Nelson Historical Society Journal, Volume 2, Issue 2, June 1967
Recent Archaeological Excavations in The Northern Part of The South Island
Recent Archaeological Excavations in The Northern Part of The South Island
With the arrival in New Zealand of European whalers, traders and settlers, the Maori, within the short span of a few years drastically changed his manner of living—readily adopting to his use European tools and weapons and sometimes leaving his former place of habitation to be closer to the coastal vantage points where the Pakeha congregated. Stone adzes were flung away in preference for steel tomahawks, jack plane blades or pieces of hoop iron. Wooden, bone and shell fish hooks were rapidly replaced by bent nails, forged spikes or real iron hooks, while European garments quickly cramped the movements of bodies used to the freedom afforded by Maori clothing. And so, in a relatively short space of time the material things of the old time Maori had become obsolete and of no further use to him. There were of course, notable exceptions to this, in the form of ornaments such as hei-tikis, and certain weapons which possessed "mana," but the frequency with which even these notable artifacts turn up in collections of Maori craftsmanship made in the early 1800's, indicates that their value to the Maori was fading.
The ever increasing tempo of inter-tribal warfare in the 18th and early 19th centuries did much also to force the abandonment of numerous pas and villages previously occupied. Sites where tribes or hapus may have lived for centuries, were abandoned for the less vulnerable positions of islands, peninsulas or hill tops.
Nature quickly covers the areas scarred by man's occupation and many sites occupied in the past by sizeable Maori populations, are no longer recognisable as such. The five sites which I wish to speak about tonight (all of them vitally important in the pre-history of Nelson) are ones which bear absolutely no superficial sign of the intense activity which must have taken place there hundreds of years ago. A walk across the famed Wairau Moa-hunter site near Blenheim is almost like walking across any stony Marlborough paddock in summer. A visit to the Heaphy River mouth on the West Coast Nikau palms, a few lupins and golden granite sand belie the presence of an Archaic Maori site beneath. The back lawn of a Tahunanui residence 30 yards from the Nelson-Richmond highway—between the fruit trees and the vegetable garden there is little to suggest the presence of one of Nelson's earliest settlements.page 6
Before examining the five sites under consideration, let's consider some of the cultural changes which are known to have taken place in New Zealand's pre-history.
Two definite cultural phases have evolved in the course of archaeological investigations throughout New Zealand, the Moa-hunter or Archaic Maori phase and the late Maori or Classic Maori phase, the latter represented by the natives observed by the early European explorers to New Zealand. The preferred technical designations are New Zealand Eastern Polynesian I for the early culture and New Zealand Eastern Polynesian II for the later culture. But for convenience it is easier to refer to the first as the Archaic Maori and the second as Classic Maori.
Our most extensive picture of the earliest Polynesians in New Zealand is obtained from the Moa-hunter or Archaic site discovered by Jim Eyles some 25 years ago at the mouth of the Wairau River. Museum work over a period of years has substantiated the collection of material originally made by Mr. Eyles and although it would be presumptious to say the artifactual representation is complete, it is certainly the most extensive in the country. Because of the durability of stone and bone artifacts, these alone have survived to give us a picture of this early culture. Stone adzes from Wairau show that the Moa-hunters had a comprehensive and varied tool kit. Large quadrangular sectioned adzes with a definite provision at the butt end for lashing to a handle, adzes of triangular cross section and the side-hafted adze are typical. Because of their close association with the moa, these people made extensive use of the bird for food and for raw material for tools, fishing gear, ornaments and clothing. There is little evidence to suggest that the Archaic Maori cultivated the kumara and the lack of weapons in Archaic sites seems to indicate that population pressures had not yet developed sufficiently to cause inter-tribal fighting.
The Classic Maori culture is typified by the simplification of the adze form, adzes often being much smaller and lacking observable provision for lashing, the type 2B adze in Duff's classification being the predominant form. Greenstone was used more extensively and the development of ornaments was climaxed with such forms as the hei-tiki and hei-matau. Fishing gear had altered considerably with the emergence of a paua faced lure shank and bait hooks were predominantly of the composite type, with wooden shank and barbed bone point. Shell fish were consumed in large quantities, the exotic kumara was cultivated and elaborate defences and earthworks were thrown up around inaccessible vantage points to protect the tribe from ever increasing enemy attacks. The decorative arts, particularly wood carving had reached a high standard of development.
Excavations at Anapai
The Anapai site is situated on the West Coast of Tasman Bay in the area controlled by the Abel Tasman National Park Board. This Archaic site comprises a small midden fronting on to the beach at present being eroded at times by the sea. Large quantities of waste argillite flakes are common, indicating considerable adze manufacture and the common occurrence of chips from completed adzes would indicate wood working on the site. The stratum is essentially granite sand and in places the occupation layer is up to four feet thick. In the lower levels of the occupation the bones of the small Northern Bush Moa, Anomalopteryx didiformis) occur, while the rest of the midden material is comprised mainly of seal, rocky shore shellfish and bones of various bush and shore birds. Fragments of bone of the Polynesian Dog (Canis familiaris) are present and it is interesting to note the occurrence of an undescribed kaka also. Mr. Ron Scarlett of the Canterbury Museum has identified this undescribed species from some other sites as well. Artifacts were not common during the execavation but the presence of typical Moa-hunter adze fragments, a typical Moa-hunter one piece fish hook and the association of moa bones with human occupation, definitely established this site as being Archaic Maori. Insufficient material was obtained for carbon dating but indications are that the site would probably be late Archaic.
Heaphy River Excavations
Situated at the mouth of the Heaphy River in North-west Nelson this is one of the most picturesque sites imaginable. The river carves its way through rugged limestone and granite country to the sea and large groves of nikau palms growing almost to the coastal edge, enhance the tropical atmosphere of the area. If one can bear with the over-sized sandflies, this is indeed a most pleasant area in which to excavate.
The site, an Archaic one, probably covered several acres originally, but considerable river erosion over the past few years has carried away much of the site. Continuing erosion threatened the rest of the site, so main excavations were carried out during 1963 and 1964.
The occupation layer varied from 10 to 24 inches thick and was composed mainly of intensely black charcoally sand, while towards the bottom of the occupation layer occurred hundreds of granite stones which had been carried from the river to serve as oven stones. Three caches of adzes were excavated from beneath this oven stone layer and many oven pits were excavated and plotted in the clean granite sand beneath the oven stones. The most interesting structures excavated at the Heaphy were three stone pavements composed of flattish slabs of granite and limestone. These pavements were of no great size; the largest being approximately page 8nine feet across. As these occurred in an area obviously used as a workshop area, their significance remained obscure until four days of continuous West Coast rain led to the conviction that the pavements were intended to keep the Archaic craftsmen and their equipment out of the granite slush. Carbon dating for the site stands at 1518 A.D. plus or minus 70 years, and this date together with the site's closer proximity to the West Coast sources of greenstone, accounts for the relatively high incidence of greenstone artifacts excavated on this Archaic site. Lure shanks were manufactured on the site and if they were used there, which is most likely, one can only admire the skill of the Archaic Maoris in guiding their canoes through the West Coast breakers. Anomolpteryx and Emeus crassus are the moas represented in the midden material from this site.
The Wairau Moa-Hunter Camp
This is indeed the most interesting Archaic site in New Zealand and the one with which the general public is most familiar. The grave offering found with burials of chiefly rank comprise some of the finest known Archaic artifacts. A great deal has been written about this site and I am sure that many of you will be familiar with Dr. Duff's book, "The Moa-hunter Period of Maori Culture." The site is a very large one, extending over 20 to 30 acres and it is the opinion of one authority that in Moa-hunter times the site was considerably larger, extending Northwards into the area now cut into by the Wairau River.
Resultant work from the excavations of 1963 and 1964 has been concerned mainly with the analysis of waste stone flakes and shell midden analysis. It is hoped that from the shell sampling it will be possible to obtain some idea of the total protein consumed on the site and perhaps an indication of population figures. An estimate has been made from the fragments of moa-egg shell found, that some 2,380 moa eggs were consumed. This would be a minimum figure, based on the average surface area of an egg as 140 square inches. From the waste flakes collected and examined, an estimated 30 tons of waste argillite litter the site. Wellman considers that in the manufacture of an average sized adze weighing 1½lbs., six pounds of waste flakes would result. This implies that some 3,000 adzes were manufactured at Wairau. These figures must not be taken too literally, but they do give some idea of the extent of the site.
The Archaic Site at Tahunanui
No two Maori sites are ever the same. Each one has its own peculiarities of stratigraphy, topography, artifact and midden composition and so on. The site at Tahunanui is no exception to this. Being on the property of relatives, it is the only site I know of where one is assured of "on-the-spot" accommodation and morning and afternoon cups of tea.page 9
The site itself is situated next to the Tahunanui Post Office and in Maori times would have been only a few yards from the sea. Levelling and filling-in on various parts of the property have destroyed the original topography, but it is also possible that these same processes have removed much of the post-Maori accumulation of sand and soil. Argillite flakes were observable in the vegetable garden and so early in 1964 a trial excavation of a five foot square was made—in the centre of the back lawn. To date some 400 square feet of the site have been excavated.
In some places the occupation layer is less than two inches below the surface and the average dept of the occupation is 6–8 inches. In spite of the comparative shallowness of the occupation layer, the density of artifacts has been quite high. As one would expect, the site's close proximity to the Maori argillite quarries of the Nelson mineral belt has meant that considerable pre-occupation with adze manufacture has taken place. Waste argillite flakes litter the occupation layer with considerable density and broken portions of "roughed-out" adzes are common artifacts. Obviously, the vast majority of adzes were being prepared for trade, or certainly for removal to another site, as few completed adzes have been excavated so far and those which have come to light have been small and somewhat insignificant, but typically Archaic in form. In the future the study of these waste argillite flakes is going to assume increasing importance as it may be possible to determine trade routes, match varying kinds of argillites to their quarry sources and possibly even determine the extent of tribal diffusions or movements in pretraditional times. The argillite which occurs at the bottom of the occupation at Tahunanui is of the light grey variety—usually called D'Urville Island argillite, but as one rises through the layer, another type, presumably from a Whangamoa or Maitai source, replaces the light grey variety. Comparison with two other sites reveals a similar change but a great deal of work needs to be done before firm conclusions can be made. Boxes of flakes from Tahunanui and Cable Bay will probably be worked on by members of the Archaeological Group this winter.
The commonest artifact at Tahunanui is the argillite drill point. So far, over 150 of these have been catalogued. Their primary use was in the manufacture of one-piece fish hooks made from moa bone. A series of holes was drilled in the central portion of the bone tab and the resultant "core" snapped out. By using this method of manufacture, considerable time was saved in filing the hook into shape. The high density of drill points indicates the manufacture of considerable numbers of fish hooks on the site and the amount of snapper, kahawai and barracouta bones present in the occupation layer indicate the effectiveness of the fishing gear manufactured. Predatory "school" fish such as kahawai and bara-couta were caught by using a lure hook—a small stone shank shaped page 10like a small fish with an unbarbed bone point attached, was trolled through the water enticing the fish to strike. Several of these characteristically Archaic lure shanks are represented.
Several of the bone fish hooks from this site have been made from moa bone and fragments or the Northern Bush Moa occur in the midden. Other creatures which have contributed to the food supply of the occupants include the Polynesian Dog, Southern Fur Seal, Weka, N.Z. Pigeon, Spotted Shag and several other sea birds.
Evidence of man made structures has been restricted to two fire pits (not haangi) from which charcoal has been obtained for future carbon dating analysis. The presence of greenstone is represented by a small polished fragment of nephrite and a finely made, well polished chisel from the bottom of the occupation layer. The latter artifact is unusual in that it has been designed as a dual purpose tool, being chisel-shaped at one end and gouge-shaped at the other. This small chisel may well be representative of the scarcity of greenstone during the Archaic period.
Rotokura — an Archaic and Classic Maori Site at Cable Bay
This site differs from the previous ones discussed, in that it represents both the Archaic and Classic phases of Maori Culture. This type of site is not yet well represented in archaeological investigations. We have Mr. R. Tobin to thank for bringing this site to my notice last year and the Archaeological sub-committee unanimously agreed to its immediate excavation in view of its importance to Nelson's pre-history. The property owner, Mr. F. D. Stuart, has been extremely co-operative and has been of considerable assistance on many occasions.
The early Maori name for the small lagoon which lies adjacent to the site is Rotokura—Red Lake—certainly a more euphonius name for the site than Cable Bay, and one too which localises the site more efficiently.
The surrounding topography—lagoon, boulder beach and hillside—has had the effect of restricting the Maori inhabitants to a semi-flat area measuring approximately 35 yards long by 20 yards wide. This has reduced considerably the scatter of occupational residue and consequently the concentration of artifacts, midden refuse and structures has been considerable. To date, approximately 550 square feet of the site has been excavated and an impressive list of artifacts has been catalogued. The natural stratum consists mainly of beach boulders and this makes for extremely difficult excavation in some areas of the site. The clean-cut work of the Heaphy excavation is almost impossible to duplicate at Rotokura.
Although no carbon dating (or any other form of dating) has yet been carried out for the site it would seem reasonable to assume, in view of the present evidence, that Rotokura was first occupied page 11some 500 to 600 years ago. The first inhabitants were undoubtedly Archaic or Moa-hunter Maoris. The first meals eaten on the site comprised at least three of the seven genera of moa. Large, often long, flake knives of argillite have been found among the moa bones. These quickly made, but effective tools, would be used for cutting up moa flesh and dismembering the Southern Fur Seal, the bones of which also occur in quantity in the earliest occupation. One piece bait hooks were made from the thick walled upper leg bones of the moa and many fragments of these hooks have been excavated. Some have been broken during the manufacture while others have been broken during use. Fishing lure shanks made from bone and a variety of stone are well represented in the Archaic layer. A variety of shellfish was eaten, but at a time when the moa and seal provided a plentiful food supply, shellfish did not form an important constituent of the Archaic Maoris' diet. The Polynesian Dog is also represented in the earliest occupation and several bones with gnawed surfaces, indicate the presence of the Polynesian Rat. Adzes were manufactured mainly from argillite, although some of greywacke occur. The occurrence of completed adze portions indicates far more than an adze manufacturing centre, but the restricted nature of the site does not support the idea of a sizeable population in Archaic times. It seems far more reasonable to assume that the site was occupied only seasonally and that the main settlement area existed somewhere else, within a radius of say, four or five miles. A cache of three argillite adzes has been found in the Archaic level and two of these are typical of the massive quadrangular adzes manufactured in Archaic times. As Archaic occupation progresses, moa and seal bones gradually diminish and artifacts previously made from green or fresh moa bone, are now being rendered in old or sub-fossil bone. It seems quite possible that the decline in moa bone supply was responsible for certain changes in the manufacture of bait fish hooks. Large pieces of bone were no longer available for the manufacture of one-piece hooks and so the hook was made from two smaller portions of bone, one piece being drilled to take a lashing, and the two pieces were then lashed together to form an all bone composite bait hook. Later still in Classic Maori times when the only sizeable bone readily available was human, the bait hook consisted of a curved wooden shank with a small barbed bone point attached. It must be pointed out that this is as yet unproven, but seems to be borne out by excavations at Rotokura.
Some 60 to 70% of the occupation at Rotokura can be assigned to the Archaic phase, and then unfortunately due to lack of artifacts, an indeterminate zone of occupation occurs, before the Classic Maori phase is represented close to the surface. It had been hoped that some light would be cast on the transitional period between Archaic and Classic cultures but this has not yet been the case. The Classic page 12phase at Rotokura is undoubtedly representative of tribal invasions from the North. Adzes are still of argillite, but are generally much smaller and are Type 2B in Duff's classification (the typical adze of the Classic Maori). Fishing gear is represented by the small barbed bone points of composite fish hooks, some of these being manufactured from dogs' teeth which require little modification. Shellfish now predominate in the middens, with a wide range of mud-flat, sandy beach and rocky shore species being represented. Bird spear points and small bird bone fragments are more numerous than in the Archaic context and small polished fragments of greenstone tools are present. Drill points are less numerous than in the Archaic occupation which is understandable in view of the change in fish hook type. European contact is represented by the presence on top of the occupation layer, of portions of the stem of a clay pipe and some very rusty pieces of an old iron "go-ashore" pot of the type commonly used by sailing ship crews.
The picture which is emerging, of the sudden termination of Maori occupation at Rotokura is indeed an interesting one. Although historically unsubstantiated, it is most likely that Te Rauparaha and his allies were responsible. The scattered fragments of a finely made patu onewa (stone club) probably testify to the vengeful destruction of a weapon which was raised against the invaders, while a few smashed fragments of human skull tell an even sadder tale of the fate of at least one defender.
Undoubtedly one of the most interesting features of the site at Rotokura is the association with man of three different kinds or genera of moa. The presence of the small Northern Bush Moa (Anomalopteryx didiformis) is not unexpected as it had previously been recorded from the Tahunanui and Anapai Archaic sites, but the presence of Dinornis torosus (a representative of the large moas) is most notable. Although remains of this large bird have occurred in occasional middens in the North and South Islands, its distribution in association with human occupation has been most limited, and until a few years ago, Dinornis was believed to be extinct prior to the arrival of the Polynesians in New Zealand. Megalapteryx, the other moa represented on the site, can be likened in many ways to a giant kiwi. This South Island bird is far more characteristic of the southern part of the island. Two other extinct birds represented in the midden material are the extinct swan and the little extinct weka.
I am indebted to Mr. R. J. Scarlett of the Canterbury Museum, for his identification of the bone material.