Ngā Tohuwhenua Mai Te Rangi: A New Zealand Archeology in Aerial Photographs
5 — Other Economic Activities
Other Economic Activities
Because of the prominence of the evidence of pits and modified soils, gardening takes a great deal-of archaeological attention. Other Māori economic activities include fishing, both marine and freshwater, hunting and gathering birds and seals (including juveniles), quarrying of stone for adzes, and cooking in ovens. In the previous chapter about horticulture, we found that aerial photography could offer unique insights in the analysis of that activity. In this chapter, with the exception of the all too rare eel channels and their setting, and the soil marks of ovens, the aerial photographic record is chiefly of value in illustration rather than offering opportunities for close analysis.
Fishing and shellfish gathering
Some evidence of pre-European fishing, principally shellfish middens, can be seen in aerial views. Middens can be seen at Ninety Mile Beach, on other beaches in the far North, and also at the dune-covered points forming the entrances to harbours such as Aotea or Mangawhai. Here, shellfish were gathered from the beach and perhaps steamed in umu (earth ovens) on the nearby dunes. The flesh was taken out of the opened shells to be dried in the sun, and the shell debris thrown into heaps. Once dried and preserved, it was taken inland for later consumption. Over time, as the sand eroded, the middens were scattered and exposed on the surface of dunes, spilling down them in drifts. These cover such an area that they can be seen from the air, exposed as white patches in the yellow-brown sand. 1 I had some problems gaining sufficient contrast between sand and the midden in bright sunlight in the few opportunities that I had for photographing middens. The photograph used here was taken under the direction of John Coster in overcast weather after rain showers on Ninety Mile Beach. Under these conditions, the rain had darkened the sand surface but not the white shell, and the contrast was marked.
Fish traps, consisting of placed-stone walls in the intertidal zone, are known on the Coromandel and Banks Peninsulas and elsewhere. 2 A stone wall was made in a favoured area in an estuary, enclosing up to half a hectare, but usually less, with the top of the wall lying just below the high-water level. Fish were trapped within as the tide receded. By far the largest example in New Zealand lies on the western side of the Onāwe Peninsula, in Akaroa Harbour, Banks Peninsula. A trap in Colville Harbour, Coromandel, 3 is still maintained by local people and is reported to catch fish.
Eeling was another important form of fishing in late pre-European and nineteenth-century Māori society. The catching of eels is normally associated with hlnaki (basket nets) and weirs, but there were other forms of mass-harvesting, such as the capture of eels as they crossed from lagoons across sand flats to the sea in the course of their migration to breeding grounds at sea. Eel channels are particularly transient in the landscape view, but can be viewed and analysed in the wider landscape setting if aerial photographs have been taken at the appropriate time. Shallow channels, known as whakamate, found in the more inland parts of the Horowhenua/ Manawatū dune country, are the oldest landscape evidence of Māori practice for catching eels.
Shell middens near Te Paki, at the northern end of Ninety Mile Beach
The beach below high-water mark was a very productive source of shellfish. This locality appears to have been particularly favoured because of the stream flowing into the beach from the small dune lake. The beach is at the bottom of the photograph, and the midden patches are about 8 m across. The shells, mainly of tuatua with some toheroa, showing as white patches against a darker yellow-brown sand (lower centre), were probably gathered in summer. The patches mark the edges of short-term settlements probably occupied in summer for this fishing. The shellfish were probably steamed in umu, extracted from the shell, dried in the sun and transported inland to more permanent settlements by the inland lakes and forests of the peninsula.
Fish trap, Onāwe, Banks Peninsula
The fish trap is marked by the distinct arched band of seaweed immediately offshore, at centre left. The seaweed is attached to the original rock wall which is probably lower in height than it was originally. Towards the shore, the stone wall appears to have been destroyed. The view is to the north-west; the fish trap is about 25 m across. On the broad ridge are some of the defences of the pā itself, to be revisited in chapter 14.
The channels, constructed for the mass-capture of eels, lie on the principal drainage outlet north-east of the former lakelet. Eels had to pass through the barrier of the dunes. The eels naturally followed moist channels. These channels did not contain running water, which would have eroded the banks. Instead they attracted the eels by offering a moist path, from which the eels could be scooped. G.L. Adkin regarded the largest of the channels, circling around the foot of the dune, top right, as the most ancient. At this date (August 1991) a regional council drain runs from the lake (out of view at bottom) parallel to the channels. The view is to the north-east.
In summer, groundwater levels are low and the lakes partially dry up. In late summer, newly matured eels begin to migrate from the lakes to the sea. They cross the barrier of the dunes and dry sand plains using natural drainage channels and lower-lying wet areas. Māori soon learnt that, by creating shallow channels across the sand barriers at strategic places, eels could be concentrated in vast numbers and caught by scooping them out of the channel. This was not a trapping routine, but more a form of mass-harvesting. Typically, the sand plains now have lower groundwater levels than 100 years or more ago, because of modern land drainage. However, channels have survived on the ground where land development has not been too intense, although these show only because the channels are slightly moister and have a different colour of grass. In chapter 3, eel channels are illustrated in such dune country (near the pā, Awamate on the Rangitīkei River), 4 but the best known example is at Tangimate, inland from Waitarere Beach, a few kilometres north of Lake Horowhenua. 5 The channels have little surface relief so that colour differences between the channels and the surrounding ground are important in photography. When I photographed these channels in light overcast weather in August 1991, the conditions were excellent, the channels showing as light green lines against the yellowing grass of the sand plain.
Not dissimilar techniques were, and still are, used at the outlets to coastal lakes today, for example at Wairewa (Lake Forsythe), Banks Peninsula. The older evidence does not survive there as it does in the Manawatū or Horowhenua. However, the location of settlements such as Orariki, near Taumutu (south end of Waihora), or the pre-European and nineteenth-century settlement, Oruaka, at Wairewa (Lake Forsythe), 6 are evidence of the importance of this and other lake fisheries. 7 Illustrated in a later chapter are other sites such as pā near Gisborne, which are located at the downstream end of suitable migration channels, partly because of the good eel fishery.
The manufacture of stone tools in Polynesia ranged from sustained periods of wide-ranging trade and great sophistication in exploitation and manufacture, to apparently crude usage of poor quality, local materials. 8 The favoured stone in Eastern Polynesia was basalt, the basic geological component of all the high islands. Finegrained, easily flaked but tough stone was especially sought after, with well-known sources in Samoa, on Pitcairn Island and on the Mauna Kea volcanic plateau, Hawaii. There it was exposed in weathered forms, in most places in spectacularly steep or otherwise difficult volcanic landforms. On Mauna Kea, it is exposed above the vegetation line at 3,500 m altitude. 9 Exploring for these sources of stone probably involved a model of a volcanic landscape with readily identified sources of stone. Polynesians appear to have applied the same insights to New Zealand, where there was a wide range of potential stone usage, with much of it actually exploited.
Two main landforms were readily identified using this model of a good stone-source landscape: the basaltic lava domes that occur from the Bay of Plenty to Northland, and the very much older geological land forms containing the serpentine 'mineral belts' of Nelson and Southland. The volcanic stone, basalt, lent itself to sophisticated stone adze production. More importantly, its presence could be readily identified from the characteristics of a volcanic landscape. The hills were high and steep-sided; and they are often associated with a basalt dome emerging from a more eroded volcanic area. 10
In the Nelson region, the distinctive vegetation cover of the mineral belts would quickly be recognised. The mineral belts are zones on the earth's continental plate where the old surface rocks, volcanic or sedimentary in origin, have been pressed into the molten part of the crust. From time to time the ancient rocks have 'oozed' back up to the surface. 11 The exterior of these blocks had been 'baked' at high temperature and pressure in the earth's deep mantle, transforming the rock so that it retained its original toughness but could be worked by chipping in the same way as the basalt. As the soil and softer rock around these hard blocks erodes, the blocks themselves are left free-standing in distinctive forms.
Two views of the same complex of argillite adze quarries at Samson Bay, Croisilles Harbour, in the Nelson mineral belt
The argillite occurred in layers on the surface of some of the rock outcrops. Both the outcrops and stone scattered on the surrounding soil were chipped into adze 'blanks' and carried out of the location. The spur is vegetated with stunted growth compared with the surrounding country, caused by the toxicity of the chemicals in the soils at this location. The vegetation has also been burnt repeatedly to gain access to the stone and, in recent times, for grazing. [Above] is a distant view to the south-east.
Umu and umu tī
Ovens exposed by wind erosion and ploughing on the river terraces at Waitaki River mouth
This area is one of the most important moa-hujitihg sites in New Zealand, dating from the thirteenth century. Carcasses were probably rafted down to a long-lived settlement here. The main river ran past the site immediately to the north, giving access to the site from the sea over a bar, but has since worked its way further north. The ovens show clearly in this photograph taken in about 1960. A crop of wheat or oats had been taken from little-ploughed land, and this photograph was taken after the oven marks had been 'freshened' by discing the crop stubble. The individual circular patches are about 6 m across while the larger area is some 250 by 50 m. The view is to the north.
Ti was amongst the tree crops cultivated by the ancestral Polynesians. In New Zealand the native tī (cabbage trees) would have been readily recognised as a potential source of food. The tī roots were prepared for consumption by cooking or steaming in a high-temperature process which converted indigestible starches to fructose (a type of sugar), with a required cooking duration of some 20-40 hours. 15 The problem of maintaining high temperatures over such a long period was solved by building very large ovens, as much as 8 m across and a metre or more deep, known as umi tī. They are widely recorded in the South Island south o Timaru, and an Otago Peninsula example is shown here. 16 The need for a starchy or sugary food derived from tī would have been greater in the south, in areas when kūmara could not be grown.
Hunting and gathering
It is impossible here usefully to illustrate a full range of Māori hunting practice. An important aspect of it, however, can be shown in aerial photographs. All hunting for subsistence purposes balances the minimisation of effort in the chase with maximising the result from a single kill. Infant animals or fledgling birds offer a good opportunity to strike this balance satisfactorily. Just be- page break fore they begin to fly, fledgling seabirds have reserves of fat and much muscle meat but are easily caught. Muttonbirds (the young of sooty shearwaters, a small petrel) are an obvious contemporary example. Albatrosses are another example. In the Chathams, the taking of albatross fledglings was a well-refined form of hunting. They were typically taken from remote rocky islets, such as the Little Sister illustrated here, where Doug Sutton was able to demonstrate seasonal visits by Moriori presumably for the purpose of taking these birds. 17
Umu tī, Otago Peninsula near Ōtākou
The umu tī are two distinct circular depressions (one with a rim), each about 3 m in diameter, in the centre of the photograph. The other smaller depressions are from tree-throw, the root plate having been forced out of the ground by the wind and the tree long rotted away. The view is to the north-east, down the slope of a hill.
Northern Royal albatross (toroa) breeding ground on the Little Sister, north-western Chatham Islands
The juvenile albatrosses show as white blobs. The eggs hatched in February; the photograph was taken in May; and the fledglings would be taken in August or September when they weighed as much as 8 kg. People who visited the island seasonally lived in caves to the east of the prominent rock knob, which is also adjacent to a permanent water supply. The landing place was on the northern rock ledges.
1 Coster (1983; 1989).
2 Photograph in Law (1982); Trotter and McCulloch (1989: 62).
3 Law (1982).
4 Cassels et al (1988).
5 Adkin (1948: 25-30,357-358); Sheppard and Walton (1983).
6 Brailsford (1981: 154-155,160-161).
7 Tau et al (1990: 40-46).
8 H.M. and B.F. Leach (1980); K. Jones (1984b).
9 Kirch (1985: 179-180).
10 Typical examples of such island areas are Samoa, parts of the Cooks, Hawaii and the isolated southerly small islands of Rapa, Easter and Pitcairn Islands.
11 G.J. Williams (1974: 143-145).
12 Teviotdale (1939: 168-169); Knight and Gathercole (1961: 133-136); Anderson (1989).
13 B.G. McFadgen (1990, pers. coram.).
14 Teviotdale (1939: 168-169).
15 H.M. Leach (1984: 20-21).
16 Knight (1966); Fankhauser (1987).
17 Sutton (1977).