- 241

For McNab's History of Northern New Zealand, as cited above, Captain James Cook's 1770 sojourn in Queen Charlotte Sound constituted an historical benchmark as far as the “demonstration” of Maori cannibalism was concerned. Given the complex European response of fascination and revulsion, and in some instances, a desire for notoriety, “proof” of anthropophagy lent itself particularly well to the Cook expeditionary records, encouraging “an orgy” of shipboard entries (Groube 1977:87), with graphic details and philosophical speculation aplenty. It is little wonder, as Fagan (1984:234, 238-9) has implied, that these earliest accounts would initiate a “formidable reputation” for Maori cannibalism.

Yet, in spite of the general assumption of cannibalism's ethnographic reality among the 18th-century Maori, it is also worth considering the challenge to such an interpretation offered by Arens (1979:18, 21):

Arens (1979:165) concluded that all of the claims of cannibalism he had investigated were characteristised by a “fleeting quality” of documentation. A recent archaeological interpretation of violence from an Illinois burial site has also introduced its findings with the following relevant concerns: - 242

From the 19th-century New Zealand documentary record, problems also exist for understanding Maori cannibalism as a phenomenon rooted in the pre-European past. Until the first decade of the 19th century, there had been little Maori interest in the acquisition of muskets. However, a growing desire for such items after 1810 escalated after the establishment of the Bay of Islands Anglican mission between 1814 and 1815. The differential northern access to knives, axes but most significantly, firearms — a pattern repeated in the selective arming of southern groups — was to have a violent and brutal impact without precedent in Maori society (Ballara 1973; Best 1904:12-15; Binney 1969:145-7, 149-50, 160-l; Burns 1980:104, 169 and passim; Colenso 1868:66-7; Shawcross 1967; Sorrenson 1979:71; Thomson 1859:1:244-66; Ulrich 1970; Vayda 1976: 87-102; for a contrary, minority view see Owens 1968:32-3 and Smith 1910:17-18).

Pre-European patterns of warfare were both transferred into and transformed within this new context (Ballara 1976; Burns 1980:171; see also Ballara 1973). However, it is often difficult to determine whether particular incidents of violence and reported cannibalism of the early 19th century musket wars were related more to the arms imbalance and a society undergoing radical and stressful change, or to pre-European cultural precedents. (Consider, for example, the contribution of the musket to the large feasts of human flesh which followed the conflicts reported in Best 1902: 72-4 and Smith 1910:105-6, 112-14, 159-60, 186-8). Furthermore, in the politicised environment of earlier 19th-century European settlement, descriptions of this new warfare are often conveyed in emotionally charged language, with much credence given to secondary reports or undocumented claims of Maori “savagery”.

Consequently, to assess cannibalism critically as a behaviour of 18th century Maori culture, this paper generally confines itself to a review of the historical record from the period before the impact of musket warfare and the permanent mission presence in the Bay of Islands. In doing so, and beyond the present concern, I do not intend to imply that all post-1815 records are without value for an understanding of “traditional” cultural patterns (on which, see Anderson 1984:266-7). Neither do I imply that all records before 1815 necessarily record a pristine precontact Maori society. For over a decade before, it is apparent that the people of the northern North Island had experienced the “cultural disturbance” of European trade items, crops, animals, diseases, and even ideologies (Owens 1968:26-7), as had some southern regions to a certain degree. There is

- 243

Figure 1. Sites and locations referred to in the text.

- 244

also evidence of significant “disturbance” during the very first New Zealand visit of the Endeavour, to be discussed below. The temporal focus of this article is intended only to minimise problems in the understanding of “traditional” ritual violence, therefore; it does not remove them.

From the published ethnographic record, one must also be careful to separate analytical descriptions relevant to historical reconstruction from speculations grounded in a contemporary cultural milieu. In some accounts an apparent mix of fact and fancy is also compounded by a lack of concern for careful documentation. As a result, the contact expeditionary accounts associated with the Cook voyages are an important focus of this study. These provide an essential base for subsequent interpretation (see above), and, frequently, a number of journal entries covering the same event. Given as well the variety of “traditional” 18th-century detail recorded and the level of relative objectivity achieved (especially for Cook and his botanist, Joseph Banks), the records of the Cook voyages are generally without parallel among the historical, Pacific-ethnographic literature.

Against the ethnographic record, this paper also reviews the interpretation of pre-European cannibalism within the history of New Zealand archaeology. In spite of the discipline's contribution to the general interpretation of the Maori past (Duff 1956; Davidson 1984), the evidence for cannibalism has not figured prominently over the last four project-intensive decades of New Zealand archaeology (Davidson 1984:137). Even so, the topic has been incorporated into a number of scenarios of culture change, while site interpretations of Maori cannibalism have appeared sporadically in the literature since the 19th century, as well as in some more recent excavation reports. As to the general propriety of applying archaeological data in what amounts to a “direct historical approach” (Carlton 1981:152-4; Gould 1980:35, 205; Steward 1942), an observation by Groube (1977:70) with respect to such cultural sequences as that of the Maori is appropriate:

The “direct historical approach” in archaeology has also been criticised, given that it has encouraged overly simplistic interpretation and even data manipulation on occasion (Charlton 1981:152-6). Approached carefully, however, such an interface (involving archaeology and either historical text or contemporary observation) may prove invaluable to a more holistic - 245 interpretation of the cultural evidence (Charlton 1981; Gould 1981; Hodder 1982; 1991).

In this light, it is the overall intention of the present study to critically review the integrity and interpretative value of the European record of “traditional” Maori cannibalism before 1815, a record defined so as to incorporate archaeology and published historical ethnography. As such, the paper does not pretend to be a comprehensive treatment of the practice itself. Whatever meaning cannibalism may have had for the people of the late 18th century can be fully extricated only from a detailed analysis of tribal tradition and the integrated social, religious, and political structures of the time, a task well outside the present focus. Beyond the question of meaning, however, this article offers a methodological contribution in the cross-disciplinary assessment of an interpretative problem that is both historical and anthropological in nature. In this regard, the present study has at least as much to say about the assumptions of early observers and New Zealand archaeologists as it does about 18th-century Maori behaviour.

One further prefatory remark should be made with respect to the sources of this essay. To confine this study to manageable proportions, the data base (with the exception of unpublished University theses and personal communications) includes published items only. For early historical records, this restriction does not significantly affect the quality of research, in part because a number of insightful accounts of early New Zealand visits were published in England between the late 18th and early 19th centuries. Even more significantly, subsequent publishing ventures (most notably those of Robert McNab, the Hakluyt Society, and the Alexander Turnbull Library), have made available a broad range of important manuscript accounts, while Salmond (1991) has reviewed in critical detail both published and manuscript sources of the first meetings between Maori and Europeans from 1642 to 1772. The archaeological data base is not so consistently represented in publication, justifying the additional incorporation of unpublished theses. With these kinds of sources, a sufficent variety of archaeological detail is available from the North and South Islands to balance the early ethnographic record and provide a relatively comprehensive overview.

The Endeavour, 1769-1770.

Early on the afternoon of January 15,1770, after reaching the north-eastern end of the South Island, the Endeavour expedition of Captain James Cook anchored at Ship Cove in the outer reaches of a Sound known to the Maori as - 246 Totaranui (renamed “Queen Charlotte's Sound” by Cook; See Fig. 1). Interaction with local people occurred almost immediately, involving several canoes from a “village” on Motuara or “Hippah Island” which Ship Cove faced (Cook journal in Beaglehole 1955:234-5; Banks journal in Beaglehole

Figure 2. Totaranui (Queen Charlotte Sound), showing locations discussed in text “Indian Cove” (now Resolution Bay) is identified from Forster 1777:1:496 and Begg and Begg 1969:14, 122.

- 247

1962:452-3, after which the date above is followed. “Hippah” is a European contraction of He pā, the generic term for a defended Maori settlement preceded by the indefinite article. The pā site was on a small islet at the south end of Motuara, now known separately as “Hippa[h] Island”; see Figure 2 and Brailsford 1981:19, 20, 21; Trotter 1987:107).

On the following day, according to Cook (Beaglehole 1955:236), “some of us went in the Pinnace into a nother Cove [probably Little Waikawa Bay; see Trotter 1987:106-7 and Figure 2] not far from where the Ship lays”. According to the first, and unofficial, published account of the Endeavour's Pacific voyage (probably by James Magra, later Matra; see Beaglehole 1955:cclvi-cclxiv), the English crew there observed a small group of Maori, who appeared to be preparing provisions. On examining “several baskets in their canoes”, to “our great surprize” roasted limbs and other human body parts were noticed. To the chronicler it was evident these had been “lately eaten”, since teeth marks were seen in the flesh “which appeared to have been recently gnawed and torn” (Anon 1771:94). From the first New Zealand contact of the Endeavour, in October of the previous year, “we had been before assured, that the inhabitants of New Zealand were cannibals from their own concurrent testimonies in many different places”, the account continued, “but had never occular demonstration of the fact until this time”.

Inquiries were made as to “how this human flesh came into their possession” (Anon 1771:95). The English crew were told that a canoe-load of 10 men and two women, “from a different district”, had been driven into the Bay, where they were attacked and killed. (One of the women had drowned while attempting to swim away.) “Their bodies were afterwards divided among them, of which the flesh we had seen was a part”. “They were not ashamed of it”, the account continued;

The published journal of Sydney Parkinson, one of two artists on board the Endeavour during the January sojourn in Totaranui, supports the anonymous journal account. Relating the observations of “some of our people…[who] went into a little cove, where one family resided”, Parkinson (1784:115) repeated the detail that “human bones…lately dressed and picked” were seen. “Six of their enemies had fallen into their hands”, Parkinson explained, four of whom “they killed and ate”, while two were drowned in attempting to escape. Several human bones were also brought on board and offered for sale, - 248 “the natives…sucking them in their mouths”, and evincing by signs “they thought human flesh delicious food”. Parkinson (1784:116) also noted that human bones were found “in the woods, near the ovens, where they used to partake of their horrid midnight repasts” (the last a colourful but unsubstantiated aside). The Maori seemed to take “pride in their cruelty”, Parkinson added, “and shewed us the manner in which they dispatched their prisoners; which was to knock them down…and then to rip them up”.

Cook's journal entry is also in general harmony with the accounts just cited:

In light of the caveats raised by Arens, one must also consider carefully Cook's remark that “there was not one of us that had the least doubt but what this people were Cannibals”, and that “the finding this Bone with part of the sinews fresh upon it was a stronger proof than any we had yet met with” (meaning, presumably, since the 1769 arrival of the Endeavour at New Zealand) (Beaglehole 1955:236). However, almost anticipating a critical response to these observations, Cook remarked that to be “fully satisfied” of what had been communicated, one of the local people was told that the bone was that of a dog, rather than a man.

Banks had also retrieved a bone of the forearm, “much in the same sate” as that obtained by Cook, on which this action was also demonstrated. “To shew us that they had eat the flesh they bit a[nd] naw'd the bone and draw'd it thro' their mouth and this in such a manner as plainly shew 'd that the flesh to them was a dainty bit” (Beaglehole 1955:236-7). As a result of these events, Cook named the feature now known as Little Waikawa Bay, “Canibal's Cove” (a name by which the cove between Ship Cove and Little Waikawa Bay is now known; see Fig. 2 and Trooter 1987:106-7).

The journal entry of Joseph Banks for this event is no less graphic, nor less certain of the fact of cannibalism. Describing (as in the anonymous account) the arrival at the cove, Banks observed the family dressing a dog for - 249 consumption, and noticed “many provision baskets”. Two human bones “pretty clean pickd” were noticed in one of these, along with “evident marks of their having been dressd on the fire”:

The Banks version confirms the basic details of the previously cited observations, differing only in the total number of people involved. (Banks understood seven people had been killed and eaten.) During March 1770, in a retrospective account written before his departure from New Zealand, Banks conceded that he had initially been “loth to give credit” to claims of cannibalism before this January incident, a demonstration of “so convincing a proof I could not withstand”. Banks then observed that the Englishmen had been assured that “all the flesh of these people” killed in the January incident had been “eat up and most of the bones thrown away, which we found to be true for in almost every cove where we landed fresh bones of Men were found near the places where fires had been made” (Beaglehole 1962:2:30-1). This last detail, which appears in Parkinson's previously cited account, is also confirmed in Cook's entry for January 19. “Some of our people found in the skirts of the wood three hip bones of Men”, wrote Cook, near a hole or “hoven”, “where the natives dress their Victuals”, which “trifleing” circumstance was taken as further proof “that this people eat human flesh” (Beaglehole 1955: 237).

What appears ostensibly to be an eye-witness claim was also amde by William Picksersgill, who wrote: “we saw one of the Bodys and two arms with flesh upon them which we saw them eat”. This was the first “Proof Possitive we have had” that these people were cannibals, Pickersgill added (in Beaglehole 1955:236, n.l, and Salmond 1991:244). Unfortunately for the demands of documentation, Pickersgill provided no further details of this primary observation. Furthermore, in spite of Pickersgill's use of the plural pronoun, one should note that this first-hand detail was not confirmed by any other writer. Indeed, from the full descriptions of January 16 (as close as anyone seems to have been to recording a contemporary incident of cannibalism), it is clear that the English crew arrived after the human flesh had apparently been consumed. Pickersgill's explicit claim, therefore, may be an embellishment of the bone-gnawing incident, as discussed above.

On the following morning, a canoe came alongside “from the Indian town” (i.e., Hippah Island) and Tupa'ia (the ship's Tahitian interpreter) “began to - 250 enquire into the truth of what we had heard yesterday and was told over again the same story” (Banks in Beaglehole 1962:1:456).

Banks (p.456) was also told that the people “expected their enemies to come and revenge the death of the 7 men”, an expectation that seemed to be reflected in the unusual quietness of activity around the settlement. On January 20, Banks noted that the old man kept his promise and brought four preserved heads, with flesh and hair intact. The brains had been removed, which led Banks to consider that these were “a delicacy here” (Beaglehole 1962:1:457). One of the crania was purchased by Banks, as it offered (in his opinion) further demonstration of cannibalism; specifically, “contusions on one side” suggesting “it had receivd many violent blows which had chippd of a part of the scull near the eye” (Beaglehole 1962:2:31).

Before leaving the 1769-70 expedition, it is worth considering several observations made before the Totaranui sojourn. According to Banks, on the very first landing at “Taoneroa” (Poverty Bay) on the east coast of the central North Island, two local boys had questioned whether the meat they were offered on board ship was human flesh. The boys “seemd ashamd of the custom, saying that the tribe to which they belonged did not use it but that another very near did” (Beaglehole 1962:1:443). Wilkinson also recorded from Anaura (“Tegadoo”) on the same coast that three boys were “afraid of being Eat by there Enimies on the North Side of Poverty Bay” (as cited in Salmond 1991:168). This ascription of cannibalism to a group other than one's own (generally at the expense of the former) is a common feature of the cross-cultural data (Arens 1979), although it must also be qualified in Wilkinson's clarification that “they [the people of “Tagadoo” generally] all agree that they kill and eat all there Enimies when ever they take any” (Salmond 1991:168).

Furthermore, the Anaura expression of shame stands in marked contrast with the attitudes expressed by the people of Totaranui in 1770. In his 1769 account, Banks also observed that the question of cannibalism had been asked constantly at each contact of the Endeavour since Anaura, and was always answered in the affirmative. In some instances, “the people have put themselves into a heat by defending the Custom”, Banks clarified (Beaglehole 1962:1:443). Collectively, these observations may hint at variant (regional and perhaps even generational) perspectives on the purported practice. However, for Banks, at least, one observation remained consistent from the North Island to Totaranui.

- 251

The Resolution and Adventure, 1773

The “Indian Cove” incident, November 1773. Cook was later to concede that his 1770 account of cannibalism in Totaranui was

This concern for documentation would underscore a significant incident which occurred during the November 1773 Totaranui sojourn of the Resolution on Cook”s second Pacific voyage. On November 23, in Totaranui, some of the officers went ashore at “Indian Cove, with a view to trade with the natives” (G. Forster 1777:1:511), this being a place where “many of the Natives generally dwelt to purchase Curiosities” (Journal of William Wales, cited in Beaglehole 1961:818;see Figure 2). One of the boat's crew happened to see the severed head of a man (“a youth about fifteen or sixteen years old”; G. Forster [1777:1:511]) and, as the other crew members looked around, they found in another place, “the Intestines Liver Lungs &c”, lying on the ground (Wales in Beaglehole 1961:818). One of the Maori impaled “one lobe of the Lungs”, and held it close to an officer's mouth, making signs for him to eat it. While, according to Wales, “it was not immediately perceived what they had been about”, and “no part of the Carcase nor even any of the Bones” were seen, he was still satisfied to report that the Maori had “just risen from feasting on the Carcase”. Pickersgill referred similarly to “the Body of which they had just eaten,” (in McNab 1914a:2:195), and G. Forster (1777:1:511) added the detail that “the natives shewed them several limbs of the body, and expressed by words and gestures that they had eaten the rest” (See also Anon. 1775:102-3).

As the officers returned to the ship with the head (which had been purchased by Pickersgill), “some Indians of a different Party were on board and were very desirous of a piece of it, which for curiosity's sake I gave them and which they greedily devour'd, before the Ships Compy” (Pickersgill's Journal, cited in McNab 1914a:2:195-6; G. Forster [1777:1:512] claimed the Maori arrived on board while the Europeans were examining the head and expressed “an ardent desire of possessing it”, while making signs that it was “delicious to the taste”). Wales (Beaglehole 1961:818) clarified that the Maori who arrived on the ship “resided in another part of the sound”, and “although in friendship with were not of the Party of whom the Head was purchased”.

- 252

Lt Charles Clerke recorded the details that one man “chearfully gave his assent” to eat the flesh, which Clerke delivered to him after giving it “a little broil upon the Grid Iron”. The man “devour'd it most ravenously, and suck'd his fingers…in raptures”. On the return of Cook, who was not on board at the time, Clerke “cut & dress'd my friend the other steak”. Cook himself acknowledged that he was “desireous of being an eye wittness to a fact which many people had their doubts about”. (Midshipman John Elliott later commented that Cook “thought this a very good opportunity to bring the Matter [of cannibalism] to a Proof”; see also Anon. 1775:103) The Captain, therefore, ordered the flesh broiled, and then watched “one of these Canibals eat it with a seeming good relish”, which had “such effect” on some of the ship's company as to cause them to vomit (Cook's journal). Wales, who returned with Cook, wrote that the “Steake” was eaten “with an avidity which amazed me”, the man “licking his lips and fingers after it as if affraid to lose the least part…of so delicious a morsel” (Clerke, Cook and wales accounts cited in Beaglehole 1961:292-3; 818; Elliott's “Memoirs” in Holmes 1984:22). Elliott recalled that the broiled flesh was eaten “instantly…with all the avidity of a Beef Steak, to the utmost horror of the Whole Quarter deck”; G. Forster (1777:1:512) also remembered that the man ate the flesh “with the greatest avidity”.

The contrived nature of this demonstration is readily apparent from these descriptions, of course. Behind this exhibition, however, it is clear that, as in the earlier Totaranui incident of January 16,1770 (Beaglehole 1955:236; 1962:1:455, 456, 2:30-1), the very serious business of warfare had been responsible for the evidence of cannibalism which the English crew had observed. From a conversation at “Indian Cove”, G. Forster (1777:1:511) recorded from the local people that they “had fought with their enemies, and had killed several of them, without being able to bring away any of the dead”, besides “this youth” who had been eaten. J. Forster (in Hoare 1982:3:427) observed further that the Maori had been in “Admiralty Bay on Sunday & fought there a battle that killed many”. These details are also confirmed by Cook (Beaglehole 1961:293-4; for the proximity of Admiralty Bay to Totaranui in the northern South Island, see Fig. 1). The loss of some of their own in battle, apparent from the women nearby who were engaged in ritual weeping and scarification, was further testimony to the reality of warfare, as was the prominent display of the youth's heart on the points of a spear attached to a canoe head, apparently as a ritual offering (Wales' Journal in Beaglehole 1961:819; Brailsford 1981:27; Forster 1777:1:511). The anonymous published account of the second voyage (probably by J. Marra; see Cook in Beaglehole 1961:961) also claimed that local Maori “gave the gentlemen to understand” that they had been at war and taken 20 of their enemies prisoner, who were “kept alive in places of security” and put to death occasionally “as they - 253 wanted to make a feast” (Anon. 1775:103-4). While contradicting Forster and Cook in the numbers of people taken, and almost certainly embellishing the details of cannibalism, 1 the anonymous chronicler at leqast agrees in the crucial detail that this incident had its origins in warfare.

The Grass Cove incident, December 1773. A more spontaneous example of cannibalism than the November shipboard exhibition would directly confront some of the English visitors in 1773. Cook left Totaranui on this second visit on November 25 of that year. He had waited for, but finally left without seeing, Captain Furneaux of the Expedition's sister ship, Adventure. Four days later, Furneaux arrived at the Sound where he would remain until December 22,1773 (Beaglehole 1961:lxxvii; Bayly's Journal in McNab 1914a:2:214-8).

On the morning of December 18, a longboat was sent in search of the Adventure's large cutter, which had not returned on the previous evening when expected. Within the Sound, the longboat finally approached a small beach adjoining Grass Cove (now identifed as Wharehunga Bay) (Fig. 2). Observing their approach, the occupants of a Maori canoe which had just been hauled up fled inland, arousing the suspicions of the Englishmen. After pulling up on to the same beach, the men of the longboat found a piece of fresh meat. Lt James Burney and Mr Fannin, the Master, were both of the opinion that this was dog flesh. (As Burney remarked later, “I still doubted their being Cannibals”.) However, after about 20 baskets lying tied up in the canoe on the beach had been opened up, “we were Soon convinced by most horrid and undeniable proofs”, Burney recorded. Some of the baskets were “full of roasted flesh & some of fern root which serves them for bread”. A severed hand was also found, with the tatooed inscription, “T.H.”, identifying Thomas Hill, one of the Forecastlemen (Burney's log in Hooper 1975:96-7; further detail from Anon 1775:92-4 and Bayly's journal in McNab 1914a:2:217).

The longboat then proceeded to Grass Cove, where a large number of Maori (“an hundred or more”; Bayly's Journal) had gathered. After shots were fired from the longboat, the Maori on shore retreated, allowing the boat's crew to land and witness “a shocking scene of Carnage & Barbarity” (Burney's Log in Hooper 1975:97). More explicitly, Bayly (McNab 1914a:2:218) reported finding “the Intrails of 4 or 5 men together with the Hearts & Lungs & 3 heads roasted, one of which appeared to be the Capt's Black serv't”. A left hand, “known [by a recently healed cut] to be Mr. Rowe's”, was also found, as were “some feet & other parts but all much defaced by roasting except one foot” (see also Anon 1775:94-7Furneaux's narrative in Beaglehole 1961:741-2).

The deaths of Furneaux's men at Grass Cove were explained by Totaranui Maori in 1777 to several chroniclers associated with Cook's third expedition - 254 as retaliatory actions. Apparently a disagreement had developed over a theft and, in the altercation which followed, two Maori were shot dead and all of the English crew killed (accounts in Andersen and Tewsley 1926:14; Beaglehole 1967:63-4; 68-9; 798-9; 998-9; McNab 1914a:2:197-9; 219-20; see also Anon. 1775:97; Beaglehole 1961:571-3; G. Forster 1777:2:456-7, 465; Hoare 1982:4:676 for the less detailed and somewhat confused accounts recorded in 1774). It was also reportedly acknowledged that the English crew members had been eaten. Thus, Burney (McNab 1914a:2:199) remarked under the date of February 24, 1777, that “one man did not scruple to acknowledge his being present and assisting at the killing and eating the Adventure's people”. Zimmerman (in Andersen and Tewsley 1926:14) also remarked that the New Zealanders “quite unreservedly” told Cook “that the men killed had afterwards been cooked and eaten” (see also Ledyard 1963:17-19 for a colourful, if secondary, account). Zimmerman also recorded that “we found the spot where the attack and murder and taken place, and where many bones were to be seen”.

Unfortunately, these records have ignored the explicit Maori perception of cannibalism's cultural content in this case. Cook, however, intent on obtaining as full an account as possible of the Grass Cove incident, believed only more firmly after this 1777 visit that the Maori lived in “perpetual apprehinsions” of attack, and that cannibalism was to be understood in the context of revenge (Beaglehole 1967:71). Cook also recorded what he had understood of the general cultural meaning of cannibalism. As expressed by Cook in the lexicon of Christian eschatology, the New Zealanders believed that “the Soul of the Man… whose flesh is eat by the Enimy, goes to Hell, that is descends down to a perpetual fire”. The soul of whoever was rescued from such a position, however, “goes to Heaven” (“the habitations of the Gods”), as do all those who die “a natural death”. In this interpretation (which, if reported exactly and literally, may also demonstrate an influence of Christian theology), cannibalism obviously constituted the ultimate revenge. There was no hint of a susbsistence interpretation from what Cook had understood by this time:

The 1773 incidents were to settle the issue of anthropophagy for all those who have left records from the Resolution and Adventure. However, a careful - 255 consideration of these and the 1770 experiences in context highlights several problems for the interpretation of Maori “cannibalism”.

The most significant of these concern observer bias and the distortion of precontact cultural integrity through interaction. The latter problem is particularly acute, as already noted, with regard to the display of “cannibalism” on board ship, November 1773. It may well be that the men considered the offer of human flesh to be a European test of, or even a challenge to, their status, mana and courage. Alternatively (or as well), there may have been a Maori expectation of further European gifts on the open demonstration of such cannibalism. In any case, the exaggerated gestures of enjoyment described in more than one account suggest an attempt to convince for whatever motive. As the earliest direct observation of “Maori cannibalism”, this incident hardly impresses as a spontaneous cultural expression or preference.

It also seems that the fascination of the Endeavour's crew for the cultural use of human remains, and cannibalism in particular, had created a cycle of supply and demand. Of the November 1773 conflict which had resulted in the deaths of the (eaten) youth and other Maori involved in the skirmish, J. Forster concluded grimly that “we are the innocent causes of the war”. According to Forster, the constant demand for “Curiosities & green Stones” had prompted the Natives in the Sound to raid their neighbours for such items (Hoare 1982:3:427; see also the similar opinion of G. Forster 1777:1:512). Whether or not the Forsters overstated the case, in light of the seeming European obssession with eaten human body parts, one cannot overlook the fact that the dismembered body of the young man lay at the cove where “many of the Natives generally dwelt to purchase curiosities” (Wales in Beaglehole 1961:818).

The remarks of Joseph Banks under date of January 24,1770, suggest that an intrusive exchange cycle involving (among other items) eaten body parts had its roots in the first contact at Totaranui.

Banks made this observation just over a week after the arrival of the Endeavour and the first recorded incident of cannibalism. English fascination and consequent demand, therefore, appear to have immediately altered the behaviour of the Maori of Totaranui, while creating a paradigm for future contact.

- 256

The further problem of bias in observation was appreciated by Banks, who reported two accounts of cannibalism from two of the ship's boats, respectively, while at Totaranui in January, 1770. Banks believed the two stories were but variants based on the misinterpretation of a single incident. From such misunderstandings, Banks remarked, “many of our gentlemen were led to conclude that thefts [of people for cannibalism] … are frequent among these Indians”. Banks added the observation that the crew “know only a few words of the language, and eating people is now always the uppermost Idea in their heads” (Beaglehole 1962:1:462; see also Salmond 1991:252). From the 1773 expedition, Wales (in Beaglehole 1961:790-1, 816, 817) documented other incidents in which misunderstandings had led the English crew to falsely assume Maori cannibalism. These situations tended to show “how far we are liable to be misled by Signs, report, & prejudice”, Wales concluded (p.791).

The problematic cultural integrity of these events aside, it appears that the men of the Cook expeditions were also unable to reach anything like a philosophical consensus with respect to the phenomenon they had encountered. This is demonstrated by the near contemporary reflections of several of these chroniclers, highlighting the difficulty of interpreting these data.

It was Banks, the careful scientific observer, who has left us the most extended reaction to the January 1770 evidence of cannibalism from Totaranui. On January 20,1770, Banks (Beaglehole 1962:1:457) observed; “Our freinds here do not seem to feel the want of…[places] of cultivation, I suppose they live intirely upon fish dogs and Enemies”. Yet, what appears to be a largely subsistence interpretation of the practice was qualified in Banks' retrospective entry in March. “That they eat the bodies of such of their enemies as are killd in war is a fact”, Banks acknowledged (Beaglehole 1962:2:30), but he was loath to “Debase human nature” as to imagine that the Maori relished human flesh “as a dainty or even look upon it as a part of common food”, an observation he reinforced with reference to the larger “order of nature” (Beaglehole 1962:2:19-20). In further explanation, Banks offered that the Maori were “implacable towards their enemies, who after having killd they eat, probably out of a princ[i]ple of revenge” (Beaglehole 1962:2:12).

On October 25, 1769, before the Totaranui sojourn, Cook remarked similarly in “notes on New Zealand”: “They eat their enimies Slane in Battell — this seems to come from custom and not from a Savage disposition” (Beaglehole 1955:539). “I firmly believe they eat the flesh of no others [than their enemies]”, Cook emphasised again in 1773, adding that this custom had “undoubtedly been handed down to them from the earliest times” (Beaglehole 1961:294). Cook remembered “one of the arguments” the Maori had offered in response to Tupa'ia's remonstratins: - 257

Cook also observed that, against Tupa'ia's arguments, the Maori had never “once owned that this custom was wrong” (Beaglehole l961:295; however, compare Banks in Beaglehole 1962:1:443 concerning the two Poverty Bay boys who seemd “ashamd” of the custom).

Unlike Banks and Cook, Pickersgill was not inclined to a more sanguine or restrained interpretation of the evidence. “These are the only People who kill their fellow creatures Puerly for the meat”, wrote Pickersgill in 1770, “which we are well Assured they do by their lying in wait one for another as a sportsman would for his game” (Beaglehole 1955:236, n.1, and Salmond 1991:244). At Totaranui in 1773, Pickersgill observed further that war parties would plunder houses and carry off “the Dead Bodies for Provision” (Holmes 1984:75). This last was apparently a theoretical observation, since no specific examples are cited. The 1770 observation stands without significant supporting evidence, since Pickersgill's claim that a young girl had been siezed and eaten “in the same bay” (Beaglehole 1955:1:236 n.1) appears to be one of the misunderstandings referred to by Banks (see discussion above and cf. Beaglehole 1962:1:462). There is no evidence that any prey being sought by a Totaranui person seen “lying in wait” in 1770 was avian rather than human, or if the later, that cannibalism was the intended end. One should also recall that Pickersgill probably exaggerated his supposedly eyewitness account (see earlier discussion).

Astronomer William Wales, initially sceptical of the reality of anthropophagy, changed his opinion after November 1773 but, unlike Cook, believed the practice was not associated exclusively with one's enemies. The consumption of flesh on board ship was not an action of the “wild Frenzy” associated with battle, Wales commented, and “those who eat the part of the head… could not know whether it belonged to a friend or Enemy” (in Beaglehole 1961:819). The amount of fish, dog and bird readily available satisfied Wales that humans were not eaten “through want of animal food”. Instead, Wales believed the practice arose “from Choice, and the liking which they have for this kind of Food”, visibly shown “in their eagerness for, and the satisfaction which they testified in eating, those inconsiderable scrapts” on board ship.

A similar argument was offered by William Anderson on board the Resolution in New Zealand in 1777. Although acknowledging that whatever gave rise to “so barbarous a practice” was now hard to determine, Anderson argued that “at present it would appear they do it merely for the sake of the delicious repast it affords”. Anderson explicitly rejected an argument of - 258 necessity or “want of food”, since cannibalism was not practised in countries “far less bountifull…than here”. He also rejected the explanation of revenge, “for these people destroy all strangers who cannot defend themselves…and consequently many who could not have created a desire of retaliation by injuring them formerly” (Beaglehole 1967:815).

G. Forster (1777:1:514-8) contributed a further philosophical variant, synthesising interpretations of vengeance, nutrition and palatability (see also the similarly expressed views of George Forster's father, John Forster [1778:325-32]). Rejecting the argument that cannibalism first developed through want of food, and reflecting contemporary social opinion, Forster wrote that the origins of the practice lay in the “strong passion” for revenge to be seen among “barbarians”. Such an “excess of passion” originally inspired the attempt to exterminate, through consumption, the very remains of the enemy. Over time, and “by degrees”, however, cannibals found the meat “wholesome and palatable”, encouraging the practice “of eating their enemies” (G. Foster 1777:1:514-7). With respect to humans consumed in New Zealand, Foster (1777:1:517-8) thus observed (in general agreement with Cook and Banks) that “they never eat their adversaries, unless they are killed in battle; they never kill their relations for the purpose of eating them; they do not even eat them if they die of a natural death, and they take no prisoners with a view to fatten them for their repast”.

The variance of these philosphical permutations (in spite of some commonalities) hardly encourages any confidence that the men of the Cook expeditions had obtained a factual, in-depth understanding of the anthropophagy they had purportedly verified. Two contemporary perspectives should also be considered here, relevant to the problem of the frequency of cannibalism, and its place within 18th-century Maori society. While in Totaranui during April, and with respect to the claim that the New Zealanders were cannibals, Bayly observed that “we could not discover anything that…even amounted to raise a conjecture of the kind had we not been prepossessed with it before, for we visited their huts both by night & by day and never saw them eat anything but fish and their bread which is made of Fern-roots” (entry for April 12, 1773, in McNab 1914a:2:204).

On June 6 of the same year, Wales also observed (against the opinion “now almost universally believe ”, that the New Zealanders were cannibals); “I have not seen the least signs of any such custom being amongst them, either in Dusky Bay or Charlotte Sound”. (Wales [in Beaglehole 1961:791; Wales p.791] dismissed the sighting by some of the Adventure's crew of a severed human head in a canoe as too inconclusive to constitute a proof of cannibalism; compare Hoare 1982:2:303, where J. Forster presents the contrary perspective from the interpretation of the Adventure's crew, with the opinion that “the killed are commonly eaten”). Bayly and Wales would both be confronted with evidence - 259 of cannibalism in New Zealand before the end of the year. However, the consistent impression of these earlier observations cannot be ignored.

Beginning with Marion du Fresne in the Bay of Islands on June 12, 1772, a number of European visitors to the northern North Island were reportedly killed and eaten in a sequence of events that continued until about 1820 (Doak 1984; McNab 1914b:71,112,115,126,147; Salmond 1991:401; Thomson 1859:1:234-8,248-51,253-4). Yet, in none of these cases is there anything like the contextual detail available from the Totaranui accounts of the 1770s. Since the North Island incidents all apparently arose from extraordinary cases of culture conflict and misunderstanding, there is the additional interpretative problem that one may not necessarily be dealing with the norms of traditional behaviour.

Given the recent work of Salmond, however, the incident involving the French expedition of the ships Mascarin and Marquis de Castries under Marion du Fresne deserves closer scrutiny. Again, the contemporary evidence for cannibalism is circumstantial. While none of the remaining crew saw Marion or his dead companions eaten, a cooked human head and a thighbone on a skewer were found in the village from which the Maori had retreated in the following altercation (Salmond 1991:401). More significantly, a 19th-century tribal account lists the names of Maori men who “witnessed these acts” (see also Salmond 1991:377). A further account has preserved a traditional meaning for the incident:

The “evil” referred to was a fishing excursion by the “foreigners” into a bay within Te Kauri's territory that had been made tapu by the death of three local men (Salmond 1991:386-7, 395, 402). As Salmond (1991:387, 395) has remarked, the consumption of fish that may well have nibbled on the bodies of these men was an act “tantamount to cannibalism”, and a physical desecration that would have angered the spirits of the deceased and the atua (gods) of Te Kauri's people (see also Thomson 1859:1:236-8). In Salmond's (1991:401) interpretation of the consequences, the bodies of the Frenchmen had been eaten “to destroy their mana and to expiate their crimes”. As a mechanism of defence and appeasement in what had been an extraordinary situation of contact, therefore, it would seem problematic to accept this incident as a common feature of Maori society.

- 260

Information on cannibalism's apparently more normative expression was provided by the Mascarin's ensign, Jean Roux, before Marion's demise. For June 1772, Roux remarked that the local people explained “very clearly” what they did with prisoners. Roux added that “several demonstrations that they gave me of this gave me every cause to think that they [finally]…ate their enemies”, as others of the crew had also concluded (Ollivier and Spencer 1985:169). One might well question, of course, the level of objectivity in, or the unstated assumptions brought to, the interpretation of these “demonstrations”. Roux further clarified that “what confirmed me most in this opinion was that a chief who understood very well what I was saying to him on the subject [were there others who didn't?] gave me to understand that after killing them they put them in the fire, and when they are cooked they eat them”. Seeing that Roux was “revolted and horrified”, the chief “began to laugh and went on reaffirming what he had just said”. One cannot overlook that the obvious gratification received by the “chief” in Roux's response may have influenced the telling.

North of the Bay of Islands, a very similar account was recorded during the December 1769 visit of the French expedition under De Surville to “Refuge Cove”, Tokerau (Doubtless Bay). First Lieutenant Pottier L'Horme was given to understand by “the chief of the village” that, “if a few enemies were left on the battlefield, they cut them in pieces and distributed them amongst themselves for eating”. It is interesting that, like Roux, L'Horme felt constrained to emphasise the accuracy of his observations, remarking that “these signs were unequivocal” (Ollivier, Hingley and Spencer 1982:129). Also like Roux, L'Horme included a description from “one of the leaders in the area” who “gave even clearer explanations”:

L'Horme was unable to say whether “they eat these frightful viands raw, or whether they cook them”, which suggests some limitation on the “clear explanation” offered by the signs. (See also the journal entry of Guillaume Labé under date of December 23,1769, where a similar communication took place, “always by signs”; Ollivier, Hingley and Spencer 1982:75-7). In this regard, it is relevant to compare the simulation of the “War-Exercise” as recorded by Wales from the South Island. According to Wales, the Maori would demonstrate how they cut off the head, legs and arms of the enemy, throw out the bowels, and “lastly shewed us they went to eating”. Yet, this - 261 might only mean, Wales commented, “that after the Engagement they refreshed themselves with some other Victuals”.

Wales also noted “that I did not see one out of the many who went through [the demonstration of] those Massacres” who “did not stop before he made the sign of eating; or that did it before some of us made the sign, as if to remind him that he had forgot that part of the Ceremony” (Wales' Journal in Beaglehole 1961:790). It seems clear from the context of these accounts (and their similarities overall) that the demonstration of the “War-Exercise” was a ritual or stylised representation. Even if the final action was intended to represent cannibalism, therefore, one cannot be sure how frequently the principle was applied to actual combat situations, or to the extent to which the enemy were eaten.

Governor King of Norfolk Island received a different perspective from “Tooke” [Tuki], one of two Maori abducted from far northern New Zealand in 1793. Remarking that “a state of constant warfare” existed between certain Maori alliances, Tuki also noted that the Maori were “not without long intervals of peace”:

Again, the ascription of cannibalism to an enemy, along with a denial (reinforced by the expression of horror) that the teller's own people ate human flesh, is reminiscent of the cross-cultural data (Arens 1979). Tuki's account may also have been influenced, in part at least, by his response to the sensibilities of the Europeans among whom he lived. In any case, King remained unconvinced (with what cause the record does not note) by Tuki's claim “that the whole of New Zealand” were not cannibals, and left his opinion that “that horrible banquet is general through both Islands” (McNab 1914a:2: 542-3).

This was not the impression of John Savage (1807:35), however, who left a record of his visit to the Bay of Islands in 1805. Although the Maori acknowledged to him they had been driven to cannibalism through “dire necessity” in times of scarcity, Savage characterised “the motive which - 262 impels them” as “vengeance”. Neither was this “passion” pursued “without limitation”; after a conquest, Savage observed, the victors were content merely to show their power to devour all of their prisoners, without actually doing so. The chief of the vanquished tribe was eaten, “but I do not believe that food is the inducement”, added Savage. In seeming contradiction, he also claimed that the introduction of the potato had probably saved many lives, “for they certainly give that root decided preference to human flesh” (Savage1807:35-6).

An observation from the visit of the Bay of Island's chief Te Pahi [“Tip-a-he”] to Sydney in 1806 as a guest of Governor King also reinforces the problem of gaining an unequivocal understanding of the behaviour, while highlighting the “us” and “them” interpretation. On the subject of cannibalism, wrote King, “we could get but little certain information”. “Tip-a-he decidedly denied the existence of such a practice in his dominions, but said it was common in Mowpah's district”. (“Mowpah”, a “chief about the River Thames, is his rival on the south”, King had noted). It is possible (as with Tuki in 1793) that this claim was influenced by an enhanced appreciation of European revulsion consequent upon Te Pahi's Sydney visit. Furthermore, “Ti-a-pe” of the northern Muri Whenua district “also said it was a practice with Tip-a-he and his subjects”. “Where truth lies I am undecided”, King concluded, “but I am of opinion, from everything I have heard and observed, that this practice most certainly prevails in New Zealand” (McNab 1908:266-7). It is interesting to compare this more-qualified view with King's earlier impression (against Tuki's description) that “that horrible banquet” was general throughout New Zealand.

Between 1814 and 1815, J. L.Nicholas accompanied the missionary Samuel Marsden to the Bay of Islands to assist Marsden in the establishment of the Anglican mission. In spite of his “satisfaction” from the “respectable testimony” of Cook that the New Zealanders were cannibals, Nicholas had resolved “to obtain the unequivocal acknowledgement of the natives themselves in this particular” (Nicholas 1817:2:62). The fact was never denied to him and, in attempting to explain the practice, Nicholas (1817:2:63) opposed what he had incorrectly believed to be Cook's opinion that cannibalism proceeded from “want of a sufficiency of food”. 2New Zealand was well capable of supplying the food requirements of its populations, it seemed to Nicholas (1817:2:63-4).

Against J. Forster's claim that human flesh was reputed to be one of the most palatable of dishes, Nicholas (1817:2:67-8) also observed, “many of them told me that they had no greater relish for this than for any other food”. Instead, Nicholas (p.68) wrote, the Maori were motivated by “a kind of superstitious revenge”, including the belief (paralleling Cook's 1777 observation from Totaranui) “that their revenge can reach beyond the grave, and that the future - 263 existence of their wretched victims must be totally annihilated, by this unnatural destruction of their mortal remains” (p.68; Nicholas may also have been aware of Cook's Totaranui remarks, as published in Cook 1784:138).

Bridging the pre- and post-mission worlds, Marsden's views are also relevant, and serve as a fitting conclusion to this section. In 1815, Marsden observed that “the New Zealanders are all cannibals”, having no idea “that this was an unnatural crime”. To Marsden's expression of abhorrence they had responded that “it had always been the custom to eat their enemies” (Elder 1932:129; see also p.285.). Yet, it is significant that, for all of the time he spent with the “New Zealanders” during this first visit, Marsden “was unable to ascertain whether they ever ate human flesh as a meal, or from choice, or in cool blood”. He did offer that “it strikes me to be only from mental gratification and in retaliation for some great injury”, adding that the “public disgrace” stamped upon the surviving relatives was similar to that endured by the families of criminals publicly executed in Europe (Elder 1932:129).

If Marsden's inability in 1815 to clarify something that had “always been the custom” seems peculiar, he was to compensate during his 1819 visit. Since these observations belong to the new era of musket warfare, one must be cautious in interpretation. However, the apparent continuity for some of his descriptions and earlier accounts (including that of Nicholas in the same region) is impressive. In response to Marsden's request to learn how the Bay of Islands Maori conducted themselves on the battlefield, and “also if they eat their enemies when killed” (Elder 1932:173), he was given to understand that cannibalism belonged to a ritual marking the conclusion of battle. At this time, a defeated chief and his wife were ceremoniously consumed by the victorious “chiefs and their wives, none of the common people being allowed to touch them as they are tabooed”. After the “sacred services” were completed, “all in common feed upon the after slain”. The dead were consumed “not so much as an object of food but as a mental gratification and to display publicly to the enemy their bitter revenge” (Elder 1932:174; see also pp.168 and 220).

To Marsden, the custom thus appeared “to have its origin in religious superstition”. “I could hear of no instance of any man ever being killed merely to gratify the appetite” (Elder 1932:169). It is also significant to note that Marsden “never discovered that the New Zealanders offered up human sacrifices to their gods upon any occasion” before the account was related to him September 1819 (p.174). A further 1819 detail from Marsden on cannibalism's meaning provides an important point of comparison with several of the pre-1815 expressions noted above:

They also believe that, by eating the flesh and drinking the blood of the departed chief, his system becomes incorporated into their system, and by that means they - 264 are secured from all danger from the departed ghost of the dead chief, and his spirit will then take up its residence in their bodies as being part of its former habitation (Elder 1932:220).

In respect to the ethnographic reality and meaning of cannibalism, the preceding discussion leaves a number of issues unresolved. At a basic descriptive level, this has much to do with the fact that no one actually witnessed the consumption of human flesh in anything like an authentic context. Cook and his crew came as close as anyone in Totaranui on January 17,1770, but, even then, they understood that all of the flesh had already been consumed. The complex mix of early Maori expressions on the subject, recorded from communities spanning much of New Zealand and both the late 18th and early 19th centuries, also creates some uncertainty. These expressions include denials, “shame”, and vigorous justification, not all of which can be easily dismissed as variables of the period of exposure to European sensibilities. As a particular case in point, the two boys who mentioned the practice at the Poverty Bay landing of the Endeavour in 1769 and “seemd ashamd of the custom” (Banks in Beaglehole 1962:1:443) stand in marked contrast with the people of Totaranui from the first contact of, and throughout, the extended 1770 visit of the Endeavour.

In spite of the diversity and contradictions of a number of these accounts, however, there are also impressive commonalities. Most fundamentally, there appears to be widespread agreement that cannibalism, although perhaps a constant threat, was not an integral component or a frequent activity associated with any of the communities visited by Europeans. There is also a general, if not complete, consensus that cannibalism was associated with motivations of revenge in conflict situations, while spiritual protection was reported as a factor in the Marion incident and Marsden's 1819 interpretation, at least. In this regard, none of the detailed, careful accounts supports a subsistence interpretation.

Post-1815 accounts would also identify an additional context for cannibalism, with the report that slaves were killed and eaten, generally on important social and ceremonial occasions (Bagnall 1957:186; Best 1924:2:555; McCormick 1966:111-7; McNab 1908:1: 573-4; Oppenheim 1973:66; Pollack 1838:1:225, 2:3-7; Smith 1910:108, 454; Thomson 1859:1:149-50; White 1874:222-57). If this was a consistent practice of the Maori in the 18th century, however, the men of the English and French expeditions were not made aware of it.

By the middle of the 19th century, it was generally assumed that cannibalism - 265 had been an integral feature of traditional Maori society. As Thomson (1859:1:141) remarked: “few persons can think of New Zealanders without thinking of cannibalism, or of cannibalism without thinking of the New Zealanders.” Consequently, no one was especially concerned to document cannibalism from the archaeological record, although apparent evidence was noted. On Christmas day, 1835, for example, Charles Darwin (1979:676) reported the find, on a small offshore island in the Bay of Islands region, of “burnt human bones strewed around an old fire place”. These, he suggested, “might have been lying there for several years”. Later in the century, Haszard (1889) reported, from two distinct places among sand-dunes at Raglan Harbour, a number of exposed Maori ovens and shell middens, about which the local Maori apparently had no traditions. Among the oven stones were “the charred remains” of human bones, while larger broken bones and “nearly complete” skeletons were described in the general vicinity. These, Haszard believed, were relics “of the old cannibal times”. Since cannibalism was well accepted, such reports seem generally to have been relegated to the status of interesting curiosities. Within the nascent discipline of archaeology in New Zealand, however, claims of early cannibalism were to assume a new significance before the close of the 19th century.

The origins of archaeology in New Zealand are linked inextricably to the debate surrounding the large, flightless moa — birds that were extinct at the time of European settlement (Anderson 1989). The earliest reports of the huge birds noted that their remains had been found in association with contemporary human activity. Richard Owen, the comparative anatomist in Britain who established the identification and first systematic description of the moa from bones sent to him, concluded that the early Polynesian settlers of New Zealand were responsible for the bird's extinction. He added: “when the source of animal food from terrestrial species was reduced by the total extirpation of the genus Dinornis…then may have arisen those cannibal practices” (Owen 1849:270).

This interpretation did not sit well with father and son Gideon and Walter Mantell (respectively), who published early site identifications and descriptions of moa remains from the Waingongoro locality, on the south-west coast of the North Island. The younger Mantell, as cited by his father (G. A. Mantell 1848:234), had dug in some mounds containing bones pointed out to him by local Maori. These were “the refuse of feasts made by their ancestors”, in which Mantell uncovered “burnt bones”, consisting of moa, dog and human, “promiscuously intermingled”. Since it was assumed that the extinction of the moa was not recent, G. A. Mantell suggested that “the practice of cannibalism [thus inferred]…will appear to have been of very ancient date, and not to have - 266 originated from the want of animal food on account of the extinction of the Moas, as Professor Owen so ingeniously and indulgently suggested” (G. A. Mantell 1848:234-5).

No stratigraphic information was recorded (or even appreciated, apparently) by Walter Mantell in his field description, and it might be wondered how far the “promiscuous” intermingling of human and moa bones was a factor of secondary disturbance. A potential problem in this regard is emphasised by Walter Mantell's further description of the Te Rangatapu sand-flat from the Waingongoro area, which was “covered with bones of men, moas and seals” after being “overhauled by the Rev. R. Taylor” (G. A. Mantell 1848:239). Even G. A. Mantell (1848:241) commented from his son's description that “some doubt” remained whether moa might have intruded into the “heaps of ancient native fires” containing moa, human and dog bone, while subsequent excavation at Waingongoro has failed to confirm a primary association of moa, ovens, and human remains (Anderson 1989:97-9, 115-6). Nevertheless, in a paper published 20 years later, Walter Mantell asserted:

The debate on the moa and their human association was to become more vigorous and critical with Julius von Haast's interpretation that the moa-hunters were an ancient, autochthonous race who did not indulge in cannibalism. They were to be distinguished from the more recent, cannibal, “Neolithic” Maori who had relatively recent roots in island Polynesia. For the sites reported by Walter Mantell, Haast (1871:78; see also p.91) pointed out the obvious interpretative problem, suggesting that “more careful and systematic researches…would prove that the Moa kitchen middens are quite distinct, and that where Maori ovens with indications of cannibalism occur, they have been formed over, near, or within those of the older race”.

Haast had relevant, negative evidence for his opinion from the moa-hunter site of Rakaia in Canterbury. Here, after “the most careful search, continued for a number of days”, not “the smallest portion of a human bone” had appeared among “the thousand fragments of bones” which had passed through his hands on the project. Haast added:

- 267

From a further visit to Rakaia, Haast himself identified evidence of “Maori” (as opposed to “moa-hunter”), occupation in the area surrounding the site. This included two human tibia, one femur and a humerus of “the same individual, a full grown man”, which had been exposed by the plough (Haast 1878:96). These, Haast (pp.96-7) reported, had had their extremities “sharply broken off, as if for the extraction of marrow”.

In what was the first recorded excavation in Polynesia with stratigraphic control (Anderson 1990:103; Duff 1963:28), Haast directed the dig at Moabone Point Cave, Sumner Road, in 1872 (Haast 1875), from which more negative evidence was available. Within the overlying, non-moa bearing “Maori or shell beds”, Haast reported but two adult male human pelvic bones and a dorsal vertebra. “Through the whole thickness of these [shell] beds”, he had found “not the least sign of any broken human bone”. Given the length of time suggested by the stratigraphic depth of the shell beds, and “the insecurity of life to which savage tribes are exposed”, Haast believed there would have been some evidence had these people been cannibals. Since “cannibalism has been practiced at least for several centuries in New Zealand”, Haast concluded that “the absence of human bones in the shell beds certainly proves that they are of considerable antiquity” (Haast 1875:74).

The focus of this 19th-century debate was, of course, the cultural association of moa-hunting. As Haast's conclusion suggests, the existence of cannibalism was never in doubt, just its precise chronological and cultural position. The absence of any evidence from the detailed post-moa stratigraphy revealed in the Sumner excavation caused little more than the minor temporal readjustment of a well-accepted scenario. Even Haast's most influential contemporary detractor accepted that “the moa-hunters of the South Island do not appear to have been cannibals”, but rather, “peaceful tribes not given to war”, in contrast with later South Island Maori (Hutton 1891:170).

Earlier 20th-century cultural scenarios were also influenced by a European retelling of the first human settlement of New Zealand which later writers have called “The Great New Zealand Myth” (Simmons 1969, 1976; Sorrenson 1979; Sutton 1985). This “Myth” identified a primal settlement of New Zealand by a Melanesian (or mixed Melanesian-Polynesian) people “of, apparently, inferior culture” (Best 1914a:73). Cannibalism, as one might expect, was as grist to the mill of these interpreters. For Best, the custom had probably been inherited by the largely Polynesian Maori from the original, “Maruiwi” settlers, who were “of a lower plane of culture” (Best 1915:439-40; see also Best 1914a). Reinforcing this cultural and racial distinction, Smith (1921:200, 211-2; 213-4) also argued that cannibalism in Polynesia generally was the result of an early connection with Melanesian peoples.

- 268

Within the 20th-century literature, however, a critical approach to the Maruiwi assumptions and a greater emphasis on material culture and archaeological evidence was to prevail in the interpretation of the Maori past (Skinner 1974; Teviotdale 1931, 1932, 1937; Williams 1937). On the main stage of the successive cultural scenarios which followed this emphasis, cannibalism appeared in various supporting roles.

Intent on demonstrating the essential cultural, and therfore racial, unity of moa-hunter and (Polynesian) Maori remains (e.g., Teviotdale 1932) interpreted a number of human postcranial fragments found among “midden refuse” of a moa-bearing deposit at Taieri River mouth in southern New Zealand as suggesting “a cannibal feast”. Two pieces of manufactured human bone “on the clay below the midden-refuse…certainly belonged to the earliest inhabitants or campers in the cave”, Teviotdale argued. “If, as seems probable, these folks were moa-hunters, then the moa-hunters made occasional use of human bone for implements” (Teviotdale 1931:90; see also 1932:101). From the region of the Papatowai site on the south-east coast, Teviotdale (1937:134-5) also quoted an earlier report that human bone had been found along with moa in a “very ancient Maori midden” which represented the “refuse of their feasts”. Even so, within the extensive excavated Papatowai moa-hunter site itself, Teviotdale (1938b:116; cf. Teviotdale 1938a:31) recorded but a single fragment of worked human tibia.

Duff (1956:15-6, see also 1949:178) vigorously opposed Teviotdale's assumptions of cultural unity between the southern moa-hunters and the late 18th century Maori. Duff believed as strongly as Teviotdale (against Smith and Best) in a consistently Polynesian identity for all of the known pre-European inhabitants of New Zealand. However, Duff also returned to a sequential model in chronological interpretation, even using Haast's cultural “moa-hunter” term to describe the earliest settlers. In Duff's view (1956), the moa-hunters had been followed by an agricultural and relatively dynamic “classic Maori” cultural form (which, in its final expression, was North Island Maori society as described by the Cook expedition of 1769). Into this scenario, Duff incorporated the negative evidence of the absence of cannibalism among moa-hunters (implicitly challenging the interpretations of cannibalism's pristine “Maruiwi” origins as well as Teviotdale's views). Based on his own understanding and experiences of moa-hunter occupation sites, Duff wrote:

There is a complete absence of any weapons, notably stone clubs of the patu type, and no evidence of the practice of cannibalism…The villages occur in open, unfortified situations, while the open burial of the dead in the village area - 269 implies the absence of fear of desecration by enemies (Duff 1956:11; see also Duff 1950:75).

Duff (1963:29) even appealed direct to Haast's observation of “the absence of cannibalism” from moa-hunter sites, a detail “since confirmed from other South Island east coast sites”. Fortified villages, warfare and cannibalism were identified as distinctive components of the intrusive, agricultural, classic Maori culture on the the same coast by Duff (1963:33, and see also discussion of Duff 1961 below).

The interpretation of Maori cannibalism as a product of population stress, resource competition, and/or subsistence needs has characterised other influential 20th-century cultural scenarios, including social-anthropological studies with an interest in past interpretation. For example, anthropologist Raymond Firth (1959:147-8) argued that cannibalism was related to the limited supply of animal flesh in pre-European New Zealand, a view shared by Maori anthropologist Te Rangi Hiroa (Sir Peter Buck; see Hiroa 1950:102, 398-401), although both men also noted an important symbolic and ceremonial component, respectively, of the practice. The subsistence interpretation has been emphasised in Vayda's (1960:69-72) ecological approach to Maori warfare. Vayda (1960:72) asserted that, whatever its symbolic component, “the fact was that human flesh served for nutriment”, which made cannibalism “a useful practice in war”. Fagan (1984:247) has offered a more balanced interpretation, allowing subsistence and (increasingly over time) ritual and symbolic components of the practice.

Cannibalism has also been discussed in Davidson's influential and comprehensive 1984 synthesis of Maori archaeology. In this regard, Davidson has exhibited both an acceptance of earlier views and the challenge of new evidence. The clear link with past scenarios is suggested in the subject's consideration in a chapter headed “subsistence economics”. From the Cook expeditionary records, Davidson also observed that “human flesh formed part of the diet of eighteenth century New Zealanders”. Yet, in an important advance beyond earlier interpretations, Davidson (1984:137) acknowledged that archaeological evidence of cannibalism was “relatively uncommon”, and that isolated human remains were difficult to interpret. She remarked that the “few occurrences” (some of which were “possible” only) reported in the archaeological record were “hardly sufficient to suggest a significant contribution to the diet, even during late prehistoric times”.

Against the orthodox view, Davidson added that cannibalism “was not entirely restricted to late sites.” Yet, consistent with earlier views, Davidson also argued that the practice was “probably more common in later times”, and was “more common in the South Island than in the North”. Consistent with earlier - 270 interpretations, she concluded that the practice was “probably more common in later times”. This general chronology is followed in the recent work of Salmond (1991:39), who remarks that, after about A.D. 1500, in the context of population pressures and resource competition (following significant bird extinctions and the decline of seal populations), “warfare apparently intensified, [and] evidence of cannibalism begins to appear”. While it has been slightly qualified, therefore, the basic association of cannibalism with later Maori culture and a period of increasing conflict and stress has not been seriously challenged since Duff, and ultimately Haast.

It is at the level of individual site and excavation reports that the assumptions of cannibalism within broader cultural scenarios can be critically assessed. The evidence for cannibalism as cited in field reports may be grouped under three general headings.

Manufactured human bone. Between excavations at the pā site of Pari Whakatau on the east coast of the South Island, and an associated coastal shell midden, Duff (1961:287-8) recorded several fishhooks and two tabs of human bone; these last were considered either “as by-products from, or prepared pieces for, hook manufacture”. In this project, Duff (1961:269) had turned his attention to the question of the archaeological “intrusion” of “classic Maori culture” into the South Island. For Duff (p.287), the two tabs were evidence of “warfare and cannibalism” characteristic of this “intrusion”. He concluded:

Such an interpretation had already been suggested in the Moa-Hunter monograph. Thus, the presence of human bone “fish-hook points etc.” (including a human cranial fragment with an incised, cuvilinear design) in a “clearly recent” north-eastern South Island coastal Maori midden had been sufficient for Duff (1956:260) to include “men” with dogs and seals as the mammals “most commonly eaten” at the site. Consistently, there are claims from historic and general ethnographic sources that the remains of people who had been eaten were sometimes retained and modified for a variety of tasks (e.g., Best 1914b:110; Elder 1932:168; McNab 1907:214; Maynard and Dumas 1937:237; Salmond 1991:401; Smith 1910:102; Tregear 1926:360). For Pari Whakatau, Duff (1961: - 271 270-3) also incorporated the evidence of local Maori tradition into his argument of warfare and cannibalism.

On closer scrutiny, the “evidence” is far from unequivocal, however. Aside from uncertainty as to the relationship of the traditional sources and the archaeological data from Pari Whakatau (Brailsford 1981:144-8), there is the complication that manufactured human bone might have been obtained from the raid of an enemy's burial area, a possibility that Duff (1956:11) himself had allowed for later Maori occupation sites. Such raids have been well documented throughout New Zealand, and the subsequent use of an individual's worked bone as a fishhook, spear point, or flute, for example, constitued a long-term insult as serious as cannibalism itself (Best 1914b:110; Colenso 1868:21; Elder 1932:161-2; Hiroa 1950:216, 426; Leach and Hamel 1978:242). In short, the desecration of an enemy's burial grounds could meet the demands of both vengeance and industry without the necessity of cannibalism.

From the expeditionary accounts of the 1770s, Orchiston (1972:94-7) has also documented the use of human remains (principally teeth) as neck and ear ornaments. These included the remains of enemies, but also those of friends worn in memorial, a use one might appropriately designate as friendly manufacture. In this last regard, and given the generally tapu nature of human remains in later Maori society, manufacture in human bone may have been intended to impart a sacred quality to a final artefact form. Thus, a multi-pronged, ornamented bone fork from the southern North Island was manufactured from a proximal end and shaft human radius, and may have had a ceremonial use in the feeding of a tapu person (Keyes 1969). The identification by Skinner (1952) from the southern South Island of a human skull bowl that may have enclosed medicinal plants to “strengthen their potency” is a further ritual example. It may be that the remains of one's own immediate kin were preferred in such ritual contexts. Voykovic (1981:74-7), for example, has explored the traditional Maori concept that the bones of the tipuna (ancestors) are “bearers of power” with important social and ceremonial implications and uses. The antiquity of this concept in New Zealand is suggested from the excavation of several of the Wairau Bar “moa-hunter” interments, where crania and long-bones had been carefully removed post-mortem. Duff (1956:40-1, 47-9, 51, 53, 55-8, 60, 95, 115) suggested that the crania may have been used as “talismans”, and that the long bone was manufactured into ornamental beads (examples of which were found at Wairau Bar) to be worn as “charms” by surviving kin.

One should also consider that human bone was generally not the dominant material of artefact manufacture for many later Maori sites in the South Island where cannibalism “may have been more common” (Davidson 1984:137). Such sites include Pari Whakatau, discussed above, as well as the late, pre- - 272 European Katiki Pt. pā site on the Otago coast. From this latter site, Trotter (1967:242, 247) recorded that “bird, human, and dog bones were used for artifact manufacture in that order of popularity”, while worked human bone represented less than 1% of the “total number” of bones found on site (see also Brailsford 1981:226-7 and fig. 214). From a 50-square-metre excavation at the coastal Long Beach site south of Katiki, spanning early and later period occupations, Leach and Hamel (1981) also noted that dog, bird and seal were the dominant bones of manufacture, with human represented only by a small point, a hook, two sawn pieces of shaft, and a “tiny fragment” with marks of sawing and snapping (pp.125, 127, 129; see also Leach and Hamel 1978).

There may, however, have been a greater industrial and ornamental use of human bone in some northern sites. After working at Oruarangi, for example, Teviotdale remarked generally that “human bone was used for all purposes far more extensively than in Otago”. Human bone chisels occurred commonly at Oruarangi, Teviotdale noted, whereas “I have found none in the south” (in Teviotdale and Skinner 1947:347-8). Davidson (1984:74) has also commented on the popularity of human bone chisels in late northern North Island sites. This marked contrast suggests the operation of a regional factor, however, such as “the comparative scarcity of greenstone in the north” (p.74; see also Davidson 1984:77), rather than a greater incidence of northern cannibalism.

Burnt Human Bone. From the excavated “archaic” layers of a midden at Whangamata, on the Coromandel Peninsula, Allo (1972:67) recorded “burnt and scattered human bones … so fragmented it was impossible to estimate sex or size”. Allo suggested that these remains were probably evidence of cannibalism, and allowed for this individual's meat weight in a reconstruction of the midden's subsistence potential. Burnt human bones have also been reported from other, generally later period Maori sites, including defended pā sites and middens where moa is absent. Publication details are generally brief or even incidental, however, and such remains are not always explicitly linked with cannibalism.

From excavations at the Tarata pā site on the south-west North Island coast, Smart (1962:181) reported that “a few shell fragments and bone, including crushed and burnt human bone, occurred in small pockets of burnt earth, especially in the material filling one of the shallow surface pits”. Burnt bone has also been reported among the remains of two individuals found in the upper, non-moa midden of the Rotokura site in Tasman Bay, northern South Island (Butts 1977:92-3; 1978:14). From the excavation of the Pari Whakatau midden site (see above), Scarlett (1960:5) added to Duff's published description of fishhooks and tabs of human bone that “a few other burnt - 273 fragments of human bone were also recovered”, all of which Scarlett considered as possible evidence of cannibalism. From “Garden Is/1”, in Chalky Inlet on the south-western South Island coast, Coutts (1972:1:101) reported fragmentary portions of a cranium excavated in a “midden dump” which did not appear to have been “smashed deliberately”. “There is some evidence of charring which suggests cannibalism”, Coutts observed, with the added caveat that “such a conclusion must be treated with caution”.

Charred human remains have also been reported for a number of sites along the Otago coast in the southern South Island. Broken cranial fragments with evidence of firing were found among midden debris on the undefended upper terrace, Karitane “B” site, while a “possibility” of cannibalism was suggested from pelvic fragments found within midden remains from the defended Karitane “A” site of the Te Wera pā and village complex at Huriawa (Leach 1969:56; personal communication 1991; see also Brailsford 1981:220-5). At the Mapoutahi pā site “a scoop-hearth packed with burnt and fragmented human bone” was excavated within the occupation limits (Anderson and Sutton 1973:110; see also 111, 117). Less than 5 kilometres to the west, burnt human bones were found within a (non-moa) midden layer of mollusca, bird and fish, at Longbeach (Dawson 1949:59; compare Skinner's [1953:400] brief note that part of a human radius and a toe bone were found in a compact Longbeach midden layer). At Little Papanui on the Otago Peninsula, Teviotdale (1935:34) reported that burnt and broken bones of two children and an adult were “scattered and mixed with the midden”, suggesting they “had been eaten”.

Although ostensibly consistent with cooking activities or disposal in an oven following consumption, there are other factors to be considered in the interpretation of burnt human remains from these sites. Within the main occupation deposit of the Takahanga site, on the north-eastern South Island coast, for example, Trotter (1974) excavated a hollow 75 cm in diameter and up to 15 cm deep. The hollow was filled with burnt, and some nearly complete human bone (including neck vertebrae), below which the soil was baked hard. Trotter interpreted this evidence as the interment-cremation of at least two people. Furthermore, Trotter (1974, 1982:100) proposed that burnt human bone, found scattered over about 10 square metres of the occupation, was evidence that the persons were largely decomposed when cremated, in a secondary interment context. Four other (unburnt) burials were also excavated, two of which had been disturbed, with the mandible, scapular and axis vertebrae only remaining of one. Trotter (1974) suggested that the disturbed burials may have been disinterred for cremation.

With respect to the larger region of the east coast South Island, Trotter (1982:100) also acknowledged that “burnt human bone is not uncommon on many sites occupied 300 to 150 years ago”. In light of the Takahanga - 274 excavations, however, Trotter suggested that this did not require an interpretation of cannibalism:

Trotter's observations are consistent with historic and archaeological accounts of cremation as a burial practice among some Maori. From the North Island, Best (1914b) documented three circumstances under which cremation was practised. These were (1) the lack of a suitable place for the final disposal of exhumed remains; (2) death of individuals outside of tribal boundaries and (3) to “stay the spread of disease”. Although cremation was “never a common racial custom among the Maori”, Best (1914b:110) asserted that the second circumstance “was common, we believe, to all tribes” where members of a war party had been slain in an enemy's territory, “lest their bodies be eaten, and their bones converted into fish hooks, etc”, (see also Campbell 1894; Hiroa 1950:426; Oppenheim 1973:68n). From an account “gathered from the northern tribes”, Best (1914b: 111-2) also acknowledged that cremation was practised in the north “as a custom”. Consistently, Taylor (1984:260) has recorded three forms of “Classic” Maori “post-cremation burial” from the Manukau area. There is also evidence that cremation was practised in the South Island. Teviotdale (1935) claimed to have excavated a cremation “pyre” … “The bones of vanquished enemies are not consumed by fire; fish hooks, flutes and other objects are made from them and are worn as trophies,” Edwardson added (McNab 1907:214). In this light and, particularly, given the options outlined by Best, it may not be possible to assign a clear archaeological interpretation in many cases.

Teviotdale (1935) claimed to have excavated a cremation “pyre” on the Otago Peninsula containing the remains of individuals who died in epidemics of the early 19th century. In 1823, Captain Edwardson described funeral rites among Maori of the far southern South Island, observing that the bones of the dead were finally taken up and burnt. “The bones of vanquished enemies are not consumed by fire; fish hooks, flutes and other objects are made from them and worn as trophies,” Edwardson added (McNab 1907:204).

In this light and, particularly, given the options outlined by Best, it may not be possible to assign a clear archaeological interpretation in many cases. For example, a complicated if uncertain sequence of events is suggested for the cranial and postcranial remains of at least two people from the upper (post-moa) extent of the Rotokura midden site mentioned above, where bone has been extensively worked and burnt (Barber 1992; for earlier reports see Butts - 275 1977:92-3, 1978:14; Millar 1967:12). Two photographs of some of this material have been published (without provenance information) in Brailsford (1981:9, [figs. 4 and 5]). This includes the small charred tab of worked cranial bone RK/1547 (fig. 5) which is far more carbonised than either of the larger, cut cranial vaults. Since intense burning would weaken bone strength and undermine a decorative aesthetic, the firing and manufacturing processes were probably independent. It is possible that the individuals were eaten originally as well, but cannibalism is not a compelling explanation for any of the events directly responsible for the present condition of the remains.

There are suggestions in the literature that bone may have been fired as the indirect result of cannibalism. Thus, Best (1902:72) cited the example of a Ngapuhi raid where the bones of those eaten were reportedly burnt to preclude their retrieval by relatives. Taylor (1984:260) has also suggested that ochred charred bones at Manukau may represent the reburial of relatives whose remains were collected from the residue of open fires after they were eaten. If one allows that the remains of high-ranking individuals were manufactured into artefacts following cannibalism, however, the carbonisation of bone would not have been desired in some cases. Furthermore, if Totaranui is any precedent, human bone that was not to be worked may as likely have been discarded around as in the oven. One should also consider the evidence from Takahanga that cremated remains had become scattered through midden deposits (Trotter 1974; as a further complication, it appears that individuals were sometimes deliberately interred within food middens). 3 At the least, therefore, one must allow several options for the archaeological interpretation of fragmentary, burnt human bone in occupation deposits. Where the remains have been found in a discrete situation and are consistently and deeply charred, deliberate cremation must be considered as a likely explanation.

In this last regard, it is enlightening to consider claims of cannibalism from the excavation report of a swamp pā site at Lake Mangakaware, in the Waikato region. Here, Bellwood (1978a: 19,36) reported “8 pieces of human femur”, which had been smashed, burnt and finally, “lavishly covered all over” with red ochre (p.36). Along with the presence of a hearth in the passageway, this suggested to Bellwood that “some rather gory ritual” had occurred. The femur had been broken, and then burnt, Bellwood added. Elsewhere from the site Bellwood reported “a cache of smashed burnt human bone, representing one individual, resting on a piece of wood”. These bones had been deliberately burnt before burial, and a wooden implement found near the cache “may have belonged to a mira-tuatini — an implement for cutting up flesh at Cannibal feasts”, Bellwood (1978a: 19) speculated.

Because the remains were so heavily charred, Bellwood acknowledged that they had been cremated, “since the degree of burning is far more than - 276 would be expected from cooking”. He conceded that “it was not clear whether the bones were in fact used for cannibalistic purposes”, but suggested that the signs of crude bevelling at one end of the long bone argued circumstantially for ritual defilement. Bellwood proposed a “possible sequence” of initial cannibalism followed by “deliberate cremation … and burial” (Bellwood 1978a:72). Elsewhere, Bellwood (1978b:411) stated less equivocally of this site that “excavations in one area inside the pā revealed clear traces of cannibalism”.

The application of red ochre, confirmed at Takahanga for burials where cremation was also practised (Trotter 1974), would seem to be more appropriate for the ritual interment of a respected individual than for an enemy or a slave who had been eaten. Although it is possible that the bones were collected from an enemy fire, their highly charred condition would seem to be more consistent with a deliberate, rather than incidental, cremation process.

A respectful burial context might not be inconsistent with the deliberate bevelling of the long bones, given the possiblity that bone might have been retained and modified by relatives for ritual use (see discussion above).

Unburnt Bone in Middens and Occupation Sites. For human bone identified in 19th-century New Zealand midden sites, the facile and convenient assumption of cannibalism was reinforced by an almost universal belief in the former reality and popularity of the practice. Questions as to the overall site context were generally not considered necessary to ask. Twentiet- century archaeology, of course, has encouraged a greater awareness of the problems of stratigraphic interpretation and association. Such problems may be particularly acute for coastal midden sites, where the processes of collapse and secondary disturbance (including interments) often complicate the stratigraphic record (Ceci 1984:64). Thus, Davidson (1984:137) has remarked that “it is sometimes difficult to tell whether isolated human remains … represent cannibalism or accidental loss during exhumation or other disturbance of burials”. The apparent scattering of cremated body parts throughout midden deposits at Takahanga, discussed above, is a case in point. It is in such a context, and against a range of possible explanations for the presence of skeletal remains in middens (including deliberate burial; see Note 1) that more recent claims of cannibalism from midden excavations must be assessed.

For several earlier 20th-century contributors to the Journal of the Polynesian Society, the presence of broken or fragmentary human bone in occupation contexts was assumed to be proof sufficient for a claim of cannibalism, especially where a spatial association with ovens could be demonstrated (George 1944:73; Sinclair 1940:142-3; Teviotdale 1931). Teviotdale, at least, - 277 left sufficient site documentation for the Taieri Mouth remains on which to reassess his interpretation of moa-hunter cannibalism. The postcranial human fragments only included rib, clavicle, three vertebrae, “one or two small pieces of femur and tibia”, and foot bones, found “scattered … among the midden refuse”. The site stratigraphy showed discrete horizons, including an upper level of sheep manure. However, the human bone was found in the inner reaches of the cave where a “fusion of midden layers” had formed (Teviotdale 1931:86, plans 1 and 2). Furthermore, as Teviotdale (1931:88) acknowledged, rabbit burrows appeared “all through and under the deposit”. Both factors complicate any cultural interpretation; certainly the “scattered position” of the bone within the midden that impressed Teviotdale and helped influence his conclusion is relatively meaningless in this context. Given the excavation of a trussed interment within the Moa-bone Point Cave midden (Haast 1875:67-8 and this paper Note 1), a burial interpretation would not be without precedent.

Some more recent archaeological suggestions of cannibalism are difficult to assess on the available documentation, although what evidence there is is generally not compelling or unequivocal. Thus, Shawcross (1967:122) has argued that a “part of one [human] body was eaten” at the Galatea Bay midden site on Ponui Island, near Auckland, on the presence of human vertebrae only from a single individual. Davidson has also cited “possible evidence” of cannibalism from “several late North Island pā sites”, including Mangakaware (discussed above), Kauri Point and Ongari. For the last, however, the evidence is no more substantial than “a few widely dispersed human bones” found among midden remains in secondary deposits within the site confines (Shawcross 1964:94).

The evidence for Kauri Point has been cited from an unpublished paper by W. Ambrose (Davidson 1984:148 n. 68, 227). By a process of elimination, it is presumably the “burials of incomplete bodies” without haunches in Davidson's (1984:137) discussion. In his report for the first Kauri Point excavation season, Golson (1961:24, 27) referred to “small irregular pits” and “a small hollow” which contained unspecified human bones within the occupation site. In the absence of more detail, one cannot comment further, except to note that the removal of body parts in a respectful post-mortem context is well documented from Wairau Bar.

Coutts (1972:1:213; table 3.39; 1982:147-8) has suggested an interpretation of cannibalism from human skeletal parts found in middens of the far southern and south-western South Island. The single fragment of cranial bone from the Sandhill Point/2 site is hardly conclusive, however. More plausible are cranial and postcranial fragments (including left and right hands and feet) found at Port Craig/3, and post-cranial fragments only (representing one - 278 female at least, aged 22-24 years), reported from Breaksea Sound/2 (Coutts 1972:2: Table 3.39); these fragments are described by Coutts (1972:1:213) as “indisputable evidence of cannibalism”. Unfortunately, sufficient stratigraphic and site details are not available to assess this claim more fully. It is relevant to the question of the frequency of cannibalism for the far south, however, that the intensive survey and sampling exercise carried out by Coutts (1972) revealed but fragmentary human remains from a few individuals only.

An identification of cannibalism has also been made for an adult male found in a cave occupation site from Cascade Cove, Dusky Sound (south-western South Island). The incomplete remains of two children and a “youth”, aged 20, were also identified (Begg and Begg 1966:113-27, 213-6), possibly in a secondary interment context. For the adult male, the fact that the teeth had been fired, the shaft of the right ulna split, and the heads of the long bones broken off, suggested “evidence of violence”, to Begg and Begg (p.216) and “that this person was the victim of a cannibal feast”. There may also be other causal factors, however, such as the breakage and scattering of bone in secondary burial, as a result of natural rock-fall (documented for this cave site after Maori occupation; see Begg and Begg 1966:113, 125), or even as a consequence of subsequent artefact manufacture (Begg and Begg [1966:119 and plate 40] report the find of a composite fishhook of human bone from the cave site, for example). As with the burnt Garden Island cranial fragments, previously cited from Coutts, the evidence of firing is problematic, rather than conclusive, especially when one considers the 1823 report of Edwardson (as cited above) that cremation was a general practice of the far southern Maori.

From several later coastal midden sites of the south-eastern South Island, Leach and Hamel (1978; 1981:122-9) have also documented scattered human bone fragments, commonly showing evidence of industrial manufacture, if not final process into an artefact form (see also earlier citations from Teviotdale and Trotter). The artefact remains are generally not associated with any clear evidence of cannibalism, however, as suggested in the authors' conclusion:

The archaeological record also provides negative evidence for the sites in Totaranui, where cannibalism was first documented during the Cook visits. Site survey work was carried out in 1979 under the direction of Brailsford (1981:250), concentrating upon the later occupation sites of first contact. During this survey, Brailsford and his team located several of the defended - 279 settlements referred to by members of the Cook expeditions, including the “Hippa[h] Island” pā site near the head of the Sound, and several features associated with Grass Cove (Wharehunga Bay) where Furneaux's crew were killed and reportedly eaten. The sites generally contained eroding midden, incorporating molluscan and fish remains, a fur-seal mandible from one site, and dog and bird bone from Hippa[h] Island. No human bone was reported during the survey, and none was found within the eroding middens (Brailsford 1981:20; 25-6 and figs. 15a-d, 33 fig. 19,35,38). From other survey work by Trotter (1987:124-8) in Totaranui in 1977 and 1982, there are also no reports of human bone either on site, or within exposed middens. Site survey and excavation work in neighbouring Tasman and Golden Bays, with a particular focus on later Maori occupation, has also failed to locate any indication of cannibalism. In fact, except for the evidence from Rotokura, there is scant evidence of human remains from any of the later, northern South Island coastal midden sites (Barber 1992; Challis 1978).

In spite of an expected emphasis on the physical data from excavations, it would seem that, with respect to cannibalism, many archaeological scenarios have resorted to inference and assumption over reliance on any particular site evidence to a surprising degree. At the level of field interpretation, the general literature lacks consistent and systematic description, as a rule, while workers dealing with fragmentary and burnt human remains have tacitly, if not explicitly, acknowledged significant difficulties in explanation.

For much of the site evidence discussed above, of course, the remains are so modified or few that one could not completely preclude the consumption of human flesh even if all of the relevant site information were available. What can be said is that the assumption of cannibalism on the basis of the site evidence is generally unconvincing and/or unnecessary. In most cases, the data are perfectly consistent with secondary inhumation, cremation, or the raiding of an enemy burial ground instead.

Here it is appropriate to ask what might constitute a minimum requirement for archaeological “proof” of cannibalism, a question that has been debated rather vigorously elsewhere (e.g., White and Toth 1991). One writer (Tregear 1926:360) has remarked that “every part of the skeleton of a great chief killed in battle” was degraded through artefact manufacture, which would not augur well for the survival of evidence. However, one must also consider the previously cited claim of Banks that “most of the bones” from the January 1770 incident were discarded (Beaglehole 1962:2:30-1), consistent with Cook and Parkinson (note especially Cook's remark that three human hip bones were found near a “hoven” in one instance; Beaglehole 1955:237; - 280 Parkinson 1784:116). Zimmerman's previously cited observation (Anderson and Tewsley 1926:14) that “many bones” were still to be seen at Grass Cove, four years after Furneaux's crew had been killed, also suggests that human remains might well be left on the landscape without disturbance. (For 19th-century accounts of eaten human bone left scattered for years on site, admittedly from the era of musket warfare, see Smith 1910:190, 352, 365).

If cannibalism occurred frequently, one would expect the commensurate survival from occupation sites of significant amounts of less utilitarian or easily detachable human remains, such as vertebrae, digits, teeth and condyles. One should also note that, in spite of the fact that robust long bones and mandibles (where appropriate) of dog, sea mammal, and bird (including moa in the earlier cultural sequence) were all commonly manufactured into artefacts, a significant proportion of recognisable robust bone still survives in many midden sites consistent with the high dietary value of these animals. If human flesh was an important subsistence component, one should expect to find butchered human fragments of some long bones and innominatum at least, perhaps exhibiting “green bone” breaks in certain instances, and representing a significant number of individuals. Given the general absence of such evidence and the fact that no more than a few individuals have been reported from any of the midden/occupation sites discussed above (and in some cases no more than one individual), an argument for widespread or subsistence cannibalism cannot be sustained. In this regard, one should consider Houghton's general observation, offered from a position of many years analysis of pre-European Maori skeletal remains from sites throughout New Zealand (Houghton 1980, 1987):

Given both the circumstantial evidence from Totaranui and the insistence of all recorders, irrespective of philosophical perspective and any initial reluctance to accept the fact, a compelling case emerges for the occasional practice of cannibalism among late 18th-century Maori communities of both the North and South Islands. Against the further weight of Maori oral tradition (Voykovic 1981:47), one would be hard pressed to argue otherwise.

From the most detailed of the early historical accounts considered above, cannibalism appears to have been a multifaceted ritual arising from conflict and associated with one or more of the meanings of revenge, status enhancement, and spiritual protection or expiation. (If a “traditional” practice, the - 281 later consumption of slaves would not necessarily have been inconsistent with the first meaning at least, as these persons were generally captured enemy.) Where a victim's mana, and power over his/her atua, were to be absorbed by the consumers (with the consequent depletion of the collective mana of the victim's kin group), one can imagine that the victim may have been substantially eaten (although not necessarily; cf. Marsden's [1992:128] account of the consumption of portions of Houtaewa's legs only, where his mana was said to reside). However, there would presumably have been less concern to thoroughly consume persons of lesser status. In these cases, and for the purposes of revenge and ritual appeasement, the act of consumption might only have involved a relatively token gesture.

From Cook to Marsden, the most detailed and authoritative of the early ethnographic accounts also consistently assert that human flesh was not a primary subsistence item. Here, one should also recall again that in the months of early Maori-European contact represented by the published 1769-1815 documentation, not one eye-witness account of anthropophagy in an authentic Maori situation survives, in spite of intense European interest in the subject. Indeed, after many weeks of contact over two expeditions, Cook and other crew members had finally to ask for a demonstration in 1773 to obtain firsthand “evidence” of cannibalism.

If humans were an important subsistence item, one should also note that the supply network would hardly have been reliable, allowing Tuki's previously cited 1793 claim that there were long periods of peace in New Zealand when trade was undertaken (Mcnab 1914a:2:542). Such a claim is consistent with the archaeological evidence of wide-ranging and varied exchange networks which operated throughout the pre-European cultural sequence (Davidson 1984:195-200; see also the account of the people of Totaranui in Beaglehole 1967:72). Furthermore, in spite of L'Horme's December 1769 observation from far northern New Zealand that “these people fight cruel wars among themselves”, and that the local chief considered the Frenchmen obligated “to help him make war on his enemy”, during “all the time” the crew stayed, “we learned of not one act of hostility from his enemies, and the tranquility that seemed to reign everywhere suggested that peace was general” (Ollivier, Hingley and Spencer 1982:129-30).

As a corollary, it would seem likely that premusket slave labour, if considered as a potential foodstore, was considerably more difficult to procure and maintain. Certainly, from the detailed, extant records of the 18th- century English and French expeditions, there are no direct references or allusions to the consumption of slaves, and the condition of slavery itself was not generally recognised. Thus, in March 1770, Banks was “inclind to beleive that these Indians give no quarter, or ever take prisoners to eat upon a future occasion” (Beaglehole 1962:2:31; see - 282 also p. 12). While this limited observation may have oversimplified the matter, Banks had obviously not recognised an enemy slave presence to the contrary. Cook expressed himself similarly in 1770 and again in 1777 (Beaglehole 1955:282;1967:71), while G. Forster (1777:1:518) denied that prisoners were kept “with a view to fatten them for their repast”. First Lieutenant Le Dez of the Marquis de Castries also noted from the Bay of Islands in July 1772: “I imagine their warfare must be cruel, because we saw neither slaves nor any signs of slavery” (Ollivier and Spencer 1985:329).

As one confronts the contrast of a more restrained interpretation of cannibalism and the assumptions of a number of the cultural scenarios considered earlier, one must consider that 19th-century descriptions of New Zealand warfare have created a distorted, popular image of pre-European Maori violence, 4 and even influenced scholarly interpretation. In this regard it is valuable to compare the account of the young man eaten after the foray described in Totaranui in 1773 with the postmusket depredations of far northern chief Hongi Hika in 1818 on the people of the eastern North Island, including the reported burning of 500 villages, the return of 2,000 prisoners of war and, in one canoe alone, about 70 heads (Elder 1932:173). Of the Totaranui situation, by contrast. Cook observed: “I think I understood them for certain that this youth was killed there and not brought away a prisoner, nor could I learn that they had brought away any more” (Beaglehole 1961:294). Although it might be argued that the Totaranui foray does not bear comparison, given the difference of scale — that, in fact, is just the point. There is no evidence that a raid with the devastating impact of Hongi's could have occurred before the introduction of muskets.

From an archaeological perspective, evidence of cremation and the industrial or utilitarian use of human remains has been documented for the North and South Islands, primarily from later Maori occupation sites. (Consistent with this last observation, Duff [1956:40-1; see also p.260] reported that “there is no evidence that the Moa-hunters… fashioned the bones (of enemies) into utensils”.) Both practices may reflect a greater incidence of conflict in the later Maori cultural sequence, as resources became less abundant and populations increased. In such a context, cremation may have been intended to discourage the expropriation of bone for enemy manufacture. For some regions of New Aealand, at least, this may also be reflected in the reported increase over time of secluded cave interments (Davidson 1984:173-7). As a corollary, it is possible that incidents of “cannibalism” also became more numerous in the later period of the Maori sequence.

Allowing this as yet unproven scenario, the absence from later (or earlier) Maori midden sites of large numbers of relatively nonultilitarian bone is still - 283 a telling indictment of any significant subsistence interpretation. In some site reports, it is also likely that evidence of cremation and/or secondary inhumation has been confused with cannibalism. However, even in the highly unlikely event that all of the human remains from midden sites could be shown to be a byproduct of consumption, the identification from each site of no more than a few individuals at most hardly argues that cannibalism was a frequent or significant feature of pre-European lifeways. In this regard, it is also fair to say that the archaeological evidence for Maori cannibalism has been exaggerated, if not misrepresented, in a number of earlier, generalised scenarios.

It may be expected that archival research within the earliest missionary, expeditionary, and Government records of New Zealand, the careful design of future archaeological projects, and the detailed comparative analysis of tribal tradition will yet address the issue of cannibalism more definitively. Certainly, the time is well overdue to revisit the vexed question of the origins and development of ritual violence in New Zealand. Beyond the racist speculations of Smith and Best, the possibility must be considered that cultural precedents for later patterns of conflict and ritual violence were introduced by the earliest Polynesian settlers (Davidson 1984:181 and Sutton 1985:5). If this is so, it may be that cannibalism was not simply a local development or a later response to demographic stress, resource depletion and increased incidents of conflict. To further clarify this point, much more needs to be known about cultural responses to what I have termed the “resource crisis” which appears to characterise the general New Zealand Maori sequence from about the 15th century A.D. (Barber 1992: Chapter 9).

To address the aetiological question beyond assumptions of earlier cultural scenarios would require the integration of archaeology and comparative ethnology, incorporating evidence from a number of relevant Polynesian sequences. A particular focus might be those Island groups with traditions of cannibalism where a relatively large number of other shared traits has suggested cultural and genetic affinities, such as New Zealand and the Marquesas (Davidson 1984:23). Furthermore, since anthropophagous assumptions in various New Zealand cultural scenarios cannot be easily reconciled with historic or archaeological data, a critical consideration of the evidence for cannibalism in other Pacific sequences, such as Fiji, the Marquesas, and Easter Island, may also be fruitful.

For the Marquesas, and from the important Hane Dune Site, for example, Kirch (1984:159) has argued that the presence of scattered, broken and frequently charred human bone in the upper levels is “irrefutable” evidence for the consumption of human flesh. In an argument reminiscent of New - 284 Zealand scenarios, Kirch has also remarked that “the institutionalization of cannibalism in late prehistoric Marquesan society must be taken as a symptom of tensions which had developed between social groups, exacerbated by dense populations and recurrent famines”. (See also Kirch 1984:277 on the development of cannibalism in the later, high-population “Decadent Phase” of the Easter Island sequence.) Kirch may well be right, of course, but, in light of alternative explanations for the presence of human bone (especially charred bone) in New Zealand midden sites, one should at least ask whether the evidence is necessarily as “irrefutable” as he has suggested. Towards resolving these questions, the critical, interdisciplinary methodology of this paper is offerred as a useful contribution.

- 285

While I accept full responsibility for the interpretations and inevitable mistakes of this article, I should like to thank Richard Walter and Michael Gunn for helpful comments on an earlier draft. My thanks also to Martin Fisher for preparing the illustrations.

• Allo, J., 1972. The Whangamata Wharf Site (N49/2): Excavations on a Coromandel Coastal Midden. Records of the Auckland Institute and Museum, 9: 61-79.
• Andersen, J. C. and U. Tewsley, (trans.). 1926 (1781). Zimmermann's account of the third voyage of Captain Cook, 1776-1780. Alexander Turnbull Library Bulletin No. 2. Wellington: Government Printer.
• Anderson, A., 1989 Prodigious Birds: Moas and Moa-hunting in Prehistoric New Zealand. Cambridge: Cambridge University Press.

- 286

• Anderson, A. J., 1984. Book review of Janet Davidson, The Prehistory of New Zealand. New Zealand Archaeological Association Newsletter, 27: 265-8.
• ——and D. G. Sutton, 1973. Archaeology of Mapoutahi Pa, Otago. New Zealand Archaeological Association Newsletter, 16:107-18.
• Anonymous [probably J. Magra/Matra], 1771. A Journal of a Voyage Round the World. London: Beckett and de Hondt.
• Anonymous [probably J. Marra], 1775. Journal of the Resolution's Voyage, in 1772, 1773, 1774, and 1775…also a journal of the Adventure's Voyage, in the Years 1772, 1773, and 1774. London: F. Newberry.
• Arens, W., 1979. The Man-eating Myth: Anthropology and Anthropophagy. New York: Oxford University Press.

• Bagnall, A. C. (ed.), 1957 [1823]. Journal of a Ten Months' Residence in New Zealand [1820] by Richard A. Cruise. Christchurch: The Pegasus Press.
• Ballara, A., 1973. Warfare and Government in Ngapuhi Tribal Society: 1814-1833. Unpublished M.A. thesis, University of Auckland.
• ——1976. The Role of Warfare in Maori Society in the Early Contact Period. Journal of the Polynesian Society 85: 487-506.
• Barber, I. G., 1992. Culture Change in Northern Te Wai Pounamu. Unpublished Ph.D. thesis. University of Otago, Dunedin.
• Beaglehole, J. C. (ed.), 1955. The Journals of Captain James Cook on his Voyages of Discovery. I: The voyage of the Endeavour 1768-1771. Cambridge: Published for the Hakluyt Society at the University Press.
• ——1961. The Journals of Captain James Cook on his Voyages of discovery. II: The Voyage of the Resolution and Adventure 1772-1775. Cambridge: Published for the Hakluyt Society at the University Press.
• ——1962. The Endeavour Journal of Joseph Banks 1768-1771. 2 vols. Sydney: The Trustees of the Public Library of New South Wales in Association with Angus and Robertson.
• ——1967. The Journals of Captain James Cook on his Voyages of discovery. III: The Voyage of the Resolution and Discovery 1776-1780. Parts One and Two [2 vols.]. Cambridge: Published for the Hakluyt Society at the University Press.
• Begg, A. C. and N. C. Begg, 1969. James Cook and New Zealand. Wellington: Government Printer.
• Bellwood, P., 1978a. Archaeological Research at Lake Mangakaware, Waikato, 1968-1970. University of Otago Studies in Prehistoric Anthropology Vol. 12. New Zealand Archaeological Association Monograph 9. Dunedin: University of Otago.
• ——1978b. Man's Conquest of the Pacific: the Prehistory of Southeast Asia and Oceania. Auckland: Collins.
• Best, E., 1902. Notes on the Art of War, as Conducted by the Maori of New Zealand …Part 2. Journal of the Polynesian Society, 11:47-75.
• ——1903. Notes on the Art of War, as Conducted by the Maori of New Zealand …Part 8. Journal of the Polynesian Society. 12:193-217.
• ——1904. Notes on the Art of War, as Conducted by the Maori of New Zealand …Part 9. Journal of the Polynesian Society, 13:1-19.

- 287

• ——1914a. The Peopling of New Zealand. Man, 14:73-6.
• ——1914b. Cremation Amongst the Maori Tribes of New Zealand. Man, 14: 110-112.
• ——1915. Maori and Maruiwi: Notes on the Original Inhabitants of New Zealand and Their Culture; on the Question of How that Culture Affected the Later-coming Maori, and on the Existence in These Isles of Customs, Arts, and Artefacts not Traceable to Polynesia. Transactions and Proceedings of the New Zealand Institute, 18:435-47.
• Begg, A. C. and N. C. Begg, 1966. Dusky Bay. Christchurch: Whitcombe & Tombs.
• Binney, J., 1969. Christianity and the Maoris to 1840: a Comment. New Zealand Journal of History, 3:143-65.
• Brailsford, B., 1981. The Tattooed Land: the Southern Frontiers of the Pa Maori. Wellington: Reed.
• Burns, P., 1980. Te Rauparaha: A New Perspective. Wellington: Reed.
• Butts, D. J., 1977. Seasonality at Rotokura, Tasman Bay: a Case Study in the Use of Faunal Identifications to Establish the Season of Occupation for an Archaeological Site. Unpublished B.A. (Hons.) thesis. University of Otago.
• ——1978. Rotokura: an Archaeological Site in Tasman Bay. Journal of the Nelson Historical Society, 3:4-17.

• Campbell, R. E. M., 1894. Cremation amongst the Maoris. Journal of the Polynesian Society, 3:184-5.
• Charlton, T. H., 1981. Archaeology, Ethnohistory, and Ethnology: Interpretive Interfaces, in Schiffer, M. B. (ed.), Advances in Archaeological Method and Theory. New York: Academic Press, Vol. 4:129-76.
• Ceci, L., 1984. Shell Midden Deposits as Coastal Resources. World Archaeology, 16:62-74.
• Challis, A. J., 1978. Motueka: an Archaeological Survey. New Zealand Archaeological Association Monograph 7. Auckland: Longman Paul.
• Colenso, W., 1868. On the Maori Races of New Zealand. Essay appended to Transactions and Proceedings of the N. Z. Institute. 1 (75pp.).
• Cook, J., 1784. A Voyage to the Pacific Ocean. Vol. 2. London: Nicol and Cadell.
• Coutts, P., 1972. The Emergence of the Foveaux Strait Maori from Prehistory: a Study of Culture Contact. 3 vols. Unpublished Ph.D. thesis, University of Otago.
• ——1982. Fiordland, in Prickett, N. (ed.), The First Thousand Years: Regional Perspectives in New Zealand Archaeology. New Zealand Association Monograph 13. Palmerston North, New Zealand: Dunmore Press, pp. 141-59.

• Darwin, C., 1979. The Journal of a Voyage in H.M.S. Beagle. Facsimile reproduction. Surrey, England: Genesis Publications in association with Australia & New Zealand Book Co.
• Davidson, J., 1984. The Prehistory of New Zealand. Auckland: Longman Paul.
• Dawson, E. W., 1949. Excavation of a Maori Burial, at Longbeach, Otago; with Notes on Associated Artifacts. Journal of the Polynesian Society. 58:58-63.
• Doak, W., 1984. The Burning of the ‘Boyd’: a Saga of Culture Clash. Auckland: Hodder and Stoughton.

- 288

• Duff, R., 1949. Moas and Man (Part 1). Antiquity, 23:172-9.
• ——1950. Moas and Man (Part 2). Antiquity, 24:72-83.
• ——1961. Excavation of House-pits at Pari Whakatau Pa, Claverley, Marlborough. Records of the Canterbury Museum, 7:269-302.
• ——1956 [1950] The Moa-hunter Period of Maori Culture. Canterbury Museum Bulletin No. 1. 2nd edit. Wellington: Government Printer.
• ——1963. Aspects of the Cultural Succession in Canterbury-Marlborough, with Wider Reference to the New Zealand Area. Transactions of the Royal Society of New Zealand, General 1:27-37.

• Elder, J. R. (ed.), 1932. The Letters and Journals of Samuel Marsden 1765-1838. Dunedin: Coulls Somerville Wilkie and A. H. Reed for the Otago University Council.
• Fagan, B., 1984. Clash of Cultures. New York: Freeman.
• Firth, R., 1959 [1929]. Economics of the New Zealand Maori. 2nd edit. Wellington: Government Printer.
• Forster, G., 1777. A Voyage Round the World,…2 vols. London: White, Robson, Elmsly and Robinson.
• Forster, J. R., 1778. Observations Made During a Voyage Round the World…London: Robinson.
• Fyfe, R., 1988. The Opua Moahunter Midden, Cape Egmont, Taranaki. Archaeology in New Zealand, 31:228-32.

• George, P., 1944. Excavation at Kaka Point, Otago. Journal of the Polynesian Society, 53:72-3.
• Golson, J., 1961. Investigations at Kauri Point, Katikati: (3) The Excavations. New Zealand. Archaeological Association Newsletter, 4(2): 18-29.
• Gould, R. A., 1980. Living Archaeology. Cambridge: Cambridge University Press.
• Groube, L. M., 1977. The Hazards of Anthropology, in M. Spriggs, (ed), Archaeology and Anthropology: Areas of Mutual Interest. British Archaeologicl Reports Supplementary Series 19. Oxford, pp. 69-90.

• Haast, J., 1872. Moas and Moa Hunters[:] Address to the Philosophical Institute of Canterbury; Additional Notes; Third Paper on Moas and Moa-hunters. Transactions and Proceedings of the New Zealand Institute, 4: 66-90; 90-4; 94-107.
• ——1875. Researches and Excavations Carried On In and Near the Moa-bone Point Cave, Sumner Road, in the Year 1872.Transactions and Proceedings of the New Zealand Institute, 7:54-85.
• Haszard, H. D. M., 1889. Notes on Some Relics of Cannibalism. Transactions and Proceedings of the New Zealand Institute, 22: 104-5.
• Hawkesworth, J., 1773. An Account of the Voyages Undertaken by the Order of His Present Majesty for Making Discoveries in the Southern Hemisphere…3 vols. London: Strahan and Cadell.
• Hiroa, Te Rangi (Sir Peter Buck), 1950 ([1949], The Coming of the Maori. 2nd Edit. Wellington: Whitcombe and Tombs for the Maori Purposes Fund Board.
• Hoare, M. E. (ed.), 1982. The Resolution Journal of Johann Reinhold Forster 1772-1775. 4 Vols. Second series 152-5. London: Hakluyt Society.

- 289

• Hodder, I., 1982. The Present Past. London: Batsford.
• ——1992. Reading the Past: Current Approaches to Interpretation in Archaeology. Second edit. Cambridge: Cambridge University Press.
• Hogg, G., 1990 [1958]. Cannibalism and Human Sacrifice. London: Hale.
• Holmes, C. (ed.), 1984. Captain Cook's Second Voyage: the Journals of Lieutenants Elliott and Pickersgill. London: Caliban Books.
• Hooper, B. (ed.), 1975. With Captain James Cook in the Antarctic and Pacific: the Private Journal of James Burney Second Lieutenant of the Adventure on Cook's second voyage 1772-1773. Canberra: National Library of Australia.
• Houghton, P., 1980. The First New Zealanders. Auckland: Hodder and Stoughton.
• ——1987. A Vigorous People: Health and Well-being, in J. Wilson (ed.), From the Beginning: the Archaeology of the Maori. Auckland: Penguin Books in Association with the New Zealand Historic Places Trust, pp. 37-42.
• Hutton, F. W., 1891. The Moas of New Zealand. Transactions and Proceedings of the New Zealand Institute. 24:93-172.

• Keyes, I. W., 1969. A Unique Ceremonial Bone Fork from Western Wellington. Dominion Museum Records in Ethnology, 2:17-19.
• Kirch, P. V., 1984. The Evolution of the Polynesian Chiefdoms. Cambridge: Cambridge University Press.

• Leach, B. F., and H. M. Leach, 1979. Burial Positions and Orientations in Palliser Bay, in B. F. and H. M. Leach (eds), Prehistoric Man in Palliser Bay. National Museum Bulletin 21. Wellington, pp. 205-13.
• Leach, H. M., 1969. Subsistence Patterns in Prehistoric New Zealand. Studies in Prehistoric Anthropology, Volume 2. Dunedin: University of Otago.
• ——and G. E. Hamel, 1978. The Place of Taiaroa Head and Other Classic Maori Sites in the Prehistory of East Otago. Journal of the Royal Society of New Zealand, 8:239-51.
• ——and J. [G.E.] Hamel, 1981. Archaic and Classic Maori Relationships at Long Beach, Otago: the Artefacts and Activity Areas. New Zealand Journal of Archaeology, 3:109-41.
• Ledyard, J., 1963 [1783]. A Journal of Captain Cook's Last Voyage. Chicago: Quadrangle.

• McCormick, E. M. (ed), 1966. Narrative of a Residence in New Zealand: Journal of a Residence in Tristan da Cunha by Augustus Earle. London: Oxford University Press.
• McKinlay, J. R., 1971. Waioneke 1968-69. New Zealand Archaeological Association Newsletter, 14:86-91.
• McNab, R., 1907. Murihiku and the Southern Islands. Invercargill: William Smith.
• ——(ed), 1908. Historical Records of New Zealand, Vol. 1. Wellington: Government Printer.
• ——(ed.), 1914a. Historical Records of New Zealand, Vol. 2. Wellington: Government Printer.
• ——1914b. From Tasman to Marsden: a History of Northern New Zealand from 1642 to 1818. Dunedin: Wilke.
• Marsden, M., 1992. God, Man and Universe: a Maori View, in M. King, (ed.), Te Ao

- 290

• Hurihuri: Aspects of Maoritanga. Auckland: Reed, pp. 117-37.
• Mantell, G. A., 1848. On the Fossil Remains of Birds Collected in Various Parts of New Zealand by Mr. Walter Mantell, of Wellington; Additional Remarks on the Geological Position of the Deposits in New Zealand which Contain Bones of birds. Quarterly Journal of the Geological Society of London, 4:225-38, 238-41.
• Mantell, W. B. D., 1868. Address by the Hon. W. B. D. Mantell, F. G. S., on the ‘Moa.’ Transactions and Proceedings of the New Zealand Institute, 1: 18-20.
• Maynard, F. and A. Dumas (F. W. Reed, transl.), 1937 (1858). The Whalers. London: Hutchinson.
• Millar, D. G. L., 1967. Recent Archaeological Excavations in the Northern Part of the South Island. Journal of the Nelson Historical Society, 2:5-12.
• Milner, G. R., E. Anderson, and V. G. Smith, 1991. Warfare in Late Prehistoric West-central Illinois. American Antiquity, 56:581-603.

• Nicholas, J. L., 1817. Narrative of a Voyage to New Zealand, Performed in the Years 1814 and 1815, in Company with the Rev. Samuel Marsden, Principal Chaplain of New South Wales. 2 vols. London: James Black and Son.

• Ollivier, O., C. Hingley, and J. Spencer, 1982. Extracts from Journals Relating to the Visit to New Zealand of the French ship St. Jean Baptiste in December 1769 under the Command of J. F. M. de Surville. Early Eyewitness Accounts of Maori Life: 1.Wellington: Alexander Turnbull Library Endowment Trust in Association with the National Library of New Zealand.
• — —and J. Spencer, 1985. Extracts from Journals Relating to the Visit to New Zealand in May-July 1772 of the French Ships Mascarin and Marquis de Castries under the Command of M.-J. Marion du Fresne. Early Eyewitness Accounts of Maori Life: 2. Wellington: Alexander Turnbull Library Endowment Trust with Indosuez N.Z.
• Oppenheim, R. S., 1973. Maori Death Customs. Wellington: Reed.
• Orchiston, D. W.,1972. Maori Neck and Ear Ornaments of the 1770s: a Study in Protohistoric Ethno-archaeology. Journal of the Royal Society of New Zealand, 2:91-107.
• Owen, R., 1849. On Dinornis. an Extinct Genus of Tridactyle Struthious Birds, with Descriptions of Portions of the Skeleton of Five Species which Formerly Existed in New Zealand (Part I). Transactions of the Zoological Society of London, 3:235-76.
• Owens, J. M. R., 1968. Christianity and the Maoris to 1840. New Zealand Journal of History. 2:18-40.

• Parkinson, S., 1784. A Journal of a Voyage to the South Seas, in His Majesty's Ship The Endeavour… London: Dilly and Phillips.
• Pollock, J. S., 1838. New Zealand: Being a Narrative of Travels and Adventures During a Residence in That Country Between the Years 1831 and 1837. 2 vols. London: Bentley.
• Prickett, N. (ed.), 1982. The First Thousand Years: Regional Perspectives in New Zealand Archaeology. New Zealand Association Monograph 13. Palmerston North: Dunmore Press.

- 291

• Salmond, A., 1991. Two Worlds: First Meetings Between Maori and Europeans 1642-1772. Auckland: Viking, Penguin.
• Savage, J., 1807. Some Account of New Zealand; Particularly the Bay of Islands, and Surrounding Country; With a Description of the Religion and Government, Language, Arts, Manufactures, Manners, and Customs of the Natives, &c. London: Murray and Constable.
• Scarlett, R., 1960. Excavations at Pari Whakatau, Claverly [sic]. New Zealand Archaeological Association Newsletter, 4(1):2-7.
• Shawcross, K., 1967. Maoris of the Bay of Islands, 1769-1840: A Study of Changing Maori Responses to European Contact. Unpublished M.A. thesis, University of Auckland.
• ——1964. Archaeological Investigations at Ongari Point, Katikati, Bay of Plenty.New Zealand Archaeological Association Newsletter, 7(2): 79-98.
• Shawcross, W., 1967. An Investigation of Prehistoric Diet and Economy on a Coastal Site at Galatea Bay, New Zealand. Proceedings of the Prehistoric Society. 33:107-31.
• Simmons, D. R., 1969. A New Zealand Myth: Kupe, Toi and the “Fleet”. New Zealand Journal of History, 3:14-31.
• —— 1976. The Great New Zealand Myth: a Study of the Discovery and Origin Traditions of the Maori. Wellington: Reed.
• Sinclair, E. D., 1940. Excavation of Cave on Okia Flat, Wicliffe Bay. Journal of the Polynesian Society, 49:137-43.
• Skinner, H. D., 1952. A ‘medicine bundle’ and two wooden bowls from inland Otago. Journal of the Polynesian Society, 61:130-5.
• ——1953. Stratification at Long Beach, Otago. Journal of the Polynesian Society, 62:400-02.
• ——(P. Gathercole, F. Leach, H. Leach, eds), 1974. Comparatively Speaking: Studies in Pacific Material Culture 1921-1972. Dunedin: University of Otago Press.
• Smart, C. D., 1962. Preliminary Report of Field Work in the Nukumaru-Waitotara Area. New Zealand Archaeological Association Newsletter. 5(3):170-84.
• Smith, S. P., 1910. Maori Wars of the Nineteenth Century: the Struggle of the Northern against the Southern Maori Tribes prior to the Colonization of New Zealand in 1840. Christchurch: Whitcombe & Tombs.
• ——1921. Hawaiki: the Original Home of the Maori. Christchurch: Whitcombe & Tombs.
• Sorrenson, M. P. K., 1979. Maori Origins and Migrations: the Genesis of Some Pakeha Myths and Legends. Auckland: Auckland University Press and Oxford University Press.
• Steward, J. H., 1942. The Direct Historical Approach to Archaeology. American Antiquity. 7:337-43.
• Sutton, D., 1985. The Whence of the Moriori. New Zealand Journal of History. 19:3-13.

• Taylor, A., 1984. Classic Maori Burial and Cremation in the Manukau Area. New Zealand Archaeological Association Newsletter. 27:256-61.

- 292

• Teviotdale, D., 1931. Notes on the Excavation of a Cave near Taieri Mouth. Journal of the Polynesian Society, 40:86-90.
• ——1932. The Material Culture of the Moa-hunters in Murihiku. Journal of the Polynesian Society, 41: 81-120.
• —— 1935. Evidence of Cremation by the Otago Maori. Journal of the Polynesian Society, 44:32-5.
• —— 1937. Progress Report on the Excavation of a Moa-hunters' Camp at the Mouth of the Tahakopa River. Journal of the Polynesian Society, 46:134-53.
• —— 1938a. Further Excavations at the Moa-hunters' Camp at Papatowai. Journal of the Polynesian Society, 47:27-37.
• —— 1938b. Final Report on the Excavation of a Moa-hunter Camp at the Mouth of the Tahakopa River. Journal of the Polynesian Society, 47: 114-8.
• —— 1939. Excavation of Maori Implements at Tarewai Point, Otago Heads. Journal of the Polynesian Society, 48:108-15.
• —— and H. D. Skinner, 1947. Oruarangi Pa. Journal of the Polynesian Society, 56:340-56.
• Thomson, A. S., 1859. The Story of New Zealand: Past and Present-Savage and Civilized. 2 vols. London: Murray.
• Tregear, E., 1926. The Maori Race. Wanganui: Willis.
• Trotter, M. M., 1967. Excavations at Katiki Point, North Otago. Records of the Canterbury Museum, 8:235-49.
• ——1974. Archaeological Excavations at Takahanga, Kaikoura. Christchurch: Canterbury Museum.
• ——1982. Canterbury and Marlborough, in N.Prickett (ed.). The First Thousand Years: Regional Perspectives in New Zealand Archaeology. New Zealand Association Monograph 13. Palmerston North: Dunmore Press, pp.83-102.
• ——1987. Settlements, Sites and Structures Around Ship Cove, in G. Barratt (ed.), Queen Charlotte Sound, New Zealand: the Traditional and European Records, 1820. Ottawa: Carleton University Press.

• Ulrich, D. U., 1970. The Introduction and Diffusion of Firearms in New Zealand 1800-1840. Journal of the Polynesian Society. 79:487-506.

• Vayda, A. P., 1960. Maori Warfare. Polynesian Society Maori Monographs No. 2. Wellington: Polynesian Society.
• ——1976. War in Ecological Perspective: Persistence. Change, and Adaptive Processes in Three Oceanian Societies. New York and London: Plenum Press.
• Voykovic, A. A., 1981. Nga Roimata O Hine-Nui-Te Po. Death in Maori Life. Unpublished M.A. thesis, University of Otago.

• White, J., 1874. Te Rou; or, the Maori at Home. London: Sampson Low, Marton, Low and Searle.
• White, T. D. and N. Toth, 1991. The Question of Ritual Cannibalism at Grotta Guattari and Comment. Current Anthropology, 32:118-38.
• Williams, H. W., 1937. The Maruiwi Myth. Journal of the Polynesian Society, 46:105-22.