Māori Cordage from Te Wao Nui a Tiriwa, Auckland.pdf

Journal of Pacific Archaeology – Vol. 8 · No. 2 · 2017 – article – Māori Cordage from Te Wao Nui a Tiriwa, Auckland, Aotearoa New Zealand Lisa Mckendry1 AbstrAct Tāmaki Paenga Hira (Auckland War Memorial Museum) holds a number of Māori archaeological textiles from cave and rockshelter sites in Aotearoa New Zealand. The textiles presented here are a cordage collection from Te Wao Nui a Tiriwa (Waitakere Ranges), Auckland. The cord fragments are manufactured with whiri (plaited) and miro (twisted) structures. The diversity of structural attributes reveals the use of a range of materials, strand forms and dimensions to manufacture cords. A range of local resources were used at all sites for plaited cords, however, the twisted cords are all made from the same plant species, harakeke (Phormium tenax, New Zealand Flax). The artefacts appear to be functional items such as lashing, binding and fishing lines. The exception is a plait made with human hair. In the main, the types of whiri and miro cords in the Te Wao Nui a Tiriwa collection are represented in other archaeological cordage assemblages in Aotearoa. This article provides comprehensive technical information which contributes to our understanding of Māori cordage technology and provides data important for future comparative textile studies. Keywords: Māori textiles, cordage, fibre plants IntroductIon Māori Cordage Textile research on archaeological collections can be heavily constrained by taphonomic issues. Fibre remains are rarely found intact which can limit access to important details, such as the original size and shape of the artefact (Norton 1990). This, in turn, impacts the certainty of identifying the raw material, the structures, or the artefact itself (Norton 1990; Smith 2014). Despite these constraints, variations in the structural attributes of textiles tell us about raw material use, technological traditions and the types of activities at sites. This paper focuses on cordage, an essential component of material culture for past Māori communities within Aotearoa New Zealand. Cordage refers to strands, cords, lines and ropes (Bernick 1998: 16). These can be made using a wide range of manufacturing methods, including knotting, twisting or plaiting plant fibres or leaf strips. Two of the main cordage forms made by Māori were twisted cords (miro) and plaited cords (whiri). This paper describes a collection of cordage from Te Wao Nui a Tiriwa (Waitakere Ranges), Auckland, in terms of the range of materials used, the strand forms and dimensions, and the functions of identified specimens, adding important data to our knowledge of Māori cordage technology and its uses in the region. Twisted cords were used for making the fishing lines and leaders that were a fundamental part of survival for Māori communities (Paulin 2007). The first European explorers of Aotearoa commented on the superiority of Māori fishing lines (Best 1986). Further, narrow twisted cords were essential for composite fishhooks, used to bind the hook to the lure and to attach the line to the hook (Leach 2006). Plaited cords were also important components for the fishing industry, such as for net-making (Best 1986), however, they were essential for land based purposes. The versatility of the plaited cord was fully exploited, used for binding cords on tools (Aranui 2006), for sandals and kawe (carrying straps) (McAra 2004) and for tu-maurea (woman’s belt) (Hiroa 1923). In addition, a plait was often used to finish woven objects such as kete (bag), whāriki (floormat) or kākahu (cloaks) (Pendergrast 1984; Aranui 2006). Plaited cords were made from a variety of plant species, including harakeke (Phormium tenax, New Zealand Flax), tī kōuka (Cordyline australis, cabbage tree), kiekie (Freycinetia banksii) and karetu (Hierochloe redolens) (Goulding 1971; McCallum & Carr 2012). Twisted cords were primarily made from processed inner harakeke fibres, known as muka (or whitau) (Best 1986; Pendergrast 2005). The Assemblage and Sites 1 649c Esdaile Road, RD8, Tauranga 3180 Email: [email protected] Submitted 16/02/2017; accepted 26/07/2017 The cordage assemblage presented here is part of a wider collection of textile fragments gathered from dry cave 44 article Journal of Pacific Archaeology – Vol. 8 · No. 2 · 2017 and rockshelter sites in Te Wao Nui a Tiriwa (Waitakere Ranges) Auckland (Figure 1). This mountain range runs north of the Manukau Harbour entrance and up the western coastline of the Auckland Region. The textiles are from six separate locations; Anawhata Q11/178, Piha: Lion Rock Q11/213 and Takatu Point Q11/223, Karekare Q11/238, Whatipu Q11/41, Q11/6 and Paratutae Island Q11/294, 295 and 296 (see Figure 1). The site settings include sheltered cliffs alongside a river valley, exposed coastal sites on the rugged Western coastline, and two islands with pā sites. Many are only accessible by rope, and most are of a small size not suitable for long term permanent occupation (Lawrence 1989). This area was actively used by Māori over a long period, with the initial occupation peaceful, consisting of a small population living primarily on the coastline and along river valleys, enjoying the abundant terrestrial and marine resources (Taua 2009). During times of war in the 16th-18th centuries the caves and rockshelters were used as refuges and after Nga Puhi raids in the 1800s decimated the population, a few dispersed groups remained to maintain a presence on the land (Taua 2009). The textiles were fossicked in the early 1900s by F. Mappin and A. Pycroft, G. Fairfield and J. Donald, W.E Browne and E. Willis and deposited in various lots at the Auckland Museum between 1920 and 1950. In addition to cordage the complete textile assemblage is made up of fishing nets and lines, baskets, cloak fragments, belts, mats, discard fibres and human hair remains. These are outside the scope of this paper and will be discussed in a future Figure 1. Map of Te Wao Nui A Tiriwa: Anawhata Q11/178, article. Here analysis focusses on the forty specimens of Piha: Lion Rock Q11/213 and Takatu Point Q11/223, Karekare cordage represented in the collection. Q11/238, Whatipu Q11/41, Q11/6 and Paratutae Island Q11/294, 295 and 296. Map by Briar Sefton. Textile Analysis The rarity of archaeological cordage finds and the small number of remains has resulted in only a few research papers on Polynesian cordage. These are primarily technical descriptions of cordage collections from locations in New Zealand such as: Lee Island (Anderson, Goulding & White 1991); Kohika (McAra 2004; Aranui 2006), Takaka (Davidson & Leach 2006); Puketoi Station, Southland and Kaitorete Spit (Smith 2014); and in wider Polynesia: Hawai’i (Summers 1990); and Rapa (Cameron 2012). A wide range of terminology is used in these reports which can hinder comparative research (Table 1). The structure of a textile is the key component in textile classifications because it is an objective attribute that exists in nearly all archaeological textiles, whether fragmentary or complete (Emery 1966; Connor 1983; Wendrich 1991; Smith & Laing 2011). The single elements making up a cord are strands and the form of the strands can vary from shredded leaf strips, leaf strips or muka (see Table 1. Terminology used in this paper, comparable terms and definitions. Miro / Twisted Hand-rolled, Plied Twisting, spinning or hand-rolling a minimum of two strands (Emery 1966). Whiri / Plaited Braided Interlacing of at least three individual strands (Emery 1966). Stripped Leaf split into two or more strips. Shredded Leaf separated into thin strips, retaining most of the epidermis (Summers 1990). Retted Leaf strips soaked in water before scraping off the epidermis to release the inner fibres (Summers 1990). Muka Whitau Inner harakeke fibre aggregates (Carr et al.2008). 45 McKendry – Māori Cordage From Te Wao Nui a Tiriwa, Auckland, Aotearoa New Zealand Table 1). There are two cordage structures represented in the currently considered collection, miro and whiri, and these have distinct structures and qualities. Miro cordage is made by twisting or hand-rolling fibres (Figure 2), and whiri cords are made from interlacing at least three individual strands (Figure 3) (Emery 1966). The structural variables are the dimensions, the number and form of the single strands, the twist direction and tightness (Table 2). The term ply is used to notate the number of strands and the final twist direction is recorded an ‘S’ or ‘Z’ twist (Summers 1990; Cameron 2012), or ‘I’ twist (Wendrich 1991) (see Figure 2). The twist tightness is the number of twists in a certain length, usually 100 mm. The other category, plaited cords, share many of the above variables. 1 2 3 3-ply plait 1 2 3 article 4 4-ply plait 1 2 3 4 5 5-ply plait Figure 3. Three-ply, four-ply and five-ply plaits. rESuLtS The plant materials represented in the assemblage have been previously identified by Goulding (1971). Goulding undertook a meticulous and comprehensive investigation of the raw materials in the Waitakere Ranges archaeological textiles and left a resource that makes an important contribution to understanding and analysing this assemblage. Scientific fibre identification techniques have since improved with the use of DNA, micro-computed tomography, SEM and polarized light microscopy (Smith, Pater- son and Lowe 2016). However, due to lack of specialised resources they were not applied in this instance. All forty artefacts were fragmented, except for the fishing line and human hair cord. The 19 miro structures were made from one plant species, harakeke. In contrast, within the 21 whiri structures, nine were made from harakeke, six from tī kōuka, five from kiekie and one from makawe (human hair). In addition, the whiri strand forms were diverse. The harakeke cords had a width range of 5–40 mm, with a mean of 15.9 mm and strands made of leaf strips, shredded leaf, and muka. The tī kōuka fragments had leaf strips and shredded strands with a width range of 12–25 mm and mean of 12.6 mm, and the kiekie strands were leaf strips, roots or unidentifiable, with a width range of 10–18 mm and mean of 17 mm. Miro Cordage S Z The plying together of fibres provides cohesion, flexibility and length (Cameron 2012). The nineteen separate miro structures are connected to six catalogue numbers (Table 3). Apart from two complete fishing lines and an anchor rope, they are short lengths of a standard width and constructed in S and Z-twist directions with variable tightness (loose <10 degrees, medium 10–25 degrees, tight I Figure 2. Twist direction S, Z, I. Table 2. The structural variables of material, dimensions, the number and form of the single strands, the twist direction and tightness. Structural Variable Miro / Twisted Whiri / Plaited Dimensions Length and width mm Length and width mm Material Species Species Number of single strands Two or more - ply Three or more - ply Form of single strands Shredded, leaf strips or muka Shredded, leaf strips or muka Twist direction Z, S, I Tightness Number of twists/100 mm 46 Number of crossings/100 mm Journal of Pacific Archaeology – Vol. 8 · No. 2 · 2017 article Table 3. Miro ‘Twisted’ Cordage Structures: AM# (Auckland Museum number), CA# (Cave Assemblage Structure number), Material species, Dimensions, Number of Strands, Width of Single Strands, Final Twist Direction, Rank of Angle of Twist (L-loose, M-medium, T-tight) and Beads. Miro Structures AM# CA# Material Species Length (mm) Width (mm) No. of Strands Width Single Strand (mm) 17107.3 PM200 European 860 6.0 2 4.0 Z L 17107.3 PM253 Harakeke 400 2.0 2 1.5 Z L two-ply Z-twist 18072 KM207 Harakeke 114 1.0 2 1.0 S M two-ply S-twist 18072 KM212 Harakeke 130 5.0 2 2.0 S M 18072 KM209 Harakeke 68 3.0 6 2.0 Z T X three-ply Z-twist (3xtwo-ply S-twist) 18072 KM210 Harakeke 94 3.0 6 2.0 Z M X three-ply Z-twist (3xtwo-ply S-twist) 18072 KM211 Harakeke 56 5.0 6 3.0 Z T X three-ply Z-twist (3xtwo-ply S-twist) 18072 KM215 Harakeke 1000 2.0 6 1.0 Z T X three-ply Z-twist (3xtwo-ply S-twist) 18072 KM215a Harakeke 1000 1.5 2 0.5 Z M X three-ply Z-twist (3xtwo-ply S-twist) 31727 KM216 Harakeke 115 1.0 2 0.5 Z T 31727 KM217 Harakeke 1000 2.0 6 1.4 Z T X three-ply Z-twist (3xtwo-ply S-twist) 31727 KM217a Harakeke 1000 1.5 6 0.5 Z M X three-ply Z-twist (3xtwo-ply S-twist) 33011.4 KM206 Harakeke 335 36.0 2 36.0 Z T two-ply Z-twist 42138 WUM201 Harakeke 100 3.0 2 2.0 S L two-ply S-twist 42138 WUM202 Harakeke 525 3.0 2 2.0 Z L two-ply Z-twist 42138 WUM203 Harakeke 760 2.0 2 1.0 Z M two-ply Z-twist 42138 WUM204 Harakeke 160 2.0 2 1.0 Z M two-ply Z-twist 42138 WUM205 Harakeke 800 2.0 2 1.0 Z M two-ply Z-twist 46374 PM246 Harakeke 305 15.0 2 8.0 Z L two-ply Z-twist >25 degrees: Emery 1966: 11). The S-twist fragments are in the form of one-ply and two-ply with medium and loose twists. The sixteen cords with a Z-twist are two-ply and three-ply with loose, medium and tight tensions. The miro cord widths range from 0.5 mm to 36 mm, however, 90% are 6 mm or less. Two cords are made with shredded fibres and the remaining are from muka. The complete or nearly complete miro artefacts include a leader and snood (AM#17107.3), an anchor rope (AM#33011.4) (Figure 4) and fishing lines (AM#18072 (Figure 5) and AM#31727). The leader is a tight two-ply Z-twist European rope with burnt ends and the snood is made with a repeated half-hitch from a loose two-ply Z-twist muka cord of 2 mm width. The anchor rope is a dense, large (36 mm wide) and tight two-ply Z-twist with a detached loop that appears to have been torn from the rope. The two complete fishing lines are manufactured in a tight three-ply Z-twist from 3 x two-ply S-twist muka cords. This structure was identified based on the presence of beads (Hurley 1979). Both lines begin at approximately 1.5 mm wide before gradually widening to a 2.5–2.8 mm wide line. The length cannot be determined due to the way the line is bundled. In addition, AM#18072 has short lengths of two-ply Z-twist, three-ply Z-twist (3 x two-ply S-twist) and two-ply S-twist muka cords and two bound 47 Final Twist Rank of Angle of Twist Bead Notation two-ply Z-twist two-ply S-twist two-ply Z-twist Figure 4: AM#33011.4 Anchor Rope from Karekare. Photograph by Tim Mackrell. cord fragments associated with it. The line AM#31727 has a fish hook shank associated with it (but without provenience). This has a fine uneven medium two-ply Z-twist line and an even finer loose two-ply Z-twist cord, both from muka, to bind the shank to the line. The cords within AM#42138 consist of various twoply S and Z-twist muka fragments with widths of 1–2 mm. They are unevenly twisted in both medium and loose ten- McKendry – Māori Cordage From Te Wao Nui a Tiriwa, Auckland, Aotearoa New Zealand article sion. The browning at the tips suggest they were burnt, and they may be from a single cord. The final two-ply Ztwist fragment (AM#46374) is unusual as it has shredded harakeke strands and is 15 mm wide. Whiri Cordage Leaf strips and fibres are plaited to produce lengths of cordage for binding or lashing purposes (Cameron 2012). The twenty-one whiri fragments are three-ply plaits of varying lengths, widths and levels of tightness (Table 4). The strand forms and materials used are diverse; harakeke, tī kōuka kiekie and makawe. The plait with the smallest width of 1 mm, AM#31729, is a rare cord made of makawe and stained with kōkōwai (red ochre) mixed with resin or oil. It is tightly plaited with each strand consisting of four strands of hair, and is a long length coiled similar to modern hand-fishing lines. The other narrow fragment, AM#47620, is loosely plaited and 5 mm wide. It is distinctive as the single strands are each made from one harakeke leaf that has been folded in half lengthways. The majority of cord fragments have a width ranging between 8–15 mm and are medium to tightly plaited using shredded leaves. The two long fragments, AM#19775.2 and AM#17107.1, are made from harakeke and broken Figure 5: AM#18072 Complete fishing line (three-ply Z-twist (3 x two-ply S-twist) from Karekare. Photograph by Tim Mackrell. Table 4. Whiri ‘Plaited’ Cord Structures: AM# (Auckland Museum number), CA# (Cave Assemblage Structure number), Material species, Dimensions, Number of Strands, Width of Single Strands, Strand Form and Rank of Crossing number/100 mm (L-loose, M-medium, T-tight). Whiri Structure AM# CA# Material Species Length (mm) Width (mm) No. of Strands Width Single Strand (mm) 17107.1 PW302 Harakeke 17114 PW309 Tī kōuka 17114 PW310 Kiekie 17114 PW312 17114 PW313 19775.2 2147 8 3 2.00 Shredded 170 15 3 7.00 Shredded M 55 18 3 7.00 Leaf Strip n/a Tī kōuka 160 12 3 6.00 Shredded M Tī kōuka 110 17 3 6.00 Shredded M PHW334 Harakeke 1490 15 3 6.00 Leaf Strip M 23887.1 AW328 Kiekie 240 10 3 4.00 Unprepared L 23887.1 AW329 Kiekie 330 10 3 4.00 Unprepared L 23887.1 AW330 Kiekie 285 10 3 4.00 Unprepared L 31729 HKW350 Human hair 500 1 3 0.25 Unprepared T 33011.5.3 KW322 Tī kōuka 155 13 3 8.00 Leaf Strip T 42138 WUW308 Harakeke 170 8 3 3.00 Shredded M 47619.1 WUW317 Harakeke 660 18 3 13.00 Shredded L 47619.2 WUW306 Harakeke 240 30 3 6.00 Shredded L 47620 WUW307 Harakeke 120 5 3 18.00 Leaf Strip n/a 47625 WUW305 Kiekie 140 15 3 7.00 Unidentifiable M 47626 WUW304 Tī kōuka 170 20 3 11.00 Leaf Strip L 47630 WUW314 Tī kōuka 960 25 3 0.50 Unknown M 49213 PW301 Harakeke 260 10 3 2.00 Leaf Strip L 49215 PW300 Harakeke 250 10 3 5.00 Leaf Strip L 2012.x.219 WUW316 Harakeke 330 40 3 20.00 Shredded L 48 Strand Form Crossing Rank M Journal of Pacific Archaeology – Vol. 8 · No. 2 · 2017 article into three and four pieces (Figures 6 & 7). They differ in that AM#19775.2 is nearly double the width of AM#17107.1 (15 mm and 8 mm), and are plaited using different techniques. AM#19775.2 is the typical flat three-ply plait, however, in AM#17107.1 each strand is twisted before being replaited, creating an irregular texture. A further three short, tightly plaited fragments, made with shredded strands are from harakeke, tī kōuka and kiekie. The three semicircular fragments of plaited kiekie root, AM#23887.1, have a loose tension, the same cord and strand width, indicating they may be from the one object. The tips of all three are stained black from probable burning. The cord fragments wider than 11 mm have a medium tension, regardless of the strand form and material species. The plaited cords in AM#17114 have a range of widths from 12–18 mm, three are made from shredded tī kōuka fibres and one (PW310) is made up of wide kiekie leaf strips. Cord AM#33011.8 consists of two plaits in a Yshape (McAra 2004: 153), where two plaits are joined, or separated. The larger plait has been laid on top of the smaller three-ply plait and both are made from tī kōuka. The wider fragments, AM#19775.2 and AM#47625 (15 mm), Figure 7: AM#.17107.1 Three-ply plait, irregular, from ParatuAM#47619.1 (18 mm), AM#47626 (20 mm), AM#47619.2/ tae Island. Photograph by Tim Mackrell. WUW306 (30 mm) and AM#2012.x.219 (40 mm) are plaited with a medium tension from shredded harakeke fibres. dIScuSSIon Materials Cordage, in a variety of sizes and forms, was vital for preEuropean Māori survival but is rarely found in archaeological contexts due to its organic nature (Anderson et al.1991; McAra 2004; Smith 2014). The rare cordage assemblage described here confirms Māori used a range of available raw materials for plaited cords, including harakeke, ti kōuka, kiekie, and makawe. The leaves used for the plaits appear, in the main, to be prepared but unprocessed, and used either as leaf strips or shredded fibre. As expected, harakeke dominates this group of cords, and, it was the sole material used in the twisted category. In addition, the harakeke leaves were further processed into muka, reflecting both the unique material qualities of this plant, and an abundance and ease of access to the raw material. The versatility of the two-ply twist and of harakeke is demonstrated by a binding cord and a rope. The narrow muka cord binds the fish hook shank to a piece of European rope, AM#31727. This demonstrates continuity in technology and practice post-contact. In contrast, the large anchor rope, AM#33011.4, is made from tightly twisted shredded harakeke (see Figure 4). The process of shredding leaves splits the blade but retains the epidermis. This is the most waterproof part of the leaf (McAra 2004) so may have ensured the anchor rope was more water resistant. The collection of miro artefacts illustrates the importance of twisted cordage for fishing related activities. Figure 6: AM#19775.2 Three-ply plait, flat, from Piha: Takatu Point. Photograph by Tim Mackrell. 49 McKendry – Māori Cordage From Te Wao Nui a Tiriwa, Auckland, Aotearoa New Zealand Miro Cords The complete fishing lines are valuable examples of cords considered to be “far superior in quality to those of Europeans at first contact” (Paulin 2007: 14). The two fishing lines are tightly and evenly twisted, indicating mastery of the material and technique. As the fishing lines are manufactured from three two-ply S-twist cords, the presence of short two-ply S-twist fragments with the fishing line, AM#18072, could indicate strands being prepared, or unravelled, from a three-ply Z-twist line (see Figure 5). The remaining cord fragments of Z-twist and bound cords may be from a stone sinker recorded by Fairfield in 1933, along with a fishing line from Karekare. The cord fragments have similar dimensions as those illustrated by Fairfield (1933: 146, fig.2) and represent the finest quality of workmanship. However, this is not evident in all the twisted cords examined. The roughly made two-ply Z-twist line attached to the fish hook shank in AM#31727 contrasts with the evenness and tightness of the fishing lines. This pattern is reflected in the fragments from AM#42138 and the snood cord from AM#17107.3 which are also uneven and loosely twisted. This shows inexperience or lack of skill in the miro technique (Best 1924), but may reflect the post-contact environment and decline of fibre-working knowledge. The relationships between these lines was lost when harvested by fossickers, and deposited in various groups over many years. In addition, this is a small collection, which limits the scope of interpretation. The rough cords could reflect a disruption in the transfer of knowledge and different passages of time, expediency or may simply be the work of a beginner. The twisted cords appear to be primarily for fishing related activities, however, it is more difficult to determine the specific uses of the plaited cords. article have been a decorative handle for a basket. The twist tightness of cords is rarely reported, yet the tightness of a plait impacts on its flexibility and therefore functionality (Cameron 2012). A pattern emerged from measuring the crossings/100 mm, or the twist tightness of the plaited cords. The wider the plait, the less tight the crossings. This is likely to be determined by the nature and size of the individual strands. The importance of considering the diameter of twisted cords when analysing twist tightness is well-reported (Emery 1966; Hurley 1979; Wendrich 1991). This appears to apply to plaits also, as a wider plait is usually made from wide, and thick strands. Therefore, less folds are required per length. This effect was also noted by Smith (2014) in plaited cords from the Southland Museum and Art Gallery. A comparison among the tension of three plaits, illustrates this effect and the influence of the raw material. Case Study The plait AM#23887, with the lowest number of crossings per 100 mm, is made from kiekie root, and the semi-rigidity and roundness of the roots prevents the strands lying next to each other, resulting in the appearance of a loose plait. However, given the nature of the material they are plaited as tightly as is possible. The widest plait, AM#2012.x.219, made from shredded harakeke, appears to be plaited tightly, however the crossing tension is loose, reflecting the density of the strands. The narrowest cord, AM#31729, has a very tight tension, over 18 crossings per 10 mm. This reflects the thinness of the strands, and the raw material, human hair. The end uses of these three cords are also likely to differ. There is little ethnographic information concerning the use of kiekie root for baskets or nets in Aotearoa. The kiekie plaits have a slight curve, indicating they may part Whiri Cords of a larger circular structure. Puketapu-Hetet (1999) describes their use for making eel baskets but does not menThe variability in plait widths and raw materials indicates tion whether they were plaited or twined. As most rigid diverse functions. The cord fragments could represent nets were twined, some with vine structures (Hiroa 1923), many different original forms; binding for composite tools, these plaits could be part of the inner structure of a net or lashing for house and boat structures, carrying loads, nets, bird snare. The wide plait could be the tie cord of a kōheke basket handles or general everyday use cords. The most (a cylindrical bag to extract juice from tutu berries), tātua common plait, the three-ply plait, was present with two (man’s belt) or kawe (burden carrier). distinct structural variations, which inform about the posIn contrast to these two textiles, early ethnographic sible use of the cords. In a typical flat three-ply plait, each accounts describe human hair as sacred and often incorstrand is folded over another, resulting in the upper and porated into a range of ritualistic acts, though specific lower sides of the leaf alternating along the plait, and a flat practices varied widely (Taylor 1855; Tregear 1904; Best surface. In contrast, as in AM#17107.1, when each strand is 1977). Human hair is very difficult to plait due to its finetwisted before being replaced in the plait, the same side of ness, and the evenness of this plait demonstrates extreme the strip faces upwards (see Figure 7). This is identifiable in competency in cord manufacturing. Previously, Lawrence harakeke leaf strips as the upper and lower surfaces differ (1989) has argued that this human hair cord is a fishing in colour and textural qualities, which remain even when line, possibly because of how it is coiled. However, both deteriorated (Goulding 1971). In addition, the twisted edge the Takatu Point, Piha and Waimamaku collections conof the leaf strip creates an irregular texture along the outer tain a three-plait human hair cord that has been used to edges of the plait. This is not suitable for lashing and may bind a separate bundle of human hair (Turbott 1947). Hu50 Journal of Pacific Archaeology – Vol. 8 · No. 2 · 2017 article man hair cords were used for fishing lines in Hawaii, but in Tahiti they were for decorative and ritualistic purposes (Turbott 1947), and in the Tuamotus they were used as necklace cords (Emory 1975). This suggests human hair cords could be manufactured for very different purposes and further research on the use of makawe is required. In the main, the types of whiri and miro cords in the Te Wao Nui a Tiriwa collection are represented in other archaeological cordage assemblages in Aotearoa. bility of raw materials or the different functions for twisted cords at South Island sites compared to the North Island. For example, the complete fishing lines and anchor rope from Karekare may reflect its abundant marine resources, and coastline suitable for line fishing. In the main, twisted cords are rare finds, as most Māori cordage fragments recovered or excavated are plaited. This is reflected in the Te Wao Nui a Tiriwa collection as there are more plaited cords than twisted. This is a similar to the Auckland War Memorial Museum cordage Archaeological Comparisons (Gould 1971), the Kohika cordage (McAra 2004; Aranui 2006) and the Southland Museum and Art Gallery cordThe fishing lines investigated here are comparable with a age (Smith 2014). However, in contrast to twisted cords, fish hook collection held at Tāmaki Paenga Hira, known as plaited cords are made from numerous raw materials, the ‘Karekare Kit’. This exceptional assemblage, fossicked forms, plait widths and tightness measures, regardless from a rock-ledge at Karekare, contains one-piece (wood of location. This demonstrates the value of the three-ply and bone) and composite fish hooks, many with intact plait, the common use of local resources across Aotearoa snood binding and leaders. Fairfield (1933) describes and and perhaps reflects their diversity of purpose. For examillustrates the lashing and binding from the Karekare Kit ple, the plaited cord in a Y form (AM#33011.8) was also as two-ply muka cords, and some of the leaders as three- represented in the Kohika assemblage (Aranui 2006: 57) ply with each cord tapering in width from 1.5–2.5 mm. and identified as the rim and handle of a basket. This cord In addition, the use of twisted muka cords for fish hook was 12 mm wide, indicating the wider cords may be the binding is as expected from ethnographic reports (Best remains of basket handles. Additional research on cordage 1986). However, the archaeological record reveals the use structures that are a component of another textile, and on of diverse materials and strand forms for the binding of larger collections is required to aid in the interpretation fish hooks. of fragmented cords. A fish hook cache from Pohara, Takaka has cordage that is “almost all of rectangular sectioned pieces of uns- References cutched strips of plant” (Davidson and Leach 2006: 187). The fibre was unable to be definitively identified, however, Adovasio, J.M & Schumacher, E. 2010. Basketry Technology: A it may be nikau (Rhopalostylis sapida) (Dr. Rod Wallace, Guide to Identification and Analysis, 2nd edition. San Franpers. comm 2016). This cache consists of one piece hooks cisco: Left Coast Press. only, which could be considered examples of Archaic fish Anderson, A.J., Goulding, J & White, M. 1991. Bark and fibre artefacts. In Anderson, A.J. & R. McGovern-Wilson (eds) Beech hooks. Yet, the archaeological record reveals a range of hook forms continuing over time, with evidence of cusForest Hunters: The Archaeology of Maori Rockshelter Sites on tomised design (Leach 2006: 114). Therefore, the use of Lee Island, Lake Te Anau, in Southern New Zealand. Auckfibre strips rather than muka may illustrate an early form land: NZAA, pp. 43–55. of binding, be the preference of the maker and user of Andersson-Strand, E., Frei, K. M., Gleba, M., Mannering, U., the hooks, or the availability of material. In addition, the Nosch, M. L & Skals, I. 2010. Old Textiles: New Possibilities. European Journal of Archaeology 13: 149–173. Pitt Rivers Museum has recently identified kiekie as the binding material on one fish hook collected by Captain Aranui, A. 2006. Fibre Use and Construction at Kohika, A WetCook (Cartwright 2013). This demonstrates Māori used a land Village in the Eastern Bay of Plenty. Masters Dissertarange of materials and structures for binding and lashing tion: University of Auckland. cordage. Bernick, K. 1998. Basketry and Cordage from Hesquiat Harbour. As with the Te Wao Nui A Tiriwa miro cordage colVictoria: British Columbia. lection, the archaeological record reveals two-ply twisted Best, E. 1924. The Maori. Volume 1. Wellington: Harry H. Tombs. cords made of shredded fibres. In contrast to the sole use Best, E. 1977. Forest Lore of the Māori, 3rd edition. Wellington: of harakeke across all the locations, evidence from South Te Papa. Island sites in Aotearoa demonstrate the use of diverse raw Best, E. 1986. Fishing Methods and Devices of the Māori, 3rd edimaterials for miro cords. The three twisted cords from the tion. Wellington: Government Printer. Southland Museum and Art Gallery are made from thin Cameron, J. 2012. Cordage from Rapan archaeological sites. In A. strips of plant material, both leaf strip and fibre aggregates, Anderson & D. Kennett (eds) Taking the High Ground, The possibly tī kōuka (Smith 2014). The twisted cords from Archaeology of Rapa, A Fortified Island in Remote Polynesia. Australia: ANU, pp. 97–104. Lee Island were made from shredded harakeke, tī kōuka and kiekie (Anderson et al.1991). Acknowledging the small Carr, D., Cruthers, N., Girvan, E and S. Scheele. 2008. Approachsample size, this may reflect the varying access and availaes for Conservators to the Identification of Plant Material 51 McKendry – Māori Cordage From Te Wao Nui a Tiriwa, Auckland, Aotearoa New Zealand Used in Māori Artefacts. Studies in Conservation 53: 252–263. Cartwright, C. 2013. Scanning Electron Microscope Identification of Fibres and Hairs from the Cook Collection at the Pitt Rivers Museum. Unpublished Typescript. Connor, J. 1983. A Descriptive Classification of Maori Fabrics: Cordage, Plaiting, Windmill Knotting, Twining, Looping and Netting. The Journal of the Polynesian Society 92(2): 189– 213. Davidson, J & Leach, F. 2008. A Cache of One Piece Fish Hooks from Pohara, Takaka, New Zealand. Terra Australis, 29: 185– 202. Drooker, P. B & Webster, L.D. 2000. Beyond Cloth and Cordage: Archaeological Textile Research in the Americas. Salt Lake City: University of Utah Press. Emery, I. 1966. The Primary Structures of Fabrics: An Illustrated Classification. Washington: The Textile Museum. Emory, K.P. 1975. Material Culture of the Tuamotu Archipelago. Pacific Anthropological Records, 22. Honolulu: Bishop Museum. Fairfield, F.G. 1933. Maori Fish-hooks from Manukau Heads, Auckland. The Journal of the Polynesian Society 42.3 (167): 45–155. Fairfield, F.G. 1937. A Necklace of Human Teeth. Journal of the Polynesian Society 46(183): 130–133. Goulding, J.H. 1971. Identification of Archaeological and Ethnological Specimens of Fibre-plant Material Used by the Māori. Records of the Auckland Institute and Museum 8: 57–101. Hiroa, Te Rangi. 1923. Maori Plaited Basketry and Plaitwork. Wellington: Government Printer. Hurley, W.M. 1979. Prehistoric Cordage: Identification of Impressions on Pottery. Washington: Taraxacum. Lawrence, J.M. 1989. The Archaeology of the Waitakere Ranges. M.A. Thesis, Anthropology, University of Auckland. Leach, F. 2006. Fishing in Pre-European New Zealand. Wellington: Te Papa Tongarewa. McAra, S. 2004. Kohika Fibrework. In G Irwin (editor) Kohika: The Archaeology of a Late Maori Lake Village in the Ngati Awa rohe, Bay of Plenty New Zealand. Auckland: University Press. McCallum, R & Carr, D. 2012. Identification and Use of Plant Material for the Manufacture of New Zealand Indigenous Woven Objects. Ethnobotany Journal 10: 185–198. Norton, R.E. 1990. Conservation of artefacts made from plant materials. In Florian, M. E, Kronkright, D.P & R. E Norton the Conservation of Artifacts made from Plant Materials. California: Getty Conservation Institute, pp. 195–286. Paulin, C.D. 2007. Perspectives of Māori Fishing History and Techniques. Tuhinga 18: 11–37. Pendergrast, M. 1984. Feathers and Fibre. Auckland: Penguin Books. Pendergrast, M. 2005. Maori Fibre Techniques: Ka Tahi; Hei Tama Tu Tama. Auckland: Reed. Puketapu-Hetet, E. 1999. Māori Weaving. Auckland: Longman. Smith, C & Laing, R. 2011. What’s in a Name? The Practice and Politics of Classifying Māori Textiles. Textile History 42(2): 220–238. 52 article Smith, C. 2014. Pre-European Māori Textiles from South Island New Zealand. PhD Dissertation: University of Otago. Splitstoser, Jeffrey C. 2012. The Parenthetical Notation Method for Recording Yarn Structure. Textile Society of America Symposium Proceedings. Paper 745. Summers, Catherine, C. 1990. Hawaiian Cordage. Pacific Anthropological Records 39. Hawaii: Bishop Museum. Taylor, T. 1855. Te Ika a Maui or New Zealand and its Inhabitants. London: William MacIntosh. Tregear, E. 1904. The Maori Race. Wanganui: Archibald, Dudingston and Willis. Turbott, O.M. 1947. Hair Cordage in Oceania. Records of the Auckland Institute and Museum 3(3): 151–155. Wendrich, W. 1991. Who is Afraid of Basketry? A Guide to Recording Basketry and Cordage for Archaeologists and Ethnographers. The Netherlands: CNWS.