Christiansen, Sofus: Work and journey to work in subsistence
agriculture - a case of cultivation of scattered areas on Rennell
Island (Mugaba). Geografisk Tidsskrift 76: 84-88, June 1, 1977.

In subsistence agriculture, area and work are usually the most important production factors. Total cultivation work is shown to be greatly increased, if journeying to work is augmented, as when cultivated areas are scattered. Formulae are devised for the increase of total work caused by transport.

On Rennell Island (Mugaba) the traditional bisettlement-based
cultivation strategy is demonstrated to be a rational means to
diminish the journey to work problem.

Sofus Christiansen, Professor, dr. phil. Geographical Institute,
University of Copenhagen, Haraldsgade 68, DK-2100 Copenhagen 0.

For a subsistence system proper, carrying capacity for population
in a given area is thought to be an important characteristic
(S. Christiansen 1976). The carrying capacity is a
property attached to the production system (which is not
necessarily regarded as static), but of course also influenced
upon by the environmental potential.

Under subsistence conditions, the main variable production input is work. Capital is usually very scarce, limiting access to tools, fertilizers, seed material etc. severely. The amount of work employed is thus largely determining output. Usually work is available in sufficient quantity in subsistence agriculture, though certain 'bottle necks' in work supply do occur especially in intensive subsistence agriculture e.g. by transplanting and harvesting in cultivation of irrigated rice.

In the formula for the calculation of carrying capacity for population: Pmax = A — where A is the area considered, y the food yield per area unit and c the food consumption per capita) work is not explicity entered. However, work is determining yields (S. Christiansen 1975), see fig. 1. If the variables are measured commonly in energy units, maximum carrying capacity is found to be achieved before maximum yields are attained, namely where output less work (input) is at maximum.

Hence a revised carrying capacity formula can be written:

Pmax =A w, where cw is the consumption resulting from the work used in production of food. Of course this means that any further increase of work in production diminishes the possible maximum population - here most easily conceived as the maximum biomass in kilogrammes sustainable by the production of the given area. It is noticed that life for the biomass thus assessed is quite drab; there is only energy available to sustain the biomass at rest (disregarding work in subsistence production). Naturally, when calculating the maximum population sustainable by a given system, other norms can be used, but they are of less evident definitions.

At least three limitations to productive work may be mentioned. One is that it takes a considerable amount of time before a subsistence farmer is trained to full efficiency, as he must know every trick of his trade. Another is that he is often under rather heavy climatic stress while working. According to P.O. Fanger (1970) thermal comfort is hardly attainable under conditions with temperatures exceeding 24°C, high humidities and almost no ventilation. This cuts

almost daily two hours out of the workday in humid tropical climates for most workers. Usually the subsistence worker depends on daylight for his activities. Including the 'noon black-out' often less than 8 hours are left as a daily maximum period of work. Hours of effective work are, however, often much shorter because of the journey to work and back. Time expenditure in this is felt to warrant a further investigation, especially considering exploitation of scattered areas.

If the efficiency of work is assumed to be uniform, the total effective work can be measured by its duration in hours, H, alone. Total transport hours are given as Ht. The total work is then Htotai =Hw+ Ht, where H_w =A• h_w. Here Ais again the given area, and h_w is the work necessary to cultivate an area unit to the point where it yields its largest net output. Necessary transport time depends on duration of a single journey, ht, as well as on the number of journeys necessary. This number is at least 5^5-, where hd is daily hours of activity. Thus the total work for the necessary cultivation is:

Slightly transformed this reads

Total work is seen to depend on the work involved in direct cultivation and in transport, which adds a fraction determined by maximum daily working hours in relation to daily time spent on transport in journeying to work.

The addition to total work caused by travelling increase
sharply as ht approaches the normal daily hours of work:

The extra workhours added to total work by various durations of ht at constant values of hd are shown in fig. 2. Even short transport may mean many extra hours. As subsistence cultivators usually travel by foot, and hd at least seasonally is very short, the incentive to minimize is strong, in table 1 the weights of added travelling is dmonstrated by application to some cases of cultivations. Graphically, the connection found can be expressed in analogue to the 'Hagerstrand time-prism' (T. Hagerstrand 1976), see fig. 3.

The problem with transport time in subsistence agriculture
is aggravated when the area for cultivation (the 'carrying
area') lies scattered on several plots:

Each of the parts i-n is assumed big enough - or isolated
enough - to be worked independently in one or more workdays.
If this is the case, the total effective workhours are:

The corresponding time necessary for traveling is then:

where r r-1 is the ratio between the work per part-area
nd — ntj
and the daily effective working hours, which equals the
number og workdays and hence the necessary journeys to
work each of the plots cultivated. If Hw +Ht are added,
an expression for the total work is arrived at:

This expression is analogous to the one arrived at earlier; travelling adds an amount of work to the effective hours which is the sum of the product of area and individual travelling time divided by the daily workhours for each of the area plots. Best localization for a settlement is evidently where transport, Ht, is at a minimum, i.e.

A settlement is, of course, only viable, if there are accessible areas enough within a distance, so that Htotal Havaiiabie. Sometimes one or more extra seasonal settlements are an economical solution to the problem of scattered resources, namely if areas can be selected so that total work for cultivation, transport and neccessary housing is less for old and new settlements together than for the original one solely.

Sometimes the scattered,utilizable areas may induce people to move between two ore more settlements. In most cases, however, moves are no doubt seasonal and from regions where areas are utilizable during different periods. Transhumance (or Senne Wirtschaft, seterdrift) in mountainous areas is thus induced by the variance of growth period with altitude. But other reasons than economy of work can also cause a pattern of regular movings. Very often social and religious factors are in fact decisive - contrary to economy of work. But, of course, the 'feasibility limit' is always valid: .Htotal = Havaiiabie.

The effect of extra transport are very pronounced when resources are scattered and travelling is by foot. Even small distances to work can be reflected by responses in the location of settlements.

An interesting example of this effect is seen on the two neighbouring Polynesian islands Bellona (Mungiki) and Rennell (Mugaba) in the Solomon Islands. The islands have almost identical cultures, at least regarding subsistence techniques, but on the small Bellona (10 x 2 km) agricultural areas are largely concentrated in the central part of the island, and only one settlement is used by each inhabitant. On the larger Rennell (85 x 10 km) agricultural areas are rather scattered both on the central plain and on the cliff rims near the coast. Field work carried out 1969 (with Torben Monberg) showed that the 18 households of the village Hatagua in western Rennell used about 180 hectares (of which 46 hectares are annually cultivated) for their sustenance. The cultivated area was composed of more than 200 plots spread within a distance of a little more than 10 kilometres from the central settlement. Traditions were that some 5 satellite settlements were seasonally employed by different people from Hatagua, see fig. 4. A similar pattern was found all over western Rennell where every central village had one or more satellite settlements. Kaangua had the coastal settlement Magautu, Nukuposa'a had Tehatumotu, Tematiga had Na'one, Honga'ubea Te'ana etc.

A calculation demonstrates that this bi-settlement strategy is essentiel for the economical use of land on Rennell. To simplify operations only two strategies have been considered, a one-settlement scheme by which all work starts from Hatagua, and a six-settlement scheme where cultivators change residences within their own properties between Hatagua and a coastal settlement according to tradition. Areas of plots within same distance zone were aggregated. This is justified because Hatagua is placed in a network of radial paths making non-radial tours insignificant. Cultivation work was calculated, estimating one fourth of the area to be annually harvested - which is possibly a rather high ratio, but reasonable

for the less work-intensive cultivations. The work per hectare was set at 1000 hours per annum; a low figure compared to Bellonese ones (S. Christiansen 1975). However cultivation on Rennell is usually considered less work-intensive than on Bellona. Daily working hours were assumed to be eight, and the extra work caused by journeying to work calculated from the formul previously found. Travelling time was found from the mean distance to each zone and a walking speed of 6 kms per hour.

From the tables A and B (table 2) the resulting total work by using one settlement and by six settlements may be compared. The total amount of work is seen not to supass the capacity of the work force, which i estimated to be arround 90,000 hours per annum (about 45 able adults at each 2000 hours). As calculations were aiming at minimum values, the difference in favour of the six-settlement strategy (more than 7500 workhours per annum) is no doubt an underestimate. As an average house including kitchen usually costs less than 200 workhours (S. Christiansen 1975), the extra costs of housing needed is easily covered by the nomadic'strategy. 'strategy. On the well-drained beaches, houses can even be built and kept at smaller expenses than normal. An essential further advantage by the 'migrating subsistence cultivation' is that access to the seasonal fishing is greatly eased. The main conclusion is, however, that the economy of the bisettlement strategy is superior, even if solely agriculture is regarded.

A brief note on terminology may be added. Whittlesey's advise to define 'shifting cultivation' as an agricultural technique resting on bush- or forest fallowing excluding anything on permanency/non-permanency of settling is justified (Whittlesey 1937). The use of the term 'migratory farming' for shifting cultivations is usually unwarranted because a fixed settlement is used. However, it seems useful to distinguish between two types of shifting cultivation: 'stationary shifting cultivation' (with permanent settlement) or 'migratory shifting cultivation' with changing settlement. To sharpen distinctions, Rennellese cultivation, which bears some resemblance to 'transhumance' in animal husbandry, may even be termed 'seasonal migratory shifting cultivation'.

Resume

Ud fra begrebet bæreevne for befolkning vises, hvorledes
arbejde indgår både som bestemmende for produktionen og
som en stofskifteudgift. Hvis sidstnævnte medtages, fås en
formel for maximum bæreevne målt i biomasse Pmax = A •
, hvor A er det producerede areal, y spiseligt udbytte
pr. arealenhed, cw arbejdsudgiften og c forbruget pr. kilo biomasse.
Fig. l viser, at maximum bæreevne opnås ved en bestemt
arbejdsindsats, som dog i subsistenslandbrug normalt
ligger under den maksimalt til rådighed værende. Udover
direkte produktionsarbejde er især transport af betydning.
Transportarbejdets betydning for spredte lodder fremgår af
de udledte formler og afflg. 2. Af flg. 3 ses, at den effektive
arbejdstid på denne måde kan blive ret begrænset.

Til slut vises det, at man selv ved korte distancer til spredte lodders dyrkning kan opnå fordel ved at benytte mere end én bopæl. På den polynesiske ø Renneil (Mugaba) er der til hver landsby inde i landet knyttet en eller flere ved kysten (fig. 4), hvortil landmændene migrerer. Besparelsen i arbejdstid er større end udgiften til ekstra bolig, hvad angår landsbyen Hatagua (tabel 2). Denne type kan evt. betegnes som 'sæsonvandrende flyttemarksbrug'.

Literature

Allan, W. (1967): The African Husbandman. London.

Christiansen S. (1975): Subsistence on Bellona Island (Mungiki).
Fol. Geogr.Dan. vol. XIII, Copenhagen.

Christiansen, S. (1976): Potential Crop Production and Carrying Capacity for Population; Basic Concepts in Cultural Geography. Congress Paper before the XXIII IGU Congress, Odessa.

FAO/WHO, 7P7s:Energy and Protein Requirements. Rome.

Hagerstrand, T. (1976): The Space-Time Trajectory Model and Its Use in the Evaluation of Systems of Transportation. Transport as an instrument for allocating space and time - a social science approach. The Inst. of Public Finance, Techn. Univ. of Vienna, No. 11. Vienna.

Whittlesey, D. (1937): Fixation of Shifting Cultivation. Econ.
Geography, vol XIII: 139-154.

»British Crown copyright. Reproduced with the permission of the
Controller, Her Britannic Majesty's Stationery Office«.