Ngā Tohuwhenua Mai Te Rangi: A New Zealand Archeology in Aerial Photographs
11 — Hawke's Bay
11
Hawke's Bay
Hawke's Bay has five main areas with a history of dense pre-European Māori settlement. From north to south, these are the Māhia Peninsula; the Wairoa River, a tributary of which runs into the hill country to Lake Wai-karemoana, and the coastal plain which runs east to Māhia Peninsula; the Hawke Bay coastal plains and the river valleys which run into the bay; the plains to the south of Hastings, including the areas around Lake Poukawa; and the coastal strip adjacent south of Cape Kidnappers along Ocean Beach. The region contains many important types of sites that can be illustrated by aerial photographs, including traditional sites, very large pā with major concentrations of pits such as Ōtātara near Taradale, pā and settlement on terrace landforms, and nineteenth-century sites including Armed Constabulary camps from the period after 1865.
Much of the northern and western part of the Hawke's Bay region is very steep and only settled in the immediate vicinity (within 1 km) of the principal rivers. However, on the Mōhaka River, sites are infrequent compared with the neighbouring Wairoa River. The paucity of open valley-floor flood plains along the Mōhaka is probably the main reason for this disparity.
Pa at Ocean Beach, just to the south of Cape Kidnappers
The site consists of a small promontory about 40 m long by 15 m wide, possibly an older dune remnant. The site is defended by a transverse ditch and bank set at the foot of the main hillslope. The pits inside the pā suggest a large area of gardening on dune soils that were stable and fertile at the time of occupation. The pits are up to 5 m long. At the time of occupation, the site would have lain amongst coastal forest both on the dunes and the hill slopes. The dune vegetation, however, would have been easily destroyed by fire. Beach-front erosion has been marked and dunes were blown inland in recent times.
Distant view of the pā at the north end of Ocean Beach
The pā (bottom left) is at the foot of the coastal face of a large hill, with dissected terrace country running north to the vicinity of Te Awanga. A low-lying ridge has been defended by a single ditch and bank with many raised-rim pits within the defended perimeter. The view is to the north-west.
Māhia Peninsula
The Māhia Peninsula, Nukutaurua mai Tawhiti, has a steep, cliffed, western side exposed to north-west winds, little settled and with few landing places. The eastern and north-eastern sides, by contrast, although exposed to the Pacific Ocean, have a number of open bays, including the long area of beach and dunes that connects the peninsula to the mainland. The north-eastern part comprises a generally narrow coastal strip backed by a high marine terrace. Headlands created by stream erosion cutting down through the high terrace typically have pā built on them. This part of the peninsula has many traditional and historical associations including the landing places of two great canoes. Tākitimu finished its voyage at Ōraka, on the north-eastern side of the peninsula, and now lies there in stone. 1 Kurahaupō, bringing the founding lineages of Rangitāne and other tribes, 2 also touched at the Māhia Peninsula.
A bay on the eastern side of the peninsula, known as Nukutaurua, lying a few kilometres south of Ōraka, was the home locality of the seer, Toiroa, who foresaw the advent of Te Kooti Arikirangi. 3 According to the Revd William Williams, in the 1840s this was the principal Māori settlement of the Table Cape vicinity. 4 A vertical aerial photograph of the Nukutaurua locality, taken in 1945, shows ample evidence of Māori settlement, although in that year it was only the site of two station homesteads. These photographs contain many sites of Māori origin but because of the clarity of the pastoral (ditch and bank fence) features they are illustrated in chapter 16 on early farming. Within a broad bay, extensive rock platforms just below the water surface extend some 150 m offshore. On the surrounding ridges were pā of relatively simple construction, following the typical East Coast model. A single ditch, occasionally double, cuts off points; or the edge of the terrace may be cut off by rectangular enclosures of ditch, reminiscent of pā on river flats in Poverty Bay. In other places, a small 'island' of the high terrace is further reinforced by steepening its sides and creating a ditch. One such 'island' is Maungakahia, a pā associated with Kahungūnu himself. 5
Wairoa River
Sites in the Wairoa River valley were surveyed and mapped by David Nevin in 1987. 6 Site types and their locations were similar to those of Uawa or Waipāoa on the East Coast, although a greater number of pā still survive on the edges of the alluvial terraces. On the ridges near the river valley are many groups of storage pits. Further inland, sites are recorded on high terraces close to the valley floor as far inland Tuai, below Lake Waikaremoana. The toe of the large landslip near Tuai, which created the lake, is curiously configured with low hillocks and large rocks. Interspersed in this largely unploughable country are a number of small well-preserved pit and terrace groups, at the limits of their inland distribution. At the Tiniroto Lakes and Reinga Falls are several pā and other positions defended by Ngāti Kahungūnu against attacks by Tūhoe in the 1830s. 7
Hawke's Bay coast
Two major pā sites of the coastal plain north and west of Napier, Heipipi and Ōtātara, are key traditional sites marking the incursion of Ngāti Kahungūnu into Hawke's Bay. Kahungūnu himself originally lived in Opotiki and then later near Waerenga ā Hika on the Gisborne plains. His departure from there, his settlement at Maungakahia on the Māhia Peninsula, and the eventual arrival of his descendants in Hawke's Bay are well detailed in a notable body of tradition. 8 Ngāti Kahungūnu attacked the local people at Heipipi and Ōtātara, the latter initially occupied by descendants of Awanuiarangi. Eventually it was held by Taraia, of Ngāti Kahungūnu.
Pōhaturoa, a Ngāti Kahungūnu pā attacked by Nga Puhi and allied Urewera forces in 1826
The Wairoa River once ran in the broad bend to the right, sweeping down to the left, but eventually broke through by the Te Reinga Falls (left of centre). The pā is on the top of the prominent hill right of centre, and is defended primarily by the steep cliffs of the old river bed. Left of the bridge is the confluence of the Ruakituri (left) and Hangaroa Rivers. The view is to the north.
Hikurangi and Ōtātara, pā in the Ōtātara Historic Reserve, near Taradale
A vertical aerial photograph of the area of the Ōtātara Historic Reserve, lying at the point where the Tūtaekurī River (bottom) leaves the hill country and spreads out over the Hawke's Bay plains. The photograph was taken in the early winter (hence the strong shadows) of 1949. The reserve actually contains two pā: Ōtātara proper, which no longer exists having been quarried away (the commencement of the quarrying can be seen at lower right), and Hikurangi, the large terraced summit. The now-destroyed pā used natural defences from the river-cut cliff, at bottom. The site complex is one of the largest single settlements known in New Zealand, matched in size only by the larger northern volcanic cone pā and some of the pā in the Waikato, Taranaki, Bay of Plenty or East Coast.
Hikurangi from the south-east
The terrace scarps had useful defensive functions, and possible defensive ditches may be obscured by subsequent erosion. Very large raised-rim pits have been dug on all the ridges. The foreground pits are up to about 8 m long by 3 m across.
Kohukete, a pā lying above an enclosed valley about 3 km inland from the Ahuriri Lagoon, Napier
The pā occupies the lightly sloping crest of a broad ridge. There appear to be few distinct transverse defences on the ridge line itself, although there are some distinct defensive scarps at centre. West of the prominent fence line, the crest of the ridge has been cultivated at some period, obscuring the features there. At the very head of the steepest western slopes are distinct lengths of ditch and bank. They are relatively unusual if defensive, and may be ditch and bank fences. If so, the suggestion is that the crest of the ridge may have been ploughed with horses at some stage, perhaps for an early-season potato crop. At the highest point of the pā is a sinuous curve of clearly defensive ditch and bank, obscured by stock damage around the fence line. The pā is about 300 m long and has a total defensive perimeter of about 700 m. The view is to the north.
In the south of the Hawke Bay coastal strip near Cape Kidnappers, there are further pā on the ridges and small platforms created on uplifted marine terraces. In the inland river valleys here are small areas with raised-rim pits, indicating gardening nearby. The pā, Tiromoana, on the Maraetōtara River near Te Awanga, occupies a sloping point created by gullies at the edge of part of the high terrace. It had spectacular surface evidence including a housefloor, raised-rim pits and lines of depressions marking the perimeter palisade line, and was excavated in 1974-75 under the direction of Lady Fox. 15 She opened several of the storage pits and terraces, cut trenches through two lines of ditch and bank defence, and excavated a major housefloor lying outside the inner defensive line. The site is probably late pre-European in age. In the immediate vicinity, there are other small areas of the high terrace with raised-rim pits showing prominently. From this point out to Cape Kidnappers, there are few pā, probably because of the lack of suitable landing places. However, south of Cape Kidnappers on low hills adjacent to the beaches there are numerous pā.
Nineteenth-century sites
Two activities have left an archaeological impact: early pastoralism and military activity. The pattern of deforestation caused by pastoralism, and re-afforestation through shrubland, is on too broad a scale to be able to be presented and analysed here. Of the mid-nineteenth-century military campaigns, there were several phases in Hawke's Bay, particularly in the north at Mōhaka, Wairoa and in the Urewera Ranges. Some Ngāti Kahungūnu joined the Kingite and Pai Mārire movements, and were involved in extensive fighting with Māori kūpapa and European forces to the north. Other Ngāti Kahungūnu joined kūpapa forces, and were prominent in the pursuit of Te Kooti Arikirangi in the late 1860s at Ngatapa and at Rotoaira on the volcanic plateau. 16
The Wairoa coastal strip was an important route for travel along the coast to and from Poverty Bay and the Māhia Peninsula. In the nineteenth century, the strategic advantages of controlling this strip were considerable. On one of the low sandy hills, near Whakakī Lagoon, there is a remarkable small redoubt (one of several in the area) and defended camp with a rifle trench perimeter, named Tokitoki, both with extensive views of what would have been a fern-covered plain and dune lands to the east. These fortifications were probably part of either the kūpapa campaign against Pai Mārire from Ngāti Kahungūnu in 1865, or the kūpapa and Colonial forces campaign against Te Kooti Arikirangi in 1868. 17 The nature of the redoubt does not suggest the relatively informal kūpapa style of fortification. It is a textbook redoubt: small in size, but rectangular and with flanking angles.
Maraetōtara River valley, with Tiromoana on the prominent point, lower left
All the sites are on a high marine terrace, with a good outlook down the valley or seaward. The valley was north-facing, and the preferred horticultural sites were elevated to avoid the risk of unseasonal frosts. On the edge of the alluvial terraces are traces of borrow-pits.
Tiromoana, a pā on the Maraetōtara River near Te Awanga
The steep natural defences of the pā and its location on the terrace edge above the river flats show clearly in this view to the north-west. Transverse defensive lines show clearly. The length of the main defended area is about 50 m. The rims of the pits and also the line of palisade postholes on the righthand (eastern) perimeter show clearly, under a covering of long dry grass. When Lady Fox excavated this site in 1974-75, she found further evidence of house floors and the nature of the palisaded defences. The major house floor, built late in the occupation of the site, was between the two lines of ditch and bank at bottom left.
Redoubt and huts near Whakakī Lagoon on the coastal strip east of Wairoa
The redoubt (at left) is a typical, rather small example, with flanking angles at opposing corners. The defended area within the banks is quite small, 5 by 7 m in plan, and from the outer edge of one flanking angle to the other is about 14 m. To the right of the redoubt is a rectangular enclosure forming a defensive perimeter for two huts, named on early survey plans as 'pa Tokitoki'. The defensive perimeter is about 20 m by 40 m. In the corners of one side are the rectangular depressions of the huts, each about 5 by 8 m in plan. Both sites were probably occupied by colonial or kūpapa forces in the period 1865-68, and as late as the 1870s. In the saddle is a clear example of ridge and furrow ploughing, surviving here because the area has never been ploughed subsequent to the period of occupation. The view is to the east.
Ohiti pā, showing a redoubt within the pre-European pā, Ngaruroro River
The four flanking-angles show distinctly in this photograph despite a heavy cover of weeds. The value of the redoubt, closed in by the defences of the pre-European pā, is doubtful. It would be a simple matter for attackers to approach the outer ditch and bank and use it as an offensive breastwork.
1 Te A. Davis and Wilson (1990: 5); Mitchell (1944: 47).
2 McEwen (1986: 10-15).
3 Binney (1990: 168-169).
4 W. Williams (1974: 415).
5 Mitchell (1944: 64, 80-82).
6 Unpublished; fieldwork contract supervised by the writer in 1987.
7 S.P. Smith (1910a: 320-368).
8 McEwen (1986: 41-44); Mitchell (1944).
9 E. Best (1927: 291-298).
10 Fox (1980); McEwen (1986: 42).
11 Bain (1992, pers. comm.).
12 Fox (1980: 235-240).
13 Fox (1982: 67-68).
14 Fox (1982: 77).
15 Fox (1978).
16 O. Wilson (1961).
17 In several phases. See Cowan (1983, Vol. 2: 129-136, 401-408).
18 Taylor (1987: 75-88); Walton (1982).
19 Cowan (1983, Vol. 2: 137-142).
20 Fox (1982: 77).