AGRICULTURAL INTENSIFICATION 

[Note: for Oct. 24^th class, read "Intro" plus section on "Swidden Systems";  for Oct. 26^th, remainder (section on "Agricultural Intensification"); no additional lecture notes for Oct. 29.7^th]

Intro

Agricultural systems come in many varieties, and can be classified many different ways

One of the most useful classifications for ecological analysis is in terms of intensification

Intensification refers to any practice that a) increases productivity per unit land area at b) some cost in labor or capital inputs

Agricultural intensification takes many specific forms, including irrigation, fertilization, use of draft animals or machinery to till soil, etc.

One important dimension of agricultural intensification = length of fallow period

(Note: "fallow" = letting land lie uncultivated for a period)

Some agricultural systems have no fallow period (instead, continuous cropping), but most do

Influential classification is that of Ester Boserup (Danish agroeconomist specializing in third-world agriculture):

Fallow Type	Cropping Period	Fallow Period
Forest fallow	1-3 years	20 years or more
Bush fallow	1-8 years	6-10 years
Grass fallow	Several years	1-2 years
Annual cropping	A few months	Less than 1 year
Multi-cropping	Continuous	None

Length of fallow found in any particular agricultural system will depend on several factors (latitude, climate, soils, etc.), and obviously some limits are hard to override (e.g., can't have multi-cropping if growing season is too short)

However, environment usually sets only broad limits, and specific fallow system that develops may often reflect adjustments made by farmers in response to various economic and social factors

Boserup's key argument is that sequence from long-fallow to shorter and shorter fallow periods is to some degree historical progression in response to population growth (see below)

SWIDDEN SYSTEMS

If Boserup idea of shortening fallow being an historical trend is valid, then the earliest and least intensive form of agriculture = long-fallow ("forest-fallow")

This system often termed swidden or slash-and-burn or shifting cultivation; common in tropics, but once much more widespread (incl. prehistoric Europe)

Typical features of swidden systems:

1) small areas of forest or brush cleared with slash & burn method (releases nutrients held in plant tissues)

2) soil prepared with dibble stick or hoe (no plowing)

3) human labor provides full work inputs (other than fire)

4) no use of fertilizers (other than wood ash)

5) no irrigation systems (rainfall watering)

6) usually multi-crop gardens, high plant diversity (e.g., up to 40 species in single garden of Hanunoo in Philippines)

7) short cropping period (1-3 yrs) followed by long fallow (>20 yrs)

8) hence most land lies fallow at any one point, requiring large amount of land per capita

Because most agronomists, economists, and anthropologists come from societies based on intensive (short-fallow) systems, swidden has often been portrayed as wasteful, ignorant, "primitive," and environmentally destructive

(In addition, swidden peoples typically self-sufficient, hence looked down on by gov't agencies, who are interested in extracting agricultural surplus for taxation or trade)

However, starting in 1950s, anthros and geographers began analyzing swidden systems carefully, and found that as practiced by indigenous or long-resident populations they are sophisticated and sustainable adaptation in many environments

First, reliance on swidden agriculture surely not a sign of environmental ignorance:

1) successful shifting often requires great knowledge of soils, plant types, and cropping regimes (e.g., Hanunoo categorize 30 soil types, 430 cultigen varieties)

2) gardens look "chaotic" to ethnocentric observers, but multi-species planting may prevent soil erosion, conserve moisture, facilitate soil nutrient use (e.g., legumes), and reduce pests (high diversity)

Second, "wasteful" only if land is scarce and suitable for more intensive forms of agriculture, but:

1) tropical soils are often thin, easily exhausted if cropped repeatedly, or even if exposed to sunlight & rain for any length of time

2) swidden systems have low yield/unit land (if we incl. fallow areas), but yield/unit labor often higher than more intensive forms (see "Intensification" section, below)

3) abandoning a swidden after 1-3 years makes sense because yield/unit labor declines sharply, due either to decreased soil fertility or to increased weeds and insect pests

Thus, it appears that swidden systems are quite adaptive at low population densities, especially in tropical areas with poor soils where shift to more intensive forms of agriculture would lead to irreversible environmental degradation

However, evidence also indicates that shortening fallow periods in response to land scarcity (induced either through population growth or unequal distribution of the land) can lead to permanent environmental degradation; this is not inevitable, however -- depends on both environmental conditions and agronomic practices

In any case, widespread evidence that shortening fallow has marked effects on both land productivity (yield/unit area) and labor efficiency (yield/unit labor time) -- topic of the following section

AGRICULTURAL INTENSIFICATION

Introduction

Just as shift from H-G systems to agricultural ones has traditionally been viewed as product of increasing knowledge and cultural "progress," so historical progression toward more intensive and productive agricultural systems has been viewed as inevitable result of increasing knowledge and technological abilities

Thus, orthodox view long held that development of intensive forms of agriculture could be attributed to major technical inventions (metal tools, terracing, the plow, oxen teams) or to increased knowledge (fertilizing, breeding of draft animals or more productive forms of crops)

However, orthodox view of agricultural "progress" has been strongly criticized:

1) Many intensive systems based on simple tools and hand labor

2) intensive systems not always adopted even when readily available (e.g., persistence of swiddeners, resistance to imported schemes of agricultural "development")

3) some instances of "reversion" to less intensive systems (discussed below)

Yet plenty of evidence that degree of agricultural intensification is correlated with population density and both technological and socioeconomic complexity

Basic contrasts between extensive & intensive forms of agriculture are summarized in following table:

Table 2. Contrasts between extensive and intensive agricultural systems.

Characteristics	Extensive Systems	Intensive Systems
Fallow length	Long	Short
Productivity	Low	High
Efficiency	High	Variable but lower
Population density	Low	High
Technology	Simple	Often complex
Fertilizing of soil	None or little	Lots
Land tenure	Communal ownership	Individual/family
Economic systems	Usually subsistence	Usually market
Sociopolitical complexity	Generally less	Generally greater

If not increased knowledge and cultural "progress," then what accounts for this pattern?

Boserup's Theory of Agricultural Intensification

Major alternative explanation offered by Boserup in her book on The Conditions of Agricultural Growth (1965)

Recall Boserup's argument that classification of agricultural systems by fallow length = historical or evolutionary progression towards increased productivity per unit land (leading to increased population density), a process termed intensification (recall definition above)

Essence of Boserup's argument can be summarized in terms of 3 axioms & 6 corollaries):

I. Environmental limits on type of agriculture are relatively "elastic" (flexible, broad)

1) Any region has broad limits of productivity ("elastic carrying capacity"), set by technology and cropping methods

2) People choose among available methods & degree of intensification (they are not strictly limited by knowledge, especially in the long run)

II. Increased production/area usually leads to decreased yield / labor

3) Most means of intensification have major costs:

These include shorter fallow (with concomitant fertility decline), use of draft animals & plowing, fertilizer application, more labor inputs (incl. child labor), mechanization, and engineering (irrigation, terracing, etc.)

4) Thus intensification generally characterized by decreased yield/unit labor (i.e., reduced labor efficiency), but increased yield/unit land (i.e., increased productivity), "all else being equal"

III. Labor minimization governs decision to intensify

5) People's labor decisions follow "Law of Least Effort"; that is, people tend to work as little as possible to satisfy their needs or desires (in this case, food needs)

6) Hence, can predict that people will only intensify when forced to do so by increasing needs or demands.  Specifically, predict:

a) Population growth ® land scarcity ® intensification (technoeconomic innovations) ® increased yields/unit area paid for by increased labor inputs/unit yield (longer work hours)

b) New economic opportunities (e.g., market for cash crops) or surplus expropriation (e.g., taxes) can also motivate intensification and greater labor costs

c) Conversely, decreased population pressure or decreased taxes/markets leads to de-intensification

To expand on points 6)b and 6)c, while Boserup emphasizes population growth as main force favoring intensification, logic of her argument suggests that demands of elites and bureaucracies for taxes and tribute, as well as new opportunities for trade and cash-cropping provided by expanding market systems, should lead to intensification just as readily as does population pressure

In any case, all of the factors predicted to cause agricultural intensification have become much more prevalent since development of politically centralized & stratified societies ("civilizations"), especially in last 500 yrs of colonialism & capitalist expansion -- which could well explain the pattern laid out in Table 2

So why aren't all agricultural systems intensive by now?  At least 3 reasons:

1) Environmental limits on intensification (climate, soils, etc.)

2) Non-agricultural limits on population growth (disease, warfare, etc.)

3) Remoteness from markets or agents of governmental expropriation

Evidence re Boserup's Theory

Several types of evidence bearing on Boserup's theory

However, in most cases we lack sufficient controls ("all else" not equal); accordingly, it may be misleading to compare yields and costs for systems in 2 different places at same time, since soil may vary, history is different, etc.

Evidence on correlation between fallow period and population density & land productivity [see Table 3]:

1) Palerm (1957) compared 3 Mexican peasant communities: swiddeners use 1.5 hectares/year/family, but need 12 hectares total over swidden cycle; short fallow option has lower yields (need 2.5 ha/yr/family) but total over fallow cycle is only 6.5 ha; irrigated land supports permanent crops with high yields, only need 0.86 ha/family

2) Comparisons of 17 rice growing societies in Asia by Hanks (1972) showed a direct relationship btwn intensification and population density, as predicted by Boserup's theory:  swiddeners had 31 people/mi.², plowers = 255/mi.², and irrigators = 988/mi.². Yield/acre cultivated about equal for swidden vs. plow, but much higher for irrigated systems; labor inputs (incl. care of animals and long term investments in canals) rise with increased intensification but so does efficiency (contrary to Boserup's prediction--though past costs of building terraces and irrigation canals not factored in)

Evidence that depopulation or decreased surplus demands leads to de-intensification:

1) Kofyar of W. Africa traditionally limited their settlements and agriculture to a high plateau, where they were safe from raids by enemy groups; have high population density (ca. 300/mi.²); colonially enforced peace allowed some Kofyar to migrate to open plains, where they reverted to swidden at low population densities; swidden gardens = less productive per acre, but average homestead has 7.7 acres (5 times as big as plateau homelands) [Netting 1977]

2) French colonists in Indochina taxed peasants; end of colonial taxation led to shift to swidden agriculture from more intensive forms of agriculture in some cases

3) Swidden methods were prevalent in early days of colonial expansion in N. and S. America, even though settlers came from European societies with intensive (short-fallow) agriculture

Some evidence on average labor inputs in relation to intensification can be seen in graph as well as the following table:

Table 4. Intensification in relation to labor time, population density, and agricultural yield.

[data from Minge-Klevana (1980) & Hames (1989); Labor time = "work outside the home," average of adult men and women; * = missing data; does not hold yields per acre, soil quality, surplus prod. for trade, taxes, etc. constant]

Conclusions

Malthusian vs. Boserupian views of relation between population and food production seem to be directly contradictory

However, recent arguments (e.g., Lee 1986, Richerson et al. 1998, Turner & Ali 1996, Wood 1998) have shown how they can be seen as complementary, each part of a more complete picture of human population-environment dynamics  [see Diagram: Malthus+Boserup]

In any case, both approaches offer major insights into ecology & dynamics of agricultural intensification, insights that are still being theoretically refined and empirically tested

References Cited

Boserup, Ester (1965) The Conditions of Agricultural Growth.  Chicago: Aldine.

Hames, Raymond B. (1989) Time, efficiency, and fitness in the Amazonian protein quest. Research in Economic Anthropology 11:43-85.

Hunt, Robert C. (2000) Labor productivity and agricultural development: Boserup revisited. Human Ecology 28(2):251-277.

Lee, Ronald D. (1986) Mathus and Boserup: a dynamic synthesis.  In The state of population theory: forward from Malthus, ed. D. Coleman and R.S. Schofield, pp. 96-103.  Oxford: Blackwell.

Hanks, Lucien M. (1972) Rice and Man: Agricultural Ecology in Southeast Asia. Chicago: Aldine.

Malthus, Thomas Robert (1798) An Essay on the Principle of Population.

Minge-Klevana, Wanda (1980) Does labor time decrease with industrialization?  A survey of time-allocation studies. Current Anthropology 21:279-98.

Netting, Robert M. (1969) Ecosystems in process: a comparative study of change in two West African societies. In Ecological Essays, ed. D. Damas, pp 102-112. Ottawa: National Museum of Canada.

Palerm, Angel (1957) The agricultural basis of urban civilization in Mesoamerica.

Richerson, Peter J. and Robert Boyd (1998) Homage to Malthus, Ricardo, and Boserup: toward a general theory of population, economic growth, environmental degradation, weatlth, and poverty. Human Ecology Review 4:83-88.

Spooner, Brian (1972) Population Growth:  Anthropological Implications.  Cambridge, MA: MIT Press.

Turner, B. L. and M. Shajaat Ali (1996) Induced intensification: agricultural change in Bangladesh with implications for Malthus and Boserup. Proceedings of the National Academy of Sciences, USA 93(25):14984-14991.

Wood, James W. (1998) A theory of preindustrial population dynamics: demography, economy, and well-being in Malthusian systems. Current Anthropology 39:99-135.