  fieldwork, a  remarkable d i f f e r e n c e i n f i s h i n g a t t i t u d e s o c c u r r e d when I n d i a n ermen went f o o d - f i s h i n g . with nets provided  By  law  t h a t the f i s h  they were a l l o w e d  to f i s h  the  caught were used f o r domestic  fishrivers consump-  tion. Here was for  a s i t u a t i o n where no one  exhibited a high  i n c r e a s i n g f i s h i n t a k e , and nobody v e n t u r e d  i t was  not very important  to take h i g h r i s k s  along on one f o o d - f i s h i n g  to the Nimpkish R i v e r , o p p o s i t e A l e r t Bay  on Vancouver I s l a n d .  Of the s i x s e t s he made, he never once took a h i g h r i s k . the f a c t t h a t more f i s h shallowness He  fished  c o u l d be had  of the r i v e r ,  l i k e a pleasant afternoon's  trout  He  acknowledged  c l o s e to the r i v e r banks but  and number of t r e e s i n the water d e c i d e d him  the middle  as  to get a l a r g e c a t c h .  Sonny Commano on the S l a v e took me trip  utility  and  against i t .  the whole a f f a i r was  fishing.  the  very much  - 180  -  I gathered from Sonny's statements, that food-fishing did not carry the same sense of urgency to get b i g catches as did commercial fishing.  There was no way  to s e l l the f i s h that was  caught, and even  with canning and s a l t i n g there was usually more than enough to consume. This surprised me because of Sonny's reputation as a commercial fisherman, as he was noted for the high f i s h i n g risks he took. But i n the food-fishing s i t u a t i o n he was operating with a d i f f e r e n t preference rating for the outcome —  increase f i s h intake^  He stated that i t was not important to him how much f i s h he caught, as long as he got some.  Discussions I had with Arthur Dick  and B i l l Scow confirmed this attitude towards food-fishing.  Nobody  took high r i s k s simply because i t was not that important to p u l l i n big  catches, as there was more than enough for domestic  consumption.  The Moturiki community development scheme and the foodf i s h i n g s i t u a t i o n had the necessary preconditions v i s - a - v i s resources and information to expect that high risks would be taken.  But i n ac-  tual fact the Moturikans turned down the prospect of ready markets and the Indian fishermen did not attempt to maximise f i s h intake. This would seem to contradict my assertions about r i s k taking. However, neither s i t u a t i o n met the required scope conditions with r e f erence to the s p e c i f i e d u t i l i t y parameter  (both instances were charac-  terised by a low u t i l i t y for increasing productivity) and are thus outside the domain of my t h e o r e t i c a l statements. rebuttal to my assertions.  They do not provide a  - 181 -  Spanish American Farmers and H y b r i d  Spanish American farmers  Corn  o f E l Cerrito i n the R i o Grande V a l l e y  of New Mexico grew c o r n as t h e i r major crop. pared of  to the h a r v e s t s o f midwestern farmers,  T h e i r y i e l d s were low comand the g e n e r a l  quality  the corn was low. In 1946 the county  e x t e n s i o n agent succeeded i n i n t r o d u c i n g  h y b r i d c o r n which gave t h r e e times The  farmers  lities half  the y i e l d o f t h a t grown t r a d i t i o n a l l y .  r e c o g n i s e d the need f o r b e t t e r p r o d u c t i o n and once the a b i -  o f the new corn were i n d i c a t e d by means o f a demonstration  the farmers  i n the community p l a n t e d the h y b r i d and each a t l e a s t  doubled h i s p r o d u c t i o n the f i r s t y e a r . q u a r t e r s o f the farmers produced h i g h y i e l d s individual  ceased  I n the f o l l o w i n g y e a r  i n the v i l l a g e had accepted  the new seed.  It  farmers. i n t r o d u c t i o n a l l the farmers  to p l a n t the h y b r i d and were u s i n g t h e i r o l d c o r n .  Table XXVII: A d o p t i o n  of H y b r i d Corn by Spanish American Farmers  Year  No. o f Farmers Participating  No. o f Farmers  1946 1947 1948  40 60 30  84 84 84  1949 1950  three-  and was w e l l w i t h i n the f i n a n c i a l means o f the  Y e t f o u r y e a r s a f t e r the f i r s t had  plot,  3 0  84 84  Yield D  0 U B L E D  - 182 -  The agent's follow up enquiry  into the eventual r e j e c t i o n of  the hybrid established that the farmers s t i l l f e l t the need for better y i e l d s , and that they wanted to increase their productivity i n order to market surplus corn (Apodaca 1952: 38). The farmers can be quite d e f i n i t e l y categorised as HU for increasing a g r i c u l t u r a l productivity.  But what was at stake was a  u t i l i t y for corn quality which had a p r i o r i t y over corn quantity. The agent eventually discovered  that the farmers and their  wives were unhappy about the way i n which the new corn made t o r t i l l a s —  which was the basic staple i n the community's d i e t .  taste were considered i n f e r i o r to the t r a d i t i o n a l corn.  The texture and Also  had a complex of symbols associated with their manufacture —  tortillas a woman's  housecraft, s o c i a l offerings and pride were at stake, as well as the necessity of eating t o r t i l l a s at every meal. Thus one can view the r e j e c t i o n of the hybrid corn not as a function of the farmers having a low u t i l i t y for increasing product i v i t y ; but as a function of their having a higher u t i l i t y for corn quality and the t o r t i l l a complex.  Some Implications  The examples I have examined indicate that a basic pattern of decision making can be expected i n d i f f e r e n t s o c i o - c u l t u r a l settings. When I examined material to support the notion that LWHU farmers take low r i s k s , I found this consideration to hold across a variety of s o c i a l structures —  v i z . Panajachel - Senapur - Malaya.  -  183 -  "Senapur, for instance, has a long-established caste system, whereas Panajachel has singular f l e x i b i l i t y i n the movement of families up and down i t s s o c i a l status scale. In Senapur the families of the p r i v i l e g e d castes, mainly the Thakur, have perpetuated their wealth, p r i v i l e g e s and s o c i a l status for many generations. In the Guatemalan community, Tax found marriages cutting across wealth lines and much mobility on the economic and s o c i a l scale; moreover the s o c i a l and economic gap separating the top and the other families has not been substantial." (Schultz 1964: 44, footnote 8). Both of these communities d i f f e r r a d i c a l l y from the Malayan example, yet risk-taking was observed  to be s i m i l a r i n a l l three cases.  More evidence to support the idea of basic patterns of decision making can be obtained from F l i e g e l et a l i a ' s study (1968).  They  used discrete sets of variables to assess the difference that agrarian structure, occupational d i v e r s i f i c a t i o n , caste structure, severity of v i l l a g e faction disputes, and r e l i g i o u s d i v e r s i t y had upon the adoption rate of a g r i c u l t u r a l p r a c t i c e s . Their hypotheses regarding the influence of agrarian structure on adoption of farm practices were not supported by the data. I t was found that neither population pressure on available land nor the extent of dominance i n land ownership by a few landlords ware related to the adoption  index.  With regard to occupational d i v e r s i f i c a t i o n i t was hypothesised that the predominance of agriculture would be associated with r e l a t i v e l y high levels of adoption of farm p r a c t i c e s . This was not supported. Similarly the hypotheses regarding caste structure were not supported.  I t was hypothesized that v i l l a g e s dominated by a s i n g l e  caste would rank high i n adoption of farm practices.  The rationale for  -  184  -  this hypothesis rests on the fact that dominant castes tend to be a g r i c u l t u r a l castes i n a g r i c u l t u r a l v i l l a g e s and that this should maximize s k i l l inputs into agriculture (1968: 65). The second variable concerning caste i s a ranking of the r i t ual status of the most numerous caste.  The hypothesis was  that v i l l a g e s  with numerically dominant castes of high status w i l l be high i n adoption of farm practices.  This contention was not supported by the data c o l l e c t e d .  The severity of v i l l a g e faction disputes was  used to measure  horizontal d i f f e r e n t i a t i o n i n v i l l a g e structure, and the hypothesis that factionalism would be negatively associated with adoption. was not supported.  Neither was  was  This  the consideration that r e l i g i o u s diver-  s i t y would have a negative e f f e c t on adoption. These findings lead the authors to conclude that s t r u c t u r a l v a r i a t i o n did not make any s i g n i f i c a n t difference to the adoption rate (1968: 69). That this can be said of India, which has long been t y p i f i e d as having every s o c i o - c u l t u r a l obstacle to development imaginable, gives some encouragement to the hard-pressed  development-planner.  These considerations have a number of implications for the way  i n which one may  view s o c i a l structure and culture from the stand-  point of analysing decision making. F i r s t of a l l structure i s not a given, an entity that exists through time.  I t i s a h e u r i s t i c device used by an analyst i n order to  solve d i f f e r e n t types of problems, (hence the many d i f f e r e n t d e f i n i t i o n s of s o c i a l s t r u c t u r e ! ) .  - 185 -  S o c i a l s t r u c t u r e can o n l y be r e l e v a n t o r s i g n i f i c a n t when i t p r o v i d e s access to r e s o u r c e s t h a t are v a l u e d . i n s t r i c t l y m a t e r i a l terms —  access  Resources can be viewed  to l a n d and markets, e t c . —  the w i d e r economic^meaning covers a l l b e h a v i o r . r e s o u r c e can o n l y be o b t a i n e d through  but  I f p r e s t i g e as a v a l u e d  a p a r t i c u l a r structure (kinship,  age o r g a n i s a t i o n ) then t h a t s t r u c t u r e i s important  to the a c t o r making  d e c i s i o n s about the r e s o u r c e p r e s t i g e . In  terms of farmers  a d o p t i n g a g r i c u l t u r a l p r a c t i c e s , i n the  v a s t m a j o r i t y of cases s o c i a l s t r u c t u r e i s not the avenue f o r access to  resources —  i n f o r m a t i o n and  technology  —  concerned w i t h  agricultural  productivity. In  I n d i a t r a d i t i o n a l s t r u c t u r e s such as c a s t e and  the p a t r o n -  c l i e n t networks are no l o n g e r the e x c l u s i v e channels f o r r e s o u r c e s sociated with a g r i c u l t u r a l productivity. and  Resources such as  i n f o r m a t i o n a r e a v a i l a b l e from sources o u t s i d e of the  networks and w h i l e the l a t t e r may  s t i l l be important  as-  technology  traditional  for certain prestige  c o n s i d e r a t i o n s , they a r e no l o n g e r the s t r u c t u r e used to g a i n access to  the r e s o u r c e s needed f o r a g r i c u l t u r a l p r o d u c t i o n . The p o i n t I would l i k e  to emphasize i s t h a t any  structure,  be i t t r a d i t i o n a l o r modern, o n l y becomes important when the r e s o u r c e s it  can p r o v i d e and d i s t r i b u t e are v a l u e d as  necessary.  T r a d i t i o n a l group norms and s t r u c t u r e s become s i g n i f i c a n t when group, a f f i l i a t i o n labour,prestige, r i t u a l  c a r r i e s w i t h i t access  to a r e s o u r c e  (land,  s t a t u s , e t c . ) t h a t an i n d i v i d u a l v a l u e s .  This  -  186  -  being the case, i t i s unlikely that an i n d i v i d u a l w i l l r i s k his access by f l o u t i n g group norms.  Given his high u t i l i t y for a resource  he i s constrained to play the game according to whatever rules and constraints are implied i n a p a r t i c u l a r access  route.  As soon as an i n d i v i d u a l obtains access to the resource he values outside of any group a f f i l i a t i o n s , then he i s less constrained to observe group sanctions, as he does not need the group to gain access to the resource he wants. This perspective i s , I submit, generalisable to a l l resources and a l l structures. Complementary to this view of structure, one can place c u l ture —  the general scapegoat for f a i l e d development plans —  s i m i l a r perspective.  in a  I would submit that c u l t u r a l values provide the  parameters within which individuals make decisions; and c u l t u r a l norms are established not through a mystical genesis but by regular patterns of aggregate decision making.  As the pattern of decision making  changes at the aggregate l e v e l , so does the norm. This view of s o c i a l structure and culture should free us from discussing them as "obstacles" or "drawbacks" to innovation and development.  They are part of the antecedent conditions to decision  making, and once one can specify the implications of these conditions then one can predict the type of decisions that w i l l be made.  - 187 -  CHAPTER 6 CONCLUSION  Introduction My i n i t i a l considerations focussed on the problem of farmers making decisions about a g r i c u l t u r a l innovations, then I progressed  to  a higher l e v e l of generality wherein I referred to individuals taking risks.  The testing procedure I used involved both laboratory and  settings, while corroborative evidence was  field  supplied from secondary  sources. From this progression I intend to draw three types of i m p l i cations —  methodological,  t h e o r e t i c a l and p r a c t i c a l , and i n doing so  intend to i l l u s t r a t e both the strengths and the weaknesses i n the approach I have used. cuss processes  Under "Methodological  Implications" I w i l l d i s -  of enquiry and types of models that can be used by eco-  nomic anthropologists.  In addition I propose a "paradigm" for research  enquiry i n order to i l l u s t r a t e the progression of the work attempted here and the required r e v i s i o n s . "Theoretical Implications" considers the way basic subject matter of anthropologists — —  may  i n which the  s o c i a l and c u l t u r a l  behavior  p r o f i t a b l y be viewed, while under " P r a c t i c a l Implications" I  w i l l be concerned with possible p o l i c y considerations suggested by this research.  - 188 -  Methodological  Implications  In his inaugural lecture at the College de France i n 1960, Levi-Strauss i n addressing  the problem of kinship analysis indicated  the i m p o s s i b i l i t y of varying at w i l l the complex relationships i n volved or apparent i n the f i e l d s i t u a t i o n (Levi-Strauss 1966). He does, however, maintain that i t i s possible to construct models of the systems we are interested i n , i n the laboratory, to determine how they would function when the basic relationships are transformed and manipulated to f i l l out every l o g i c a l p o s s i b i l i t y . Thus we would obtain two series of phenomena — results and f i e l d examples.  laboratory  The process then becomes one of succes-  sive t r i p s from the laboratory to the f i e l d , f i e l d to the laboratory to find a f i t . But as i n so many of his endeavours Levi-Strauss for a p a r t i c u l a r procedure are rarely made e x p l i c i t .  reasons  He does not  expound on the nature of models, nor on this process as a testing device f o r theory. However, the exercise he proposes seems meaningful mainly i n terms of theory construction and a precedence for laboratory work.  With  reference to decision theory considerations, i n the laboratory one can set  up and control a series of l o g i c a l p o s s i b i l i t i e s for individuals  taking risks which would allow one to systematically manipulate a set of incentive conditions and a set of external constraints with respect to a given decision task. type of model-building  This arrangement provides  the basis f o r the  that Levi-Strauss believes i s important.  - 189 -  I maintained (in Chapter 2) that the laboratory and f i e l d are p a r t i c u l a r research settings i n which my assertions about r i s k - t a k i n g behavior could be tested and compared.  This assumption was conditional  upon the same scope conditions being met i n each s e t t i n g .  The scope  conditions placed an i n d i v i d u a l decision-maker within parameters of wealth and subjective u t i l i t y for some outcome, information and incentive conditions for any r i s k ; and my propositions predicted the type of decision strategy used. In the laboratory and the f i e l d settings these conditions were met, but where I anticipate objections i s with respect to the operationalisation of variables such as wealth and information into low and high categories. Can a $1 wealth category be s t r i c t l y sure of low wealth for A l e r t Bay fishermen?  comparable to my mea-  Not d i r e c t l y , but I am  not going to argue that exactly the same scope conditions applied i n the two situations with regard to measurement of parameters.  I  w i l l argue that the wealth and information gradients r e l a t i v e to the f i e l d and laboratory s i t u a t i o n are comparable as each was calculated from a baseline of the lowest category having the property of not being able to afford a loss i n resources. The t h e o r e t i c a l propositions are translated into d i f f e r e n t substantive settings v i a d i f f e r e n t operational d e f i n i t i o n s , because I am not concerned with wealth or information per se but with of these variables r e l a t i v e to p a r t i c u l a r contexts.  gradients  - 190 -  Where there i s a wealth gradient, information gradient and a decision involving r i s k —  r e l a t i v e to a p a r t i c u l a r s i t u a t i o n —  these  are the scope conditions I specify must be present i n a l l cases. The methodology I have espoused provides necessary  links  between farmers accepting innovations, fishermen deciding where to f i s h and sophomore students gambling.  I am trying to say something general  about the sorts of risks people w i l l take when they value an end highly. I am i n fact dealing with a means (decision) ends (goal) relationship. The end can vary —  maximise productivity, increase r i t u a l  cleanliness, or promote family s o l i d a r i t y .  What I am submitting i s  that any action decided upon towards that end can be analysed i n terms of incentive conditions, information processing and u t i l i t y preferences. The c a l c u l a t i o n of the incentives, both p o s i t i v e and negative can take place on one or more dimensions —  but that this c a l c u l a t i o n  i s made i s the major assumption of my thesis.  On Models and Research Cycles  Anthropologists have long argued among themselves and with economists regarding the range of a p p l i c a b i l i t y of economic theory. I do not propose to go any further into the controversy as I stated e a r l i e r (Chapter 1) that I view economic theory as a set of axioms which have standard interpretations i n a market context, but which are uninterpreted i n a non-market context.  - 191 -  This implies that the " r e a l " question i s whether or not empirical interpretations can be given to the axioms of economic theory so that they cover a domain wider than that of the market, and i s a plea for getting on with the job of constructing and testing models instead of talking about putative domains of  relevance.  The progression of the work attempted here has demonstrated the usefulness of borrowing analogies from economics.  Using a decision  theory format and modifying i t to include s i t u a t i o n a l constraints I have shown that a s i m i l a r type of decision making occurs i n d i f f e r e n t substantive  contexts.  The model I b u i l t was to give one confidence  not a perfect f i t , but adequate enough  i n the methodological assumptions that were fun-  damental to i t s construction.  The lack of congruence i n testing pro-  cedures meant that i n some cases i t was propositions i n a l l contexts. s i t u a t i o n a l contexts  not possible to test a l l the  However this i s a function of adverse  and inadequate operationalisation of variables  and does not cast doubt on the methodological assumptions used to construct the model. Perhaps the best way  to assess the merits of the approach  attempted here i s to suggest an i d e a l research paradigm and measure my work against t h i s .  In addition to the implications this may  have  for enquiry i n economic anthropology i t should suggest necessary revisions as far as the model attempted here i s concerned.  - 192 -  I think that much of the debate i n economic  anthropology  could be rendered superfluous i f i t were possible to get the proponents i n both camps to adhere to a research paradigm that has served science extremely well (Greer:  1969).  This paradigm can be characterised as a cycle having  five  d i s t i n c t steps: (1)  select a p a r t i c u l a r problem; without how  (2)  hard he t r i e s —  a problem no one —  can graduate beyond imprecise g e n e r a l i t i e s .  the problem i s conceptualised i n such a way solution.  no matter  as to suggest a t r i a l  A model i s constructed which places the problem within  a larger frame of reference. (3)  the t r i a l s o l u t i o n i s tested, l o g i c a l l y and empirically.  (4)  the t e s t t r e s u l t s are evaluated against what was  expected, which  leads to (5)  the extension and editing of the model, and the generation of  new  problems. The problem I was  concerned with was  i n terms of individuals  making decisions under conditions of uncertainty and l o s s . a model of risk-taking which predicted the way  I constructed  individuals would make  decisions subject to constraints of information, wealth and subjective u t i l i t y for the outcome the decision i s about. The laboratory test indicated that my statements were a p p l i cable to a domain narrower than I had expected.  - 193 -  (1)  I t indicated that increments i n information — odds and payoffs were s p e c i f i e d —  after the range of  had no s i g n i f i c a n t influence  upon decision-making. (2)  I t also indicated that some individuals from low wealth categories would take risks for low payoffs when the prospect and amount of loss were low, and the prospect of the payoff was high.  (I had  asserted that this category of i n d i v i d u a l would not go f o r low payoffs under any circumstances.) On the credit side, however i t showed unequivocally that low wealth individuals did have a d e f i n i t e preference f o r low r i s k strategies, as the exceptions i n (2) above s t i l l indicated a preference f o r low r i s k s .  Those individuals'who opted f o r marginal payoffs  only did so when the remaining incentive conditions were strongly i n their favour.  In other words low payoff bets were taken by some low  wealth individuals when i t was highly l i k e l y that they would be successf u l (high P ) and when the cost factors were n e g l i g i b l e (low P , low $ ). w ± 1 Increments i n wealth from low to medium were accompanied by s i g n i f i c a n t increases i n high r i s k taking,though high wealth individuals did not take s i g n i f i c a n t l y d i f f e r e n t types of r i s k to medium wealth individuals.  I think, however this result may be due rather to an i n -  adequate d i s t i n c t i o n between upper wealth levels than to a uniform pattern of r i s k taking across medium and high wealth l e v e l s . The f i e l d test suggested that further constraints on my model were necessary, as i t did not adequately predict the investment risks  - 194 -  taken by a group of Japanese and Yugoslav g i l l n e t t e r s .  The  insertion  of the supplementary proposition with respect to s o c i a l marginality and r i s k taking accounted for the higher than expected l e v e l of r i s k taking. In empirical terms, one of the l o g i c a l p o s s i b i l i t i e s i n the laboratory — t h a t  of high information —  did not occur, so I was  forced to deal with only two levels of information i n terms of strategy risks.  Also the assumptions of r e l a t i v e weighting applied to the  wealth and education parameters, which were fundamental to the i n f o r mation c i r c u i t s model, were not supported i n some cases. Despite  these constraints the sample of g i l l n e t t e r s from  A l e r t Bay indicated general support for my  assertions about r i s k taking  behavior i n terms of the predicted r e l a t i o n between r i s k taking, u t i l i t i e s , information and wealth. The evidence marshalled from secondary sources was  of a gen-  e r a l corroborative nature and two of the contrary cases were shown to l i e outside my specified hybrid corn i n New  scope conditions.  Mexico —  However the third case  —  i l l u s t r a t e d the necessity of considering  the constraints imposed upon decision making by c o n f l i c t i n g  utilities.  The New  exhibited  Mexico example showed that while a high u t i l i t y was  for increasing corn y i e l d , there was  a higher u t i l i t y attached  quality which eventually led to the rejection  to corn  of the corn hybrid.  suggests that rankings of u t i l i t i e s associated with p a r t i c u l a r  This  types  of outcomes w i l l have to be made i n order to specify the dominance of one u t i l i t y over another.  - 195 -  Editing Editing of the model can be done i n two ways.  F i r s t l y , as  a r e s u l t of the tests, one can s t i p u l a t e more precisely the range of a p p l i c a b i l i t y of the model; and secondly one can o f f e r revisions of the basic statements. Proposition 1':  "Individuals w i l l show a preference  for low r i s k s t r a -  tegies with respect to decisions involving the outcome to which HU i s attached." Conditions:  LWHU.  The experimental test showed that the LW/H.Inf category did not show the anticipated preference position 1' predicted. empirical counterpart  for low r i s k strategies that pro-  However the fact that I was  unable to find an  to the LW/H.Inf p o s s i b i l i t y suggests (a) that  i t ;is not to be found i n the r e a l world and  (b) that i t s nonoccurrence  i s a function of the effects of intervening variables (such as education) which prevents the configuration from occuring. In fact the information c i r c u i t s model used the parameter of education  i n such a way  to high information.  that no low wealth i n d i v i d u a l ever had access  As this part of the model was  strongly supported  by the f i e l d test, i t seems reasonable to s t i p u l a t e that the experimental result i s not a rebuttal to the proposition, as i t deals with a l o g i c a l p o s s i b i l i t y which has no empirical referent. The f i e l d test strongly supported proposition 1'.  Under  conditions of low information low wealth fishermen took 44% more low  - 196 -  risks than medium wealth fishermen, while high information produced a 54.3% difference i n the proportion  of low risks taken by low wealth  as compared to medium wealth fishermen.  These figures show conclusively  that low r i s k strategies are preferred by low wealth fishermen. The discussion of the A g r i c u l t u r a l Revolution i n Asia was i n terms of peasant farmers being offered high y i e l d , low cost production factors.  The Director General of F.A.O., A.H. Boerma  stated i n the 1969 annual report that the food problem facing the world i n the near future i s more l i k e l y to be surpluses than starvation (FAO 1970) .  The underdeveloped  areas of the world increased their a g r i c u l -  t u r a l production by 5% i n 1967 and 2% i n 1968 with estimates f o r 1969 showing a continuation of the trend. This s t a r t l i n g change from F.A.O.'s prophesies of doom and famine (as recently as 1967)has been effected by breakthroughs i n a g r i c u l t u r a l technology, especially i n new strains of r i c e and grains. As i t would be naive to expect peasant farmers to have changed d r a s t i c a l l y i n three years, I would submit that they are operating on the same strategy that they have been forced to use f o r centuries — low r i s k . Development agencies have generally accepted the a g r i c u l t u r a l revolution as a boon and a w i n d f a l l ; and very l i t t l e has been done i n analysing why i t happened i n the late 1960's and not before.  I would  submit that i t was only i n the late 1960's that development planners could o f f e r low r i s k production factors to peasant farmers.  - 197 -  In India and the P h i l i p p i n e s , hybrid grains and r i c e have been adopted wholesale simply because a farmer does not have to take a high r i s k to invest i n them.  The opportunity to invest i n these  new  crops constitutes a low r i s k strategy for the farmer, as his costs are minimised through subsidies, and the incentive conditions of high yields reassures him, at the very worst, of a harvest bigger than that which can be obtained by using t r a d i t i o n a l  varieties.  Thus proposition 1' can be accepted as v a l i d . Proposition 1':  "Individuals w i l l show a preference for low r i s k s t r a tegies with respect to decisions involving the outcome to which HU i s attached." Conditions:  Proposition 2':  LWHU.  "Individuals w i l l show an increased tolerance for high r i s k strategies with respect to the outcome to which HU i s attached, as wealth increases".  The experiment indicated that s t a t i s t i c a l l y s i g n i f i c a n t i n creases i n r i s k taking did not occur between medium wealth and high wealth categories, (though I suggested that this could be attributed to an inadequate d i s t i n c t i o n between the upper wealth l e v e l s ) .  Sig-  n i f i c a n t increases i n high r i s k taking were observed between low  and  medium wealth l e v e l s . In the f i e l d , fishermen exhibited an increased tolerance for high r i s k strategies as wealth increased, though comparison was between medium wealth and high wealth categories as I only had  difficult two  - 198 -  fishermen i n the l a t t e r category.  However the fact that these two  fishermen never took low risks while medium wealth operators did, gives some support to the o v e r a l l contention of the proposition. The evidence from secondary sources indicated that wealth operates as a general predictor f o r r i s k taking.  Bauer (1946) found  that r i s k s taken on replanting and newplanting of rubber trees i n Malaya were d i r e c t l y related to the wealth levels of the growers concerned. F l i e g e l et a l i a (1968) showed that the rate of adoption of a g r i c u l t u r a l technology i n Indian v i l l a g e s was p o s i t i v e l y and s i g n i f i c a n t l y related to measures of affluence. This general support for proposition 2', however, even when taken i n conjunction with the fact that high wealth fishermen took more high risks than medium wealth fishermen does not permit one to say with confidence that r i s k taking increases beyond the medium l e v e l . while the f i e l d test and evidence from secondary sources does not contradict the proposition, too much weight should not be attributed to a sample of two and evidence of a general nature. So, while I tend to think that the experimental results may be a function of inadequate d i s t i n c t i o n s between the upper wealth l e v e l s , there i s i n s u f f i c i e n t evidence from other sources to show that the pattern of wealth v i s - a - v i s r i s k taking suggested by the experiment could be disregarded. These considerations lead to a revised proposition: Proposition 2":  "Individuals w i l l show an increased tolerance for high r i s k strategies, with respect to the outcome to which HU i s attached as wealth increases from low to medium." Conditions:  HU  - 199 -  Proposition 3':  "Individuals w i l l show an increased tolerance for high r i s k s t r a t e g i e s , with respect to the outcome to which HU i s attached, as information increases."  The most interesting result from the experiment was the threshold point of information v i s - a - v i s r i s k taking.  Once the range  of odds, payoffs and costs had been s p e c i f i e d (Medium Information i n the experiment) further increments i n information did not produce s i g n i f i c a n t changes i n patterns of r i s k taking.  This implies that a l -  though uncertainty i s an important part of r i s k taking, a f t e r a c e r t a i n point has been reached uncertainty can continue to be lowered with no resulting change i n decision making. It could be argued that again the experimental  distinctions  between the upper information levels were weak, but as I pointed out e a r l i e r (page  84 ) there i s evidence  from other experimental  results  to indicate that high information i s d i s t i n c t l y d i f f e r e n t to medium information. The difference can be attributed to the fact that high i n f o r mation affects low wealth subjects so that they show?.-: s i m i l a r kinds of preferences for low risks,as do MW/H.Inf subjects, while medium information operated on low wealth subjects to produce a s i g n i f i c a n t l y d i f f e r e n t pattern of r i s k taking to that exhibited by MW/M.Inf subjects If the experimental  d i s t i n c t i o n was inadequate then there would be no  difference i n the effect that high information and medium information h on low and medium wealth l e v e l s .  Thus i t i s reasonable  to accept the  threshold effect of information increments upon r i s k taking as v a l i d .  - 200 -  Unfortunately the f i e l d test d i d not provide me with a high information category analoguous to the experimental one. With regard to decisions on f i s h i n g strategies there was no e f f e c t i v e way that i n formation could be increased f o r the fisherman beyond specifying the range of odds, payoffs and costs for any a l t e r n a t i v e . Thus I was unable to see whether increments i n information a f t e r the threshold point, i d e n t i f i e d i n the laboratory, made any difference to decision strategies employed. Information i n terms of investment risks on technology was concerned with measuring the number of information c i r c u i t s an i n d i vidual had access to, and did not control f o r cut o f f points i n s p e c i fying the four r i s k f a c t o r s . As i n the previous proposition, the evidence from secondary sources was of a general rather than a s p e c i f i c nature, which means that the experimental  result with regard to information and r i s k tak-  ing can be used to revise proposition 3' as follows: Proposition 3": "Individuals w i l l show an increased tolerance for high r i s k s t r a t e g i e s , with respect to the outcome to which HU i s attached, as information increases to the point where the range of r i s k factors i s s p e c i f i e d . " Conditions:  HU.  - 201 -  Intervening Variables  (1) E d u c a t i o n  I s t a t e d e a r l i e r t h a t there had to be some i n t e r v e n i n g v a r i a b l e which p r e v e n t e d  the LW/H.Inf c o n f i g u r a t i o n from o c c u r r i n g and  p o i n t e d o u t t h a t the i n f o r m a t i o n c i r c u i t s model v i a the parameter o f e d u c a t i o n was the o f f e n d i n g v a r i a b l e . T h i s model v i s u a l i s e d technology  access  to i n f o r m a t i o n c i r c u i t s about  as a f u n c t i o n o f w e a l t h and e d u c a t i o n ,  on technology  and t h a t r i s k  was a f u n c t i o n o f i n f o r m a t i o n c a t e g o r y .  r e l a t i v e weighting  of wealth v i s - a - v i s  graded sequence o f c i r c u i t s .  education  taking  I assumed a  i n order  to g e t a  T h i s assumption was n o t always:borne  out by t h e f a c t s . For i n s t a n c e i n the f i e l d  test  the MW/MEd category  ends up  on c i r c u i t e which p r e d i c t s medium i n f o r m a t i o n .  Of the seven  i n t h i s category  t h r e e had medium  f o u r had h i g h i n f o r m a t i o n w h i l e  fishermen  i n f o r m a t i o n as p r e d i c t e d . The  t h r e e fishermen  i n the MW/LEd category  a l l had medium  i n f o r m a t i o n i n s t e a d o f low i n f o r m a t i o n as p r e d i c t e d . The  only other  " d e v i a n t " from the model was a LW/MEd  fisher-  man who had medium i n f o r m a t i o n i n s t e a d o f t h e low t h a t was p r e d i c t e d . But as f o u r o t h e r fishermen  i n this  category  mation, I do not f e e l t h a t a r e v i s i o n  had the p r e d i c t e d  i s called  infor-  f o r i n this section  of the model. These c o n s i d e r a t i o n s suggest a new o r d e r i n g o f c i r c u i t s w i t h regard  to i n f o r m a t i o n category  as below:  - 202 -  From e we ean s t i p u l a t e that there i s a 4 i n 7 chance (approximately p=0.6) of the i n d i v i d u a l having access to high information, and a 3 i n 7 chance (approximately p=.04) of an i n d i v i d u a l having access to medium information. Using the same type of p r o b a b i l i t y statements,  the medium  r i s k taking category can be eliminated (I only measured low and high risks i n the f i e l d ) and we can state that for any i n d i v i d u a l with medium access to information c i r c u i t s there i s a 0.25 chance that he w i l l take high risks and a 0.75 chance that he w i l l take low r i s k s .  This  i s based on the fact that of eight medium information fishermen s i x took low investment risks while only two took high r i s k s .  7  - 203 -  Intervening Variables (2) Social Marginality  During fieldwork evidence was  collected that supported  the  supplementary proposition r e l a t i n g considerations of s o c i a l marginality to r i s k taking.  I observed that Japanese and Yugoslav g i l l n e t t e r s took  far higher investment risks then one would expect, and that their i n vestment i n technology was  considerably more than the A l e r t Bay  collec-  t i v i t y of high r i s k investors. Although I only collected information on f i v e Japanese boats and three Yugoslav boats, for the skippers concerned I was establish that their ethnic separation was  able to  s o c i a l l y reinforced by  higher frequencies of i n t e r a c t i o n with people of the same ethnic group then with outsiders. This and other considerations of language and residence seemed to meet the conditions of s o c i a l marginality s p e c i f i e d i n the section on supplementary propositions.  The s i z e of the sample i s too small to  draw d e f i n i t e conclusions, but on the results of my observations I would l i k e to o f f e r an a d d i t i o n a l proposition, though of a tentative nature. Proposition 4':  "The proportion of individuals p r e f e r r i n g high r i s k strategies, with respect to an outcome to which HU i s attached,  w i l l be higher i n s o c i a l l y marginal groups  than i n non-marginal groups; a l l other things being equal."  - 204 -  Scope C o n d i t i o n s :  The  The  Utility  Parameter  c o n d i t i o n s s p e c i f i e d as b e i n g  r e l e v a n t to my t h e o r e t i c a l  c o n s i d e r a t i o n s p l a c e d an i n d i v i d u a l d e c i s i o n maker w i t h i n parameters of w e a l t h and  subjective u t i l i t y  i n c e n t i v e c o n d i t i o n s f o r any  f o r some outcome, i n f o r m a t i o n  risk.  In Chapter 5 the examination of c o n t r a r y on the whole they l a y o u t s i d e p a r t i c u l a r they d i d not one  low  met,  cost production T h i s was  cases showed t h a t  the scope c o n d i t i o n s of my  r e f e r to the s p e c i f i e d HU  case ( t h a t of h y b r i d corn i n New  HU parameter was  and  concern, i n  parameter.  However,  Mexico) i n d i c a t e d t h a t w h i l e  the farmers e v e n t u a l l y r e j e c t e d a h i g h  yield/-  farmer. due  to a h i g h e r  u t i l i t y being  p l a c e d on  corn  quality  than on q u a n t i t y , which i m p l i e s t h a t the s t i p u l a t i o n of  for  outcome" as a scope c o n d i t i o n has  any  the  f i c a t i o n i n terms of o r d e r i n g of u t i l i t y  "HU  to c a r r y an a d d i t i o n a l q u a l i priorities.  T h i s can be done by s t a t i n g t h a t the scope of my t h e o r e t i c a l c o n s i d e r a t i o n s a p p l i e s to "any which i s not dominated by  New  (a)  utilities  attached  i s attached,  to o t h e r  and  outcomes."  Problems  To l i s t is  outcome to which HU  the new  a major task i n i t s e l f the  problems generated by —  the t e s t i n g procedure  however the more immediate tasks  t e s t i n g procedure i t s e l f .  The  immediate task i s to  b e t t e r i n d i c e s and measures i n o r d e r  to p r o v i d e  concern: construct  a l e s s clumsy  tool  - 205 -  than the one used here.  The use of better sampling techniques  would allow me access to parametric techniques which could provide weightings f o r the e f f e c t d i f f e r e n t variables have upon r i s k taking; the attempt to axiomatise considerations of s o c i a l marginality, information, resources,  u t i l i t i e s and r i s k taking so that a l l  the statements can be derived from a set of basic hypotheses; the i n c l u s i o n of a wider set of decision making s t r a t e g i e s .  I  only dealt with high r i s k and low r i s k strategies (maximise gain and minimise loss r e s p e c t i v e l y ) , whereas calculations of minimax strategies would provide a measure for medium r i s k .  This would  give a more extensive set of statements with regard to r i s k taking and provide a better f i t f o r actual decision making s i t u ations; the testing of the revised hypotheses i n other situations involving risk; the generation of a new cycle of laboratory and f i e l d tests. Levi-Strauss' dictum that research endeavour should involve successive trips from the laboratory to the f i e l d , f i e l d to the laboratory  thence from the  (Levi-Strauss, 1961) implies that the next  stage of the research begun here l i e s i n the laboratory.  To  this end I intend to run another decision making experiment which modifies research.  to a considerable degree the design used i n this  - 206 -  This i s p a r t l y due to the need for better operational d e f i n i t i o n s , but also i s concerned with the fact that a d i f f e r e n t set of propositions w i l l be tested.  The duplex gamble w i l l be  used to manipulate the incentive conditions (payoffs, costs, and p r o b a b i l i t i e s ) attached to d i f f e r e n t bets.  The use of experi-  mental s a l a r i e s to play for actual stakes to simulate  realistic  conditions and the subjective evaluations of each gamble's a t t r a c tiveness proved to be worthwhile.  With regard to the l a t t e r  aspect I would test to see what kind of association exists between the evaluation of the attractiveness of d i f f e r e n t bets and decision made whether or not to play a p a r t i c u l a r bet.  the  This  w i l l offer a check on the r e l a t i o n between evaluation scores and decisions to play bets, and w i l l permit me  to assess whether  the lack of high r i s k taking by some subjects i s related to d i f f e r e n t subjective evaluations of the bets  considered.  The major modifications involve the o p e r a t i o n a l i s a t i o n of the wealth and information parameters.  I was  not sure that the  d i s t i n c t i o n made between upper wealth levels was  adequate.  The  wealth gradient w i l l s t i l l be calculated i n terms of the lowest category having  the property of being unable to afford losses.  By demarcating the low wealth l e v e l at $2, medium at $4 and high at $8, I w i l l be able to map t e r v a l scale.  the wealth categories into an i n -  The medium and high wealth figures are exponen-  t i a l functions of the low wealth category  taken as a base.  - 207 -  (LW=2 ; MW=2 ; HW=2 )  By assigning a measure of 1 to low wealth,  2 to medium wealth and 3 to high wealth, I can map gradient into an i n t e r v a l scale.  the wealth  The assumption of a d d i t i v i t y  and t r a n s i t i v i t y required for an i n t e r v a l mapping are met  by  considering each wealth l e v e l as an exponential function to the base 2.  In theoretical terms I w i l l construct a set of proposi-  tions which assert that r i s k taking i s an exponential function of wealth. In order to assess more p r e c i s e l y the i n t e r r e l a t i o n between wealth and information with regard to r i s k taking, I intend to permit subjects to buy information about the gambles.  They  w i l l be presented with a vague description of each bet and w i l l be able to purchase two types of information.  For 50$  the  range of any one of the four incentive conditions can be bought; for $1 the precise s p e c i f i c a t i o n of the appropriate payoff or p r o b a b i l i t y w i l l be given. The mandatory $1 betting stake w i l l mean that the low wealth subjects ($2) can only spend $1 on information as they require the remaining  d o l l a r to bet with.  The maximum amount that a  subject can spend on information i s $4 and the minimum i s $0. The f i n a n c i a l outlays made on information enables one to construct an i n t e r v a l scale for information corresponding to the amount paid out by the subject.  directly  This transformation rests  on the premise that subjective u t i l i t i e s attached to small monetary amounts correspond  to the objective value of the amount. (Edwards,  Lindman, P h i l l i p s  1965)  - 208  The  -  advantage of mapping these parameters i n t o i n t e r v a l  s c a l e s i s t h a t i t allows  me  access to p a r a m e t r i c techniques which  p e r m i t a c a l c u l a t i o n of the weights t h a t can be incentive conditions high  and  low  risk  and  a t t r i b u t e d to  s i t u a t i o n a l c o n s t r a i n t s with regard  the purpose would not  simply  be  to see  i f risk  i s i n the p r e d i c t e d d i r e c t i o n but whether or not of w e i g h t i n g s are a t t a c h e d f u n c t i o n a l to r i s k t a k i n g . much more r i g o r o u s main d i f f i c u l t i e s respect The  to  taking.  T h i s would then have i m p l i c a t i o n s f o r the subsequent t e s t as  the  taking  s i m i l a r types  to the c o n s t r a i n t s s t i p u l a t e d as This  c o n s i d e r a t i o n would i n v o l v e a  o p e r a t i o n a l i s a t i o n of v a r i a b l e s , and foreseen  field  i n subsequent f i e l d  the  t e s t s are w i t h  to t h i s problem o f measurement. next experiment w i l l  still  d e a l w i t h b e t s as  unique,  d i s c r e t e events, though subsequent experiments would be to i n c o r p o r a t e  designed  the e f f e c t s of p r i o r winnings.and l o s s e s on  de-  c i s i o n making. The  c h o i c e of s u b j e c t s  c o u l d a l s o be v a r i e d .  One  obvious  domain f o r t e s t i n g i s f i s h e r m e n themselves, though o t h e r p a t i o n a l groups — be  fertile  policemen, s t o c k h o l d e r s ,  avenues to e x p l o r e  must remain constant,  and  compare.  bartenders —  would  One  that  parameter  however, i s the u t i l i t y parameter.  may  appear d e s i r a b l e to examine the e f f e c t s t h a t a low  for  an outcome would have upon r i s k t a k i n g .  a c o n t r a d i t i o n i n terms as one would no  occu-  But  l o n g e r be  It  utility  t h i s would  be  dealing with  - 209  risk.  -  Risk taking r e f e r s to behaviour i n s i t u a t i o n s where there  i s a desirable goal and a lack of c e r t a i n t y that i t can be attained, with attendant p o s s i b i l i t i e s of loss (p. 42).  Therefore,  i f subsequent experiments are to be concerned with t e s t i n g theo r i e s of r i s k taking they n e c e s s a r i l y involve a high u t i l i t y for the outcomes examined. From some of my e a r l i e r remarks about the experimental  sit-  uation as a t e s t i n g device (p. 54, 9 2 f f , 198) i t could be i n f e r r e d that I attach more weight to the s i g n i f i c a n c e of laboratory res u l t s than I do to f i e l d r e s u l t s . I should s t a t e that t h i s i n ference i s not j u s t i f i a b l e , as I regard the laboratory as an equivalent t e s t i n g ground to the f i e l d .  My concern i s to show  i t s p o t e n t i a l s i g n i f i c a n c e as a t o o l f o r enquiry i n anthropology. The value of any t e s t —  be i t i n the laboratory or the f i e l d  —  l i e s i n the construction of the research design and not i n the i n f a l l i b i l i t y of e i t h e r context. T h e o r e t i c a l Implications The immediate t h e o r e t i c a l i m p l i c a t i o n i s that I have shown that i n d i v i d u a l s take r i s k s i n s i m i l a r s o r t s of ways i n divergent  sub-  s t a n t i v e ins tances. This implies that e f f e c t i v e c r o s s - c u l t u r a l analysis requires a process of a b s t r a c t i o n at a l e v e l that does not have s i t u a t i o n a l boundaries.  The operational d e f i n i t i o n s of the abstract concepts i s  the mechanism by which s i t u a t i o n a l boundaries are t r a n s l a t e d i n t o measurable parameters relevant to the model.  - 210  -  But I must emphasize that the abstract conceptualisation has to be free from the s i t u a t i o n a l referent. The obvious extension of my assertions about risk-taking i s into the exchange theory of Homans (1958), Belshaw (1965), Blau  (1964),  the communication theory of Levi-Strauss (1967) and the transactional models proposed by Barth  (1966).  I categorised the individuals i n my thesis according to a wealth s t r a t i f i c a t i o n and was  concerned with decisions with respect  to s t r i c t l y material outcomes —  f i n a n c i a l increments.  To permit wider  generality, instead of r e f e r r i n g to wealth, I could refer to the possession of a p a r t i c u l a r resource (prestige, r i t u a l status, control of sanctions, etc.) and analyse decision-making source i n order to increase i t .  i n terms of r i s k i n g that re-  Considerations of information, u t i l i t y  and incentive conditions would apply, though they would involve d i f ferent operational d e f i n i t i o n s . Costs and payoffs would have to be calculated along s o c i a l as w e l l as economic dimensions,  and given that the measurement problems  could be overcome, one could construct a predictive model of r i s k taking with regard to p a r t i c u l a r resources and p a r t i c u l a r ends. Further implications are with regard to the way the t r a d i t i o n a l subject matter?of.anthropology  may  i n which  p r o f i t a b l y be  viewed. I argued e a r l i e r (Chapter 5) that s o c i a l structure i s a h e u r i s t i c device used by the observer and i s not a given.  I stated  - 211 -  further that s o c i a l structure exists for any actor i n a s o c i a l system when i t provides access to resources  that he values and needs.  Resources can be viewed i n terms of material commodities such as land and money but I am also r e f e r r i n g to intangible such as status, l e i s u r e , power and s e c u r i t y . and needed resource  resources  I f "status" as a valued  can only be obtained through a p a r t i c u l a r structure  then that structure i s important to the actor making decisions about the resource  "status".  From this standpoint  of resources  and actors making decisions,  s o c i a l structure can be viewed as the outcome of choice behavior.  Regu-  l a r patterns of choice behavior at the aggregate l e v e l define the s i g nificance of p a r t i c u l a r structures such as caste, or kinship groups i n terms of the resources desired resources  they can control and d i s t r i b u t e . Once the  can be obtained by alternative means and  sufficient  individuals decide to use those means, then the importance of caste or kinship a f f i l i a t i o n wanes and new  structures emerge as a function  of the change i n patterns of choice behavior. From this viewpoint i t would seem that decision making provides a basic building block for the study of s o c i a l behavior. Supplementary to this view of s o c i a l structure i s the way which one could regard culture.  Instead of the role of culture as a  residual category i n analysis, one could put i t into a more precise perspective by stating that c u l t u r a l values provide the parameters within which individuals make decisions.  in  - 212  Ethnographic cultures natives  then permit  -  descriptions  one  of p a r t i c u l a r v a l u e systems  to d e s i g n a t e  t h a t are a p p r o p r i a t e  the c u l t u r a l l y p e r c e i v e d  and  alter-  to p a r t i c u l a r d e c i s i o n s i t u a t i o n s and  demarcate the p r i n c i p l e s which are d e t e r m i n a t e (or seem to be) choosing between the a l t e r n a t i v e s .  to  for  T h i s i m p l i e s t h a t the major  task  f o r the a n t h r o p o l o g i s t i s the i d e n t i f i c a t i o n of these c o n s t r a i n t s . I would s t a t e t h a t men to cope w i t h  culturally  use  culturally  d e f i n e d problems but  a p p r o p r i a t e means  the assumption behind  a  d e c i s i o n t h e o r e t i c approach i s t h a t the means chosen are chosen r a tionally.  Men  a c t i n order  to get the t h i n g s they v a l u e and  through the r a t i o n a l use of the r u l e s and and  c u l t u r a l environments.  The  through time and  the argument made here i s t h a t the p r o c e s s  Practical  of t h e i r p h y s i c a l  r u l e s , rewards, r e s o u r c e s  p e r c e p t i o n of them can change and vary  a constant  resources  need  and man's  space, but  of r a t i o n a l c a l c u l a t i o n i s  cross c u l t u r a l v a r i a b l e .  Implications  The p o s s i b l e p r a c t i c a l i m p l i c a t i o n s of the work attempted here can perhaps most r e a d i l y be seen i n the a r e a of p l a n n i n g development i n the T h i r d World. t h e s i s i s concerned w i t h  the m i c r o - p r o c e s s  For i n s t a n c e M i c h a e l and with  T h i s i s on the grounds t h a t  for this  of d e c i s i o n making.  L i p t o n (1969) i n a review of Adelman  Thorbecke (1966) notes t h a t P a r t I of the book which i s concerned "Development Theory" bears no  "Development  Planning".  r e l a t i o n to P a r t I I which i s  titled  - 213 -  This does not surprise him as he states "Can one expect app l i e d macro-economics to be properly linked with theory when we are so dismally lacking i n a micro-economics of development or, indeed, i n tested propositions about the decision procedures of farmers, who p r i s e two-thirds  com-  of economic decision-takers i n underdeveloped countries  and produce h a l f their output?" (1969:  641)  He comes out strongly i n favour of concentrating  research  and f i e l d energy on the analysis of micro-processes and of the authors i n this volume states " I f even 10% of this talent were devoted to the testing of micro-economic hypotheses — decisions —  as regards for example, peasant  the next book of this type might be r e a l l y useful" (1969:  645) . However, Lipton may micro-studies  have overstated his plea for exclusive  as they are a beginning for systematic enquiry and  an end i n themselves.  Methodologically  not  micro-theory i n terms of l i m i -  ted problem areas and a concern with formalisation i s a l l that one e f f e c t i v e l y handle at the moment and not lose track of rigour. be emphasised, however, that micro-analysis  can  I t should  i s only a b u i l d i n g block  for macro-theory. Although my s p e c i f i c considerations are i n terms of a microtheory of r i s k taking, i n the context of farmers and a g r i c u l t u r a l i n novations i t could be linked to a model of macro-eaonomic development i n terms of the following assumptions: (1)  an increase i n risk-taking with respect to new  factors of produc-  tion by peasant farmers w i l l lead to an increase i n productivity;  - 214 -  (2)  increased productivity leading to surplus conditions i s one of the keys to sustained economic growth because (a)  with an a g r i c u l t u r a l surplus the a g r i c u l t u r a l production sector of an economy provides productive inputs for other sectors of the economy,  (b)  this increases incomes and r u r a l purchasing power and i n turn produces higher demands for consumer goods from the i n d u s t r i a l sector, with attendant increases i n investment  and  savings; ultimately this leads to the creation of a s i t u a t i o n of self-sustaining growth; (3)  the other key to the development problem i s that surplus a g r i c u l t u r a l labour be relocated i n a productive capacity i n some other sector of the economy —  i n d u s t r i a l or commercial.  In this way my considerations could be t i e d i n with macrodevelopment models of the nature of Ranis and Fei's "Development of the Labour Surplus Economy" (1965). In addition to this t i e i n with macro-models, there i s an urgent s t r a t e g i c reason for concentrating on micro-processes gard to implications for development p o l i c y .  with re-  This i s because, by  and  large, development planners have very l i t t l e information and knowledge to go on when they come to formulate plans. The information from the tests I have conducted should i n d i cate to the planner certain constraints and l i k e l y outcomes for p o l i c y . With the knowledge that poor farmers cannot be induced to take high  -  215  -  risks on new a g r i c u l t u r a l practices, and that the present a g r i c u l t u r a l revolution i n Asia i s a function of providing low-risk production factors, the planner knows (more than he did) the parameters he has to work with. In a s t r a t e g i c sense my considerations could be regarded as a l i n k between theory and the p r a c t i c a l world of development planning. Conclusion What I have attempted  to do i n this thesis does not coincide  with what I would have l i k e d to have done.  Measurement problems denied  me the l e v e l of rigour I thought desirable, but I consider this to be at least a step i n the right d i r e c t i o n . The procedure of using laboratory tests as a guideline f o r fieldwork may  seem somewhat novel, but I must confess that i n other  d i s c i p l i n e s , the procedures I have used are long since standardized. However, i f this thesis i s to be a contribution to the growing formal t r a d i t i o n i n economic anthropology, i t s e x p l i c a t i o n of the process of decision making should enable one to have confidence i n the methodological assumptions that were fundamental to both model construction and testing procedures.  - 216 -  BIBLIOGRAPHY A  Adelman I. and Thorbecke E. 1966 Theory and Design of Economic Development. Press. Apodaca A. 1952  Johns Hopkins  Corn and Custom — The Introduction of Hybrid Corn to Spanish American Farmers i n New Mexico. In Spicer E.H. (ed.): Human Problems i n Technological Change. N.Y.: Russell Sage Foundation. B  Barth F. 1966 Bauer, P.T. 1946  Models of Social Organisation. No. 23.  R.A.I. Occasional Papers,  Report on the Rubber Smallholdings i n Malaya. Colonial Research Publications No. 1.  H.M.S.O.  Beal, G.M., Rogers E.M., Bohlen, J . 1957 V a l i d i t y of the Concept of Stages i n the Adoption Process. Rural Sociology 22: 166-168. Belshaw, C.S. 1965 T r a d i t i o n a l Exchange and Modern Markets. C l i f f s , N.J.: Prentice H a l l ; Blau, P. 1964  Exchange and Power i n Social L i f e .  N.Y.: Wiley.  Buchler, I.R. and Nutini H.G. (ed.) 1969 Game Theory i n the Behavioral Sciences. Pittsburgh Press. Burling, R. 1962  Englewood  University of  Maximisation Theories and the Study of Economic Anthropology. American Anthropologist 64:802-821.  - 217 -  C  Cancian, F. 1967  Cook, S. 1966  Copp, J.H. 1958  S t r a t i f i c a t i o n and Risk Taking: A Theory Tested on A g r i c u l t u r a l Innovation. American S o c i o l o g i c a l Review 32: 912-927. The Obsolete 'Anti-Market' Mentality: A Critique of the Substantive Approach to Economic Anthropology. American Anthropologist, 68: 323-345.  Toward Generalisation i n Farm Practise Research. Sociology 23: 103-111.  Rural  D Dalton, G. 1961 1969  Economic Theory and Primitive Society. pologist. 63: 1-25.  American Anthro-  Theoretical Issues i n Economic Anthropology. Anthropology, February 1969.  Davenport, W. 1960 Jamaican Fishing: A Game Theory Analysis. versity Publications i n Anthropology, 59.  Current  Yale Uni-  Department of Fisheries of Canada 1969 Fisheries S t a t i s t i c s of B r i t i s h Columbia 1968.  E  Economist 1967 Edwards, W. 1954  Food for India.  Economist, May 17.  The Theory of Decision Making. Psychological B u l l e t i n 51: 380-417. Edwards, W., Lindman, H., and P h i l l i p s , L.D. 1965 Emerging Technologies for Making Decisions. In New D i rections i n Psychology I I . T. Newcomb (ed.), N.Y.: Holt, Rinehart and Winston.  - 218 -  F.A.O. 1967  Annual Report of the Food and Agriculture Organisation of the United Nations. 1965-1966 H.M.S.O.  1970  Annual Report of the Food and Agriculture Organisation of the United Nations. 1968-1969-H.M.S.0.  F i r t h , R. 1939  Primitive Polynesian Economy. Kegan Paul.  London:  Routledge and  1964a  Essays on Social Organisation and Values.  1964b  Social Prerequisites of Economic Growth.  London'. Athlone Press, UNESCO/55/38.  F l i e g e l , Roy, Sen and K i v l i n 1968 A g r i c u l t u r a l Innovations i n Indian V i l l a g e s . ;HyderabadInstitute for Community Development Studies. Foster, G.M. 1962  Goodfellow, 1939 Gouldner, 1960 Greer, S. 1969  T r a d i t i o n a l Cultures and the Impact of Technological Change. N.Y.: Harper.  D.M. P r i n c i p l e s of Economic Sociology.  London:  Routledge.  A.W. The Norm of Reciprocity. American S o c i o l o g i c a l Review 25: 161^178. Logic of Social Enquiry.  Aldine.  G r i l i c h e s , Z. 1957 Hybrid Corn: An Exploration i n the Economics of Technological Change. Econometrica 25: 501-522. Gross, N.C. The Diffusion of a Culture T r a i t i n Two Iowa Townships. 1942 Ames, Iowa State College.  - 219 -  Gross, N.C. and Taves, M.J. 1952 Characteristics Associated with Acceptance of Recommended Farm Practices. Rural Sociology 17: 321-327.  H Hayden, H. 1954  Moturiki: A P i l o t Project i n Community Development. London, Oxford University Press.  Herskovits, M. 1940 The Economic L i f e of Primitive Peoples. Knop f. 1952  Economic Anthropology.  New York:  S o c i a l Change.  London.  New York:  Knopf.  Hogbin, I. 1958  Watts:  Homans, G.C. 1958  Social Behavior as Exchange. American Journal of S o c i o l ogy 63: 597-606. Hsieh, Sam-Chung 1968 New Outlook for Asian Agriculture. International Development Review September 1968. Hoselitz, B.F. and Moore, W. 1966 I n d u s t r i a l i s a t i o n and Society.  UNESCO:  Mouton.  K  Kogan, N. and Wallich, M. 1967 Risk Taking as a Function of Situation, Personality and Group. In New Directions i n Psychology I I I . Newcomb, T. (ed.). N.Y.: Holt, Rinehart and Winston.  L  Leach, E.R. 1961  Rethinking Anthropology.  London, Athlone Press.  - 220 LeClair, E.E. 1962 Economic Theory and Economic Anthropology. Anthropologist 64: 1179-1203.  American  LeClair,E.E. and Schneider, H.K. 1968 Economic Anthropology: Readings i n Theory and Analysis. Holt, Rinehart and Winston. Levi-Strauss,. C. 1966 Scope of Anthropology. 1967 Lipton, M. 1969  Current Anthropology 7,2: 112-123.  P r i n c i p l e of Reciprocity. In Elementary Structures of Kinship. Aldon: Second Edition. Review of Theory and Design of Economic Development, by Adelman I. and Thorbecke E. Economic Journal LXXIX: 641-645. M  Marsh, C P . 1955  Moerman, M. 1968  Nature 1970  and Coleman, A. The Relation of Farmer Characteristics to the Adoption of Recommended Farm Practices. Rural Sociology 20: 289296.  A g r i c u l t u r a l Change and Peasant Choice i n a Thai V i l l a g e . Berkeley: University of C a l i f o r n i a Press.  Food Production-Shortage to Surplus?  von Neumann,J. and Morgernstern, 0. 1953 Theory of Games and Economic Behavior. v e r s i t y Press.  Orans, M. 1968  Maximising i n Jajmaniland. 875-895.  225; 5227: 9-10.  Princeton Uni-  American Anthropologist 70:  o  - 221 P Polanyi, K. 1957  The Economy as Instituted Process. In Trade and Market i n the Early Empires. Polanyi, K. Arensberg, CW. and Pearson, H.W. (ed.), New York: Free Press.  Prattis, J.I. 1967 The Implications of Technological Change i n a Theory of Community Development. I n s t i t u t e of S o c i a l Anthropology, Oxford University. Unpublished B . L i t t . d i s s e r t a t i o n . R  Ranis, G. and F e i , J . 1965 Development of a Labour Surplus Economy. I l l i n o i s : Irwin. Robbins, L. 1935 Rogers, E.M. 1962  Homewood,  An Essay on the Nature and Significance of Economic S c i ence. London: Macmillan. The D i f f u s i o n of Innovations. Glencoe.  New York:  Free Press of  Ryan, B. and Gross, N. 1943 Diffusion of Hybrid Seed Corn i n Two Iowa Communities. Rural Sociology 8: 15-24. S'  Schultz, T.W. 1963 The Economic Value of Education. Press.  N.Y.:  Columbia University  1964  Transforming T r a d i t i o n a l Agriculture.  Yale University Press,  1965  Economic C r i s i s i n World Agriculture. Michigan Press.  University of  Slovic, P. 1968  Relative Importance of P r o b a b i l i t i e s and Payoffs i n Risk Taking. Research B u l l e t i n , V o l . 7, No. 7. Oregon Research I n s t i t u t e , Eugene.  - 222 -  Slovic, P. 1969  Spicer, E.H. 1952  Tax, S. 1953  D i f f e r e n t i a l Effects of Real vs. Hypothetical Payoffs upon Choices Among Gambles. Oregon Research I n s t i t u t e , Eugene.  Human Problems i n Technological Change. Foundation.  Penny Capitalism. No. 16.  Russell Sage  Smithsonian I n s t i t u t i o n , Publication  - 223 -  APPENDIX I  Duplex Gambles Master Sheet  :t No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27  EV 0 -4.875 0 -0.15 -1.475 -1.375 2.275 -0.65 -5.675 2.2 0.65 3.575 -6.175 6.175 0.15 2.775 0 5.675 -3.575 0.65 -0.65 -2.2 1.375 1.475 -2.275 4.875 -2.775  $ w  P w  7.5 7.5 3.5 3.5 7.5 1 3.5 1 3.5 7.5 1 7.5 1 7.5 1 3.5 1 7.5 1 7.5 1 3.5 3.5 3.5 3.5 7.5 1  .55 .2 .55 .2 .2 .55 .85 .2 .2 .55 .85 .55 .2 .85 .85 .85 .55 .85 .55 .2 .85 .55 .55 .85 .2 .85 .2  1 7.5 7.5 3.5 1 3.5 3.5 3.5 1 7.5 3.5 1 1 7.5 1 3.5 1 1 3.5 7.5 1 7.5 7.5 .1 7.5 3.5 7.5 3.5  P, 1 .55 .85 .55 .85 .85 .55 .2 .85 .85 .55 .2 .55 .85 .2 .2 .2 .55 :2  .55 .85 .2 .55 .55 .2 .85 .2 .85  The expected values of the 27 duplex gambles range from a high of +$6,175 to a low of -$6,175. Only three gambles are " f a i r " — 1,3 and 17 — i n that their expected values are zero. This minimal concern with " f a i r " bets r e f l e c t s the fact that my purpose i n using the duplex gambling design i s not to have the two u t i l i t y functions sum to zero, but i s to manipulate and vary the four incentive conditions of the bets. This i s i n order to assess the significance of each component f o r r i s k taking and also to present d i s c r e t e alternatives for the subjects to choose between.  224  APPENDIX I I  INSTRUCTIONS TO  Ref/L.Inf  I would l i k e to thank you The  experiment you  SUBJECTS* f o r h e l p i n g me  i n this  experiment.  are about to p a r t i c i p a t e i n c a l l s  f o r you  to make e v a l u a t i o n s aboud d i f f e r e n t gambles t h a t w i l l be p r e s e n t e d  to  you. Although  the i n f o r m a t i o n g i v e n to you w i l l not be  each gamble can be v i s u a l i s e d win  as h a v i n g  ($ ) ; the chance of w i n n i n g w  chance of l o s i n g  f o u r components, an amount to  (C ); an amount to l o s e ( $ , ) ; and w L  L e f t hand d i s c r e p r e s e n t s  t a i n amount. (60% chance of w i n n i n g $4);  two  d i s c s as below:  the chances of winning a c e r the r i g h t hand d i s c  the chances of l o s i n g a c e r t a i n amount (40% chance of l o s i n g Each gamble has  *  a  (C^).  I t can be r e p r e s e n t e d by v i s u a l i s i n g  The  specific,  represents $5).  these f o u r components.  T h i s appendix c o n t a i n s t h r e e s e t s of i n s t r u c t i o n s to s u b j e c t s — low i n f o r m a t i o n , medium i n f o r m a t i o n and h i g h i n f o r m a t i o n c a t e g o r i e s r e s pectively. The only d i f f e r e n c e i n the i n s t r u c t i o n sheets i s i n terms of the i n f o r m a t i o n about the b e t s which i s d i s p e n s e d to s u b j e c t s . The process of p l a y i n g , and i n s t r u c t i o n s w i t h r e f e r e n c e to e v a l u a t i n g and choosing b e t s to be p l a y e d remains the same i n each case.  - 225 -  The experimental task i s for you to decide which of the gambles i n the questionnaire you would play for r e a l stakes.  In order to  do this you w i l l be given a salary at the beginning of each session. Your salary may  be small, $1; moderate $3; or high $5.  By a random  assignment your salary w i l l be $1, $3, $5 i n this session. You w i l l need your salary to bet with, with the r e s u l t that you may  end up making a l o t of money, none at a l l , or somewhere i n  between. There i s a f i n a n c i a l obligation placed upon you i n that for any bet you decide to play, you w i l l have to pay the experimenter $1 out of any possible winnings. This means that i f you decided to play a p a r t i c u l a r bet i n fact won  $10 you would receive $10 - $1, i . e . $9.  tion remains constant  and  This $1 obliga-  for every bet you decide to play, and you  can  regard i t as the stake that i s required for each bet. To make i t clear what you have to do, examine the bet below. You w i l l see a verbal description of a gamble and a series of Bet A:  Some people have t r i e d this bet.  •ft  1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate i t on a degree of attractiveness scale? -5 1 1 1  3.  questions.  0 .  1  1  1  1  1  1  +5 1  Would you bet on this gamble? The f i r s t question concerns your evaluation of the gamble.  The second question refers to your rating of the gamble.  The  indif-  - 226 -  ference point at zero represents a situation where the gamble appears neither a t t r a c t i v e nor unattractive. The third question c a l l s for your decision. The description of Bet A does not specify any of the four components of the gamble.  However for your information the range of  the four components i s as follows: ( $ ) : Amount to win has a range from $0 - $10. ( C ) : chance to win has a range from 07, - 1007°. ($•£_): amount to lose has a range from $0 - $10. ( C L ) : chance of losing has a range from 07. - 1007.. w  W  As you go through the questionnaire, regard each gamble as d i s c r e t e , i n that you w i l l be considering each one i n terms of having your salary i n your hand and to pay $1 for every bet played. To determine your earnings for taking part i n the experiment, those gambles that you decided to bet on i n the questionnaire, you w i l l in actual fact play for r e a l stakes. Say for instance you decided to play Bet A. Bet A:  Some people have t r i e d this bet, and that this corresponded to a gamble (on the master sheet that the experimenter has) that had a  607o chance of winning $6; 807o chance of losing $4.  (You w i l l r e a l i s e that the description "Some people have t r i e d this bet", could allude to a variety of combinations such as 207. chance of winning $10; 507, chance of losing $4. OR 507o chance of winning $5; 507, chance of losing $5. OR 407. chance of winning $3; 607. chance of losing $2,  - 227 -  and so on.  However there i s only one such combination on the master  sheet.) To come back to Bet A "Some people have t r i e d this bet" which corresponds on the master sheet to 607. chance of winning $4; 807> chance of losing $4. To determine how you would do on this gamble, you w i l l spin a pointer on the d i s c to determine your winnings. stops i n the shaded area w i l l you win $6.  Only i f the pointer  (The shaded area covers 607>  of the t o t a l disc and accurately portrays the chance element.)  To ascertain your losses, you w i l l again spin the pointer on the d i s c .  You w i l l lose $4 only i f the pointer stops i n the shaded  area, which represents 807, of the d i s c .  - 228 -  Your earnings from each bet w i l l consist of any p o s i t i v e amount of money gained minus the $1 that you have to pay the experimenter for playing the bet. For instance, i f you played the gamble above, and spun the pointer and i t stopped i n the 607> chance of winning area, you w i l l have won $6. I f you then spin the pointer and i t stops outside the 807o chance of losing area, you w i l l not lose $4; so your net gain i s $6 $0, i . e . $6. Subtracted from this sum i s the $1 stake necessary  to play  So you would receive $6 - $1, i . e . $5.  the bet.  Your t o t a l earnings w i l l consist of any p o s i t i v e amount you may have accumulated.  To stay i n the "game" however you must always  have at least a $1 stake before you can play the next bet. This means that i f you won $7 on the f i r s t gamble played, but l o s t your stake on the next one you played -- then you would be out of the game. i s necessary  This i s because you would not have the $1 stake that  to enable you to play any subsequent bets. ARE  Ref: Bet 1:  THERE ANY QUESTIONS  L.Inf A NUMBER OF PEOPLE HAVE TRIED THIS BET; A FEW OF WHOM HAVE BEEN POOR GAMBLERS. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  - :2t9 -  Bet 2: THOSE PEOPLE WHO  HAVE TRIED THIS BET RARELY DO WELL.  1. 2.  Do you think this i s an a t t r a c t i v e gamble? , How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  Bet 3: OF THOSE WHO HAVE TRIED THIS BET, SOME HAVE BEEN POOR GAMBLERS, SOME HAVE BEEN GOOD ONES. '  1. Do you think this i s an a t t r a c t i v e gamble? 2. How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1 3.  Bet 4:  scale?  Would you bet on this gamble?  SOME PEOPLE HAVE TRIED THIS BET 1. 2.  Do you think this i s an attractive gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 5: OF THE FEW PEOPLE WHO HAVE TRIED THIS BET, SOME HAVE BEEN POOR GAMBLERS.  Bet 6:  1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  THIS BET IS TRIED BY PEOPLE WHO ARE GENERALLY CONSIDERED TO BE POOR GAMBLERS. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  - 230 _  Bet 7: QUITE A FEW PEOPLE HAVE TRIED THIS BET, SOME HAVE BEEN GOOD GAMBLERS. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 8: ONLY A FEW PEOPLE HAVE TRIED THIS BET, SOME HAVE BEEN POOR GAMBLERS. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 9: VERY FEW PEOPLE HAVE TRIED THIS BET, ALL HAVE BEEN POOR GAMBLERS. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 10: BOTH GOOD AND POOR GAMBLERS HAVE TRIED THIS BET. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 11: SOME PEOPLE HAVE TRIED THIS BET. 1. 2.  Do you think this i s an attractive gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  231 -  Bet 12: SOME GOOD GAMBLERS HAVE TRIED THIS BET. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 13: VERY FEW PEOPLE HAVE TRIED THIS BET. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 14: OF THE PEOPLE WHO HAVE TRIED THIS BET, MOST ARE CONSIDERED TO BE GOOD GAMBLERS. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 15: A FAIR NUMBER OF PEOPLE HAVE TRIED THIS BET, A FEW OF WHOM HAVE BEEN POOR GAMBLERS. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 16: SOME GOOD GAMBLERS HAVE TRIED THIS BET. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  - 232 -  Bet 17: A NUMBER OF PEOPLE HAVE TRIED THIS BET. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  Bet 18: MANY PEOPLE HAVE TRIED THIS BET, SOME OF WHOM HAVE BEEN GOOD GAMBLERS. 1. 2.  3.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale -5 0 +5 1 1 1 1 1 1 1 1 1 1 1 Would you bet on this gamble?  Bet 19: THERE HAVE BEEN SOME PEOPLE WHO HAVE TRIED THIS BET, THOUGH THEY ARE CONSIDERED TO BE POOR GAMBLERS. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  Bet 20: OF THE PEOPLE WHO HAVE TRIED THIS BET THERE HAVE BEEN MORE POOR GAMBLERS THAN GOOD GAMBLERS. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  Bet 21: FEW PEOPLE HAVE TRIED THIS BET, THOSE WHO HAVE ARE CONSIDERED TO BE POOR GAMBLERS. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  Bet 22: POOR GAMBLERS SEEM TO BE THE PEOPLE WHO HAVE TRIED THIS BET THE MOST. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness seal -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  Bet 23: A NUMBER OF GOOD GAMBLERS HAVE TRIED THIS BET. 1. 2.  Do you think this i s an attractive gamble? How would you rate this on a degree of attractiveness s;ale -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  Bet 24: OF THE PEOPLE WHO HAVE TRIED THIS BET, SOME HAVE BEEN POOR GAMBLERS. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness seal -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  Bet 25: GOOD GAMBLERS HAVE RARELY TRIED THIS BET. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness seal -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  Bet 26: QUITE A NUMBER OF PEOPLE HAVE TRIED THIS BET. 1. 2.  Do you think this i s an attractive gamble? How would you rate this on a degree of attractiveness seal -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  - 234-; -  Bet 27: GOOD GAMBLERS HAVE STAYED AWAY FROM THIS BET. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  Ref/M.Inf  scale  INSTRUCTIONS TO SUBJECTS  I would l i k e to thank you for helping me i n this experiment. The experiment you are about to participate i n c a l l s for you to make evaluation about d i f f e r e n t gambles that w i l l be presented to you. The gamble has four components, an amount to win ( $ ) ; the w  chance of winning ( C ) ; an amount to lose w  ; n d a chance of loosing a  (CL).  I t can be represented by v i s u a l i s i n g two d i s c s :  - 235 -  The  l e f t hand disc represents  the chances of winning a c e r t a i n  amount (60% chance of winning $4); the right hand d i s c represents  the  chances of losing a certain amount (40% chance of losing $5). Each gamble has  these four components.  The experimental task i s for you to decide which of the gambles i n the questionnaire you would play for real stakes.  In order to  do this you w i l l be given a salary at the beginning of each session. Your salary may  be small, $1, moderate $3, or high $5.  By a random  assignment your salary w i l l be $1, $3, $5, i n this session. You w i l l need your salary to bet with, with the r e s u l t that you may  end up making a l o t of money, none at a l l , or somewhere i n  between. There i s a f i n a n c i a l obligation placed upon you i n that for any bet you decide to play, you w i l l have to pay the experimenter $1 out of any possible winnings. This means that i f you decided i n fact won  $10 you would receive $10  tion remains constant  to play a p a r t i c u l a r bet and  - $1, i . e . $9.  This $1 obliga-  for every bet you decide to play.  To make i t clear what I want you to do, examine the bet low.  be-  You w i l l see a verbal description of a gamble, and a series of  questions: BET A:  GOOD CHANCE of HIGH WINNINGS, FAIR CHANCE of MODERATE COSTS. (C ) ($ ) (C ) ($ ) w  w  L  L  - 236. -  1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate i t on a degree of attractiveness 0 -5 +5 very indifference very unattractive attractive  3.  Would you bet on t h i s gamble?  scale?  The f i r s t question concerns your evaluation of the gamble. The second question refers to your rating of the gamble.  The i n d i f f e r -  ence point at zero represents a s i t u a t i o n where the gamble appears neither a t t r a c t i v e nor unattractive. The third question c a l l s for your decision. The verbal descriptions of each component cover a range of values as follows: A GOOD CHANCE of winning or losing has a range of 70% to 100% A FAIR CHANCE of winning or losing has a range of 40% to 70% A POOR CHANCE of winning or losing has a range of 0% to. 40% HIGH winnings or losses range from $5 to $10 MODERATE winnings or losses range from $1 to $5 LOW winnings or losses range from $0 to $1 (These ranges are written on the blackboard for your reference.) This means that Bet A:  GOOD CHANCE OF HIGH WINNINGS; FAIR  CHANCE OF MODERATE COSTS could be either (1) 100% chance of winning $10; 40% chance of losing $1.50 OR  (2) 71% chance of winning $5; 69%  chance of losing $4.50. You w i l l note that one verbal description covers a wide range of p o s s i b i l i t i e s .  (2) i s very d i f f e r e n t to (1), yet they are both  covered by the same verbal description.  - 237. "  As you go through the questionnaire, regard each gamble as d i s c r e t e i n that you w i l l be considering each one i n terms of having your salary i n your hand and being obligated to pay the experimenter $1 for every bet played. To determine your earnings for taking part i n the experiment, those gambles that you decide to bet on i n the questionnaire, you w i l l i n actual fact play for r e a l  stakes.  Say for instance, you decided on the gamble:  ( i n the questionnaire)  to bet  FAIR CHANCE OF HIGH WINNINGS; HIGH CHANCE OF MODERATE  LOSSES and that this verbal description meant:  60% chance of winning  $10; 807> chance of losing $2. To determine how you would do on this gamble, you w i l l spin a pointer on the d i s c to determine your winnings. stops i n the shaded area w i l l you win $10.  Only i f the pointer  (The shaded area covers  607> of the t o t a l disc and accurately portrays the chance element.)  To ascertain your losses, you w i l l again spin the pointer on the d i s c .  You w i l l lose $2 only i f the pointer stops i n the shaded  area, which represents  807o of the d i s c .  - 238. -  Your earnings from each bet w i l l consist of any positive amount of money gained minus the $1 that you have to pay the experimenter for playing the bet. For  instance i f you played the gamble above, and spun the  pointer and i t stopped i n the 60% chance of winning area, you w i l l have won $10. I f you then spin the pointer and i t stops i n the 80% chance of losing area, you w i l l lose $2; so your net gain i s $10 - $2, i . e . $8. Subtracted from this sum i s the $1 f i n a n c i a l obligation placed upon you by the experimenter. . So you would receive $7. Your total earnings w i l l be calculated from the results of a l l the gambles you decided to play. ARE THERE ANY QUESTIONS?  Ref: Bet  M.Inf 1:  FAIR CHANCE OF HIGH PAYOFF, FAIR CHANCE OF HIGH LOSSES 1. 2.  Do you think this i s an attractive gamble? How would you rate this on a degree of attractiveness s;ale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  - :239 -  Bet 2: POOR CHANCE OF HCGH PAYOFF, GOOD CHANCE OF HIGH LOSSES. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  Bet 3: FAIR CHANCE OF MODERATE PAYOFF, FAIR CHANCE OF MODERATE LOSSES. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 4: POOR CHANCE OF MODERATE PAYOFF, GOOD CHANCE OF LOW LOSSES. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 5: POOR CHANCE OF HIGH PAYOFF, GOOD CHANCE OF MODERATE LOSSES. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 6: FAIR CHANCE OF LOW PAYOFFS, FAIR CHANCE OF MODERATE LOSSES. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  - 240; -  Bet 7: GOOD CHANCE OF MODERATE PAYOFFS, LOW CHANCE OF MODERATE LOSSES. 1. 2.  Do you think this i s an a t t r a c t i v e gambe? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 8: LOW CHANCE OF LOW PAYOFFS, HIGH CHANCE OF LOW LOSSES. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 9: POOR CHANCES OF MODERATE PAYOFFS, GOOD CHANCE OF HIGH LOSSES. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you r a t e this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 10: FAIR CHANCE OF HIGH PAYOFFS, FAIR CHANCE OF MODERATE LOSSES. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 11: GOOD CHANCE OF LOW PAYOFFS, LOW CHANCE OF LOW LOSSES. 1. 2.  Do you think this i s an attractive gamble? How would you rate this on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  - 24:1 -  Bet 12: FAIR CHANCE OF HIGH PAYOFFS, FAIR CHANCE OF LOW LOSSES. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness seal -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  Bet 13: LOW CHANCE OF LOW PAYOFFS, HIGH CHANCE OF HIGH LOSSES. 1. 2.  Do you think this i s an attractive gamble? How would you rate i t on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 14: GOOD CHANCE OF HIGH PAYOFFS, LOW CHANCE OF LOW LOSSES. 1. 2.  Do you think this i s an attractive gamble? How would you rate this on a degree of attractiveness seal -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  Bet 15: GOOD CHANCE OF LOW PAYOFFS, LOW CHANCE OF MODERATE LOSSES 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness seal -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  Bet 16: GOOD CHANCE OF MODERATE PAYOFFS, LOW CHANCE OF LOW LOSSES. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate i t on a degree of attractiveness -5 0 +5 1 1 1 . 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  - 242 -  Bet 17: FAIR CHANCE OF LOW PAYOFFS, FAIR CHANCE OF LOW COSTS 1. 2.  Do you think this i s an a t t r a t i v e gamble? How would you rate i t on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 18: GOOD CHANCE OF HIGH PAYOFF, LOW CHANCE OF MODERATE LOSSES 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate i t on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 19: FAIR CHANCE OF LOW PAYOFFS, FAIR CHANCE OF HIGH LOSSES 1. '2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  '3.  Would you bet on this gamble?  Bet 20: POOR CHANCE OF HIGH PAYOFFS, GOOD CHANCE OF LOW LOSSES 1. 2.  Do you think this i s an a t t r a c t i v e gamble? ... How would you rate this on a degree of attractiveness scale -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on t h i s gamble?  Bet 21: GOOD CHANCE OF LOW PAYOFFS, LOW CHANCE OF HIGH LOSSES. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on t h i s gamble?  - 243 -  Bet 22: FAIR CHANCE OF MODERATE PAYOFFS, FAIR CHANCE OF HIGH LOSSES. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness seal -5 0 +5 1 1 1 1 1 1 . 1 1 . I l l  3.  Would you bet on this gamble?  Bet 23: FAIR CHANCE OF MODERATE PAYOFFS, FAIR CHANCE OF LOW LOSSES. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate i t on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 24: GOOD CHANCE OF MODERATE PAYOFFS, LOW CHANCE OF HIGH LOSSES. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this i n a degree of attractiveness seal -5 0 + 5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  Bet 25: LOW CHANCE OF MODERATE PAYOFFS, GOOD CHANCE OF M3ERATE LOSSES. 1. 2.  Do you think this i s an attractive bet? How would you rate i t on a degree of attractiveness -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  scale?  Bet 26: GOOD CHANCE OF HIGH PAYOFFS, LOW CHANCE OF HIGH LOSSES. 1. 2.  Do you think this i s an attractive gamble? How would you rate this on a degree of attractiveness seal -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  - 244 -  Bet 27: LOW CHANCE OF LOW PAYOFFS, HIGH CHANCE OF MODERATE LOSSES. 1. 2.  Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale? -5 .0 +5 1 1 1 1 1 1 1 1 1 1 1  3.  Would you bet on this gamble?  Ref/H.Inf  INSTRUCTIONS TO SUBJECTS I would l i k e to thank you for helping me i n t h i s experiment. The experiment you are about to p a r t i c i p a t e  i n c a l l s for you  to make evaluations about d i f f e r e n t gambles that w i l l be presented to you. The gamble has four components,  an amount to win ( $ ) ; the w  chance of winning ( C ) ; an amount to lose ($L) ; and a chance of losing W  (CL).  I t can be represented by v i s u a l i s i n g two discs  below:  The l e f t hand disc represents the chances of winning a certain amount (607o chance of winning $4) ; the r i g h t hand disc represents the chances of losing a certain amount (407> chance of losing $5).  - 245 -  Each gamble has these four components. The experimental  task i s for you to decide which of the gam-  bles i n the questionnaire you would play for r e a l stakes.  In order to  do this you w i l l be given a salary at the beginning of each session. Your salary may  be small, $1; moderate $3; or high $5.  By a random  assignment your salary w i l l be $1, $3, $5, i n this session. You w i l l need your salary to bet with, with the r e s u l t that you may  end up making a l o t of money, none at a l l , or somewhere i n be-  tween. There i s a f i n a n c i a l o b l i g a t i o n placed upon you i n that for any bet you decide to play, you w i l l have to pay the experimenter $1 out of any possible winnings. This means that i f you decided to play a p a r t i c u l a r bet and i n fact won  $10 you would receive $10 - $1, i . e . $9.  This $1 obliga-  tion remains constant for every bet you decide to play. To make i t clear what you have to do, examine the bet below. You w i l l see a verbal description of a gamble, and a series of questions. Take s p e c i a l note of the f i r s t question as this enables you to get more information about the gamble. Bet A:  GOOD CHANCE of HIGH WINNINGS, FAIR CHANCE of MODERATE COSTS. <w> C  1. 2.  <V  < L> C  (  Which two of the four components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble?  V  - 246 -  3.  How would you rate i t on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble? The f i r s t question permits you to gain more information about  the gamble.  For instance, i f you want to know what the chances of l o -  sing (C^) and the amount to- win ( $ ) are, ask the experimenter and he w  w i l l give you the precise figure, e.g. 60% chance of l o s i n g ; 4$ to win. The second question concerns your evaluation of the gamble. The third question refers to your rating of the gamble. The indifference point at zero represents a s i t u a t i o n where the gamble appears neither a t t r a c t i v e nor unattractive. The fourth question c a l l s for your decision. The verbal description of each component covers a range of values as follows: A GOOD chance of winning or losing has a range of 707. to 100% A FAIR chance of winning or losing has a range of 40% to 707. A POOR chance of winning or losing has a range of 0% to 407> HIGH winnings or losses range from $5 to $10 MODERATE winnings or losses range from $1 to $5 LOW winnings or losses range from $0 to $1. (These ranges are written on the blackboard for your reference). This means that Bet A:  GOOD CHANCE of HIGH WINNINGS, FAIR  CHANCE of MODERATE COSTS could be either (1)  1007, chance of winning $10, 407. chance of losing $1.50 OR  (2)  717, chance of winning $5 , 697. chance of losing $4.50.  You w i l l note that one verbal description covers a wide range of possibilities.  (2) i s very d i f f e r e n t to (1), yet they are both covered by  the same verbal description.  - 247 -  As you go through the questionnaire, regard each gamble as d i s c r e t e i n that you w i l l be considering each one i n terms of having your salary i n your hand and being obligated to pay the experimenter $1 for  every bet that you play. To determine your earnings for taking part i n the experiment,  those gambles that you decide to bet on i n the questionnaire, you w i l l i n actual fact play for r e a l  stakes.  Say for instance, you decided on the gamble:  ( i n the questionnaire)  to bet  FAIR CHANCE of HIGH WINNINGS; HIGH CHANCE of MODERATE  LOSSES and that this verbal description meant: 60% chance of winning $10; 80% chance of losing $2. To determine how you would do on this gamble, you w i l l spin a pointer on the d i s c to determine your winnings. stops i n the shaded area w i l l you win $10.  Only i f the pointer  (The shaded area covers  607o of the t o t a l d i s c and accurately portrays the chance element.)  To ascertain your losses, you w i l l again spin the pointer on the d i s c .  You w i l l lose $2 only i f the pointer stops i n the shaded  area, which represents  807> of the d i s c .  '248-  $  L  = ?2  Your earnings from each bet w i l l consist of any p o s i t i v e amount of money gained minus the $1 that you have to pay the experimenter for playing the bet. For instance, i f you played the gamble above, and spun the pointer and i t stopped i n the 60% chance of winning area, you w i l l have won  $10. I f you then spin the pointer and i t stops i n the 80% chance  of losing area, you w i l l lose $2; so your net gain i s $10 - $2, i . e . $8. Subtracted from this sum i s the $1 stake necessary to play the bet.  So you would receive $7. Your t o t a l earnings w i l l consist of any positive amount you  may  have accumulated.  To stay i n the "game" however, you must always  have at least a $1 stake before you can play the next bet. This means that i f you won $7 on the f i r s t gamble played, but lost $8 on the next one you played -- then you would be out of the game. This i s because you would not have the $1 stake that i s necessary to enable you to play any subsequent bets. ARE THERE ANY QUESTIONS?  - 249 -  Ref:  H.Inf  Bet 1: FAIR CHANGE OF HIGH PAYOFF, FAIR CHANCE OF HIGH LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gambles? How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  Bet 2: POOR CHANCE OF HIGH PAYOFF, GOOD CHANCE OF HIGH LOSSES 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4. Would you bet on this gamble? Bet 3: FAIR CHANCE OF MODERATE PAYOFF, FAIR CHANCE OF MODERATE LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an attractive gamble? How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this, gamble?  Bet 4: POOR CHANCE OF MODERATE PAYOFF, GOOD CHANCE OF LOW LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  250  Bet 5: POOR CHANCE OF HIGH PAYOFF, GOOD CHANCE OF MODERATE LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  Bet 6: FAIR CHANCE OF LOW PAYOFFS, FAIR CHANCE OF MODERATE LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an attractive gamble? How would you rate this on a degree of attractiveness scale? -5 0 + 5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  Bet 7: GOOD CHANCE OF MODERATE PAYOFFS, LOW CHANCE OF MODERATE LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1' 1 1 1 1  4.  Would you bet on this-gamble?  Bet 8: LOW CHANCE OF LOW PAYOFFS, HIGH CHANCE OF LOW LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  - 251 -  Bet 9: POOR CHANCE OF MODERATE PAYOFFS, GOOD CHANCE OF HIGH LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  Bet 10: FAIR CHANCE OF HIGH PAYOFFS, FAIR CHANCE OF MODERATE LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 T 1 1 1 1 1 1  4.  Would you bet on this gamble?  Bet 11: GOOD CHANCE OF LOW PAYOFFS, LOW CHANCE OF LOW LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  Bet 12: FAIR CHANCE OF HIGH PAYOFFS, FAIR CHANCE OF LOW LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  - 252'  Bet 13: LOW CHANCE OF LOW PAYOFFS, HIGH CHANCE OF HIGH LOSSES. 1. 2. 3.  4.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate i t on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1 Would you bet on this gamble?  Bet 14: GOOD CHANCE OF HIGH PAYOFFS, LOW CHANCE OF LOW LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? • How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  Bet 15: GOOD CHANCE OF LOW PAYOFFS, LOW CHANCE OF MODERATE LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an attractive gamble? How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  Bet 16: GOOD CHANCE OF MODERATE PAYOFFS, LOW CHANCE OF LOW LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale? -5 0 + 5 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  Bet 17: FAIR CHANCE OF LOW PAYOFFS, FAIR CHANCE OF LOW COSTS. 1.  Which 2 of the 4 components of the bet do you want precise information about? 2. Do you think this i s an a t t r a c t i v e gamble? 3. How would you rate i t on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 4. Would you bet on this gamble?  - ?53 -  Bet 18: GOOD CHANCE OF HIGH PAYOFFS, LOW CHANCE OF MODERATE LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an attractive gamble? How would you rate i t on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  Bet 19: FAIR CHANCE CF LOW PAYOFFS, FAIR CHANCE OF HIGH LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness seal -5 0 +5 l l l l l l l l l l ' l  4.  Would you bet on this gamble?  Bet 20: POOR CHANCE OF HIGH PAYOFFS, GOOD CHANCE OF LOW LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness seal -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  Bet 21: GOOD CHANCE OF LOW PAYOFFS, LOW CHANCE OF HIGH LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness seal -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  - 25,4 -  Bet 22: FAIR CHANCE OF MODERATE PAYOFFS, FAIR CHANCE OF HIGH LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  Bet 23: FAIR CHANCE OF MODERATE PAYOFFS, FAIR CHANCE OF LOW LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate i t on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4. Would you bet on this gamble? Bet 24: GOOD CHANCE OF MODERATE PAYOFFS, LOW CHANCE OF HIGH LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate this i n a degree of attractiveness scale -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  Bet 25: LOW CHANCE OF MODERATE PAYOFFS, GOOD CHANCE OF MODERATE LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate i t on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  - '255. -  Bet 26: GOOD CHANCE OF HIGH PAYOFFS, LOW CHANCE OF HIGH LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4.  Would you bet on this gamble?  Bet 27: LOW CHANCE OF LOW PAYOFFS, HIGH CHANCE OF MODERATE LOSSES. 1. 2. 3.  Which 2 of the 4 components of the bet do you want precise information about? Do you think this i s an a t t r a c t i v e gamble? How would you rate this on a degree of attractiveness scale? -5 0 +5 1 1 1 1 1 1 1 1 1 1 1  4. Would you bet on this gamble? C 0 M M E N T S:  Table IV:  Cell 1 Subjects  SR FR  VR  Cell 2 SR FR  Cell 4  Cell 3 1  VR  SR FR  Bets Categorised Per Subject  VR  SR FR  Cell 5  VR  SR FR  Cell 7  Cell 6  VR  SR FR  VR  SR FR  Cell 9  Cell 8  VR SR FR  VR  SR FR  VR M  1  2  2  2  4  1  —  3  2  —  7  6  7  1  —  8  4  6  6  4  —  7  4  —  8  7  2  —  7  2  —  8  4  —  3  —  7  2  8  5  3  4  5  4  4  4  5  7  7  3  —  8  5  2  5  3  6  1  —  7  6  2  3 —  —  8  3  —  5  7  3  7  4  4  2  4  4  —  O I—I  8  4  1  6  2  —  8  2  —  8  4  —  5  8  2  —  8  5  2  —  5  8  4  —  7  3  —  4  —  5  1  —  7  4  —  8  4  1  1  —  8  3  —  6  5  1  8  4  —  1-—  8  1  —  8  2  —  8  2  1  o 3 o  4—-  7  1  —  7  6  —  8  3  —  w  5  —  X  H o CO  1  —  6  —  4  1  —  7.3 7  3  —  8  2  —  7  l-tl  o  Hi  8 9 10 Total  0  3  25  41  SR = s l i g h t l y risky  1  3  —  —  4  1  —  7  2  —  8  4  —  8  6  1  —  6  1  —  8  1  —  8  2  —  7  4  —  7  1  —  8  5  1  5  0  —  6  1  —  7  3  —  8  3  —  8  4  —  8  3  —  8  5  —  46  3  —  65 16  —  74 23  —  74 27  —  — 75 33  2  77 3.7  5  68 23  X! T3 fD l-i HB fD 0 fT fa  tf Co  H CO  cr FR = f a i r l y risky VR = very risky  M fD  co  I  < M M  ro Ui ON  - 257 -  Table V:  Frequency of Bets Played Per C e l l  Bet Type  S.R. Bets  F.R. Bets  V.R. Bets  Bet 7 10 12 14 16 18 23 26 1 3 4 11 15 17 20 21 24 2 5 6 8 9 13 19 22 25 27  Cell 1  Cell 2  Cell 3  Cell 4  Cell 5  Cell 6  Cell 7  Cell 8  Cell 9  4 1 2 8 2 2 6  10 2 4 10 1 7 6 3  8 2 8 10 5. 7 5 2  9 3 10 10 8 10 7 8  8 10 10 10 8 10 8 10  10 9 10 10 8 10 10 7  7 8 8 10 7 10 8 10  10 9 10 10 9 10 8 9  10 9 10 10 10 10 8 10  ,1 1  1 3 1 7 3 1 1 1 5  4 4  3 2  4 3 1 •6 6 2 4 2 5  6 6 4 7 5 1 2  1  4  1  1  .-  -  2 . --  1 2  1  --  -  --  --  -  --  ---  -  -  -  -  --  5 1 4 1 3  --  8 4 2 1 .4  5 3 1 3 1 4  6  - 258 -  Table VI: Raw Data Matrices Wealth  Information  LW  MW  HW  Ll  25  41  46  Ml  65  74  74  HI  68  75  77  LW  MW  HW  Ll  0  3  3  Ml  16  23  27  HI  23  33  37  LW  MW  HW  Ll  0  0  0  Ml  0  0  0  HI  0  2  5  Using the formula X.. = 1 J  (a) SR Bets  (Slightly risky)  (b) FR Bets  (Fairly risky)  (c) VR Bets  (Very risky)  ij x 100 the raw data • (N ).. x (N , ).. I ij rb i j v  T  are corrected f o r unequal c e l l frequency giving three new data matrices to which a fourth has been added which combines the two top levels of riskiness into one category. (X i s the c e l l entry i n the i * t  of possible bets taken i n that c e l l ;  1  row and j ^ i  s  column; B i s ; the number  the number of possible  bets that could have been taken i n that c e l l where N^. i s the number of individuals i n the c e l l and  i s the number of risky bets i n that c e l l . )  - 259 -  Table VII:  Corrected Data Matrices  LW  MW  HW  Ll  31.3  51.3  57.5  Ml  81.3  91.3  HI  85  93.8  91.3  (a) SR  96.3  LW  MW  HW  Ll  0  0  0  Ml  0  0  0  HI  0  2  5  (b) VR  LW  MW  HW  Ll  0  3.3  3.3  Ml  17.8  25.6  30  HI  25.6  36.4  41  LW  MW  HW  Ll  0  3.3  3.3  Ml  17.8  25.6  30  HI  25.6  38.4  46  (c) FR  FR (d) and VR  - 26.0 -  APPENDIX IY .  Chi Square. Median Tests on Propositions 1, 2 and 3. Proposition One Cells 1 and 2 The n u l l hypothesis to be tested i s that the median amount of losses w i l l be the same for each c e l l . Table l a Subject No. 1 2 3 4 5 6 7 8 9 10  C e l l 1: Bets Played  Losses: Cell 1  14,23 14,23 7,14,18,23 7,12,14,23 10,16 14,23 7,14,16 7,12,14,23 14,18  2 2 9 6.5 4.5 2 5.5 6.5 4.5 0  C e l l 2: Bets Played  Losses: Cell 2  7,11,14,18,23 7,12,14 7,14,18,23,26 7,12,14,18,23 7,10,14,18,23 7,14,23 3,7,12,14,16,18,23,26 7,12,14,18 3,7,10,14,26 7,14,18  10 5.5 16.5 11 12.5 5.5 22 9 19 8  These results give a median of 6.5. When the median i s found, one sets up a table of observed frequencies for each c e l l .  This i s done by determining the number of  losses above and below the o v e r - a l l median, for each c e l l . Observed Frequencies Cell 1  Cell 2  Above Median  1  Below Median  9  2  11  10  10  20  8  9  - 261 -  The expected frequencies have 507., of the cases above the median and 507o of the cases below the median for both c e l l s ; v i z . 5. Expected Frequencies Cell 1  ? X'' =  Above Median  5  5  Below Median  5  5  2 (o-e) e  k 1  Cell 2  2 where X = chi square; k = no. of categories; o =  observed frequency i n a category; and e = expected frequency i n a category.  In this the number of df = K - 1 = 1 therefore a correction has to be  made for continuity.  This simply involves subtracting .5 from each  absolute difference between an e and an o. The formula used i s X = /(o-e) - . 5 7 i — 2  2  k  e  X  2  2 "2. 2 = /(1-5) - . 5 7 + /(8-5) - . 5 7 + /(9-5) - . 5 / + /(2-5) - .57 2  = 3.5  2  + 2.5  2  + 3.5  2  + 2.5  2  5 = 7.46 which i s s i g n i f i c a n t at the (.01) l e v e l and enables us to reject the n u l l hypothesis,  i . e . there are very s i g n i f i c a n t differences  between the medians of both c e l l s w.r.t. losses.  - 262 -  Chi Square Median Test on C e l l s 4 and 5  Table lb Subject No.  Cell 4 Bets Played  Cell4: Losses  1  7,12,14,16,18.26  17.5  2  7,12,14,16,18,23  11  3  7,11,12,14,16,18, 20,23,26 7,11,12,14,15,16, 18,20,23,26 10,11,12,14,16,18, 26 1,3,7,12,14,16,18, 20,23,26 7,10,11,12,14,16, 18,21,23,24 7,11,12,14,16,18, 20,23,26 7,12,14,18,23,24, 26 7,10,12,14,18,24, 26  20.5  4 5 6 7 8 9 10  24 18.5 30.5 30.5 20.5 25 22.5  Cell 5 Bets Played 7,10,11,12,14,16, 18,26 3,4,10,12,14,18,21, 23,24,26 7,10,11,12,14,15,16, 18,26 7,10,11,12,14,15,16, 18,23,24,26 7,10,12,14,18,23,26 1,7,10,11,12,14,16, 18,23,24,26 5,7,10,11,12,14,16, 18,23,26 7,10,11,12,14,15,16, 18,20,23,24,26 7,10,11,12,14,16,18, 23,26 3,10,12,14,16,17,18, 23,24,26  C e l l 5: Losses 22 37 25.5 34 21 38 26.5 33 25.5 30.5  These results have a median of 25. Observed Frequencies  Expected  Cell 4  Cell 4  Cell 5  Cell 5  Frequencies  Above Med.  2  8  10  Above Med.  5  5  Below Med.  _8  _2  10  Below Med.  5  5  10  10  20  X  2  =  k  /(o-3) - .5/ i = =_ e  2  = /(2-5) - .5_7 + 7(8-5) - .57 + /(8-5) - .5_7 + /(2-5) - .5_7 , _ . 2  2  2  =5.00 which i s s u f f i c i e n t l y high to reject the n u l l hypothesis at  - 263 "  C e l l s 7 and 8  Table l c Subj ect No. 1  Cell 7 Bets Played  C e l l 7: Losses  42.5  3  5,7,10,11,12,14,16, . 17,18,20,26 1,7,10,11,12,14,15, 16,18,20,23,24,26 12,14,18,23,26  4  10,12,14,18,26  16.5  5  14,16,18,23,24,26  22.5  6  1,7,10,11,12,14,16, 18,23,24,26 7,10,11,12,14,16,18, 23,26 7,10,12,14,15,18,23, 26 7,10,11,14,16,18,21, 23,24,26 3,7,10,12,14,16,18, 20,23,24,26  38  2  7 8 9 10  27.5  16.5  23 24.5 44.5 37.5  Cell 8 Bets Played  Cell 8 Losses  1,5,7,10,12,14,15, 16,17,18,20,23,26 1,7,10,12,14,16,18, 20,23,26 3,7,10,12,14,16,18, 23,24,26 1,3,7,10,12,14,16, 18,20,23,24,26 7,11,12,14,15,16, 18,20,21,23,26 3,4,7,8,10,11,12, 14,15,18,23,24 7,10,11,12,14,15, 16,18,23,26 1,7,10,11,12,14,15, 16,17,18,21,24,26 7,10,11,12,14,16,18, 26 7,10,11,12,14,15,16, 18,23,24,26  38.5 30.5 37 41.5 31.5 31 26.5 49 22 34  These r e s u l t s give a median of 31.25. Observed Frequencies  Expected Frequen<  Cell 7  Cell 8  Cell 7  Cell 8  Above M  4  4  10  Above M  5  5  Below M  _6  _j4  10  Below M  5  5  10  10  20  X  2  =  k /(-o-e) . .5/2  = /(4-5) -.bj  1  + 7(6-5) - .57  5 5  5 5  5  5  2  + /(6-5) - .5/ + 7(4-5) - .5/ 2  5  5  - 264 -  = .2 which means the n u l l hypothesis cannot be rejected. plies  This im-  that at high levels of information there i s no s i g n i f i c a n t d i f -  ference between the way LW and MW operators take r i s k s .  Proposition 2 The n u l l hypothesis being tested i s that the median amount of gains w i l l be the same for each i cell. C e l l s 1, 2, 3 Table 2a Subj ect No.  C e l l 2: Gains  C e l l 1: Gains  1 2 3 4 5 6 7 8 9 10  11 11 22 22 11 11 14.5 22 15 0  23 18.5 28.5 29.5 29.5 14.5 44 26 29.5 18.5  C e l l 3: Bets Played  Cell 3 Gains  12,14,16,18,23 7,12,14,18 7,12,14,18 7,10,12,14,18,23,26 14,18,23 3,7,12,14,16,23,26 7,14,16,23 4,7,12,14,18 4,7,10,12,14,16,18 7,12,14,16,18  29.5 26 26 44.5 18.5 36.5 18 29.5 40.5 29.5  These r e s u l t s give a median of 22. Expected Frequencies  Observed Frequencies Cell 1  Cell 2  Cell 1  Cell 3  Cell 2  Cell 3  Above M  0  7  8  15  Above M  5  5  5  Below M  10  3  _2  15  Below M  5  5  5  10  10  10  30  df  =3-1=2  - 265 -  X  2  =  \  (o-e) e  = (o-5) 5  2  + (7-5) 5  2  + (8-5) 5  2  2  + (10-5) 5  + (3-5) 5  2  2  + (2-5) 5  2  = 5 + 4 + 9+ 5+ 4 + 9 5 5 5 5 = 15.2 which i s high enough to reject p = .01 l e v e l , i . e . there are s i g n i f i c a n t  the n u l l hypothesis at the  differences between the c e l l s  as wealth varies with information constantly low.  Test between 2 and 3 median = 29 ObS. Freq. Cell 2  Cell 3  Above M  4  6  10  Below M  _6  _4  10  10  10  20  Exp. Freq.  X  2  =  k  i  5  5  5  5  /(o-e) - .57  2  5 2  = 1.5 + 4 5 = 1.8 - not  significant.  Test between 1 and 2 median = 21.3  - 266 -  Obs. Freq. Cell 1  Cell 2  Above M  3  8  11  Below M  _7  __2  _9  10  10  20  Exp. Freq.  X  2  =  5  5  5  5  k /(o-e) i e  •ll  2  .2 = 1.5 + 2.5 + 1.5 + 2.5 5 5 5 5 2  - 3.4  2  2  must r e j e c t the n u l l hypothesis, though this i s close enough  to 3.8 to point to a s i g n i f i c a n t difference.  Chi Square Median Test on C e l l s 4, 5, 6 Table 2b Subject No. " 1 2 3 4 5 6 7 8 9 10  C e l l 4: Gains 37 33 40 41 42 63 46 49 40.5 44.5  C e l l 5: Gains 44.5 62.5 46.5 53.5 44.5 60 52.5 61 49 51.5  C e l l 6: Bets Played 7,10,12,14,15,16,18,23,24,26 1,3,7,10,12,14,17,18,23,24,26 1,3,7,10,11,12,14,15,16,18,23,26 7,10,11,12,14,16,18,23 1,7,10,11,12,14,16,17,18,23,24,26 7,11,12,14,18,23 7,10,11,12,14,16,18,23 3,7,10,11,12,14,15,16,18,21,23,26 7,10,11,12,14,15,16,18,23,26 3,7,10,11,12,14,16,18,23,24,26  These results give a median of 49.  C e l l 6: Gains 49 60 61 49 60 30.5 41.5 54.5 50 56  _ 267  Observed Frequencies Cell 4  Cell 5  Expected  Cell 6  Above M  1  5  6  Below M  _9  _5  10  10  Cell 4  Frequencies Cell 5  Cell 6  12 Above M  5  5  5  _4  18  5  5  5  10  30  Below M  df = 3 - 1 = 2  e = (1-5) 5  2  + (5-5) 5  2  + (6-5) 5  2  + (9-5) 5  2  + (5-5) 5  2  + (4-5) 5  2  = lji + 0 + l + l_6 + 0 + l_ 5 5 5 5 = 6.8 which i s s u f f i c i e n t l y high to r e j e c t the n u l l hypothesis at p = .05. /N.B.  Firther tests between 4 and 5, and then between 5 and 6 show  that for 4 and 5 the n u l l hypothesis i s rejected at p = 0.01, but between 5 and 6 the n u l l hypothesis i s not rejected.  This means that  there are s t a t i s t i c a l l y s i g n i f i c a n t differences i n tolerance f o r r i s k between individuals i n c e l l s 4 and 5, but not between individuals In c e l l s 5 and;^/  C e l l s 4 and 5 median = 46  - 26.8 -  Obs. Freq. Cell 4  Exp. Freq.  5  Cell 4  5  Above Med.  2  8  5  5  Below Med,  8  2  5  5  /(2-5) - . 5 ) + 7(8-5) 2  = 2.5 5  ;72 .51  x 4  Z  = 5.00 C e l l s 5 and 6 median = 52 Obs. Freq. Cell 5  6  4  5  Above Med.  Exp. Freq. Cell 5  Below Med. X  2  = ((4-5)  .5) •+ 0 + ((4-5)  ,5)  2  6  5  5  5  5  + 0  = .1  Chi Squ. Median Test on c e l l s 7, 8, 9 Table 2c Subject No.  C e l l 7: Gains  C e l l 8: Gains  1 2 3  56.5 68.5 33.5  76.5 63 59  4 5 6 7 8 9 10  37.5 33 60 49 45.5 53.5 67.5  70 51 50.5 50 59.5 45.5 53.5  C e l l 9: Bets Played 7,11,12,14,16,18,20,26 1,4,7,10,12,14,16,18,23,24,26 1,3,5,7,8,10,11,12,14,16,18,20,23,24> 26 1,4,7,10,11,12,14,16,18,26 3,5,7,10,11,12,14,15,16,13,23,24,26 1,3,7,10,11,12,14,15,16,18,23,26 1,5,7,10,12,14,15,16,18,23,26 1,3,6,10,12,14,16,18,23,24,26 3,4,5,7,10,11,12,14,15,16,18,23,24,26 4,7,10,11,12,14,15,16,17,18,23,24,26  C e l l 9: Gains 44.5 62.5 78.5 56.5 64.5 61 64 62.5 68 58  -  2.6.9 -  These numbers g i v e a median of 58.5. Observed F r e q u e n c i e s Cell 7  Cell 8  Expected  Cell 9  Cell  7  Frequencies Cell  8  Cell 9  Above M  3  5  7  15  Above M  5  5  5  Below M  _7  _5  _3  15  Below M  5  5  5  10  10  10  30  df  =3-1=2  use X  2  J (o-e) e  =  = (3-5) 5  2  2  + (5-5) 5  2  + (7-5) 5  2  + (7-5) 5  2  + (5-5) 5  2  + (3-5) 5  2  = 4 + 0 + 4 + 0 + 4 + 0 + 4 5 5 5 5 = 3.2 which i s not h i g h enough f o r us t o r e j e c t the n u l l i.e.  the d i f f e r e n c e s i n t o l e r a n c e f o r r i s k  statistically  i n the 3 c e l l s  significant.  C e l l s 7 and 8 median =53.5 Obs. F r e q .  Exp. F r e q .  Cell 7  8  Above M  4  5  5  5  Below M  6  5  5  5  X  2  Cell  7  8  = ((4-5) - . 5 ) + 0 + (4-5) - . 5 + 0 = 0 . 1 2  C e l l s 8 and 9 median =59.5  2  hypothesis,  i s not  270  Obs. Freq Cell 8 Above M  3  Below M  7  X  2  9 7 3  = ((3-5) - . 5 ) = 1.5 5  Z  Exp. Freq. Cell 8  9  Above M  5  5  Below M  5  5  2  + 4  = 1.8  Proposition 3 . The n u l l hypothesis being tested i s that the median amount of gains i s the same for each case. Table 3a Subject No.  Gains: C e l l 1  1 2 3 4 5 6 7 8 9 10  Gains: C e l l 4  11 11 22 22 11 11 14.5 22 15 0  Gains: C e l l  37 33 40 41 42 63 46 49 40'.,5 40..5  56.5 68.5 33.5 37.5 33 60 49 45.5 53.5 67.5  These r e s u l t s give a median of 37.5 Observed Frequencies  Expected  Cell 1  Cell 4  Above M  0  7  7  Below M  10  _3  _3  10  10  10  10  30  Cell 7  Frequencies  Cell 1  Cell 4  14 Above M  5  5  5  Below M  5  5  5  Cell 7  - 271 df = 3 - 1 = 2 X  2  Z  2 = (c-e) Z  (0-5) + (7-5) + (7-5) + (10-5) 2  2  = 5 + _4 + 4. + 5 + 5 5  2  2  + (3-5) + (3-5) 2  2  _4+_4 5 5  = 13.5 which i s high enough to reject  the n u l l hypothesis at p = .01. 2  [N.B.  A c h i squ. test on c e l l s 4 and 7 (median = 44.5) gives an X  1.8 which means we cannot reject  the n u l l  hypothesis.]  C e l l 1 cf C e l l 4 median = 20.5 Obs. Freq. 1  Exp. Freq•  4  1  4  Above M  0  10  5  5  Below M  10  0  5  5  X  2  = 4.5  2  x 4  5 = 16.1  C e l l 4 cf C e l l 7 median = 44.5 Obs. Freq.  Exp. Freq•  Above M  3  7  5  5  Below M  7  3  5  5  X  2  = ((3-5) - , 5 ) x 4 5 2  = .45 x 4 = 1.80 •  of  - 272 Chi Square Median Test on c e l l s 8, 5, 2 The n u l l hypothesis being tested i s that the median amount gains w i l l be the same f o r each c e l l . Table 3b Subject No.  Cell 2  Cell 5  Cell 8  1 2 3 4 5 6 7 8 9 10  10 5.5 16.5 11 12.5 5.5 22 9 19 8  64.5 62.5 46.5 53.5 44.5 60 52.5 61 49 51.5  76.5 63 59 70 51 50.5 50 59.5 45.5 53.5  A cursory glance at the entries i n c e l l s 2 and 5 indicates the n u l l hypothesis w i l l be rejected at p = 0.01 as the largest entry i n c e l l 2 i s smaller than the smallest entry i n c e l l 5. C e l l 2 cf C e l l 5 . median  =22 Obs. Freq;  Exp. Freq.  Above M  0  10  5  5  Below M  10  0  5  5  X  2  = 4.5 5  2  x 4 = 16.1  C e l l 5 cf C e l l 8 median = 49  - 273 -  Obs. Freq.  Exp. Freq.  2  5  8  Above M  0  6  9  5  5  5  Below M  10  4  1  5  5  5  X  2  =5^+1+16+5^+1+16 5  5  5  5  5  =10+6.8=16.8.  5  Between c e l l s 5 and 8, the median i s located at 53. Observed Frequencies Cell 5  Cell 8  Above M  4  6  Below M  _6 10  ~  5  _4  10  5  5  10  20  5  = 2(4-5) - .5T  2  5  Cell 8  5  k  x  Cell 5 10 Above M  df = 2-1 = 1 X = /(o-3) 2  Expected Frequencies  Below M  .57  2  ~  + 7(6-5) - .5_7 + /T6-5) - .5_7 2  5  2  5  + 7(4-5) -  .57  2  5  = .2 which i s too low to reject the n u l l hypothesis, i . e . there are no s t a t i s t i c a l l y s i g n i f i c a n t differences between c e l l s 5 and 8.  Chi Square Median Test on c e l l s 3, 6, 9 The n u l l hypothesis being tested i s that the median amount of gains w i l l be the same f o r each c e l l .  - 274 Table 3b S u b j e c t No.  Cell 3  Cell 6  Cell 9  1 2 3 4 5 6 7 8 9 10  29.5 26 26 44.5 18.5 36.5 18 29.5 40.5 29.5  49 60 61 49 60 30.5 41.5 54.5 50 56  44.5 62.5 78.5 56.5 64.5 61 64 62.5 68 58  These s c o r e s g i v e a median at  49.5  Observed F r e q u e n c i e s  Expected  Cell 3  Cell 3  Cell 6  Cell 9  Frequencies Cell 6  Cell 9  Above M  0  6  9  15  Above M  5  5  5  Below M  IP.  _4  _1  15  Below M  5  5  5  10  10  10  30  df  =3-1=2  X  =  (c-e) e  ±  = (0-5) 5  2  + (6-5) 5  = 5 +1+16 5  2  + (9-5) 5  2  + (10-5) 5  2  + (4-5) 5  2  +  (1-5) 5  2  + 5+1+16  5  5  5  = 16.8 which i s h i g h enough to r e j e c t  the n u l l h y p o t h e s i s  at p =  .01  level. [The  c h i square median t e s t was  done on c e l l s 6 and 9 and gave 1.8 which  i s not s i g n i f i c a n t enough to r e j e c t the n u l l  Cell 3 cf Cell 6 median = 41  hypothesis.]  - 275 -  Obs. Freq. Cell 3  Exp. Freq. 6  Above M  1  9  5  5  Below M  9  1  5  5  X  2  = ((1-5) - 5 ) x 4 = 3.5 5 5 2  2  x 4 = 9.7  C e l l 6 cf C e l l 9 median = 58 Obs. Freq.  Exp. Freq  Above M  3  7  5  5  Below M  7  3  5  5  X  2  = ((3-5) - . 5 ) 5 = 1.5 5  2  = 1.8.  x 4  2  - 276 -  APPENDIX V:  Scale for Wealth S t r a t i f i c a t i o n for A l e r t Bay G i l l n e t t e r s  Unit  LW  MW  HW  2,500*  +3,000  +6,000  + 1 dependent  3,000  +3,000  +6,000  + 2 dependents  3,500  +3,000  +6,000  + 3 dependents  4,000  +3,000  +6,000  +4 dependents  4,500  +3,000  +6,000  + 5 dependents  5,000  +3,000  +6,000  + 6 dependents  5,500  +3,000  +6,000  + 7 dependents  6,000  +3,000  +6,000  + 8 dependents  6,500  +3,000  +6,000  1 ingle Man  [*  Annual Income from a l l sources, not just  fishing.]  CD  ADDENDUM The concern with methodological issues i n this thesis i s largely due to the view I have of the p o t e n t i a l development of economic anthropology.  A viewpoint which i s not as yet widely accepted.  However,  I am a l i v e to the fact that my concern with "new d i r e c t i o n s " has a number of d i f f i c u l t i e s , as the methodology of moving from one s i t u a t i o n to another creates a number of problems with regard to conceptualization and data c o l l e c t i o n . In this addendum I intend to discuss b r i e f l y a number of problems dealing with l e v e l of analysis, comparability of data c o l l e c t e d in d i f f e r e n t contexts, and l e v e l of data analysis, r e s p e c t i v e l y . From an i n i t i a l substantive concern with Third World farmers making decisions on a g r i c u l t u r a l innovations  I move to a higher  level  of generality and make statements about the way i n which p a r t i c u l a r individuals w i l l take r i s k s .  The constraints upon r i s k taking  specified at this second l e v e l of generality then require operational d e f i n i t i o n i n s p e c i f i c substantive contexts to assess whether p a r t i c u l a r categorized individuals take r i s k s i n the manner predicted. As part of the cat-egorization process a state of nature conceptualization was used whereby every decision maker could be c l a s s i f i e d according to t h e i r location on two axes -- (a) subjective u t i l i t y and,  (b) wealth (with regard to a stated outcome.)  This  typology placed a decision maker i n one of nine c e l l s , and was a necessary prerequisite to ordering the phenomena dealt with i n the dissertation.  Only three of these c e l l s were considered  MW/HU, HW/HU) as the d e f i n i t i o n of r i s k used precluded  (LW/HU, consideration  (ii) of any c e l l other than High U t i l i t y .  (Risk taking refers to behavior  i n s i t u a t i o n s where there i s a desirable goal and a lack of c e r t a i n t y that i t can be attained, with attendent p o s s i b i l i t i e s of loss.) Thus the entire typology was not used, and i n a sense could not be used, as the task of explicating the c h a r a c t e r i s t i c s of every c e l l would require more than one t h e s i s .  This does not, however, deter  from the value of a state of nature c l a s s i f i c a t i o n , as this typology could be used for a v a r i e t y of problems i n addition to those of r i s k taking. Data c o l l e c t i o n was  i n terms of operationally defining the  scope conditions i n d i f f e r e n t contexts.  The scope conditions  considered  here place an i n d i v i d u a l decision maker within parameters of resources and subjective u t i l i t y with regard to some outcome, information incentive conditions for any r i s k .  and  The propositions predict the  type of decision strategies that would be employed f o r given values of the above parameters.  Certain problems existed i n the measurement  of incentive conditions.  In the laboratory a l l four incentive  conditions were measured i n each decision s i t u a t i o n , whereas i n the f i e l d I measured u t i l i t y functions rather than discrete incentive conditions. attached  (A u t i l i t y function i s the p r o b a b i l i t y times the value  to the outcome). However, this was  s t i l l be adequately tested.  not a drawback as the I was  propositions'could  simply concerned with two  things.  Given that the u t i l i t y function for gains ( i . e . Ps ,Us) attached  to  f i s h i n g i n area A i s greater than that f o r f i s h i n g i n area B, what does the fisherman do -- f i s h A or f i s h B.  The second aspect i s , i s his  (iii) decision to f i s h A or B a function of the r e l a t i v e tolerance f o r losses (the negative u t i l i t y functions). From these two considerations I could i n f e r whether a fisherman took a High or a Low Risk i n each decision s i t u a t i o n . Comparability between f i e l d and laboratory contexts was effected by establishing that the objective nature of the duplex gambles corresponded c l o s e l y to the subjective evaluation of them (in ranking the 27 duplex gambles into categories of Low, Medium and High Risk there was 80% agreement among a sample of subjects). was an important point as the incentive conditions recorded f i e l d corresponded to each fishermans unique payoff  This  i n the  matrix.  On a more general level', comparability i s v a l i d provided i t can be shown that the relevant scope conditions are present.  Risk i s  defined i n abstract terms as a decision s i t u a t i o n characterized by desirable goals, uncertainty as to t h e i r achievement and a prospect of loss.  To explain v a r i a t i o n i n r i s k I use the parameters of information,  wealth and incentive conditions.  Any s i t u a t i o n (whether i f be an  investment i n a larger boat, a decision to put out a net or to take a p a r t i c u l a r bet) as long as i t has the above c h a r a c t e r i s t i c s can be defined as r i s k . The relevance of the explanatory parameters i s assessed'in terms of the operational d e f i n i t i o n of the parameter with regard to context.  This means that the operational d e f i n i t i o n of information  with respect to investment r i s k s w i l l be d i f f e r e n t to the operational d e f i n i t i o n of information with respect to strategy r i s k s .  I emphasize  that this i s a substantive difference and not an a n a l y t i c one.  (iv) In terms of data analysis, the "new  d i r e c t i o n s " f o r economic  anthropology implied i n this work suggest a battery of formal analysis that I do not possess.  My own mathematical expertise i s limited.  Perhaps my methodological ambitions outstripped my command of the r e q u i s i t e technologies. I would submit, however, that I have proceeded consistent with a concern with "new  i n a manner  d i r e c t i o n s " i n economic anthropology.  In methodological terms I have demonstrated of borrowing analogies from economics.  the usefulness  Also the testing procedure  employed has implications f o r the manner i n which anthropologists may conduct enquiry, as i t suggests that laboratory contexts are as legitimate a source of v e r i f i c a t i o n f o r anthropological problems as f i e l d contexts. It i s from these two considerations that I o f f e r a test case for methodology i n economic anthropology.  - 14 -  Though the d i s c u s s i o n has moved from c h o i c e to exchange i t r e s t s on the same assumption of c a l c u l u s o f m a x i m i z a t i o n a l i t y i n choice. s i o n theory  and  ration-  I t a l s o assumes t h a t the t o o l s of economics —  and m a x i m i z i n g models —  can a l s o be the t o o l s of  deci-  the  anthropologist. These are the " i n p r i n c i p l e " p o s s i b i l i t i e s , and a number o f a n t h r o p o l o g i s t s have endorsed the i d e a of u s i n g f o r m a l d e c i s i o n t h e o r y to s o l v e a n t h r o p o l o g i c a l problems ( B u c h l e r and N u t i n i 1969; But the r e s u l t has  Davenport 1960).  u s u a l l y e i t h e r been a d i s c u s s i o n of what  ought to be done by a n t h r o p o l o g i s t s i n u s i n g f o r m a l models or a s u r f e i t ' i of mathematics a t the expense of r e l e v a n c e  to a n t h r o p o l o g y .  between a n t h r o p o l o g i c a l d a t a and f o r m a l models has  A balance  to be k e p t c l e a r  i n t h a t more s t r i c t l y a n t h r o p o l o g i c a l c o n s i d e r a t i o n s s e r v e to i d e n t i f y the c o n s t r a i n t s and parameters w i t h i n w h i c h d e c i s i o n models can  be  applied . T h i s means t h a t e t h n o g r a p h i c systems and c u l t u r e s p e r m i t s  one  d e s c r i p t i o n of p a r t i c u l a r  to d e s i g n a t e  a l t e r n a t i v e s t h a t are a p p r o p r i a t e  the c u l t u r a l l y  perceived  to p a r t i c u l a r d e c i s i o n s i t u a t i o n s ,  and to demarcate the p r i n c i p l e s w h i c h are d e t e r m i n a t e ( o r seem to for  choosing And  be)  between the a l t e r n a t i v e s . i t i s i n t h i s way  t h a t the a n t h r o p o l o g i s t , w i t h h i s  d i t i o n a l t o o l k i t of a n a l y s i n g v a l u e systems and ways of doing  value  culturally  appropriate  t h i n g s , can make an enormous c o n t r i b u t i o n t o the  c u l t u r a l study of d e c i s i o n making.  tra-  cross-  - 19 -  CHAPTER 2 THE PROBLEM AND  ITS CONCEPTUALIZATION  Introduction From my i n i t i a l substantive concern with Third World farmers I generate a number of propositions which l i n k the decisions made by a subsistence farmer about a g r i c u l t u r a l innovations to his state of resources, incentive conditions attached to an innovation, and information and u t i l i t i e s  attached to the outcome "increase i n productivity."  From these statements s p e c i f i c to peasant farmers, I generalise to three statements about individuals and r i s k taking.  In the interests  of parsimony i t would perhaps be better to start with the general statements and proceed immediately to testing procedures.  However, given the  lack;;of precedents for the type of methodology implemented here and the explicational nature of my research, I think i t i s necessary to show the step-by-step progression from one l e v e l of generality to another. I t i s also important to i l l u s t r a t e the source from which my  statements  originate.  Farmers and Innovations  The i n i t i a l substantive problem I am dealing with i s the exp l i c a t i o n of the factors that influence farmers' decisions on a g r i c u l t u r a l innovations. decision theory,  The conceptual approach I intend to use i s that of  go both my substantive and conceptual concerns are  within limited boundaries.  - 33 -  Social M o b i l i t y  By s o c i a l mobility I am r e f e r r i n g to an index based on role recruitment  p r i n c i p l e s , i . e . a s c r i p t i o n and achievement.  This rests  on a conceptualisation of roles as slots or positions in a s o c i a l structure for which individuals may  compete.  index i s with the rules of competition  My  concern with this  which define the flow or c i r c u -  l a t i o n of persons between positions. The  index measures the degree to which 'achievement role  s l o t s ' are present in a p a r t i c u l a r society.  Thus a society, with a  high index of s o c i a l mobility indicates that a greater number of people can compete for roles than in a society with a low s o c i a l mobility index. The follows.  implications of s o c i a l mobility for innovation are as  I f we v i s u a l i s e the competition  for a p a r t i c u l a r role-as re-  quiring the attainment of c h a r a c t e r i s t i c s (a  n) ; i t follows that  persons aspiring to that role have to acquire those c h a r a c t e r i s t i c s . Societies low in s o c i a l mobility are characterised by exclusion rules which prevent everyone i n the population prerequisites for p a r t i c u l a r r o l e s .  from acquiring the necessary  Societies high in s o c i a l mobility  do not have these b a r r i e r s . The  c h a r a c t e r i s t i c s (a  expected behavior patterns and, and can be acquired  n) may  consist of prestige symbols,  for want of a better word, 'entry fees';  i n a variety of ways.  I f prestige symbols have to  be attained, this means that an i n d i v i d u a l has to involve himself i n  

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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0103965/manifest

Comment

Related Items