Frank Cancian, “Stratification and Risk-Taking: A Theory Tested on Agricultural Innovation,” American Sociological Review 32, No. 6 (Dec. 1967), 912-927.

---

Notes

• 913 – Cornell University – “Much of the work on this paper was done during the tenure of a Postdoctoral Foreign Area Fellowship in Latin American Studies. My own data on Zinacantan were gathered during the summer of 1965 under a grant from the Morrison Fund of Stanford University. Previous work on Zinacantan was done under the Havard [sic] Chiapas Project which is sponsored by NIMH . . ..”
• 913-914 – “It is usually asserted that rich farmers adopt new farming practices more readily than do poor ones. The question of the linearity or nonlineaerity of the relationship is normally left open, but in the absence of detailed data to the contrary, the most common assumption is that the relationship is approximately linear — the wealthier the farmer the more likely his is to adopt an innovation.”
  • Explicit statement of assumption of relationship between wealth and receptivity to innovation
• 914 – “In this paper I will try to show that: (1) the relationship between wealth and innovation is more complex than is usually assumed; (2) insofar as adoption depends on willingness to take economic risks, the relation between the variables is negative rather than positive; and (3) empirical studies will show the relation to be curvilinear rather than linear.”
• “A theory which predicts an inverse relationship between wealth and early adoption will be presented, and then modified to take account of: a. Factors that counteract the relationship predicted by the theory; and b. Factors that tend to make the relationship curvilinear.”
• “This view, and the assumptions on which it is based, may be generalized and stated in a simple ‘theory’ of stratification and risk-taking:// Propositions:// In all societies, individuals (or classes of individuals) will be ranked from high to low in direct relation to their possession of any resource valued in the society.// Individuals in all societies would rather be high than low on any such ranking.// Definitions:// Risk is a charcteristic of sitatuations of exchange in which the rate of return on investment of resources is uncertain: the greater the uncertainty, the greater the risk.// For individuals of different ranks, equal risk is represented by investment of an equal proportion of the total resources of each individual under conditions of equal uncertainty.// Conditions: 1. All risks (uncertain investments of resources) are perfectly divisible.// 2. Knowledge is equally spread over all ranks.// 3. The risk necessary to maintain present rank is equal, as a proportion of total resources, for individuals of all ranks.// 4. No individual can suffer total loss of resources from loss on a single risk.// 5. No individual has so many more resources than the next lower relevant individual, or category of individuals, that he is completely protected from loss of rank.”
  • What does the attempt to restructure status in society along mater-economic lines, ignoring social and cultural status elements, say about Cancian, et al.’s underlying assumptions
  • Does this represent an attempt to laboratize the human wilds, to control all the variables and conditions?

 

• “These conditions often will not be met in natural situations. The implications of their not being met are the focus of the discussion in the sections on modification of the theory.”

• “Given the foregoing propositions, definitions, and conditions, it follows that individuals of higher rank will risk less than individuals of lower rank.”

• 914 – “Despite this relatively formal statement, this assertion does not pretend to be a highly formalized theory in the sense that the general hypothesis about rank and risk-taking is derivable from the initial propositions. The general hypothesis is, rather, an interpretation of the second proposition based on the idea that insofar as an individual holds high rank, he will act to conserve it, and insofar as an individual holds low rank, he will act to improve it. The conditions ‘protect’ this interpretation from some of the more obvious situations which would make it empirically false. While the general hypothesis, together with the two major modifications which follow, moves in the direction of theory in the more formal sense, this paper is intended as an exploration of the complexities involved rather than as a final statement of such a theory. Emphasis has been given to this first of the three elements of the approach because it is the key to the difference between the overall interpretation and more usual ones.”

 

• • So much going on here

 

• “The ‘theory’ has been stated in very general terms and is meant to apply to any area in which people are ranked by possession of a resource, and in which change in rank is related to risk of that resource. While the discussion from this point on often will be stated in terms of agriculture, I hope that the observations and hypotheses will have applicability in other domains, and that parallel modifications and parallel hypotheses might eventually be developed for those domains. Hopefully, generalization might move in the direction of resources other than economic ones, for example, social positions and power; and, while it is not provided in this statement, observations about the individual as the actor might be generalizable to firms and other supra-individual actors.”

• To what extent does this paper play into later conversations about the ‘Risk Society’, as an attempt to develop a general model about the relationship between risk and innovation. Note, especially, that this model was calibrated on examples of agricultural innovation, and was designed to be tested against scenarios of agricultural innovation

• “In the application of the theory to agriculture, it is assumed that wealth is the valued resource and that early adoption of a new agricultural practice is a risk, for the results it will produce are uncertain. Thus, the theory predicts that the inclination of a farmer to innovated decreases as his wealth increases. This inverse relationship between wealth and early adoption of innovations will be called the ‘inhibiting effect’ of wealth.”

 

• “Modification of the Theory: the Facilitating Effect. If the conditions stated above are met, the relationship of wealth and adoption of agricultural innovations will be linear and negative. However, in most natural situations (including agricultural) it is highly unlikely that conditions 1 and 2 will be met.”
• “It is also highly unlikely that agricultural innovations are perfectly divisible so that every farmer can afford to adopt in proportion to his wealth.”
• “Other factors, like the absolute cost of obtaining information about innovations, and the impracticality of certain innovations (like a tractor) on small farms may be seen as restraining the poorer or smaller farmer in a similar manner.”
  • So with these modifications and acknowledgements of messiness, what we have in this paper is the creation of TWO models, one an ideal, and the other an attempt to systematize deviations from the ideal, with the acknowledgement that the ideal is never found in reality.
• 914-15 – “It is also improbably that knowledge will be spread equally over all wealth ranks in an agricultural community. More probably, the richer a farmer is, the more information he will have.”
• 916 – “We may assume that when a new practice has been adopted by a substantial proportion of a population, knowledge about it will spread, and, consequently, the risk involved in adoption will be lowered. At this point, the practice is no loner an innovation, and inclination to risk is no longer a major element in the decision to adopt. Thus, the adoption process may be seen as having at least two stages: Stage 1 in which inclination to risk is important; and Stage 2 in which inclination to risk is substantially less important. The overall theory, since it is heavily dependent on arguments about inclination to risk, is principally applicable to Stage 1 of the adoption process.”
• 916n9 – “That is, risk based on incomplete knowledge of what is known by technicians about the innovation will be lowered. Risk due to lack of total technical control, climate and similar factors may remain constant, as it does for even very old agricultural techniques.”
• 919 – [SECTION: DATA AND TESTS]
• “Data from seven studies will be used to test the hypotheses developed above. One case is my own study of Zinacantan, Mexico, which will be described in more detail. Data for four other cases were kindly provided by scholars whose publications did not include the detail needed to test the hypotheses. The final two cases came from a publication and an M.A. thesis.
• 919n12 – “Acknowledgement is also due: The W. K. Kellogg Foundation, the North Carolina AES and the Cooperative Extension Service, North Carolina State College; and the Department of Agricultural Economics AES, University of Illinois, the Food and Agriculture Organization, the International Christian University, Tokyo, Japan, and the Ministry of Agriculture, Government of Japan.”
• 920 – locations of samples – Mexico, North Carolina (x2), Wisconsin, Iowa, Japan, Kentucky
  • Do the variations in locations mean anything to the applicability of the generalizability of the model?
• “The theory was originally developed during work on Zinacantan.”
• “Zinacantan is a community of 7,650 (1960 census) Tzotzil-speaking Maya Indians.”
• “The sample (N=93) includes all married males, 27 years old and older, who are independent corn farms and live in the hamlet of Apas. Amount of corn seeded on rented lands in the low, hot country near Zinacantan is used as the measure of the wealth variable.”
• “Sale of corn (surplus over family needs) to receiving centers recently established by a government-created agency was used as the adoption variable. . . . [B]y the 1964 seed season 48 of the 93 farmers in the sample were selling at least part of their crop there.”

 

• The innovation being examined is the sale of surplus to a receiving center — whether this is used as an approximation of the adoption of other ‘technologies’ is unknown

 

• 926 – “In this paper the obvious importance of the facilitating effect as a modification of the original theory is simply affirmed. No attempt is made to add to what is already known about it — except, of course, that the demonstrated importance of the curvilinear and inhibiting effects reduces the importance the facilitating effect has appeared to have. That is, insofar as the facilitating effect represents that theoretical basis for the commonly reported positive, and presumably linear, relation between wealth and adoption of agricultural innovations, the present study suggests drastic modification of its status.”

 

• “Conclusions: The Overall Theory. On the whole, the predications derived from the overall theory have been confirmed. The relation of wealth and adoption of new agricultural practices clearly has both curvilinear and negative aspects. On the other hand, the overall theory itself is a conglomeration of three effects derived in different ways from first principles on different levels of generality. Further testing on agriculture and other domains may result in substantial changes in the overall theory.”

• “First principles” “Further testing”

• The Original Theory. The support for the applicability of the original theory in the middle ranks is, it seems to me, impressive enough to warrant further research. Direct tests may be possible in laboratory situations. If other tests in natural situations require the extensive modifications that were necessary in the application to agriculture, the original theory may be, in the long run, more valuable as a statement of conditions that must be attended to than as a basis for direct predictions about natural situations. Its ultimate status will, of course, also depend on its applicability to situations where the relevant resource is social position or some other ‘non-material’ thing valued by individuals living as members of a society.”

• Scrutinize what “laboratory conditions” means in this case

 

• 927 – “This suggests that theories of risk-taking may be more important than theories of information diffusion for the study of the earliest stages of the process of adoption of new agricultural practices.”