Agriculture Reference
In-Depth Information
From these agent-based theories and models emerged a perspective where net-
works of farmers and other stakeholders interact and technological innovation arises
from that interaction (Röling and Wagemakers 1998; Biggs and Matsaert 2004; Sayer
and Campbell 2004). Building on the work of Coughenour and Chamala (2000) and
Ekboir et al. (2002), Swenson and Moore (2009) demonstrated the influence of net-
works and social relations on decision-making and technology adoption by review-
ing case studies on the development of conservation agriculture, and found that
field-specific conditions and unique farm and household characteristics are only part
of the explanation for adoption of conservation agricultural practices. Successful
adaptation of these practices appears to involve vast networks of relationships that
reinforce certain sets of knowledge, beliefs, and behaviors (Davis 2013). Swenson
and Moore (2009) focused on the critical relationship between actors and ecologi-
cally specific development problems within complex networks as they resolve sys-
temic issues of input supply and delivery mechanisms, on-farm labor and biophysical
adaptation, and reliable output markets.
Further analyzing the effect of social relations on decision making, Henrich
(2001) suggested that payoffs are only part of what drives adoption; biased
cultural transmission figures prominently in the process. People make invest-
ment decisions even in the face of evidence arguing against such behavior (e.g.,
Baudron et al. 2011; Andersson and Giller 2012). Henrich (2001) argued that
Rogers' classification of adopters based on degrees of innovativeness does not
result in the S-shaped curve (Figure 14.2) that defines the diffusion process.
Instead, he noted that biased cultural transmission was the predominant force in
the process of diffusion. Biased cultural transmission used information that may
include but certainly goes beyond the innovation-evaluation information relevant
to the payoffs of any particular technology adoption. Henrich (2001) builds on
Boyd and Richerson's (1988) work, which distinguishes between “learners” who
acquire behavior through a process of experimentation, and “imitators” whose
behavior is acquired through social learning. Henrich attached this concept to
Rogers' classification of adopters based on innovativeness by categorizing indi-
viduals as innovators and imitators. His analysis of the S-curve of adoption (see
Figure 14.2) indicated that imitators are copying ideas and practices not directly
related to an analysis of costs and benefits.
Henrich (2001) argues that purely environmental learning, where decisions are
based on a cost-benefit analysis, creates R-shaped diffusion models (Figure 14.3), or
graphs where the inflection point, the moment of maximum rate of growth, occurs at
time zero. In other words, in most diffusion situations, which are best represented by
S-curves, adoption is slow until a threshold level is achieved. At this inflection point,
a threshold level of adoption has been achieved that results in a much higher rate of
diffusion. Up to this point, the rate of adoption is slow. “Long-tailed” S-curves (Figure
14.2) demonstrate the slowness of initial adoption relative to the rate at which individu-
als imitate previous adopters. Rogers (1995) recognizes that a critical mass must be
achieved to make individuals more open to adoption. Henrich explains this through
Boyd and Richerson's (1985) description of conformist transmission, which indicates
that the more frequently a trait appears in the population, the more valuable it seems.
