Archaeology in
the Hauraki Region:
A Summary

THE HAURAKI TREATY CLAIMS

VOLUME 3

Hauraki Maori Trust Board

Archaeology

in the Hauraki Region

a summary

LOUISE FUREY

HAURAKI MAORI TRUST BOARD

1997

First published in 1997 by
Hauraki Maori Trust Board
PO Box 33, Paeroa
Aotearoa New Zealand

ISBN 1-877198-01-3

© Hauraki Maori Trust Board

This report was commissioned by the Hauraki Maori Trust Board
as part of its Waitangi Tribunal Claim research programme.
Any views expressed and condusions drawn are those of the author.

All rights reserved.
No part of this publication may be reproduced,
stored in a retrieval system, or transmitted in any form or by any means,
electronic, mechanical, including photocopying, recording or otherwise,
without the prior permission of the publisher.

Typeset by Wordset Enterprises Limited, Wellington
Printed by GP Print, Wellington, New Zealand

FOREWORD

The Hauraki Treaty Claim project has examined the nature and extent of the interaction of Maori with the Crown in the Hauraki tribal territory during the nineteenth and twentieth centuries. The claims, together with the research and supporting evidence are set out in a series of eleven volumes. These are presented to the Waitangi Tribunal to support the Hauraki case.

The history of colonisation in Hauraki—the social and economic deprivation endured by those who have gone before us and their years of responsible protest—has not been told before. These volumes, the foundation of the Hauraki case, will forever rewrite our nation's history books contributing, only now, a Maori perspective to the history of this region.

We began this project four years ago with a multi-disciplinary team approach. Louise Furey was part of this team contributing an archaeological perspective, to set the scene and examine Maori society in Hauraki before the influx of European settlers.

Louise Furey's Archaeology in the Hauraki Region provides a comprehensive summary of the economy that existed then and the places where that activity occurred. The maps depicting the distribution of archaeological sites clearly demonstrate a significant focus on the coastal marine area. The use of other natural resources, in particular, minerals is also well documented.

The Hauraki treaty claims are a consequence of the Crown's actions after it signed the Treaty of Waitangi. Louise Furey's report will significantly support the Hauraki case in the debate that will inevitably surround the Hauraki claims. I take this opportunity to thank Louise Furey for her contribution to this project.

No reira, noho ora koutou.

T.J. McEnteer

Claims Manag

Hauraki Maori Trust Board

III

CONTENTS

FOREWORD   III

PREFACE   VII
INTRODUCTION   

1. SITE DISTRIBUTION   5

The data   6

Types of sites recorded   7

Pa   7
Middens   10

Storage pits   12
Terraces   13

Garden sites                              14

Artefact manufacturing places                        14

Other sites                              15

Burials   16

Pattern of settlement   16

Case study—Waihou River   17

2. CHRONOLOGY   18

3. SUBSISTENCE ECONOMY   23

Fish   23

Shellfish   26

Marine mammals   26
Dogs, rats and reptiles   27

Moa and other birds   31

Vegetable and forest foods   33

Summary   33

4. STONE RESOURCES AND MINERALS   23

Obsidian   35

Basalt   38
Greywacke   41

Chert   42

Other stone   42

Kokowai   43

Non-stone materials   43

5. HISTORIC ACCOUNTS   44

(

ARCHAEOLOGY IN THE HAURAKI REGION

APPENDIX I

Excavated Hauraki sites   57

General summaries of excavations   59

Bibliography of excavation reports   59

APPENDIX II

Radiocarbon dates for Hauraki sites   68

APPENDIX III

Scientific names of plants and animals   75

PREFACE

My name is Louise Furey. I have an MA in Anthropology from the University of Auckland. My thesis was on archaeological site recording on the Coromandel Peninsula. Between 1981 and 1984 I was employed by Historic Places Trust and NZ Forest Service on short-term contracts to carry out site surveys and excavations prior to site modification as a requirement of the Historic Places Act. From 1984–87 I was employed by NZ Forest Service as archaeologist responsible for management of archaeological sites in exotic forests on the Coromandel Peninsula and, on dissolution of the Forest Service, transferred to the Department of Conservation Waikato Conservancy, again taking responsibility for heritage sites in the Coromandel region. I was also Coromandel region filekeeper for NZ Archaeological Association's site recording scheme, the national data base of recorded archaeological sites. Between 1988–91 I had a change in career with the births of my two daughters although I continued to write papers for publication in archaeological journals. Since 1991 I have been employed by Auckland Museum under Lottery and Foundation for Research Science & Technology grants to research the recent history of artefacts recovered from the Hauraki pa, Oruarangi. This work culminated in a description of Maori material culture published as a Bulletin of Auckland Institute and Museum. A similar, recently completed, project has focussed on the archaeology and material culture of Mt Carmel, Houhora, one of the first landfall sites in Northland. Latterly I have been working as an archaeological consultant, again mainly on the Coromandel Peninsula.

VII

INTRODUCTION

People have lived in Hauraki for about 800 years. During that time they have left their mark on the landscape both physically and in place names.

The Hauraki region reaches out from the nucleus of the Coromandel Peninsula and Hauraki Plains to include the western shore and islands of the Hauraki Gulf and into the Tamaki-Waitemata area (Fig. 1). An extensive coastline, a mountainous backbone and the swampy plains make this an environmentally diverse region and very rich in resources. Over time the Hauraki iwi have utilised the advantages of these different areas for gardening, hunting birds, harvesting kai moans, and for defence.

In the last 40 years there have been a number of archaeological excavations, making the wider Hauraki area one of the more intensively studied regions in Aotearoa (Fig. 2 and Appendix I). The attention has, however, been focussed on smaller areas within the region, such as the western Bay of Plenty, Coromandel east coast, Waihou River, Tamaki River and Motutapu Island, and on specific topics such as initial colonisation by East Polynesians, gardens in East Tamaki, and the undefended settlements of Motutapu Island and Tamaki.

Archaeologists sift through the physical evidence of the past to construct a broad picture of settlement and use of the landscape. The interpretations presented here are derived from excavation but are supplemented by early European eye-witness accounts and journals of missionaries and travellers. Land Court records are another important source of information: in establishing their rights to a land block, claimants often named occupation sites and resources used. Also, where appropriate, artefact collections have been used to interpret activities on specific sites.

This report is organised under broad topics to provide a picture of pre-European Maori life in the Hauraki area, outlining the pattern of settlement, food resources used and the subsistence economy as it can be reconstructed from archaeology This is complemented by, and contrasted with, historic accounts which demonstrate the extent to which European influence made an impact on traditional Maori society.

ARCHAEOLOGY IN THE HAURAKI REGION

FIGURE I: THE HAURAKI REGION

2

Introduction

FIGURE 2: SITES EXCAVATED

3

ARCHAEOLOGY IN THE HAURAKI REGION

Archaeology is limited as a tool to interpret the past. Specific events are rarely identifiable and the people who lived at a particular place, or their tribal affiliations, cannot be determined by archaeological methods. Broad patterns of activity can be identified in settlements. Archaeology can provide quantitative information on economy but only a partial picture is obtained for several reasons: generally only the bones, stone and shell survive to indicate what people ate, or as examples of technology. Radiocarbon dating provides a general indication of chronology, probably to within 100 years, but is unable to be precise on when a site was occupied.

Throughout this paper there are references to 'early' (Archaic) and 'late' (Classic), separating Maori culture into two broad periods of time. These are somewhat arbitrary as no date can be set for a change from one period to another. For a time after initial colonisation the settlers were culturally more East Polynesian than Maori but change in technology, social organisation and settlement, language and cultural practices occurred over the centuries prior to European arrival.

4

I. SITE DISTRIBUTION

Archaeologists have been actively working in the Hauraki region since the mid-1950s, carrying out excavations and recording the surface evidence of Maori occupation on the landscape. The information has been incorporated into the site recording scheme of the NZ Archaeological Association (NZAA), the national database on archaeological sites. The NZAA files are the basis of the lists of sites, called the County Inventories and published by the Historic Places Trust, to inform territorial local authorities, development and planning agencies, and the public of the presence of sites.

Site recording began in the late 1950s, using students and interested amateurs to identify the locations of past Maori and European occupation where there was physical surface evidence remaining. After the implementation of the Historic Places Amendment Act 1975, which provided statutory protection for all archaeological sites, the Historic Places Trust began sponsoring site surveys. The number of records incorporated into the site files increased steeply for about eight years until the NZ Historic Places Trust withdrew funding. By the early 1980s the Hauraki area was one of the better surveyed regions in New Zealand. Little systematic survey has been carried out since that time, and new records being added into the file are from small area surveys usually carried out because of a specific development threat.

The overall distribution of sites, shown in Fig. 3, indicates intensive use of the landscape. Not all of the sites, represented by dots on the map, were occupied at the same time but reflect the pattern of settlement over some 500–600 years. A small number of sites on the east coast of the Coromandel Peninsula were clearly occupied within the first hundred years of Maori settlement of New Zealand. The full range of site types occupied at any one time by any one group of people is unknown but hints are provided from Land Court records and by using European observations of settlements in the pre-1840 period.

While site surveys provide information on where people lived on the landscape and patterns emerge from the distribution of occupation evidence, interpretation of the past is largely derived from excavations. The ability to use some of this information is however limited by poorly reported work or by the lack of substantial information about an excavation, by the small area of the site excavated, and by a focus on one particular aspect of information.

5

ARCHAEOLOGY IN THE HAURAKI REGION

FIGURE 3: DISTRIBUTION OF RECORDED MAORI SITES

Note: Dotted line shows Hauraki boundary.

The data

There are approximately 6000 recorded Maori sites in the Hauraki region. This does not represent all the evidence for several reasons including incomplete coverage, and damage to sites through erosion, forestry, farming practices or development. The total number of sites, or places where evidence of occupation is present on the landscape, can probably

6

Site Distribution

realistically be doubled. However not all the sites still exist so the total number will never be known. In addition people often lived in favoured places again and again.

Sites are often difficult to find, especially where shell middens or obvious changes to the ground surface are absent. Consequently evidence of occupation in inland areas, or the use of the forested areas, is severely under-represented on the site distribution map.

Types of sites recorded

The sites are classified according to the evidence visible on the surface. Commonly recorded site types include pa (fortified or defended sites), middens (rubbish heaps including shell, bone, stone and charcoal), terraces (artificially levelled areas), storage pits (semi-subterranean structures for kumara and other foods), stone rows, heaps and mounds usually associated with clearing soil for gardens and for marking out garden plots, cultivated soil where the natural soil profile has been modified by repeated digging over, ovens (cooking sites including oven stones, charcoal and stained soil), working floors where stone tools have been made or prepared, rock carvings, rockshelters and caves which were used as occupation sites or for burials, botanical evidence such as wild taro, find spots where artefacts have been found but there is no other evidence of occupation, canoe landings (area on a foreshore cleared of stones) and human burials.

These site characteristics can appear singly or in combinations. The site records only identify the evidence which is visible on the surface. If excavation takes place, subsurface features such as storage pits, postholes, stone working areas or ovens may become apparent and expand on the interpretation.

Each of the main site types will be described, outlining the evidence uncovered by archaeology.

Pa

There are approximately 700 recorded pa within the Hauraki region (Fig. 4). These are usually on a headland, a ridge or spur end with natural slope defences supplemented by either a single or double ditch and high inner bank enclosing a living space and often having secondary lateral ditches or steepened scarps. Palisade posts provided extra security. Size varied considerably from very small, only large enough for a small family group, through to very large involving numerous terraces and complex defences. Many, but not all, pa were occupied more than once, with modifications to the defences or the layout of the pa.

On the Waihou and Piako Rivers pa were built on flat land adjacent to the river, taking advantage of naturally raised levees in an otherwise poorly drained landscape. The flatland pa were surrounded by large palisade posts enclosing the pa but may also have had ditches and banks such as those on Oruarangi and Paterangi. Parkinson (1784:106)

ARCHAEOLOGY IN THE HAURAKI REGION

FIGURE 4: RECORDED PA SITES

who accompanied Captain Cook on his first voyage also commented that the mud at low tide provided a form of defence from the river side.

In the Tamaki area the volcanic cones of Maungarei (Mt Wellington), Otahuhu (Mt Richmond) and Te Apunga o Tainui (McLennan Hills), Matanginui (Green Mt), Ohuiarangi (Pigeon Mt), Te Puke o Tara (Smales Mt), and also Motukorea (Browns Island), Takapuna (North Head) and Takarunga (Mt Victoria) were defended by a combination of steepened scarps and palisades on the terrace edges, and a ditch and bank s stem on the rim of the crater.

8

Site Distribution

Yet other sites likely to fit into the category of defended sites have little surface indication of their function. A site beside the Tamaki River (R11/1506) when excavated was found to have been entirely surrounded by a fence. The missionaries from the Royal Admiral in 1801 described a fenced village for about 100 people on the foreshore near Tapu (Missionaries' Journal ms:12). Yet another description of a fenced village, Whakatiwai in 1836, reported in Wade (1977:29–30), can probably be attributed to Rev. Henry Williams although Wade refers to the writer as 'a much esteemed individual'.

Wakatiwai (sic) consists of a quadrangular enclosure of about four hundred feet along the front, by about two hundred in depth; the fence composed of stakes of all sorts and sizes, varying in height from ten to twenty feet, driven into the ground about a couple of inches apart, and having, at intervals, large posts, of which the upper part is rudely shaped in the semblance of a human figure. ... Outside, in all directions, are large mounds of cockle shells. The entrance is by an opening about two feet from the ground, just wide enough for one person; a block of wood driven into the earth serving as a step. Within the enclosure, without the slightest attention to order or plan of any kind, are scattered, about a hundred rush huts and hovels, some with fenced enclosures and some without; with occasionally a small patch of potato ground. In all the enclosures, and elsewhere about the pa, are store houses, consisting of a platform raised upon stakes ten or fifteen feet high, on which are deposited the potatoes, &c., and here and there are poles on which are hung bundles of dried fish, scattering their fragrance abroad. In all directions the ground is covered with cockle-shells, fragments of potato baskets, and other remains; whilst its surface is everywhere broken into little pits and hillocks, occasioned by the constant formation of ovens for cooking. Pigs are seen wandering about, grubbing up the abandoned ovens ...

Excavations of pa at Kauri Point and Ongare Point in the Bay of Plenty, Sarah's Gully near Opito and Castor Bay on Auckland's North Shore all show complex histories with undefended living sites being overlaid at a later time by pa complete with ditch and bank, and palisades. In the case of Kauri Point, the pa in the final stage of use was smaller than the original. Excavation of the pa in Station Bay, Motutapu revealed a change in use over successive occupations from an undefended settlement with kumara store pits to a defended storage site to a fortified site without storage, reflecting the different pressures placed on the people of Motutapu Island.

Not all pa were occupied on more than one occasion. For instance Kauri Point, Birkenhead had very ephemeral evidence, and was occupied for only a short time. By contrast, Oruarangi may have been occupied semi-continuously for a considerable time with the site being enlarged on at least two occasions, until by the early 19th century it enclosed nearly five acres of flat land (Furey 1996).

In the early 16th century ditch and bank pa began to make an appearance on the landscape. However it is likely that conflict between groups of people dates back to the first settlement. Weapons were used in East Polynesia, the homeland area, so it can be inferred behaviour involving warfare and skirmishing was also brought here by the first settlers. No recognisable weapons have been recovered from sites of an early age in New Zealand. The earliest form of defence may have been simple fences around settlements, or refuge sites in the bush, as the evidence for conflict is rare.

9

ARCHAEOLOGY IN THE HAURAKI REGION

Some pa were constructed in the 19th century. The almost constant warfare in the 1820s and 1830s meant people needed to protect themselves. A pa built at Ongare Point by Ngai Te Rangi was attacked by Taraia of Ngati Tamatera in 1842 (Shawcross 1964), and Opita on the Waihou River was re-built in 1842 in anticipation of reprisals for Taraia's actions against Ngai Te Rangi at Ongare Point (Phillips 1994: 412). There is mention in the Land Court records of other pa constructed at this time, for example, on Waiheke (reported in Monin 1992), and Te Kari on the Waihou River re-fortified in 1831 (Phillips

1994:377)

It is apparent from the Land Court records and from European eye-witness accounts that the pa was often the focus of settlement in an area, with an undefended village nearby and gardens scattered through the area. John Nicholas (18171: 394) described several villages on the Wharekawa coast (on the western side of Firth of Thames) in 1815, one of which may be Whakatiwai, with the palisaded pa on the ridge above and gardens on hillslopes in the vicinity At Matatoki, Oruarangi was identified in the Land Court records as the principal pa with Paterangi a satellite pa, and kainga and gardens in the vicinity (Hauraki MB 63:133). By contrast, Captain Cook in Mercury Bay described the well organised Wharetaewa pa with numerous houses, and very little evidence of occupation outside of the pa, although there were people in different parts of the wider bay. (B eaglehole 1955:198-99).

Middens

Middens are the rubbish heaps of a settlement. There are approximately 4300 recorded sites containing midden (Fig. 5). They are made up mainly of shell, but may also contain small amounts of fish and other bone, charcoal, oven stones and other debris such as seeds. Shell deposits are highly visible in disturbed ground and are often the only indication that an occupation site is present. The amount of shell can vary from a few kitfuls to several cubic metres. For instance the middens at Kauri Point pa and Ongare Point pa were just under two metres in depth.

By sorting the midden into its various components and identifying shell and bone material, archaeologists can interpret diet and preferences for food although the information has many limitations.

Because shell middens are highly visible, the absence of this site type from some areas appears significant but can be explained by environmental conditions. Thus a comparison of Fig. 5 with the overall site distribution map (Fig. 3) shows a general absence of shell midden at the northern end of the Coromandel Peninsula and along the upper west coast north of Colville. By contrast almost all of the Coromandel east coast sites contain shell. On the Wharekawa coast middens are generally only present in the coastal sites. Some of these absences can be explained by distance from sources of estuarine and sandy shore shellfish (such as the rocky Moehau coastline on Coromandel) but shellfish have been carried some distance in other places, for example, to Raupa near Paeroa, and cockles are present in middens on Slipper Island.

I0

Site Distribution

FIGURE 5: RECORDED SITES WITH MIDDEN

On close examination the middens vary enormously in content. Some contain the every day debris of the settlement, with stone flakes, oven rake-out, shells, bone fragments and broken oven stones, but others are very specialised. For instance from a midden at Galatea Bay on Ponui Island only the head bones of snapper were recovered indicating the bodies had been taken elsewhere (Shawcross 1967). Seasonal camps where fish were caught, deheaded, dried and taken away to another settlement were probably common around the Hauraki coast.

11

ARCHAEOLOGY IN THE HAURAKI REGION

Storage pits

Storage pit depressions in the ground indicate semi-subterranean roofed food stores were present in many undefended sites as well as pa. Often there are one or two pits in association with the same number of terraces but in some areas large numbers of pits are present over some distance along a ridge or spur. For instance in the Colville area, up to 33 pits are to be found in one site. By contrast, on the opposite coast of the Coromandel Peninsula, at Waikawau Bay, pits occur singly. This suggests that in some places there was a greater community emphasis on storage of food, while the small sites with single pits may be associated with small family groups responsible for their own food supply. Over 1700 sites with storage pit depressions have been recorded.

Many storage pits have been excavated in the Hauraki region. Sites at Opito and nearby Sarah's Gully had rectangular pits with small bin pits dug into the side walls and into the floor, dated to the 13th and 14th centuries. This is amongst the oldest evidence for storage in New Zealand, displaying sophistication in what was apparently a newly developed technique for storing kumara. A number of excavated pits have features such as external perimeter drains for keeping water out of the stores, or for removing water from within pits where internal drains and tunnels to the outside slope were constructed. There appears to be no regional pattern to the use of these features. Pits not visible on the surface are often uncovered during excavations, having been completely and deliberately filled in and the level surface used for another purpose.

The most common form of pit is rectangular. Rua, or underground pits with a narrow circular entrance at ground level, and pits dug into a scarp or slope are rare in the Coromandel area although they are common in the western Bay of Plenty.

Like middens, variation is also evident in the distribution of storage pits. It is noticeably apparent from Fig. 6 that few sites with pits have been recorded on the east coast of the Coromandel Peninsula south of Opito. This may be due to several factors which have not yet received any archaeological investigation, ranging from erosion of the soft volcanic ash soil infilling pit depressions through to a change in type of settlement. Excavations at Kauri Point and Ongare Point revealed numerous storage pits and an extremely complex pattern of infilling and reconstruction, very little of which was apparent at ground level. Pits were also uncovered in two sites at Whangamata during excavation. The inference is that pit storage was not carried out in the lower eastern Coromandel area to any great extent in the 17th and 18th centuries.

The absence of storage pits in sites on the Hauraki Plains can be easily explained by the unsuitable storage conditions in the poorly drained ground. However above ground storage structures were noted by European visitors to the area (Wilson ms). Food grown in the naturally fertile soil around the river settlements may also have been taken to higher ground where pit storage would have been more appropriate. This is apparent on Fig. 6 where numerous pits have been recorded in the low hills bordering the Hauraki Plains.

12

Site Distribution

FIGURE 6: RECORDED SITES WITH PITS

Terraces

Terraces often contain evidence of houses, cooking and storage. They vary in size from small flat areas of approximately 3m x 2 m up to 30 m x 10 m which might have had several houses and associated features. Terrace construction on some sites would have required a large labour input. By contrast small sites with two or three terraces may have been occupied by a small whanau.

13

ARCHAEOLOGY IN THE HAURAKI REGION

TABLE 1: EXCAVATED HOUSES

Place   Site No.   Size of house (in metres)

Motutapu Is   R10/31   1.8 × 2.8

Fisher Road   R11/887-9   2.8 × 6.0 (incl. porch of 1.4 m on south side)

Tamaki River   R11/1506   3.5 × 4.0 (porch 0.8 m wide on south side)

3.4 × 2.8

3.5 × 3.0

? × 2.8

Westfield   R11/898   7.2 × 3.2

Puriri   T12/318   4.0 × 4.0 (door to east)

Raupa   T13/13   10.5 × 6.25

Whangapoua T11/644   4.0 × 2.75 (door to east)

   3.0 × 2.75

One such excavated terrace on Motutapu Island had three separate and different periods of use. In the first occupation there was an open courtyard with a house and two pits around the edge. Some time later when all evidence of the house and pits was covered over, the terrace was a convenient place to sit and make stone adzes. At another time, hangi were dug. Houses were probably on the adjacent terraces (Leahy 1970, 1972).

Few houses have been excavated. They vary from temporary shelters to well constructed houses (see Table 1). Often there has been so much re-building and re-organisation of a settlement the posthole outline of individual houses is difficult to find among many apparently randomly patterned postholes.

Garden sites

Garden sites are difficult to find unless there are stone walls and mounds such as on the volcanic fields of East Tamaki and Otahuhu. In other places shallow parallel channels or ditches running down hillslopes may also indicate garden plots. Excavation may reveal that sand or gravel had been added to the clay soil to make it lighter, for example, at Rocky Bay on Waiheke Island or on Kauri Point pa, Bay of Plenty (Law 1975).

Artefact manufacturing places

Artefact manufacturing places such as stone workshops where adzes were shaped by flaking blocks of stone, and grinding grooves for shaping and sharpening adzes, are not common. Stone workshops are situated on various beaches on the east coast of the Coromandel Peninsula, and on Motutapu Island, Rakino Island and the shores of the Tamaki River. A number of the settlement sites from the 13th to 15th centuries have considerable evidence for the making of artefacts such as ornaments, fishhooks and adzes. Sites from later in time with similar evidence are very few and include Oruarangi and some of the other Waihou sites. The lack of flakes from later sites can be attributed to the use of different stone materials which could not be worked by the traditional

14

Site Distribution

method of flaking. Greywacke was too hard to flake easily and accurately and so bruising the stone with a hammerstone became the main way by which adzes were shaped. The by-product of bruising is a fine powder which cannot be detected in the ground.

Other sites

Other sites, including some which fall into the categories of art and ritual, are difficult to find and interpret. Symbols and stylistic faces carved into the rock in caves or on large boulders represent another dimension to decoration known more widely from portable artefacts and wood carvings. Rock carvings are known from a number of locations: Ongare Point, Whiritoa, Papa Aroha, Coromandel and Kaihere to the west of the Hauraki Plains. At Flaxmill Bay, Whitianga there is a face carved into the rock on the edge of a pool within a stream bed. Obsidian flakes were found within the pool (Law 1966). There are undoubtedly a number of other sites like this that may have had ritual significance.

A different type of site also considered to have ritual or tapu associations was the swamp at Kauri Point. Here a small wooden enclosure had hundreds of broken hair combs made of wood and thousands of obsidian flakes along with many other wooden objects such as spears, musical instruments, gardening tools, wooden containers, plus cordage and gourds (Shawcross 1976). The site is dated to the early 16th century and later.

FIGURE 7: FISHTRAP AT SANDY BAY

15

ARCHAEOLOGY IN THE HAURAKI REGION

Other site types which are rarely found because of the limited archaeological evidence are related to the use of the foreshore: canoe landing areas which are only apparent if stones have been cleared away; stone fishtraps of which only four are recorded from Colville Bay and Port Charles on the Coromandel Peninsula (Fig. 7) and Browns Island; and eel weirs.

Burials

Burials have been found in a number of locations, including beach dunes. It was a common East Polynesian practice to bury people near or in the settlement, a custom which was brought to New Zealand. Over time however there was a shift to burying people in hidden places away from the villages, and to secondary internment after appropriate ceremonies had been conducted. Burials have been found during excavations, for example, Raupa at Paeroa and Station Bay on Motutapu Island, but in the majority of cases the burials post-date the time the site was lived on.

Pattern of settlement

Recorded archaeological sites are mainly within one kilometre of the coastline except in major river valleys where sites can be found some distance inland. Although the map distribution is partly a reflection of where archaeologists have been, several extensive site surveys on the Coromandel Peninsula failed to find evidence of occupation more than two kilometres from the coast (Furey 1987). There was undoubtedly use of the inland areas for hunting birds and rats, for extracting stone and for routes between areas, but occupation sites may have been fewer in number, and difficult to detect in the absence of terraces or shell middens.

From the recorded sites the settlement pattern can be broadly reconstructed as it related to the economic cycle. Small middens and terrace sites suggest temporary occupation by whanau on a seasonal basis while gardening, gathering forest foods or other products. Temporary fishing camps are also apparent. This type of site is present throughout history. In the early period (up to 1500 AD) some sites, for example, Sarah's Gully, may have acted as base camps or semi-permanent villages at which a wide range of activities were carried out including gardening, food storage and burial of the dead. Associated with the villages were hunting camps which may have only been occupied for a few weeks and where a limited range of activities were carried out. Later, pa became the focus of settlement, to be identified with, and for protection if needed. However some pa were only used in a very limited way and could not be termed semi-permanent defended villages. People continued to live in semi-permanent villages or in small whanau-based camps, probably on a seasonal basis while tending gardens or using localised resources.

16

Site Distribution

Case study—Waihou River

Few areas within the Hauraki region have been studied in sufficient detail to allow description of the settlement pattern over time. One such area is the lower Waihou River where archaeology has been used in conjunction with Land Court records to outline an occupation sequence based on political expansion of Marutuahu tribes and environmental change over time (Phillips 1994).

There is no direct evidence yet for pre-1450 AD settlement in the Waihou area but it is implied by the presence of sites on the east coast Coromandel and islands of the Hauraki Gulf. From about 1450 AD undefended settlements were situated on natural shell banks near the Waihou River mouth, evident from the lower occupation level of Oruarangi and forest clearance at Hurumoimoi (Kirikiri). Between about 1550–1650 AD pa were constructed, the river banks were gardened and undefended settlements were situated on elevated levees beside the river.

The arrival of Matutuahu tribes into the Hauraki area around 1650–1700 AD led to changes in the settlement pattern which coincided with environmental changes in the lower Waihou River, and resulted in adaptations to living sites and an increase in the number of pa. There is evidence for a slight rise in sea level, resulting in settlements being raised up to 700 mm by the addition of shell mined from natural shell banks. Expansion southwards along the river necessitated construction of pa to provide protection. The first occupation of Raupa was around this time. Over the next 50 to loo years more pa were built in the area reflecting warfare and competition both internally and externally. By the end of the 18th century Marutuahu were a powerful political force, with a fearsome reputation in the Far North (Collins 1798). Internal political power may have been controlled by an ariki whose influence extended over an area of the river. In 1801 Oruarangi was the residence of such a chief (Missionaries' Journal ms); in 1820 an unnamed chief at Raupa also controlled a stretch of the river, as did Te Puhi and Te Horeta near the mouth of the Waihou (Elder 1932:255).

The environment in the early 19th century continued to deteriorate, with bush clearance in the hill country to the east of the Waihou River leading to erosion and heavy sedimentation in the tributaries and the main river. The Whakarewa Channel on the eastern side of Tuatahi Island adjacent to Oruarangi became progressively silted up until by the 1870s it was completely dry land.

17

2. CHRONOLOGY

The age of a site is determined by radiocarbon dating which measures the time at which the sample being analysed died. Radiocarbon dating will not give a calendar date, rather it indicates a range within which there is a high statistical probability that the sample ceased living. By choosing samples which correlate closely with the archaeological activity, for example, shellfish in a midden or twigs from a fireplace, an age range for that activity is obtained. Because of the short time-scale and the lack of distinctive markers in Maori culture, there is little in the greater part of New Zealand's settlement history which can be used to assess the age of sites without resorting to radiocarbon dating techniques. The exception is the presence of distinctive artefact types closely resembling East Polynesian styles and therefore assigned to the early period of settlement of New Zealand. The remains of sea mammals, moa and other extinct birds in a site also contribute to a classification of 'early' (pre-1500 AD).

The settlement of Hauraki, in keeping with dates for other regions in New Zealand, is condensed into a relatively short time-scale of about 500-600 years before European arrival. While there have been previous claims made for occupation of 1000 years, there is no evidence to validate the assertions. Anomalous radiocarbon dates of an early age can be individually discounted on the basis of the type of wood used (often driftwood used in fires, long-lived species or heartwood posts) and in several instances old results on shell suggest that natural beach deposits rather than food remains have been dated. The reliable dates all fit comfortably after 1250-1300 AD (Fig. 8). The Polynesian settlers explored and settled most areas within a century. Although the earliest dates for settlement are found on the east coast of the Coromandel Peninsula and islands of the Hauraki Gulf, the Hauraki Plains have evidence of occupation at Oruarangi in the 15th century and the Tamaki area was also well known at that time. No sites of an early age are known from the North Shore and northern coast, but occasional finds of adzes of an early type suggest activity here also.

A model of settlement based on initial rapid exploration of the country has been proposed (Anderson 1991). Thus major stone resources of basalt, argillite, chert and obsidian are found in all known early sites. After initial exploration people may have become more locally focussed and developed territorial rights, for example, the

18

Chronology

Coromandel people worked basalt and made adzes for trade to other communities. The size of the founding population is unknown but according to various theories ranged from a few dozen to hundreds of people in a deliberate colonising expedition after an initial reconnaissance voyage. This is not an unlikely scenario—two way voyaging has been documented in the archaeological record in East Polynesia and Mayor Island obsidian has been found in a site dated to the 13th century on Raoul Island in the Kermadec Group (Anderson and McFadgen 1990). This raises the possibility that voyagers did return to the homeland with news of an unoccupied land mass rich in resources. It is also possible there were multiple settlements of New Zealand over several hundred years—a pearl shell trolling lure found in an archaeological excavation at Tairua, dated to at least 100–150 years after initial settlement, was an imported item as pearl shell does not grow here (Green 1967).

Dates for the building of pa, after 1500 AD, are also similar to dates in other regions. Sites at Kauri Point (Bay of Plenty), Sarah's Gully Pa and Castor Bay all have at least one undefended occupation prior to the defences being constructed.

The Tamaki area had an extensive population by the late 16th-early 17th century. Several settlements among the gardens are all dated to around the same time and the initial activity on Maungarei was also probably in the late 16th century (Davidson 1993).

Calibrated radiocarbon dates are presented in Appendix II.

19

FIGURE 8: RADIOCARBON AGE RANGES OF SITES

20

continued on p 22

21

FIGURE 8: RADIOCARBON AGE RANGES OF SITES (continued)

KEY

s = shell

c = charcoal

br = bracken fern gr = grease

be = bone collagen

22

3. SUBSISTENCE ECONOMY

Marine foods played an important role in the Maori economy. Fish from open sea, rocky shore and estuarine environments are found in middens, along with large quantities of shellfish. Approximately 70% of recorded sites have visible shell midden although this is an under-estimate of the true value of seafood.

Detailed analysis of the food waste in middens allows the economy to be reconstructed at a general level. While lists of bird and fish species found in excavations are produced, exact numbers of individuals or species are not possible for several reasons. These include:

1. processes of decay. Bones of eel and flounder and the cartilaginous skeleton of ray and sharks are rarely found in sites yet references in historic accounts and Land Court records emphasise their importance. Inanga and fresh water fish have left no trace whatsoever. Food preservation techniques such as potting of birds may have led to softening of bone and subsequent rapid decay. The full range of vegetables and plant foods is also unable to be identified.

2. the absence of certain bones or parts of the body which are integral to identification of species, particularly relevant where the heads of fish have been removed at another place.

3. changing techniques in analysis of midden samples. Over time archaeologists have become more aware of the information which can be recovered from middens. In the early years of archaeological investigation, little or no midden material was analysed and certainly not quantified. Detailed examination of several middens has demonstrated the importance of preserved foods and summer and hunting camps.

Fish

A wide range of species of fish was caught by various means—line fishing, trolling, spears, nets and traps. The older sites have a wider variety of species than the more recent sites (Fig. 9 and Table 2). This is partly due to the location of the older sites on the east coast of the Coromandel Peninsula. The excavated sites of a more recent age tend to be from the more sheltered waters of the Waitemata Harbour and lower Firth of

23

ARCHAEOLOGY IN THE HAURAKI REGION

Thames. Snapper occurs in all sites regardless of age or location. However blue cod, barracouta, kahawai, leather jacket and labrids are present in over half of the early sites but only kahawai and Elasmobranchs (sharks and rays) are present in more than one third of later sites. Under-represented species include flounder and eel. These species were mentioned repeatedly in Land Court cases at Kauaeranga and in the lower Waihou River area (for example, Hauraki MB 4:217-29 and Hauraki MB 58:123; Hauraki

MB 60:150).

There is little evidence for deep off-shore fishing. The bones of hapuku (one individual) have been recognised only in the Port Jackson site although hapuku holes are traditionally known in Hauraki. Whale meat was probably obtained from beach strandings. Sharks are known to be important historically (for meat and oil from the liver) but there is little evidence archaeologically. Large wooden hooks found in some Waihou sites were probably used to catch shark.

Small fishing camps were established through the Gulf area in summer. One such site at Galatea Bay on Ponui Island had many bones from the heads of snapper but few vertebrae, indicating the dried fish were taken elsewhere (Shawcross 1967). The Sunde site on Motutapu Island had similar evidence but small snapper were eaten on site and fish with an estimated length of more than 6o cm long had the heads cut off and were taken away from the site (Nichol 1988).

FIGURE 9: NUMBER OF SITES WITH EACH FISH SPECIES

24

25

ARCHAEOLOGY IN THE HAURAKI REGION

Shellfish

Analysis of middens shows a change in preference for shellfish species over time. For instance at Tairua, the lower layer (early 1400s) had rocky shore shellfish but the upper layer (early 1700s) had sandy beach species with pockets of rocky shore species perhaps suggesting isolated meals. At the Sunde site, rocky shore species dominate the lower midden under the Rangitoto Ash but pipi and cockle account for about 95% of shells in the upper or later midden, possibly explained by changes in the local beach environment as a result of the eruption of Rangitoto. Very detailed examination of the contents of the Sunde midden below the Rangitoto Ash on Motutapu found kina were the most important by meat weight, contributing more than other shellfish combined (Nichol 1988).

The first people in an area gathered the large rocky shore shellfish, probably following similar practices used in the Polynesian homeland on the extensive coastal platforms. The most commonly eaten rocky shore shellfish were mussel, cats eye, cooks turban, white rock shell, oysters and limpet.

Generally speaking there has been a trend over time away from the rocky shore and towards greater exploitation of estuarine and sandy beach environments. While the early sites have mainly rocky shore species with some sandy shore species, late sites are almost all cockle, pipi and tuatua with small numbers of gastropods such as Amphibola and Cominella. This is probably because the soft shore environment was more able to sustain regular harvesting.

Estimates of meat weight from various foods in middens indicate shellfish was a minor component in the diet of the early settlers. Fish and sea mammals such as seals formed the bulk of the protein eaten.

Shellfish commonly represented in middens include cockle, pipi, tuatua, oyster, paua, mussel, scallop, cats eye, Amphibola spp., Cominella spp., dog cockle, cooks turban and white rock shell. In some sites the shells of oyster, dog cockle, paua and cooks turban have been used to make ornaments and fishhooks.

Marine mammals

Evidence of sea mammals occurs only in sites dated 400–700 years ago (see Table 3). The remains of fur seals, sea lions and elephant seals are present in a number of sites. The bones of pups and juveniles indicate there were breeding colonies of seals in the vicinity of some of these sites. Winter hunting is suggested from the juvenile bones at Opito, Sarah's Gully and Tairua. Fur seals contributed enormously to the meat supply in some sites—for instance a minimum of 17 fur seals were eaten at N40/3 at Opito, eight at Sarah's Gully and six at Tairua (Smith 1981). With breeding colonies gone, the importance of seal declined but it is likely opportunistic killings of seals in northern waters continued.

z6

Subsistence Economy

TABLE 3 : MARINE MAMMALS AND OTHER SPECIES IN SITES

P = present

Fishhooks made from whalebone are found in several early sites, and later sites in the Waihou area have patu and other large objects made from whalebone. At Oruarangi a number of bone hair combs have been fashioned from dolphin or pilot whale jaw bones. Undoubtedly the bone attributed to large whales came from fortuitous strandings but dolphins and possibly pilot whales may have been hunted for food and for their bones.

Dogs, rats and reptiles

Dog (kuri) bones have been found in many sites. Usually there are the remains of only one dog but at Whangamata Wharf site there were 17 from one occupation, 19 at Sarah's Gully from several occupations and N40/3 at Opito had 19 from one occupation (Smith

27