INFORMATION-BASED FARMING

Ricardo J Salvador rjsalvad at IASTATE.EDU
Tue Sep 29 18:05:40 EST 1992


In article <1992Sep12.210032.1713 at NeoSoft.com>, claird at NeoSoft.com (Cameron
Laird) writes:
> In article <1992Sep11.183716.1512 at u.washington.edu>
toby at milton.u.washington.edu (Toby Bradshaw) writes:

> >Along these lines, should we ban tractors because
> >they are expensive and give the large farmer higher returns than someone
> >who uses a mule to plow and a hoe to cultivate?

> Without trivializing at all the topical questions
> you raise, please note that the historical process
> of mechanization does *not* necessarily procede in
> the direction you imply here.  A number of studies
> (none of which I now have handy) of US operations
> of the early 20th century concluded that tractors
> gained ascendancy on small farms first; in crude
> terms, mechanical power is more fungible, and re-
> quires less (!) specialized care.  Teams of horses
> held out longest on farms with multiple employees,
> and of course where there were "sentimental" reasons.
> I think it would surprise you even today how expensive
> a good draft ox is, compared to the delivered price of
> a small mechanized platform.

I was curious about this a few years ago and calculated a few
estimates of my own.  I utilized a well-documented system (Barret et al.
1982.  Animal traction in eastern Upper Volta: A technical, economic,
and institutional analysis.  Michigan State Univ., Int.'l Development
Papers, No.4) and computed that, energetically, an ox costs 592 MJ per
937 MJ foodstuff produced (sorghum grain), or 63.2% of the human
metabolizable E produced.  The equivalent calculation for a maize system
yields an efficiency of 45%.  Such calculations need to be used with care.
They do not include other aspects of the utility of oxen in a subsistence
system, however these data are applicable to the example discussed here
since they are restricted to that aspect of mechanized production systems
(traction) that is directly comparable to the utility provided by tractors. 

> Related, but distinct, question:  do winners, or losers,
> adopt new technologies?
  [...]
> Which corn-and-bean growers
> in Indiana adopted minimum-till first?  I've met plenty
> of people who think they know the answers to these
> questions, reasoning from first principles, but I haven't
> yet seen a convincing, a-situational, theoretical framework 
> sufficient to these topics.

As an avid partisan of the view that early agriculturists were most 
definitely NOT innovators, but rather people who were trapped in 
situations that demanded the adoption of subsistence methods that
were less desirable than previously viable alternatives (gathering,
hunting), I agree that the related question you pose is important.
I think a useful way to approach the theoretical framework you
propose is to assume that humans in the early neolithic behaved in
just the same way that humans behave today, meaning that decisions 
are by and large made on the basis of short range criteria, are
self-interested, and that exchange patterns and behavior are non-
altruistic.  You may be interested in an interesting simulation developed
here by an economist and two mathematicians who have modified Axelrod's
Prisonner's Dilemma to deal with this issue (cf., Stanley, Ashlock &
Tesfatsion. 1992. Iterated Prisoner's Dilemma with Choice and Refusal,
Economic Report Series No. 30, Dept. of Economics, Iowa State Univ.).

The simulation procedes by allowing initially random interaction between
"agents" (modelled as finite automata) with "genetically determined"
behavior" (their behavior is defined by a "genome" and this "genome" can
mutate).  The question asked of the simulation is: "under what conditions
might repeated actions among non-related and non-altruistic agents lead
to reciprocal cooperation?"  A key aspect of the mechanism of the
simulation is that "score" is kept by awarding greater chance of survival
to paired agents who consistently cooperate than to "singles" who "defect" from
cooperative relationships.  Interesting behaviors evolve in the simulation.
There is for instance the "tit-for-tat" individual who does exactly what
is done to her; there are those who either always cooperate or always
defect, there are those who are forgiving and will forgive one defection
from a partner, but not two ("tit-for-two-tats") and there are parasites
such as the "ripoff artist" who take advantage of "tit-for-two-tat"
individuals by cooperating in response to cooperation OR first defection,
but defect at the second incidence of cooperativity and yet manage to
reenter the good graces of "tit-for-two-tats" by cooperating when defected
from and then behaving in "tit-for-tat" fashion.  The point is that
a complex ecology evolves from this scenario and one sees what appear to
be succesful cooperative strategies developing even though agents are
basically "looking out for themselves."  I find this simulation useful
to humanize for students the question "why did we become agricultural?" in
a course on World Food Issues.  If you're interested, Stanley may be 
contacted via e-mail as: stanley at iastate.edu; Ashlock as: danwell at iastate.edu;
and Tesfatsion as: tesfatsi at iastate.edu.

-Ricardo
   



More information about the Plantbio mailing list